JP4130149B2 - Burst demodulation method, burst demodulation device, program medium storing burst demodulation method, and burst transmission device - Google Patents

Description

[0001]
[Industrial application fields]
The present invention relates to a burst demodulation method, a burst demodulation device, a program medium, and a burst transmission device that demodulate a digitally modulated burst signal.
[0002]
[Prior art]
As a first conventional example related to a burst demodulator, there is a type in which clock recovery is divided into two types: clock recovery in a preamble and subsequent clock recovery. As this first conventional example, there is one shown in Patent Document 1. The first conventional example will be described with reference to the drawings.
[0003]
FIG. 18 is an overall configuration diagram of the burst mode demodulator described in Patent Document 1. In FIG. The clock phase estimation circuit (which takes part of the operation performed by the clock recovery unit in the present invention) in the figure is a clock recovery code (the same signal as the preamble in the present invention).
Is used to estimate the clock phase at the initial burst (same meaning as the sample timing in the present invention). The clock phase error polarity detection circuit (which takes part of the operation performed by the clock recovery unit in the present invention) is a burst synchronization word unit (between the preamble and the data region in the present invention) following the clock recovery code. The polarity of the clock phase error may be detected using the clock phase error (which may be considered to be present or included in the data area), and clock synchronization (corresponding to clock recovery in the present invention) is established. However, this method has the following problems.
[0004]
(A-1) First, the first problem will be described.
The clock phase estimation circuit has a very large circuit scale and is not suitable as a circuit for actually constructing a burst demodulation device. FIG. 19 of Patent Document 1 shows a configuration diagram of a clock phase estimation circuit. Here, a circuit having a large scale such as a multiplier is used. Generally, a multiplier that multiplies a variable and a constant can be realized with a small circuit. However, a multiplier that multiplies a variable and a variable used here has a very large circuit scale. It is not suitable as a circuit for actually constructing a burst demodulator. It is possible to adopt another configuration such as replacing the whole of FIG. 19 with a ROM, but the circuit scale becomes large because high accuracy is necessary anyway.
[0005]
In addition, it is assumed that the circuit scale of the clock phase estimation circuit 22 can be reduced by using some technique such as reducing the precision of multiplication. In this case, a certain amount of coarse clock synchronization is performed during the period of the clock recovery code, and is completed during the subsequent period of the burst synchronization word part. At this time, a long time is required until the clock synchronization is completed. That is, in the first conventional example, clock synchronization cannot be completed at high speed.
[0006]
(A-2) The second problem will be described.
In the first conventional example, the clock phase estimation circuit always estimates the clock phase (sample timing) using the same method every symbol during the period of the clock recovery code. This method is greatly affected by noise. For example, if a large amount of noise is generated during the last clock phase estimation, the clock phase may be greatly mistaken and clock synchronization may be completely achieved during the subsequent burst synchronization word. Thus, the first conventional example is susceptible to noise and cannot complete clock synchronization at high speed.
[0007]
As a second conventional example relating to a burst demodulator, there is one that performs clock recovery using the result of integrating a certain signal. A second conventional example is shown in Patent Document 2 and is called a clock recovery circuit. This second conventional example will be described.
[0008]
The second conventional example uses the π / 4 shift QPSK modulation method and utilizes the fact that the carrier envelope has a baud timing frequency component. This is specialized for a clock recovery circuit for a burst demodulator. This part of the mechanism can be applied to a general burst demodulator. This feature is that the integration circuit integrates the signal used for clock phase estimation. As a result, the problem of being easily affected by noise, which is the second problem in the first conventional example, can be solved. That is, the influence of noise can be averaged by integration.
[0009]
However, this integration creates new problems. Until the integration is completed, the clock recovery circuit in the second conventional example supplies a low-accuracy clock to another circuit, so that the operation accuracy of the other circuit is lowered, and the operation of the entire burst demodulator is completed. The accuracy is reduced.
[0010]
In order to solve this, it is only necessary to stop the integration and estimate a new clock phase (sample timing) for each symbol. However, as with the second problem in the first conventional example, it is affected by noise. The problem of becoming easier occurs.
[0011]
Thus, in the second conventional example, the operation accuracy of the entire burst demodulator cannot be increased while reducing the influence of noise in the clock recovery circuit. Therefore, the second conventional example cannot complete the burst pull-in at high speed.
[0012]
As a third conventional example relating to a burst demodulator, there is one that uses a delayed signal when performing clock recovery and carrier recovery. As this third conventional example, there are those shown in Patent Document 3 and Patent Document 4. The third conventional example shown in these two documents performs exactly the same operation during the preamble period. The third conventional example will be described with reference to the drawing of Patent Document 3.
[0013]
FIG. 20 is an overall configuration diagram of a burst signal demodulating apparatus in the third conventional example. After the burst detection means (which plays a role corresponding to the signal presence / absence determination unit in the present invention) detects a burst, the bit timing extraction means (plays a role corresponding to the clock recovery unit in the present invention). 1), the bit timing is extracted (corresponding to clock recovery and sample timing in the present invention). Thereafter, the carrier recovery means (which plays a role corresponding to the carrier recovery unit in the present invention) estimates the carrier frequency and phase, and uses the result to calculate a complex multiplier (corresponding to the complex multiplier in the present invention). Performs demodulation.
[0014]
In this operation, the third conventional example has the following characteristics.
(B-1) First, the first feature will be described.
The time required for the burst detection means to detect the burst and the input of the bit timing extraction means are stored in the first delay means (not present in the present invention). As a result, even if it takes a long time for the burst detection means to detect the burst, the bit timing extraction means can be used from the beginning of the preamble. For this reason, the third conventional example is capable of high-speed clock recovery, but on the other hand, the first delay means is required, so that the circuit scale becomes large.
[0015]
(B-2) Next, the second feature will be described.
The time required for the carrier recovery means of FIG. 20 to estimate the carrier frequency and phase is stored in the third delay means 8 (not present in the present invention). Thereby, even if it takes a long time for the carrier recovery means to estimate the carrier frequency and phase, the complex multiplier can be used from the beginning of the preamble. For this reason, the third conventional example is capable of high-speed carrier reproduction. However, since the third delay means 9 is necessary, the circuit scale becomes large.
[0016]
(B-3) The third feature will be described.
First, the operation of the burst detection means will be described with reference to FIG. FIG. 21 is a block diagram of the burst detection means. The burst detection means measures the received signal power from the inverse modulation means to the third envelope detection means, compares it with the threshold value by the comparison means, and if the received signal power is greater than the threshold value, there is a burst ( It corresponds to the presence of a signal in the present invention).
[0017]
However, this method has the following problems. That is, the timing at which the burst is determined to be different depends on the magnitude of the transmission power and the magnitude of the reception power caused by the difference in the magnitude of attenuation in the transmission path.
[0018]
For example, if there are 20 symbols in the preamble, the first 5 symbols gradually increase from power 0 and rise completely at the 8th symbol. A certain received signal has a very large power and may exceed the threshold at the third symbol. In addition, since a certain received signal has very small power, the threshold may be exceeded at the sixth symbol.
[0019]
Here, the description will be supplemented with reference to FIGS. FIG. 14 (a) shows the state of burst pulling when the reception power is low because the transmission power is low or the attenuation in the transmission path is large. FIG. 14B shows the state of the burst pull when the reception power is large because the transmission power is large or the attenuation in the transmission path is small. In FIGS. 14A and 14B, the received signal gradually rises from power 0 and advances completely by 8 symbols. On the way, it is determined that “signal present”. Here, it is assumed that “signal present” is determined based on the power “b”. The preamble length is 20 symbols.
[0020]
In the third conventional example, since the burst pull-in is started after it is determined that the signal “has a signal” at the power “b”, the entire 20-symbol preamble cannot be used for the burst pull-in. For example, in FIG. 14A, since the received power is small, the first six symbols are “no signal” and cannot be used. In FIG. 14B, the received power is large, which is advantageous over FIG. 14A, but the first three symbols are still “no signal” and cannot be used.
[0021]
As described above, as the first problem (B-3-1) mentioned in the third feature of the third conventional example, a part of the preamble cannot be used for the burst pull-in. There is a possibility that the burst pull-in cannot be completed.
[0022]
Next, the second problem (B-3-2) mentioned in the third feature in the third conventional example will be described. In carrier reproduction and clock reproduction, it is expected that efficient processing can be performed by changing the processing method finely depending on the timing of the preamble. For example, the first 10 symbols take a large range and perform rough control instead, and then take a small range from the 11th symbol and perform fine control instead. It is expected that processing can be completed at high speed by performing such processing.
[0023]
However, this third conventional example has a negative influence on the detailed processing method in carrier reproduction and clock reproduction because the timing at which the preamble is determined to be different depends on the difference in received power. Processing takes time.
[0024]
For example, when the threshold is exceeded at the third symbol, the third symbol is recognized as the head of the preamble. If the threshold is finally exceeded at the sixth symbol, the sixth symbol is recognized as the head of the preamble. In this case, the processing method cannot be changed finely depending on the timing of the preamble. As a result, in the third conventional example, the circuit of the delay means increases. Further, when there is no delay means, a long preamble is required.
[0025]
As described above, since the third conventional example requires a lot of delay means, the circuit scale becomes very large, and it is not suitable as a circuit for actually constructing a burst demodulator. Since the burst demodulator is actually configured, if the delay means is reduced, a long preamble is required, and the burst pull-in cannot be completed at high speed.
[0026]
As described above, according to the first to third conventional examples, the conventional burst demodulation device cannot complete the burst pull-in at high speed. Therefore, it is necessary to add a long preamble to the head of the burst in the burst transmission device on the transmission side. As a result, the overhead is increased and the throughput is decreased.
[0027]
Focusing on this point, description will be given of the fourth to sixth conventional examples related to the burst demodulator and the burst transmitter.
[0028]
As a fourth conventional example related to the burst demodulator and the burst transmitter, there is one that can be set to a long preamble of 512 symbols. As the fourth conventional example, there is one that complies with the standard “DOCSIS 1.0” shown in Non-Patent Document 1 and the standard “DOCSIS 1.1” shown in Non-Patent Document 2. These standards are international standards for cable modems.
[0029]
In the fourth conventional example, the length of the preamble is determined by selecting from 0 to 512 symbols in the QPSK mode and 0 to 256 symbols in the 16QAM mode. However, generally, a long preamble of 30 symbols or more is set.
[0030]
For example, the Cisco DOCSIS center unit sets the length of the preamble as follows by default.
The “request” packet is 64 bits in QPSK, that is, 32 symbols.
The “initial” packet is 128 bits in QPSK, that is, 64 symbols.
The “station” packet is 128 bits in QPSK, that is, 64 symbols.
The “short” packet is 16QAM and 144 bits, that is, 32 symbols.
A “long” packet is 160 bits in 16 QAM, that is, 40 symbols.
[0031]
The DOCSIS center unit sets the length of the preamble as follows by default when the QPSK mode is fixed.
The “request” packet is 64 bits, that is, 32 symbols.
The “initial” packet is 128 bits, that is, 64 symbols.
The “state” packet is 128 bits, that is, 64 symbols.
The “short” packet is 72 bits, that is, 32 symbols.
A “long” packet is 80 bits or 40 symbols.
[0032]
The DOCSIS center section sets the preamble length by default as follows when the 16QAM mode is fixed.
The “request” packet is 128 bits, that is, 32 symbols.
The “initial” packet is 256 bits or 64 symbols.
The “state” packet is 256 bits, or 64 symbols.
The “short” packet is 144 bits, that is, 32 symbols.
The “long” packet is 160 bits or 40 symbols.
As described above, the fourth conventional example requires a long preamble.
[0033]
As a fifth conventional example relating to the burst demodulator and the burst transmitter, there is one which is set to a long preamble of 144 symbols. As the fifth conventional example, there are those according to the standards “IEEE802.11” and “IEEE802.11b” shown in Non-Patent Document 3. These standards are international standards for wireless LAN.
[0034]
According to Non-Patent Document 3, in these standards, a long preamble of 144 bits in BPSK, that is, 144 symbols is used in the long preamble. In addition, in the case of a short preamble, “preamble + header” having a half time width is used with respect to the “preamble (144 bits) + header (48 bits)” time width in a long preamble. Although the specific number of symbols of the preamble at the time of the short preamble is not described, even if the number of bits of the preamble is reduced without reducing the number of bits of the header, a long preamble of 48 bits, that is, 48 symbols is used. Is done. As described above, the fifth conventional example requires a long preamble.
[0035]
As a sixth conventional example relating to a burst demodulator and a burst transmitter, there is one that handles symbols having different time widths. As the sixth conventional example, there is one according to the standard “IEEE802.11a” shown in Non-Patent Document 4. This standard is also an international standard for wireless LAN as in the fifth conventional example.
[0036]
According to Non-Patent Document 4, a short preamble of 12 symbols is fixedly used in this standard. The first 10 symbols are symbols with a short time width (0.8 μsec), and the last two symbols are symbols with the same time width (4 μsec) as the data area following the preamble. By using a symbol having a shorter time width than a symbol in the data area, it is possible to perform a burst pull-in with high accuracy, thereby enabling a burst pull-in with a short preamble of 12 symbols.
[0037]
However, in this method, since symbols having different time widths need to be handled, the circuit is complicated on both the transmission side and the demodulation side. As described above, the sixth conventional example requires a complicated circuit in order to enable burst pull-in with a short preamble.
[0038]
The prior art has been described using the first to sixth conventional examples. As described in the third conventional example, the burst pull-in method can be changed finely depending on the timing of the preamble. Therefore, it is expected that efficient processing can be performed. However, in the first to sixth conventional examples shown above, the specific method is not clear. If the processing method is not changed precisely depending on the timing of the preamble, a long preamble is required to perform the burst pull-in.
[0039]
[Patent Document 1]
Patent Publication "Patent No. 2753485 Burst Mode Demodulator"
(Page 3-4, FIG. 1 and FIG. 3)
[Patent Document 2]
Patent Gazette “Patent No. 2940895 Clock Recovery Circuit”
(Page 3-4, Fig. 1)
[Patent Document 3]
Patent Publication “Japanese Patent Publication No. 7-44576 Burst Signal Demodulator”
(Page 2-4, Fig. 1 and Fig. 5)
[Patent Document 4]
Patent Gazette "Japanese Patent Publication No. 7-20146 Burst Signal Demodulator"
(Page 3-5, Fig. 1 and Fig. 5)
[Non-Patent Document 1]
Data-Over-Cable Service Interface Specifications Radio Frequency Interface Specificatio SP-RFI-I05-991105, page 26
[Non-Patent Document 2]
Data-Over-Cable Service Interface Specifications Radio Frequency Interface Specificatio SP-RFIv1.1-I03-991105, page 26
[Non-Patent Document 3]
Masahiro Morikura, Hideaki Matsue, “Trends in IEEE802.11 Compliant Wireless LAN”, November 2001, IEICE Transactions B Vol.J84-B No.11, pp. 1920-1922, FIG.
[Non-Patent Document 4]
Same as above, pages 1923-1925, FIG. 9, FIG.
[0040]
[Problems to be solved by the invention]
As described above, the conventional burst demodulator cannot complete burst at a high speed, and therefore requires a long preamble. Therefore, it is necessary to add a long preamble to the head of the burst in the burst transmission device on the transmission side. As a result, the overhead is increased and the throughput is decreased.
[0041]
The present invention has been made in view of the above-described conventional problems, and is intended to realize a burst decoding apparatus that uses a short preamble or detects a preamble signal at an early stage and can perform high-speed burst pull-in. Objective.
[0042]
[Means for solving the problems]
The invention of claim 1 of the present application inputs a received signal consisting of a preamble and a subsequent data area, completes the burst pull-in by the end of the preamble, demodulates the data area, and performs the original processing. A burst demodulation method for outputting a signal, wherein a digital received signal obtained by digitally converting the received signal is input, and presence / absence information indicating whether or not the modulated signal is included in the digital received signal The presence / absence determination step of outputting a signal, the digital reception signal and the presence / absence information are input, a burst pulling method is switched based on the presence / absence information, burst pulling and demodulation are performed, and the original signal is And a digital demodulation step for outputting, wherein the signal presence determination step inputs the digital reception signal and measures the probability of including the modulated signal. A certainty information outputting step for outputting the result as certainty information, an increasing amount outputting step for inputting the certainty information, measuring the increasing amount and outputting the result as the certainty increasing amount, and the certainty A presence / absence information output step of inputting an increase amount, obtaining a degree of stability when the digital reception signal rises and stabilizes, and outputting the degree as the presence / absence information. .
[0043]
According to claim 2 of the present application, in the burst demodulation method according to claim 1, the presence / absence information is information indicating at least two states of no signal and presence of signal, and the presence information output step includes The presence / absence information is set to “no signal” when the property increase amount is greater than “a” (a is a positive value), and the presence / absence information is set to “no signal” when the presence / absence information one time before is not signaled. Is characterized by having a signal.
[0044]
According to a third aspect of the present invention, in the burst demodulation method according to the first aspect, the presence / absence information is information indicating at least three states of signal absence, signal instability, and signal stability, and the signal presence / absence determination step. Further includes a decrease amount output step for inputting the certainty information, measuring the amount of decrease, and outputting the result as the certainty decrease amount, and the presence information output step includes the certainty increase amount and the certainty amount output step. When the certainty decrease amount is input and the presence / absence information is output, and the certainty increase amount is larger than a (a is a positive value), the presence / absence information is made unstable, and the certainty When the decrease amount is larger than b (b is a positive value), the presence / absence information is absent, and when the certainty increase amount is smaller than a and the certainty decrease amount is smaller than b, the presence / absence before one time point. If the information is unstable, the presence / absence information It is characterized in that the signal stability.
[0045]
The invention of claim 4 of the present application is the burst demodulation method of claim 3, wherein the signal instability of the presence / absence information has n state values of 1, 2,..., N (n is a natural number). The signal instability of state value k (k is a natural number equal to or less than n) is represented as signal instability k. In the presence / absence information output step, if the certainty increase amount is larger than a, the presence / absence information before one time point is If the signal is unstable, the value of k is incremented.
[0046]
The invention of claim 5 of the present application is the burst demodulation method of claim 3, wherein the signal stability of the presence / absence information has m state values of 1, 2, ..., m (m is a natural number), The signal stability of the state value j (j is a natural number equal to or less than m) is expressed as signal presence j, and the presence / absence information output step is performed when the certainty decrease amount is smaller than b and the certainty increase amount is smaller than a. If the presence / absence information before the time point is unstable, the presence / absence information is set to “1”, and if the presence / absence information before one time point is stable, the value of j is incremented.
[0047]
According to a sixth aspect of the present invention, in the burst demodulation method according to the first aspect, the correction digital signal is obtained by digitally converting the received signal and correcting it by burst pulling, and the digital demodulation step includes: A burst pull-in step for inputting the correction digital signal and the presence / absence information and outputting a new pull-in execution variable; and an original signal output step for inputting the correction digital signal and outputting the original signal. The burst pulling step performs a burst pulling operation by obtaining the pulling execution variable so that the state of the correction digital signal becomes an ideal state based on the characteristics of the preamble, and A different method is used depending on the presence / absence information.
[0048]
According to a seventh aspect of the present invention, in the burst demodulation method according to the sixth aspect, the burst pull-in step outputs difference information for obtaining difference information based on a difference between the state of the corrected digital signal and the ideal state. An adjustment value output step for obtaining an adjustment value of the pull-in execution variable from the difference information; and a pull-in execution variable output for obtaining a new pull-in execution variable from the adjustment value and the previously determined pull-in execution variable The adjustment value output step is configured to increase the adjustment value for the difference information having the same value when the presence / absence information indicates instability and the presence / absence information indicates stability. It is characterized by making it smaller when it is.
[0049]
The invention according to claim 8 of the present application is the burst demodulation method according to claim 6, wherein the correction digital signal is subjected to a process of sampling the received signal based on at least timing information, The burst pulling step includes a clock recovery step for inputting the corrected digital signal and the presence / absence information and outputting the timing information, and the clock recovery step is a method for obtaining the timing information based on the presence / absence information. Is characterized by switching.
[0050]
According to a ninth aspect of the present invention, in the burst demodulation method according to the eighth aspect, the correction digital signal is generated by oversampling the received signal twice, and the clock recovery is performed. The step holds 3 samples of the corrected digital signal, and the signal amplitude of one of the first and third samples is A, the other is -A, and the second sample is 0, so that the ideal state is obtained. A timing difference information output step for obtaining the timing information based on a difference between a state of a corrected digital signal obtained by the timing information and the ideal state, and obtaining the timing information based on a zero cross method; A timing adjustment value output step for obtaining a timing adjustment value from the difference information; and the timing adjustment value and the timing obtained previously. A timing information output step for obtaining new timing information from the timing information, wherein the timing adjustment value output step uses the timing adjustment value for the timing difference information of the same value, and the presence / absence information is invalid. It is large when it shows stability, and it is small when the presence / absence information shows stability.
[0051]
According to a tenth aspect of the present invention, in the burst demodulation method according to the sixth aspect, in the digital demodulation step, the digital reception signal and phase information which is one of the pull-in execution variables are input, and the digital demodulation step is performed. A complex multiplication step of performing phase adjustment by complex multiplication of the phase information on the received signal and outputting as the corrected digital signal is further provided, and the burst pulling step includes the corrected digital signal and the presence / absence information. And a carrier reproduction step for outputting the phase information. The carrier reproduction step switches the method for obtaining the phase information according to the presence / absence information.
[0052]
According to an eleventh aspect of the present invention, in the burst demodulation method according to the tenth aspect, in the carrier reproduction step, the correction digital signal is input, and the phase of the correction digital signal on the IQ plane is alternated for each symbol. Phase difference information output step for obtaining phase difference information based on a difference between the state of the corrected digital signal and the ideal state; A phase adjustment value output step for obtaining a phase adjustment value from information; and a phase information output step for obtaining new phase information from the phase adjustment value and the phase information obtained last time. Is larger than the phase difference information of the same value when the presence / absence information indicates instability, and the presence / absence information indicates stability. It is characterized in that smaller.
[0053]
According to a twelfth aspect of the present invention, in the burst demodulation method according to the sixth aspect, in the digital demodulation step, the digital reception signal and amplitude information which is one of the pull-in execution variables are input, and the digital demodulation step is performed. The method further includes a multiplication step of performing amplitude adjustment by multiplying the reception signal by the amplitude information and outputting the corrected digital signal, and the burst pulling step includes the correction digital signal and the presence / absence information. A gain adjustment step of inputting and outputting the amplitude information is provided, and the gain adjustment step switches a method of obtaining the amplitude information according to the presence / absence information.
[0054]
According to a thirteenth aspect of the present invention, in the burst demodulation method according to the twelfth aspect, in the gain adjustment step, the correction digital signal is inputted, and the amplitudes of I and Q of the correction digital signal take reference values. The amplitude information is obtained so as to be in an ideal state, an amplitude difference information output step for obtaining amplitude difference information based on a difference between the state of the corrected digital signal and the ideal state, and an amplitude adjustment value from the amplitude difference information An amplitude adjustment value output step for obtaining the amplitude information output step for obtaining new amplitude information from the amplitude adjustment value and the previously obtained amplitude information, and the amplitude adjustment value output step has the same value. The amplitude adjustment value with respect to the amplitude difference information is increased when the presence / absence information indicates instability, and is decreased when the presence / absence information indicates stability. It is characterized by.
[0055]
According to a fourteenth aspect of the present invention, in the burst demodulation method according to the first aspect, the certainty information output step measures the power of the received signal to obtain the certainty information. is there.
[0056]
The invention according to claim 15 of the present application stores any one of the burst demodulation methods according to claims 1 to 14.
[0057]
According to the sixteenth aspect of the present invention, a received signal consisting of a preamble and a data area following the preamble is input, the burst pull-in is completed by the end of the preamble, and the data area is demodulated to obtain the original. A burst demodulator for outputting a signal, wherein a digital received signal obtained by digitally converting the received signal is input, and presence / absence information indicating whether or not the modulated signal is included in the digital received signal A signal presence / absence determining unit that outputs the digital received signal and the presence / absence information, and a burst pull-in method is switched based on the presence / absence information to perform burst pull-in and demodulation, and the original signal is A digital demodulator that outputs the signal, and the signal presence / absence determination unit inputs the digital received signal, measures the probability of including the modulated signal, and ensures the result. A certainty information output unit that outputs the information, an input for the certainty information, an increase amount output unit that measures the amount of increase and outputs the result as the certainty increase amount, and the certainty increase amount And a presence / absence information output unit that obtains a degree of stability when the digital reception signal rises and stabilizes and outputs the degree as the presence / absence information.
[0058]
According to a seventeenth aspect of the present invention, in the burst demodulator according to the sixteenth aspect, the presence / absence information is information indicating at least two states of absence of a signal and presence of a signal, and The presence / absence information is set to “no signal” when the property increase amount is greater than “a” (a is a positive value), and the presence / absence information is set to “no signal” when the presence / absence information one time before is not signaled. Is characterized by having a signal.
[0059]
According to an eighteenth aspect of the present invention, in the burst demodulator according to the sixteenth aspect, the presence / absence information is information indicating at least three states of signal absence, signal instability, and signal stability, and the signal presence / absence determination unit. Further includes a decrease amount output unit that inputs the certainty information, measures the decrease amount, and outputs the result as the certainty decrease amount, and the presence / absence information output unit includes the certainty increase amount and the When the certainty decrease amount is input and the presence / absence information is output, and the certainty increase amount is larger than a (a is a positive value), the presence / absence information is made unstable, and the certainty When the decrease amount is larger than b (b is a positive value), the presence / absence information is absent, and when the certainty increase amount is smaller than a and the certainty decrease amount is smaller than b, the presence / absence before one time point. If the information is unstable, the presence / absence information is And it is characterized in Rukoto.
[0060]
According to a nineteenth aspect of the present invention, in the burst demodulator according to the eighteenth aspect, the signal instability of the presence / absence information has n state values of 1, 2, ..., n (n is a natural number). , The signal instability of the state value k (k is a natural number equal to or less than n) is represented as signal instability k, and the presence / absence information output unit indicates that the presence / absence information before one time point is If the signal is unstable, the value of k is incremented.
[0061]
The invention of claim 20 of the present application is the burst demodulator according to claim 18, wherein the signal stability of the presence / absence information has m state values of 1, 2, ..., m (m is a natural number), The signal stability of the state value j (j is a natural number less than or equal to m) is represented as signal presence j, and the presence / absence information output unit is configured to output 1 when the certainty decrease amount is smaller than b and the certainty increase amount is smaller than a. If the presence / absence information before the time point is unstable, the presence / absence information is set to “1”, and if the presence / absence information before one time point is stable, the value of j is incremented.
[0062]
According to a twenty-first aspect of the present invention, in the burst demodulator according to the sixteenth aspect, the correction digital signal is obtained by digitally converting the received signal and correcting it by burst pulling. A burst pulling unit that inputs the correction digital signal and the presence / absence information and outputs a new pull execution variable, and an original signal output unit that inputs the correction digital signal and outputs the original signal. The burst pull-in unit performs a burst pull-in operation by obtaining the pull-in execution variable so that the state of the correction digital signal becomes an ideal state based on the characteristics of the preamble, and A different method is used depending on the presence / absence information.
[0063]
According to a twenty-second aspect of the present invention, in the burst demodulation device according to the twenty-first aspect, the burst pull-in unit outputs difference information for obtaining difference information based on a difference between the state of the corrected digital signal and the ideal state. An adjustment value output unit for obtaining an adjustment value of the pull-in execution variable from the difference information, and a pull-in execution variable output for obtaining a new pull-in execution variable from the adjustment value and the pull-in execution variable obtained last time The adjustment value output unit is configured to increase the adjustment value for the difference information having the same value when the presence / absence information indicates instability and the presence / absence information indicates stability. It is characterized by making it smaller when it is.
[0064]
According to a twenty-third aspect of the present invention, in the burst demodulator according to the twenty-first aspect, the correction digital signal is subjected to a process of sampling the received signal based on at least timing information, The burst pulling unit includes a clock recovery unit that inputs the corrected digital signal and the presence / absence information and outputs the timing information, and the clock recovery unit obtains the timing information based on the presence / absence information. Is characterized by switching.
[0065]
According to a twenty-fourth aspect of the present invention, in the burst demodulator according to the twenty-third aspect, the correction digital signal is generated by oversampling the received signal twice, and the clock reproduction is performed. The unit holds 3 samples of the corrected digital signal, and the signal amplitude of one of the first and third samples is A, the other is -A, and the second sample is 0, so that an ideal state is obtained. The timing information is obtained based on a zero-cross method, and a timing difference information output unit for obtaining timing difference information based on a difference between the state of the corrected digital signal obtained by the timing information and the ideal state, A timing adjustment value output unit for obtaining a timing adjustment value from timing difference information, the timing adjustment value and the timing information obtained last time; A timing information output unit for obtaining new timing information, and the timing adjustment value output unit indicates the timing adjustment value with respect to the timing difference information having the same value, and the presence / absence information indicates instability. When the presence / absence information indicates stability, it is reduced.
[0066]
According to a twenty-fifth aspect of the present invention, in the burst demodulation device according to the twenty-first aspect, the digital demodulation unit inputs the digital reception signal and phase information which is one of the pull-in execution variables, It further comprises a complex multiplier for performing phase adjustment by complex multiplication of the phase information on the received signal and outputting the corrected digital signal, and the burst pull-in unit includes the corrected digital signal and the presence / absence information. And a carrier reproducing unit that outputs the phase information. The carrier reproducing unit switches the method for obtaining the phase information according to the presence / absence information.
[0067]
According to a twenty-sixth aspect of the present invention, in the burst demodulator according to the twenty-fifth aspect, the carrier reproduction unit inputs the correction digital signal, and the phase of the correction digital signal on the IQ plane is alternated for each symbol. Phase difference information output unit for obtaining phase difference information based on a difference between the state of the corrected digital signal and the ideal state, and the phase difference A phase adjustment value output unit for obtaining a phase adjustment value from information; and a phase information output unit for obtaining new phase information from the phase adjustment value and the phase information obtained last time. The phase adjustment value for the phase difference information having the same value is increased when the presence / absence information indicates instability, and is decreased when the presence / absence information indicates stability. Than is.
[0068]
According to a twenty-seventh aspect of the present invention, in the burst demodulator according to the twenty-first aspect, the digital demodulation unit inputs the digital received signal and amplitude information which is one of the pull-in execution variables, and The signal processing apparatus further includes a multiplication unit that performs amplitude adjustment by multiplying the reception signal by the amplitude information and outputs the corrected digital signal, and the burst pulling unit includes the correction digital signal and the presence / absence information. A gain adjusting unit that inputs and outputs the amplitude information is provided, and the gain adjusting unit switches a method of obtaining the amplitude information according to the presence / absence information.
[0069]
According to a twenty-eighth aspect of the present invention, in the burst demodulator according to the twenty-seventh aspect, the gain adjusting unit inputs the correction digital signal, and the amplitudes of I and Q of the correction digital signal each take a reference value. The amplitude information is obtained so as to be in an ideal state, an amplitude difference information output unit for obtaining amplitude difference information based on a difference between the state of the corrected digital signal and the ideal state, and an amplitude adjustment value from the amplitude difference information An amplitude adjustment value output unit for obtaining a new amplitude information from the amplitude adjustment value and the previously obtained amplitude information, and the amplitude adjustment value output unit has the same value. The amplitude adjustment value with respect to the amplitude difference information is increased when the presence / absence information indicates instability, and is decreased when the presence / absence information indicates stability.
[0070]
The invention of claim 29 of the present application is characterized in that, in the burst demodulator of claim 16, the certainty information output unit measures the power of the received signal to obtain the certainty information. is there.
[0071]
The invention of claim 30 of the present application is a burst demodulator using the burst demodulation method according to any one of claims 1 to 14, or the bar according to any one of claims 16 to 29. A burst transmission device for transmitting a signal to a burst demodulation device, which transmits a burst signal composed of a preamble and a data area following the preamble, wherein the preamble is less than 30 symbols, The time width is the same as the time width of one symbol in the data area.
[0072]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the burst demodulator according to the present invention will be described with reference to the drawings.
(Embodiment 1)
FIG. 1 is a block diagram showing the overall configuration of a burst demodulator according to the present invention. Received signal 1 is input to the burst demodulator. The received signal 1 is a burst signal in which the main signal is 16QAM modulated, and is composed of a leading 20-symbol preamble followed by a main signal, that is, a data area. The preamble is composed of an alternating signal such as BPSK. The alternating signal is the same as that described in detail in the first conventional example.
[0073]
The digitized version of the received signal 1 is a digital I signal and a digital Q signal. The digital I signal 2 is an I component on the IQ plane (horizontal axis I, phase plane of the vertical axis Q), and the digital Q signal 3 is a Q component on the IQ plane. 4 is timing information, which indicates one of the values 0 to 359. Reference numeral 5 denotes an original signal. 6 is presence / absence information, and has an integer value of 0 or more. 0 indicates “signal absent”, 1-10 indicates “signal unstable”, and 11 or more indicates “signal stable”.
[0074]
Here, “no signal” indicates a determination result that there is no modulated signal. “Signal instability” indicates a determination result that the rise of the modulated signal may start, but the signal may not be stable and may be noise and cannot be determined. “Signal stability” indicates a determination result that a modulated signal exists. Reflecting the above meaning, “signal absence” and “signal instability” are also referred to as “no signal” below. “Signal stability” is also referred to as “signal present”.
[0075]
The sampling unit (SP) 51 inputs the received signal 1 and the timing information 4, samples the received signal 1 based on the timing information 4, and outputs a digital I signal 2 and a digital Q signal 3. . The digital demodulator (DEM) 52 inputs the digital I signal 2, the digital Q signal 3, and the presence / absence information 6, and outputs the original signal 5 and the timing information 4. The signal presence / absence determination unit (SIG) 53 inputs the digital I and Q signals 2 and 3 and outputs the presence / absence information 6.
[0076]
FIG. 2 is a block diagram showing an internal configuration of the sampling unit 51. Reference numeral 101 denotes an AD clock. Reference numeral 102 denotes a digital reception signal, which is obtained by digitizing the reception signal 1. Reference numerals 103 and 104 denote digital I and Q reception signals, respectively, which are obtained by converting the digital reception signal 102 into a baseband.
[0077]
The AD clock unit (ADCLK) 151 outputs the AD clock 101. The AD converter (A / D) 152 receives the received signal 1 and the AD clock 101, samples the received signal 1 with the AD clock 101, digitizes it, and outputs it as a digital received signal 102.
[0078]
A digital down converter (DDC) 153 inputs the digital reception signal 102 and the AD clock 101, down-converts the carrier frequency of the digital reception signal 102 and converts it to a baseband, and also converts the data. The data is converted into data of 2 samples per symbol, that is, double oversampled data, and the I component and Q component are converted into the digital I received signal 103 and the digital Q received signal, respectively. 104 is output.
[0079]
Filters (FIL) 154 and 155 have a waveform shaping function. The digital I reception signal 103 and the digital Q reception signal 104 are input to the filters (FIL) 154 and 155, respectively, and filtering processing is performed based on the timing information 4, respectively. 3 is output.
[0080]
The filters 154 and 155 hold 360 filter coefficients corresponding to 360 sample timings, and select and use the filter coefficients according to the value of the timing information 4. When the value of the timing information 4 increases by 1, the filter coefficient that advances the sample timing once is selected. When the value of the timing information 4 decreases by 1, a filter coefficient that delays the sample timing once is selected.
[0081]
FIG. 3 is a block diagram showing an internal configuration of the signal presence / absence determination unit 53. 301 is a power value. Reference numeral 302 denotes delay power. 303 is the maximum power. The power measurement unit 351 is a certainty information output unit that receives the digital I and Q signals 2 and 3, measures the received power, and outputs the received power as a power value 301. The delay unit 352 is an increase amount output unit that receives the power value 301 and outputs it as a delay power 302 after being delayed by one symbol. The maximum value storage unit 354 receives the power value 301 and the presence / absence information 6. When the presence / absence information 6 indicates “no signal”, the maximum value storage unit 354 holds the power value 301 as the maximum power 303 and the presence / absence information 6 indicates “signal not present”. When “stable” or “signal stable” is indicated, if the power value 301 is larger than the maximum power 303, the power value 301 at that time is held as the maximum power 303. The determination unit 353 is a presence / absence information output unit that receives the power value 301, the delay power 302, and the maximum power 303 and outputs the presence / absence information 6 as follows.
[0082]
If the power value 301 is greater than the delay power 302 by a or more (a is a constant), it is determined that the received signal is in a transition period, and the presence / absence information 6 is “signal unstable”, that is, one of the values 1 to 10. Here, if the presence / absence information 6 before one time point is “no signal”, it is assumed that the reception signal has started rising, and the value is 1. On the other hand, if the presence / absence information 6 one time before is “signal unstable”, it is assumed that the rising operation of the received signal is continuing and the value is incremented. Also, if the presence / absence information 6 before one point in time is “signal stable”, it is determined that “signal stable” until then is an error, the received signal has just started to rise, and the value is set to 1. .
[0083]
If the maximum power 303 is larger than the current power value 301 by b or more (b is a constant), it is determined that the received signal is in a transition period, and the presence / absence information 6 is set to “no signal”, that is, a value of 0. In other cases, that is, when the power value 301 is not greater than the delay power 302 by a or more and the maximum power 303 is not greater than the power value 301 by more than b, the received signal rises even during the transition period. It is determined that the transition period is not falling, and the presence / absence information 6 is set as follows.
[0084]
If the presence / absence information 6 before one point in time is “no signal”, “no signal” continues, that is, the value is 0.
Further, if the presence / absence information 6 one time before is “signal unstable”, the transition period in which the reception signal rises is finished, and “signal stable”, that is, the value is 11.
Further, if the presence / absence information 6 one time before is “signal stable”, that is, a value of 11 or more, the value is incremented by assuming that it is still stable.
[0085]
FIG. 4 is a block diagram showing the internal configuration of the digital demodulator 52. Reference numerals 201 and 202 denote amplitude-adjusted I and Q signals, respectively, which are amplitude-adjusted digital I and Q signals 2 and 3, respectively. 203 and 204 are obtained by adjusting the phase of the amplitude-adjusted I and Q signals 201 and 202, respectively, and are also referred to as corrected digital signals. Reference numeral 205 denotes an original signal point, which takes one of values 0-15.
[0086]
Reference numeral 207 denotes amplitude information. 208 is phase information, which is a pull-in execution variable for burst pull-in. The value of the phase information 208 is 0 degrees, a positive value is clockwise on the IQ plane, a negative value is counterclockwise on the IQ plane, and the unit of the numerical value is “degree”. For example, if the value of the phase information 208 is “−30”, the content is 30 degrees counterclockwise on the IQ plane. The multipliers 251 and 252 both input the amplitude information 207, input the digital I and Q signals 2 and 3, respectively, and perform amplitude adjustment by multiplying the respective signals by the amplitude information 207 to adjust the amplitude. Output as the completed I and Q signals 201 and 202.
[0087]
The complex multiplier 253 receives the amplitude-adjusted I and Q signals 201 and 202 (hereinafter referred to as Ia and Qa) and phase information 208 (hereinafter referred to as θ) and performs phase adjustment of Ia × cos θ + Qa × sin θ. A completed I signal 203, that is, a corrected digital I signal (hereinafter referred to as I) and a phase adjusted Q signal 204 of Qa × cos θ−Ia × sin θ, that is, a corrected digital Q signal (hereinafter referred to as Q) are output. Is.
[0088]
The signal point determination unit 254 receives the phase-adjusted I and Q signals 203 and 204, compares the signal with 16QAM signal points on the IQ plane, determines which signal point it belongs to, and is a number from 0 to 15 This is output as the original signal point 205 shown. FIG. 13 shows the correspondence between 16QAM signal points on the IQ plane and values of 0 to 15.
[0089]
The parallel-serial conversion unit (P / S) 255 is an original signal output unit that inputs the original signal point 205, converts it into a serial signal, and outputs it as the original signal 5. The burst pulling unit 257 receives the phase adjusted I and Q signals 203 and 204 and the presence / absence information 6 and outputs the timing information 4, the phase information 208 and the amplitude information 207.
[0090]
FIG. 5 is a block diagram showing the internal configuration of the burst lead-in portion 257. As shown in FIG. The clock recovery unit (CLK) 451 inputs the phase adjusted I and Q signals 203 and 204 and the presence / absence information 6 and outputs the timing information 4. The carrier reproducing unit (CAR) 452 inputs the phase-adjusted I and Q signals 203 and 204 and the presence / absence information 6 and outputs the phase information 208. The gain adjusting unit (AGC) 453 inputs the phase adjusted I and Q signals 203 and 204 and the presence / absence information 6 and outputs the amplitude information 207.
[0091]
FIG. 6 is a block diagram showing the internal configuration of the clock recovery unit 451. Reference numeral 501 is timing difference information, and 502 is timing adjustment information. The timing difference information output unit 551 inputs phase-adjusted I and Q signals 203 and 204 (denoted as I and Q) and holds I and Q by three samples. The three samples mean IQ at one point, one sample before, and two samples before, and 0 to 2 are added after each character. Every symbol, ie every 2 samples,
(I0−I2) × I1 + (Q0−Q2) × Q1
And this value is output as timing difference information 501.
[0092]
The timing adjustment value output unit 552 receives the timing difference information 501 and the presence / absence information 6 and creates the timing adjustment information 502 as follows.
When the value of the presence / absence information 6 is 0, the value of the timing adjustment information 502 is, for example, an integer value of the timing difference information 501 × 40.
When the value of the presence / absence information 6 is 1, the value of the timing adjustment information 502 is set as an integer value of the timing difference information 501 × 30.
When the value of the presence / absence information 6 is 2 to 10, the value of the timing adjustment information 502 is set as an integer value of the timing difference information 501 × 20.
When the value of the presence / absence information 6 is 11, the value of the timing adjustment information 502 is set as an integer value of the timing difference information 501 × 10.
When the value of the presence / absence information 6 is 12 to 20, the value of the timing adjustment information 502 is set as an integer value of the timing difference information 501 × 5.
[0093]
The timing information output unit 553 receives the timing adjustment information 502 and outputs the timing information 4 obtained by subtracting the timing adjustment information 502 from the timing information 4 one symbol before.
[0094]
FIG. 7 is a block diagram showing an internal configuration of the carrier reproducing unit 452. Reference numeral 601 denotes phase difference information, and reference numeral 602 denotes phase adjustment information. The phase difference information output unit 651 receives the phase-adjusted I and Q signals 203 and 204 (denoted as I and Q), obtains I 2 -Q 2 for every symbol, that is, every 2 samples, A value is output as phase difference information 601.
[0095]
The phase adjustment value output unit 652 receives the phase difference information 601 and the presence / absence information 6 and outputs the phase adjustment information 602 as follows.
When the value of the presence / absence information 6 is 0, the value of the phase adjustment information 602 is an integer value of the phase difference information 601 × 10.
When the value of the presence / absence information 6 is 1, the value of the phase adjustment information 602 is set as an integer value of the phase difference information 601 × 5.
When the value of the presence / absence information 6 is 2 to 10, the value of the phase adjustment information 602 is set as an integer value of the phase difference information 601 × 4.
When the value of the presence / absence information 6 is 11, the value of the phase adjustment information 602 is set as an integer value of the phase difference information 601 × 3.
When the value of the presence / absence information 6 is 12 to 20, the value of the phase adjustment information 602 is set as an integer value of the phase difference information 601 × 2.
[0096]
The phase information output unit 653 receives the phase adjustment information 602, and outputs the phase information 208 obtained by subtracting the phase adjustment information 602 from the phase information 208 one symbol before.
[0097]
FIG. 8 is a block diagram showing the internal configuration of the gain adjustment unit 453. Reference numeral 701 denotes amplitude difference information, and reference numeral 702 denotes amplitude adjustment information. The amplitude difference information output unit 751 inputs the phase-adjusted I and Q signals 203 and 204 (denoted as I and Q), calculates I ^ 2 + Q ^ 2-2 for each symbol, that is, every two samples, and this value. Is output as amplitude difference information 701.
[0098]
The amplitude adjustment value output unit 752 receives the amplitude difference information 701 and the presence / absence information 6 and outputs the amplitude adjustment information 702 as follows.
When the value of the presence / absence information 6 is 0, the amplitude adjustment information 702 is set as an integer value of the amplitude difference information 701.
When the value of the presence / absence information 6 is 1, the amplitude adjustment information 702 is set to an integer value that is 1/5 of the amplitude difference information 701.
When the value of the presence / absence information 6 is 2 to 10, the amplitude adjustment information 702 is set to an integer value of 1/6 of the amplitude difference information 701.
When the value of the presence / absence information 6 is 11, the amplitude adjustment information 702 is set to an integer value that is 1/7 of the amplitude difference information 701.
When the value of the presence / absence information 6 is 12 to 20, the amplitude adjustment information 702 is set to an integer value that is 1/10 of the amplitude difference information 701.
[0099]
The amplitude information output unit 753 inputs the amplitude adjustment information 702, and outputs the amplitude information 207 obtained by subtracting the amplitude adjustment information 702 from the amplitude information 207 one symbol before.
[0100]
The operation of the burst demodulation apparatus according to the first embodiment configured as described above will be described. Here, the description is divided into three as follows. First, as a first explanation (E-1-1), an outline of the operation will be described along the flow of signal processing from the received signal 1 to the original signal 5. Next, as a second explanation (E-1-2), an explanation will be given of the operation focusing on the burst pull-in operation. Finally, as the third explanation (E-1-3), an explanation will be given focusing on how the presence / absence information 6 is switched to “signal absent”, “signal unstable”, and “signal stable”. .
[0101]
(E-1-1)
The following operations are performed inside the sampling unit 51 shown in FIG. The AD converter 152 samples and digitizes the received signal 1 at the rising edge of the AD clock 101 and outputs it as a digital received signal 102. The digital down converter 153 down-converts the carrier frequency of the digital received signal 102 and converts it to a baseband. The digital down-converter 153 performs data decimation and converts it into data of 2 samples per symbol, that is, double-oversampled data. Are output as a digital I reception signal 103 and a digital Q reception signal 104, respectively.
[0102]
The filters 154 and 155 both receive the timing information 4, receive the digital I received signal 103 and the digital Q received signal 104, respectively, perform a filtering process, and output the digital I signal 2 and the digital Q signal 3. This filtering process is an operation combining waveform shaping and sample timing adjustment, and is an operation related to “clock recovery” which is one of the burst pull-in operations. This will be described in the second operation description (E-1-2).
[0103]
The following operations are performed inside the digital demodulator 52 shown in FIG. Multipliers 251 and 252 perform amplitude adjustment by multiplying digital I and Q signals 2 and 3 by amplitude information 207, respectively, and output them as amplitude adjusted I and Q signals 201 and 202, respectively. This amplitude adjustment is an operation related to “gain adjustment” which is one of the burst pull-in operations. This will be described in the second operation description (E-1-2).
[0104]
The complex multiplication unit 253 performs phase adjustment on the IQ plane by complexly multiplying the set of amplitude-adjusted I and Q signals 201 and 202 (Ia, Qa) by the phase information 208 (θ), and each phase-adjusted I , Q signals 203 and 204 are output. The phase-adjusted I and Q signals 203 and 204 are also referred to as corrected digital signals. Specifically, a phase adjusted I signal 203 (hereinafter referred to as I) of Ia × cos θ + Qa × sin θ and a phase adjusted Q signal 204 (hereinafter referred to as Q) of Qa × cos θ−Ia × sin θ are obtained. . For example, if (Ia, Qa) = (1.37, 0.37) and θ = −30,
I = 1.37 × cos (−30) + 0.37 × sin (−30) = 1.0,
Q = 0.37 × cos (−30) −1.37 × sin (−30) = 1.0,
It becomes.
[0105]
This phase adjustment is an operation related to “carrier regeneration” which is one of the burst pulling operations. This will be described in the second operation description (E-1-2). The signal point determination unit 254 receives the phase-adjusted I and Q signals 203 and 204, compares them with 16QAM signal points on the IQ plane, determines which signal point they are, and has a value of 0 to 15 The indicated value is output as the original signal point 205. The correspondence between the 16QAM signal points on the IQ plane and the values of 0 to 15 follows FIG. For example, if the phase-adjusted I and Q signals 203 and 204 are 1, 1 respectively, the original signal point 205 is 3. If the phase-adjusted I and Q signals 203 and 204 are −1/3 and −1, respectively, the original signal point 205 is 9. However, the correspondence between the 16QAM signal points on the IQ plane and the values of 0 to 15 is only required to match between the transmission side and the reception side, and is not limited to this FIG. can do. The parallel-serial conversion unit 255 receives the original signal point 205, converts it into a serial signal, and outputs it as the original signal 5. As described above, the burst demodulator according to Embodiment 1 demodulates the received signal 1 and outputs the original signal 5.
[0106]
(E-1-2)
Next, an explanation will be given focusing on the burst pull-in operation. As the first explanation (E-1-2-1) in the second explanation, explanation about “clock recovery” will be given.
[0107]
(E-1-2-1)
In the sampling unit 51 shown in FIG. 2, the filters 154 and 155 both receive the timing information 4 and also receive the digital I and Q received signals 103 and 104 and perform filtering to perform digital I and Q signals 2 and 3 respectively. Is output. The filters 154 and 155 switch the way of selecting the filter coefficient held therein by the timing information 4 and adjust the sample timing. In the sample timing adjustment, the timing of thinning out the data in the digital down converter 153 is changed, and the central data between the symbol change point and the next change point is output. It is intended to adjust. In the case of double oversampling, the center sample and the sample at the changing point appear alternately. Here, the thinning-out timing of data thinning in the digital down converter 153 is fixed, and the equivalent processing is performed by selecting the filter coefficient. The mechanism will be described in detail in the first conventional example. Has been.
[0108]
This sample timing adjustment is an operation related to “clock recovery” which is one of the burst pull-in operations, and can be used for demodulation of all digitally modulated signals regardless of differences in amplitude modulation, phase modulation, etc. Needed.
[0109]
Here, the sample timing adjustment will be described with reference to FIGS. 9 and 10A to 10C. FIG. 9 is a signal point arrangement diagram of the preamble. It is assumed that the black spot and the lower left black spot in this figure are alternately moved for each symbol (every 2 samples). FIG. 10 (a) is a timing chart showing the time variation of the amplitude in the I direction or the Q direction. The arrows indicate the digital I and Q signals 2 and 3 when the sample timing adjustment is completed. It can be seen that the center sample and the sample at the changing point appear alternately, and 1,0, -1,0... Are repeated.
[0110]
Also, FIGS. 10B and 10C show how the sample timing adjustment is not completed. It can be seen that the position of the arrow is deviated from the normal position, and the amplitude does not become 1,0, -1,0. The filters 154 and 155 complete such sample timing adjustment by the end of the preamble, and prepare for the demodulation of the data area.
[0111]
Note that the filters 154 and 155 adjust the sample timing. The method is to switch the method of selecting the filter coefficient held therein by the timing information 4 and perform filtering. For example, if the value of the timing information 4 is the same as one symbol before, the same filter coefficient as that one symbol before is selected and the same sample timing is obtained.
[0112]
Further, if the value of the timing information 4 is larger than the previous symbol, the filter coefficient is selected in the direction of advancing the sample timing, and if smaller, the filter coefficient is selected in the direction of delaying. The magnitude of the change is set to 1 degree while the value of the timing information 4 changes by 1.
[0113]
The operation of creating the timing information 4 is called “clock recovery”, and is performed by the clock recovery unit 451 shown in FIG.
[0114]
The operation of the clock recovery unit 451 will be described with reference to FIG. The timing difference information output unit 551 obtains a difference between the ideal state and the current state in clock recovery based on the zero cross method, and uses the value as timing difference information 501. Here, the ideal state in clock recovery is a state in which the set of (I0, Q0) and (I2, Q2) indicates the signal point in FIG. 9 and the set of (I1, Q1) is a symbol change point. And At this time, if there is no noise, the signal amplitudes of I0 and Q0 are both “1”, the signal amplitudes of I1 and Q1 are both “0”, and the signal amplitudes of I2 and Q2 are both “−1”. Alternatively, the signal amplitudes of I0 and Q0 are both “−1”, the signal amplitudes of I1 and Q1 are both “0”, and the signal amplitudes of I2 and Q2 are both “1”. This is indicated by a square mark in FIG. The reason why the timing difference information output unit 551 calculates (I0−I2) × I1 + (Q0−Q2) × Q1 and sets this value as the timing difference information 501 is based on the above contents.
[0115]
For example, in the ideal state, the timing difference information 501 is
(1 + 1) × 0 + (1 + 1) × 0 = 0,
Or
(-1-1) × 0 + (-1-1) × 0 = 0,
Thus, the difference from the ideal state is 0.
[0116]
FIG. 10B shows a case where the clock recovery is delayed by π / 4 from the ideal state. This is because the signal amplitudes of I0 and Q0 are both “0.7”, the signal amplitudes of I1 and Q1 are both “−0.7”, and the signal amplitudes of I2 and Q2 are both “−0.7”. Or, the signal amplitudes of I0 and Q0 are both “−0.7”, the signal amplitudes of I1 and Q1 are both “0.7”, and the signal amplitudes of I2 and Q2 are both “0.7”. Where
(0.7 + 0.7) × (−0.7) + (0.7 + 0.7) × (−0.7) = − 1.96,
Or
(−0.7−0.7) × 0.7 + (− 0.7−0.7) × 0.7 = −1.96,
Thus, there is a delay from the ideal state.
[0117]
FIG. 10C shows a case where the clock recovery is advanced by π / 4 from the ideal state. This is because the signal amplitudes of I0 and Q0 are both “−0.7”, the signal amplitudes of I1 and Q1 are both “−0.7”, and the signal amplitudes of I2 and Q2 are both “0.7”. Or, the signal amplitudes of I0 and Q0 are both “0.7”, the signal amplitudes of I1 and Q1 are both “0.7”, and the signal amplitudes of I2 and Q2 are both “−0.7”. Where
(−0.7−0.7) × (−0.7) + (− 0.7−0.7) × (−0.7) = 1.96,
Or
(0.7 + 0.7) × 0.7 + (0.7 + 0.7) × 0.7 = 1.96
This indicates that there is progress from the ideal state.
[0118]
The timing adjustment value output unit 552 outputs the following values.
When the value of the presence / absence information 6 is 0, that is, when the signal is absent (hereinafter referred to as state 1), an integer value obtained by multiplying the timing difference information 501 by 40 is used as the timing adjustment information 502.
When the value of the presence / absence information 6 is 1, that is, the first symbol that becomes “signal unstable” (hereinafter, state 2a), the timing adjustment information 502 is an integer value obtained by multiplying the timing difference information 501 by 30. .
When the value of the presence / absence information 6 is 2 to 10, that is, from the second signal to the tenth symbol (hereinafter referred to as state 2b) after becoming “signal unstable”, the timing difference information 501 is multiplied by 20 times. The numerical value is used as timing adjustment information 502.
When the value of the presence / absence information 6 is 11, that is, the first symbol that becomes “signal stable” (hereinafter referred to as state 3a), an integer value obtained by multiplying the timing difference information 501 by 10 is used as the timing adjustment information 502.
When the value of the presence / absence information 6 is 12 to 20, that is, from the second signal to the tenth symbol (hereinafter referred to as state 3b) after “signal stability”, an integer value obtained by multiplying the timing difference information 501 by five. Is the timing adjustment information 502.
[0119]
The timing information output unit 553 outputs the timing information 4 obtained by subtracting the timing adjustment information 502 from the timing information 4 one symbol before. For example, in the ideal state, the timing difference information 501 is 0, so the timing adjustment information 502 is 0. Therefore, the timing information 4 is the same as the previous one symbol.
[0120]
When the clock recovery is delayed by π / 4 from the ideal state, the timing difference information 501 is -1.96, so the timing adjustment information 502 is -78 in state 1, -59 in state 2a, and -in state 2b. 39, -20 in state 3a, -10 in state 3b. Therefore, the timing information 4 is 78 larger in the state 1, 59 in the state 2a, 39 in the state 2b, 20 in the state 3a, and 10 in the state 3b than one symbol before.
[0121]
If the clock recovery is advanced by π / 4 from the ideal state, the timing difference information 501 is 1.96, so the timing adjustment information 502 is 78 in state 1, 59 in state 2a, 39 in state 2b, and state 3a. Is 20 and in state 3b is 10. Therefore, the timing information 4 is smaller than that of the previous symbol by 78 in state 1, 59 in state 2a, 39 in state 2b, 20 in state 3a, and 10 in state 3b.
[0122]
Here, it returns to operation | movement description of the filters 154 and 155 in the sampling part 51 again. For example, in the ideal state, the value of the timing information 4 is the same as one symbol before, so the same filter coefficient as that one symbol before is selected and the sample timing is not changed.
[0123]
If the clock recovery is delayed by π / 4 from the ideal state, the timing information 4 is 78 symbols in state 1, 59 in state 2a, 39 in state 2b, 20 in state 3a, 20 in state 3b, one symbol before. It is 10 bigger. In this case, the filter coefficient is selected so that the sample timing is advanced by 78 degrees in state 1, 59 degrees in state 2a, 39 degrees in state 2b, 20 degrees in state 3a, and 10 degrees in state 3b.
[0124]
When the clock recovery is advanced by π / 4 from the ideal state, the timing information 4 is 78 symbols in state 1, 59 in state 2a, 39 in state 2b, 20 in state 3a, and 20 in state 3b from the previous symbol. 10 smaller. In this case, the filter coefficient is selected in a direction to delay the sample timing by 78 degrees in state 1, 59 degrees in state 2a, 39 degrees in state 2b, 20 degrees in state 3a, and 10 degrees in state 3b.
[0125]
As described above, the clock recovery unit 451 performs different clock recovery processing depending on the value of the presence / absence information 6. Specifically, when the value is small, the clock recovery process is performed with coarse pull-in adjustment with an emphasis on the pull-in speed, and when the value is large, the clock recovery process is performed with fine pull-in adjustment with an emphasis on the pull-in accuracy.
[0126]
As the second explanation (E-1-2-2) in the second explanation, explanation will be given regarding “gain adjustment”.
[0127]
(E-1-2-2)
The following operation is performed in the digital demodulator 52 shown in FIG. The multipliers 251 and 252 both receive the amplitude information 207 and also input the digital I and Q signals 2 and 3, and perform amplitude adjustment by multiplying the respective signals by the amplitude information 207. Q signals 201 and 202 are output. This amplitude adjustment is an operation related to “gain adjustment” which is one of the burst pull-in operations, and is particularly essential in digital modulation including information in amplitude components such as ASK and QAM. It is not always necessary for digital modulation in which information is included only in the phase component and information is not included in the amplitude component, such as PSK.
[0128]
Here, the amplitude adjustment will be described with reference to FIGS. FIG. 11A is a signal point arrangement diagram of the preamble after the amplitude adjustment is completed. FIG. 11B is a signal point arrangement diagram of a preamble for which amplitude adjustment has not been completed. FIG. 11C is a signal point arrangement diagram of the data area in which the amplitude adjustment is completed. FIG. 10D is a signal point arrangement diagram of the data area where the amplitude adjustment is not completed.
[0129]
In FIGS. 11 (b) and 11 (d), if the original (1, 1) signal point is anyway, it will be closer to the (1/3, 1/3) signal point, and the original signal will be correctly displayed. It turns out that it is in the state which cannot guide. Therefore, the multipliers 251 and 252 complete the amplitude adjustment by the end of the preamble, and prepare for the demodulation of the data area in the state as shown in FIGS.
[0130]
Note that the multipliers 251 and 252 perform the amplitude adjustment, but only multiply the amplitude information 207. For example, if the value of the amplitude information 207 is the same as one symbol before, the same value is multiplied, so the amplitude adjustment is the same as one symbol before. Also, if the value of the amplitude information 207 is larger than one symbol before, the amplitude adjustment is changed in a direction to increase the amplitude from one symbol before, and if smaller, the amplitude adjustment is changed in a direction to decrease. The operation of creating the amplitude information 207 is called “gain adjustment” and is performed by the gain adjustment unit 453 inside the burst pull-in unit 257.
[0131]
Therefore, the operation inside the gain adjustment unit 453 will be described with reference to FIG. The amplitude difference information output unit 751 obtains the difference between the ideal state and the current state in gain adjustment, and sets the difference value as the amplitude difference information 701. Here, the ideal state in gain adjustment is the case where the set of phase-adjusted I and Q signals 203 and 204 (denoted as I and Q) is (1, 1) or (-1, -1). , (I, Q) indicate the signal points in FIG. However, the phase may be shifted as long as the amplitude is the same. That is, (I, Q) = (cosA + synA, cosA−synA) may be satisfied (A is an arbitrary phase angle). The reason why the amplitude difference information output unit 751 obtains I ^ 2 + Q ^ 2-2 and sets this value as the amplitude difference information 701 is based on the above contents.
[0132]
For example, in the ideal state, the phase difference information 701 is
(cosA + sinA) ^ 2 + (cosA-sinA) ^ 2-2
= (1 + 2 · cosA · sinA) + (1-2 · cosA · sinA) -2 = 0
Thus, the difference from the ideal state is 0.
[0133]
Consider a case where the gain is halved (hereinafter referred to as gain 1/2), such as (I, Q) = (0.5, 0.5). In the state shown in FIG.
(0.5 × (cosA + sinA)) ^ 2+ (0.5 × (cosA−sinA)) ^ 2-2
= 0.25 (1 + 2 · cos A · sin A) + 0.25 (1-2 · cos A · sin A) -2
= -1.5,
It becomes.
This indicates that the gain is smaller than the ideal state.
[0134]
Also, when the gain is 1.5 times (hereinafter referred to as the gain 1.5 times state) such as (I, Q) = (1.5, 1.5),
(1.5 × (cosA + sinA)) ^ 2+ (1.5 × (cosA−sinA)) ^ 2-2
= 2.25 (1 + 2 · cosA · sinA) +2.25 (1-2 · cosA · sinA) -2
= 2.5,
It becomes.
This indicates that the gain is larger than the ideal state.
[0135]
The amplitude adjustment value output unit 752 uses the amplitude difference information 701 as the amplitude adjustment information 702 as it is when the value of the presence / absence information 6 is 0, that is, when the signal is absent (state 1). Further, when the value of the presence / absence information 6 is 1, that is, the first symbol (state 2a) that has become “signal unstable”, the amplitude difference information 701 is multiplied by 1/5 as amplitude adjustment information 702. Also, when the value of the presence / absence information 6 is 2 to 10, that is, from the second signal to the tenth symbol (state 2b) after “signal instability”, the amplitude difference information 701 is multiplied by 1/6. Is amplitude adjustment information 702. When the value of the presence / absence information 6 is 11, that is, the first symbol (state 3a) that has become “signal stable”, the amplitude difference information 701 is multiplied by 1/7 as amplitude adjustment information 702. Also, when the value of the presence / absence information 6 is 12 to 20, that is, from the second signal to the tenth symbol (state 3b) after becoming “signal stable”, the amplitude difference information 701 is multiplied by 1/10. Amplitude adjustment information 702 is assumed.
[0136]
Then, the amplitude information output unit 753 outputs the amplitude information 207 obtained by subtracting the amplitude adjustment information 702 from the amplitude information 207 one symbol before. For example, in the ideal state, the amplitude difference information 701 is 0, so the amplitude adjustment information 702 is also 0. Therefore, the amplitude information 207 is the same as the previous symbol.
[0137]
In the gain 1/2 state, the amplitude difference information 701 is -1.5, so the amplitude adjustment information 702 is -1.5 in state 1, -0.3 in state 2a, and -0.25 in state 2b. In the state 3a, -0.21 and in the state 3b, -0.15. Therefore, the amplitude information 207 is 1.5 larger than the previous symbol by 1.5 in state 1, 0.3 in state 2a, 0.25 in state 2b, 0.21 in state 3a, and 0.15 in state 3b.
[0138]
In the gain 1.5 times state, the amplitude difference information 701 is 2.5. Therefore, the amplitude adjustment information 702 is 2.5 in the state 1, 0.5 in the state 2a, 0.41 in the state 2b, and 0.4 in the state 3a. 0.36 and 0.25 in state 3b. Accordingly, the amplitude information 207 is smaller than the previous symbol by 2.5 in state 1, 0.5 in state 2a, 0.41 in state 2b, 0.36 in state 3a, and 0.25 in state 3b.
[0139]
Here, the operation of the multipliers 251 and 252 will be described again. For example, in the ideal state, the value of the amplitude information 207 is the same as that one symbol before, so there is no change in amplitude adjustment. That is, the digital I and Q signals 2 and 3 are multiplied by the same value as one symbol before.
[0140]
In addition, in the gain 1/2 state, the amplitude adjustment is changed in a direction in which the amplitude is larger than the previous symbol because the value of the amplitude information 207 is larger than the previous symbol. For example, if the amplitude-adjusted I and Q signals 201 and 202 are obtained by multiplying the digital I and Q signals 2 and 3 by 0.5 one symbol before, this time, the state 1 is 2.0 and the state 2a is It is multiplied by 0.8, 0.75 in state 2b, 0.71 in state 3a, and 0.65 in state 3. This is a change of the gain adjustment in the outward direction by 1.5 in state 1, 0.3 in state 2a, 0.25 in state 2b, 0.21 in state 3a, and 0.15 in state 3b. Means that
[0141]
Further, in the gain 1.5 times state, the value of the amplitude information 207 is smaller than one symbol before, so the amplitude adjustment is changed in a direction to make the amplitude smaller than one symbol before. For example, when the digital I and Q signals 2 and 3 multiplied by 10 one symbol before are used as amplitude-adjusted I and Q signals 201 and 202, this time is 7.5 in state 1 and 9 in state 2a. .5, 9.59 in state 2b, 9.64 in state 3a, and 9.75 in state 3b. This is a change of the gain adjustment in the center direction by 2.5 in state 1, 0.5 in state 2a, 0.41 in state 2b, 0.36 in state 3a, and 0.25 in state 3b. Means that
[0142]
As described above, the gain adjustment unit 453 performs a gain adjustment process that varies depending on the value of the presence / absence information 6. Specifically, when the value is small, a gain adjustment process is performed in which the adjustment of the pull-in focusing on the pull-in speed is rough. When the value is large, a fine gain adjustment process is performed with a fine adjustment of the pull-in with emphasis on the pull-in accuracy.
[0143]
As the third explanation (E-1-2-3) of the second explanation, explanation on “carrier reproduction” will be given.
[0144]
(E-1-2-3)
The complex multiplier 253 shown in FIG. 4 receives the amplitude-adjusted I and Q signals 201 and 202 and the phase information 208 and performs complex multiplication of the I and Q sets by the phase information 208 to obtain a phase on the IQ plane. Adjustments are made and output as phase-adjusted I and Q signals 203 and 204, respectively. This phase adjustment is an operation related to “carrier reproduction” which is one of the burst pull-in operations, and is particularly essential in digital modulation including information in phase components such as PSK and QAM. It is not always necessary in digital modulation that includes information only in the amplitude component and does not include information in the phase component like ASK.
[0145]
Here, the phase adjustment will be described with reference to FIGS. FIG. 12A is a signal point arrangement diagram of the preamble after the phase adjustment is completed. FIG. 12B is a signal point arrangement diagram of a preamble for which phase adjustment has not been completed. FIG. 12C is a signal point arrangement diagram of the data area after the phase adjustment is completed. FIG. 12D is a signal point arrangement diagram of the data area where the phase adjustment is not completed.
[0146]
In FIGS. 12B and 12D, the signal point of (1, 1) is close to (1/3, 1), and it can be seen that the original signal cannot be correctly derived. Therefore, the complex multiplication unit 253 completes the phase adjustment by the end of the preamble, sets the state as shown in FIGS. 12A and 12C, and prepares for the demodulation of the data area.
[0147]
Note that the complex multiplier 253 performs the complex adjustment of the phase information 208, although it performs the phase adjustment. For example, if the value of the phase information 208 is the same as that of one symbol before, the same value is subjected to complex multiplication, so that the phase adjustment is the same as that of one symbol before. Further, if the value of the phase information 208 is larger than one symbol before, the phase adjustment is changed in a direction in which the phase is turned clockwise from one symbol before, and if it is smaller, the phase adjustment is turned in a counterclockwise direction. The operation of creating the phase information 208 is called “carrier reproduction” and is performed by the carrier reproduction unit 452 of the burst pull-in unit 257.
[0148]
The operation inside the carrier reproduction unit 452 will be described with reference to FIG. The phase difference information output unit 651 obtains the difference between the ideal state and the current state in carrier reproduction, and uses this difference value as phase difference information 601. Here, the ideal state in carrier reproduction is the case where the set of phase-adjusted I and Q signals 203 and 204 (denoted as I and Q) = (1, 1) or (−1, −1), In this state, (I, Q) indicates the signal point in FIG. However, as long as the phases are the same, the amplitudes may be shifted. That is, (I, Q) = (A, A) or (−A, −A) may be sufficient (A is an arbitrary value). The reason why the phase difference information output unit 651 calculates I ^ 2-Q ^ 2 and sets this value as the phase difference information 601 is based on the above contents. For example, in the ideal state, the phase difference information 601 is A ^ 2-A ^ 2 = 0, indicating that the difference from the ideal state is zero.
[0149]
Hereinafter, in order to simplify the description, let us consider a state in which the amplitude deviation is not taken into consideration, that is, the case of A = 1. FIG. 12B shows the case where carrier reproduction is deviated counterclockwise by π / 8 from the ideal state (hereinafter referred to as −π / 8 state). This is the case when the signal amplitude of (I, Q) is (0.5, 1.3) or (-0.5, -1.3). in this case,
0.5 ^ 2-1.3 ^ 2 = -1.44,
Or
(−0.5) ^ 2 − (− 1.3) ^ 2 = −1.44,
Thus, there is a deviation in the counterclockwise direction from the ideal state.
[0150]
Further, consider a case where carrier reproduction is deviated clockwise from the ideal state by π / 8 (hereinafter referred to as + π / 8 state). This is the case when the signal amplitude of (I, Q) is (1.3, 0.5) or (-1.3, -0.5). in this case,
1.3 ^ 2-0.5 ^ 2 = 1.44,
Or
(-1.3) ^ 2-(-0.5) ^ 2 = 1.44,
Thus, there is a deviation in the clockwise direction from the ideal state.
[0151]
The phase adjustment value output unit 652 in FIG. 7 outputs the following values.
When the value of the presence / absence information 6 is 0, that is, when the signal is absent (state 1), an integer value obtained by multiplying the phase difference information 601 by 10 is used as the phase adjustment information 602.
Further, when the value of the presence / absence information 6 is 1, that is, the first symbol (state 2a) that becomes “signal unstable”, an integer value obtained by multiplying the phase difference information 601 by 5 is used as the phase adjustment information 602.
When the value of the presence / absence information 6 is 2 to 10, that is, from the second signal to the tenth symbol after the “signal instability” (state 2b), an integer value obtained by multiplying the phase difference information 601 by four is used as the phase. The adjustment information 602 is used.
When the value of the presence / absence information 6 is 11, that is, the first symbol (state 3a) that has become “signal stable”, an integer value obtained by multiplying the phase difference information 601 by 3 is used as the phase adjustment information 602.
In addition, when the value of the presence / absence information 6 is 12 to 20, that is, from the second signal to the 10th symbol (state 3b) after “signal stability”, the phase adjustment is performed on an integer value obtained by doubling the phase difference information 601. Information 602 is assumed.
[0152]
The phase information output unit 653 outputs the phase information 208 obtained by subtracting the phase adjustment information 602 from the phase information 208 one symbol before. For example, since the phase difference information 601 is 0 in the ideal state, the phase adjustment information 602 is 0. Therefore, the phase information 208 is the same as the previous symbol.
[0153]
In the −π / 8 state, the phase difference information 601 is −1.44, so the phase adjustment information 602 is −14 in state 1, −7 in state 2a, −6 in state 2b, −4 in state 3a, In state 3b, it is -3. Therefore, the phase information 208 is 14 larger in the state 1 than in the previous symbol, 7 in the state 2a, 6 in the state 2b, 4 in the state 3a, and 3 in the state 3b.
[0154]
In the + π / 8 state, since the phase difference information 601 is 1.44, the phase adjustment information 602 is 14 in the state 1, 7 in the state 2a, 6 in the state 2b, 4 in the state 3a, and 3 in the state 3b. . Therefore, the phase information 208 is smaller by 14 in state 1, 7 in state 2a, 6 in state 2b, 4 in state 3a, and 3 in state 3b than one symbol before.
[0155]
Here, the description returns to the operation of the complex multiplier 253 again. For example, in the ideal state, since the value of the phase information 208 is the same as that of the previous symbol, there is no change in phase adjustment. That is, the amplitude-adjusted I and Q signals 201 and 202 are complex multiplied by the same value as the previous symbol.
[0156]
In the −π / 8 state, since the value of the phase information 208 is larger than one symbol before, the phase adjustment is changed in the direction in which the phase is shifted clockwise from one symbol before. For example, if the digitally-adjusted I and Q signals 2 and 3 are shifted 15 degrees clockwise one phase before as phase-adjusted I and Q signals 203 and 204, this time is 29 degrees in state 1 and in state 2a The shift is 22 degrees, 21 degrees in state 2b, 19 degrees in state 3a, 18 degrees in state 3 and clockwise. This means that the phase adjustment has been changed clockwise by 14 degrees in state 1, 7 degrees in state 2a, 6 degrees in state 2b, 4 degrees in state 3a, and 3 degrees in state 3b compared to the previous symbol. To do.
[0157]
In the + π / 8 state, since the value of the phase information 208 is smaller than one symbol before, the phase adjustment is changed in a direction in which the phase is shifted counterclockwise from the previous symbol. For example, if the digitally adjusted I and Q signals 2 and 3 shifted 15 degrees clockwise one symbol before are used as phase adjusted I and Q signals 203 and 204, this time, once in state 1 and in state 2a 8 degrees, 9 degrees in state 2b, 11 degrees in state 3a, and 12 degrees in state 3b are shifted clockwise. This means that the phase adjustment has been changed counterclockwise by 14 degrees in state 1, 8 degrees in state 2a, 9 degrees in state 2b, 11 degrees in state 3a, and 12 degrees in state 3b compared to the previous symbol. To do.
[0158]
As described above, the carrier reproduction unit 452 performs different carrier reproduction processing depending on the value of presence / absence information. Specifically, when the value is small, the carrier regeneration process is performed with a coarse pull-in adjustment that emphasizes the pull-in speed, and when the value is large, the carrier regeneration process is performed with a fine pull-in adjustment that emphasizes the pull-in accuracy.
[0159]
As described above, as shown in the first to third descriptions (E-1-2-1) to (E-1-2-3) in the second description, the burst pull-in unit 257 has the phase adjusted I , Q signals 203 and 204 and presence / absence information 6 are input, and timing information 4, phase information 208 and amplitude information 207 are output. The burst pull-in unit 257 performs different burst pull-in operations depending on the value of the presence / absence information 6. Specifically, when the value is small, the pull-in adjustment focusing on the pull-in speed performs coarse burst pull-in operation, and when the value is large, the pull-in adjustment focusing on pull-in accuracy performs fine burst pull-in operation. .
[0160]
As described in the third conventional example, the burst pulling that has been performed after it is determined as “signal present” in the prior art is “signal present”, that is, “ Not only when the value of “signal stable” and presence / absence information 6 is 11 or more, but also when the signal is “no signal”, that is, “signal absent” and “signal unstable”, and the value of presence / absence information 6 is 10 or less Thus, the burst pull-in can be completed at high speed. Further, since the burst pull-in operation is finely switched according to the value of the presence / absence information 6, the burst pull-in can be further completed at high speed.
[0161]
Finally, as the third explanation (E-1-3), an explanation will be given focusing on how the presence / absence information 6 is switched to “signal absent”, “signal unstable”, and “signal stable”. .
[0162]
(E-1-3)
The signal presence / absence determination unit 53 measures received power and creates presence / absence information 6 from the amount of change in received power as follows. If the amount of increase in received power is larger than the threshold “a”, it is determined that the received signal is about to rise, and the presence / absence information 6 is set to “signal unstable”, that is, one of values 1 to 10. Here, when “signal absence” or “signal stable” changes to “signal unstable”, when the value is 1 and “signal unstable” continues, the value is incremented. If the amount of decrease in received power is greater than the threshold “b”, it is determined that the received signal is about to fall, and the presence / absence information 6 is set to “no signal”, that is, the value is zero.
[0163]
In cases other than the above, that is, if the increase amount of the received power is not larger than the threshold value “a” and the decrease amount is not larger than the threshold value “b”, the following is performed.
That is, if “signal unstable” is one point before, it is determined that the rising of the received signal is completed, and the value of presence / absence information in “signal stability” is set to 11.
If one signal before is “signal stable”, it is determined that the signal is still stable, and the value of presence / absence information in “signal stability” is incremented.
If one signal before is “no signal”, it is determined that there is no signal, and the value of presence / absence information in “no signal” is set to zero.
[0164]
The burst pulling unit 257 performs different burst pulling operations depending on the value of the presence / absence information 6. Specifically, when the value of the presence / absence information 6 is small, a pull-in operation with coarse pull-in adjustment focusing on the pull-in speed is performed, and when the value is large, the pull-in adjustment focusing on pull-in accuracy is fine. Pull in.
[0165]
Here, in the conventional technology, the signal presence / absence determination operation performed by the signal presence / absence determination unit 53 is performed not from the amount of change in power but from the value of power itself. However, the following features can be obtained by using the amount of change in power as in the present embodiment. The characteristics will be described with reference to FIGS. 15 (a) and 15 (b).
[0166]
FIG. 15 (a) shows the state of burst pulling when the reception power is low because the transmission power is low or the attenuation in the transmission path is large. FIG. 15 (b) shows the state of burst pulling when the reception power is large because the transmission power is large or the attenuation in the transmission path is small. In FIGS. 15A and 15B, the received signal gradually rises from power 0 and advances completely several symbols later. In the prior art, as shown in FIG. 14, the received signal is determined to have a burst while gradually rising from power 0, and the point in time when it is determined that there is a burst in (a) and (b). Different. As described in the third conventional example, this method has the following problems.
[0167]
Normally, in the burst pull-in, the processing method is finely changed depending on the timing of the preamble. Also in this embodiment, rough control is initially performed in clock recovery, carrier recovery, and gain adjustment, and then gradually switched to fine control. However, according to the prior art, the timing at which the preamble is judged to be different differs depending on the difference in received power, which adversely affects the detailed processing method in carrier recovery and clock recovery. Takes time.
[0168]
In the present embodiment, presence / absence information 6 is created not from the received power itself but from the amount of change in received power. Therefore, it is determined that the signal is unstable near the start of the preamble rise, and the signal is stable near the complete rise, and the value of the presence / absence information 6 is accurately determined at any timing of the preamble regardless of the magnitude of the received power. Therefore, it is possible to obtain a characteristic that processing does not take time without adversely affecting a fine processing method in carrier reproduction and clock reproduction.
[0169]
As described above, in this embodiment, it is determined that there is a burst at a constant timing of the preamble regardless of the magnitude of the transmission power and the magnitude of the reception power caused by the difference in the magnitude of attenuation in the transmission path. Since it is possible to perform fine burst pull-in, burst pull-in can be completed with a short preamble.
[0170]
In the present embodiment, the determination unit 353 of the signal presence / absence determination unit 53 uses the threshold values a and b as constants when creating the presence / absence information 6 using the threshold values a and b. Also good. For example, when the value of the presence / absence information 6 is small, the threshold value is set to be small so that the presence / absence information 6 changes sensitively to fluctuations in received power before the signal is received, and the value of the presence / absence information 6 is large. In some cases, by setting a large threshold, malfunctions due to noise can be suppressed as much as possible after the signal has been stabilized.
[0171]
In this embodiment, only the 10 symbols after “signal stability” is determined, and the fine adjustment of the pull-in with emphasis on the pull-in accuracy is performed, but this need not be the case. . Even after exceeding 10 symbols after it is determined that the signal is stable, the burst pulling operation may be performed by the following method.
[0172]
In the first embodiment regarding clock recovery, the pull-in operation based on the characteristics of the preamble can be continued in the data area. In the preamble, the set of (I0, Q0) and (I2, Q2) points to the signal points in FIG. 9 described above, and in the ideal state where the set of (I1, Q1) is a symbol change point, I0, Q0 The signal amplitudes are all “1”, the signal amplitudes of I1 and Q1 are both “0”, the signal amplitudes of I2 and Q2 are both “−1”, or the signal amplitudes of I0 and Q0 are both The signal amplitudes of “−1”, I1, and Q1 are all “0”, and the signal amplitudes of I2 and Q2 are both “1”. From this,
(I0−I2) × I1 + (Q0−Q2) × Q1 = 0,
Used to be.
[0173]
Even in the data area, if the data is random, the average is zero. Further, if the pull-in operation is performed only when the set of (I0, Q0) and (I2, Q2) is point-symmetric on the IQ plane, in the ideal state, (I0, Q0) = (A, B), (I1 , Q1) = (0,0), (I2, Q2) = a value close to (−A, −B) (A and B are arbitrary numbers),
(I0−I2) × I1 + (Q0−Q2) × Q1 = 0
Therefore, the accuracy can be further increased.
[0174]
Regarding carrier reproduction and gain adjustment, the phase information 208 and the amplitude information 207 are compared by comparing the amplitude and phase of the original signal point 205 on the IQ plane with the amplitude and phase of the phase adjusted I and Q signals 203 and 204. The pull-in operation can be continued in the data area.
[0175]
In this embodiment, the clock recovery unit 451 creates timing difference information 501 from the phase-adjusted I and Q signals 203 and 204, and multiplies the timing difference information 501 from the timing information 4 one symbol before. This timing information 4 is used. At this time, the timing adjustment unit 552 multiplies the timing difference information 501 to obtain the timing adjustment information 502. At this time, the value to be multiplied according to the value of the presence / absence information 6 is switched to 40, 30, etc. May be other values. For example, 35 may be used instead of 40.
[0176]
The present invention is not limited to such details, and the clock recovery unit 451 performs different clock recovery processing depending on the value of the presence / absence information 6 by switching the value to be multiplied according to the value of the presence / absence information 6. . Specifically, when the value is small, the clock recovery process is performed with coarse pull-in adjustment with an emphasis on the pull-in speed, and when the value is large, the clock recovery process is performed with fine pull-in adjustment with an emphasis on the pull-in accuracy.
[0177]
In this embodiment, the carrier reproducing unit 452 creates phase difference information 601 from the phase-adjusted I and Q signals 203 and 204, and subtracts the product of the phase difference information 601 from the phase information 208 one symbol before this time. The phase information 208 was obtained. At this time, the phase adjustment value output unit 652 performs a multiplication process on the phase difference information 601 and switches the value to be multiplied to 10, 5 or the like according to the value of the presence / absence information 6 when the phase adjustment information 602 is obtained. This value may be another value, for example, 15 instead of 10.
[0178]
The present invention is not limited to such details, and the carrier reproduction unit 452 performs carrier reproduction processing that varies depending on the value of presence / absence information by switching the value to be multiplied according to the value of presence / absence information 6. Specifically, when the value is small, the carrier regeneration process is performed with a coarse pull-in adjustment that emphasizes the pull-in speed, and when the value is large, the carrier regeneration process is performed with a fine pull-in adjustment that emphasizes the pull-in accuracy.
[0179]
In the present embodiment, the gain adjustment unit 453 creates amplitude difference information 701 from the phase-adjusted I and Q signals 203 and 204, and subtracts the product of the amplitude difference information 701 from the amplitude information 207 one symbol before this time. Amplitude information 207. At this time, the amplitude adjustment unit 752 multiplies the amplitude difference information 701 to obtain the amplitude adjustment information 702. At that time, the value to be multiplied is switched to 1, 1/5, etc. according to the value of the presence / absence information 6. . This value may be another value, for example, 1/2 instead of 1.
[0180]
The present invention is not limited to such details, and the gain adjustment unit 453 performs gain adjustment processing that varies depending on the value of the presence / absence information 6 by switching the value to be multiplied according to the value of the presence / absence information 6. . Specifically, when the value is small, the gain adjustment process is performed with a rough pull-in adjustment that emphasizes the pull-in speed, and when the value is large, the gain adjustment process is performed with a fine pull-in adjustment that emphasizes the pull-in accuracy.
[0181]
As described above, conventionally, since the value of the presence / absence information 6 did not accurately represent the timing of the preamble, the mechanism of switching the value to be multiplied according to the value of the presence / absence information 6 did not function well. In the present invention, since the value of the presence / absence information 6 accurately indicates the timing of the preamble, the mechanism of switching the value to be multiplied according to the value of the presence / absence information 6 works well, and the burst pull-in is completed at high speed. can do.
[0182]
In this embodiment, the clock recovery unit 451 creates timing difference information 501 from the phase-adjusted I and Q signals 203 and 204, and multiplies the timing difference information 501 from the timing information 4 one symbol before this time. The timing information is 4. However, this need not be the case. For example, a list may be provided, a value is directly derived from the timing difference information 501, and the value may be subtracted from the timing information 4 one symbol before.
[0183]
Further, although the clock recovery unit 451 uses the phase-adjusted I and Q signals 203 and 204 to generate the timing difference information 501, this need not be the case. For example, the amplitude-adjusted I and Q signals 201 and 202 are different in phase on the IQ plane as compared with the phase-adjusted I and Q signals 203 and 204, so that the timing difference information 501 can also be created using this. . The digital I and Q signals 2 and 3 are different in amplitude from the amplitude-adjusted I and Q signals 201 and 202, and may be used.
[0184]
Further, although the clock recovery unit 451 handles the data of 2 samples per symbol, that is, the data oversampled twice, as the digital received signal 102, it may not be so. The reason for this is to obtain the difference between the ideal state and the current state in clock recovery based on the zero-cross method and to obtain timing difference information 501. For example, 4 × oversampled data includes 2 × oversampled data, so zero crossing is possible using 4 × oversampled data. The method can be used.
[0185]
In this embodiment, carrier reproducing section 452 creates phase difference information 601 from phase-adjusted I and Q signals 203 and 204, and multiplies this by subtracting from phase information 208 one symbol before. The current phase information 208 is used. However, this need not be the case. For example, a list may be provided, a value is directly derived from the phase difference information 601, and the value may be subtracted from the phase information 208 one symbol before.
[0186]
Further, although the carrier reproducing unit 452 creates the phase difference information 601 using the phase-adjusted I and Q signals 203 and 204, this need not be the case. In general, a configuration in which feedback is applied to fine control is suitable, but it is not necessary for rough control. Therefore, when performing a pull-in operation with coarse pull-in adjustment, amplitude adjusted I, Q The phase difference information 601 can also be created using the signals 201 and 202. The digital I and Q signals 2 and 3 are different in amplitude from the amplitude-adjusted I and Q signals 201 and 202, and may be used.
[0187]
In the present embodiment, the gain adjustment unit 453 creates amplitude difference information 701 from the phase-adjusted I and Q signals 203 and 204, and subtracts the product of the amplitude difference information 701 from the amplitude information 207 one symbol before this time. Amplitude information 207. However, this need not be the case. For example, a list may be provided, a value is directly derived from the amplitude difference information 701, and the value may be subtracted from the amplitude information 207 one symbol before.
[0188]
Moreover, although the gain adjustment part 453 produced the amplitude difference information 701 using the phase adjusted I and Q signals 203 and 204, it may not be so. In general, a configuration in which feedback is applied to fine control is suitable, but it is not necessary for rough control. Therefore, when performing a burst pull-in operation with a rough pull-in adjustment, the phase difference information 601 can be created using the amplitude-adjusted I and Q signals 201 and 202. The digital I and Q signals 2 and 3 are different in amplitude from the amplitude-adjusted I and Q signals 201 and 202, and may be used.
[0189]
Although the first embodiment has been described with the configuration in which processing is executed only by hardware, this need not be the case. For example, a DSP can be used as hardware, and most of the processing can be executed by software. By replacing each part with “... Step”, software that performs the same operation can be realized.
[0190]
(Embodiment 2)
FIG. 16 is a block diagram showing a configuration of a burst demodulation device according to Embodiment 2 of the present invention. This burst demodulating device is obtained by replacing all the blocks in FIG. 2 and the filters 154 and 155 in FIG. 1 with a DSP in the first embodiment shown in FIGS. The same components are denoted by the same reference numerals and detailed description thereof is omitted.
[0191]
The DSP 851 inputs the digital I and Q reception signals 103 and 104 and outputs the original signal 5. FIG. 17 is a signal processing diagram of the DSP 851. Reference numeral 901 (FIL) denotes a filter step. 902 (SIG) is a signal presence / absence determination step. Reference numeral 903 (GAIN) denotes a gain adjustment multiplication step. 904 (CAR1) is a complex multiplication step. Reference numeral 905 (CLK) denotes a clock recovery step. 906 (CAR2) is a carrier reproduction step. 907 (POINT) is a signal point determination step. 908 (P / S) is a parallel-serial conversion step.
[0192]
The filter step 901 performs the same processing as the filters 154 and 155 in FIG. The signal presence / absence determination step 902 performs the same processing as the signal presence / absence determination unit 53 of FIG. The gain adjustment multiplication step 903 performs the same processing as the gain adjustment unit 453 in FIG. 3 and the multiplication units 251 and 252 in FIG.
[0193]
The complex multiplication step 904 performs the same processing as the complex multiplication unit 253 in FIG. The clock recovery step 905 performs the same processing as the clock recovery unit 451 in FIG. The carrier reproduction step 906 performs the same processing as the carrier reproduction unit 452 of FIG. The signal point determination step 907 performs the same processing as the signal point determination unit 254 of FIG. The parallel-serial conversion step 908 performs the same processing as the parallel-serial conversion unit 255 of FIG. The burst decoding apparatus according to the second embodiment configured as described above performs the same operation as the burst decoding apparatus according to the first embodiment.
[0194]
The present invention includes a program describing the above burst demodulation method and a program medium for recording the program.
[0195]
If the burst pulling method of the present invention is used, burst pulling can be performed satisfactorily even if the length of the preamble is less than 30 symbols. By making the length of the preamble less than 30 symbols, the throughput can be improved.
[0196]
As shown in the sixth conventional example, if the time width of one symbol is changed between the preamble and the data area, the length of the preamble is set to less than 30 symbols without using the burst pull-in unit of the present invention. However, in this method, since it is necessary to handle symbols having different time widths, the circuits are complicated on both the transmission side and the demodulation side. According to the present invention, even if the time width of one symbol is the same in the preamble and the data area, the length of the preamble can be made less than 30 symbols.
[0197]
The present invention relates to a burst transmission apparatus for transmitting a burst having a preamble of less than 30 symbols to the burst pull-in unit, and the time width of one symbol being the same as the time width of one symbol in the data area. Is included.
[0198]
The burst pull-in method of the present invention can be executed even if the time width of one symbol is changed between the preamble and the data area, and the throughput can be improved by using the present invention. it can. Therefore, the burst pull-in method and the burst pull-in part of the present invention are not limited to the case where the time width of one symbol is the same in the preamble and the data area.
[0199]
In the first and second embodiments, when the “signal is absent”, the pull-in operation is performed with coarser adjustment of the pull-in with more emphasis on the pull-in speed than when the “signal is unstable”. Good. For example, if the received power is 0, there is no effect even if the burst pull-in operation is stopped or moved. Also, if the received power is less than the noise level, there is almost no effect even if the burst pull-in operation is performed.
[0200]
Taking the above into consideration, the burst pull-in operation may be stopped during “no signal” to reduce power consumption. Further, it may be stopped during “signal absent” to “signal unstable”, stopped in the initial period of “signal unstable”, or stopped in the initial period of “signal stable”. Even in such a case, it is determined that there is a burst at a certain timing of the preamble regardless of the difference in received power caused by the difference in the magnitude of transmission power or the attenuation in the transmission path. -Strike retraction can be performed.
[0201]
Further, although the three states of “signal absence”, “signal instability”, and “signal stability” have been shown as the state of the received signal, this need not be the case. Conventionally, the presence / absence of a signal is determined simply by whether or not the value of the power exceeds a threshold value.However, as in the embodiment, the presence / absence of a signal is determined by using an increase in power. It is possible to perform fine burst pull-in.
[0202]
Further, it is not necessary to use power for the determination of the presence or absence of a signal. The timing difference information 501, the phase difference information 601, and the amplitude difference information 701 have a value greater than 0 in the ideal state and greater than 0 otherwise. Using this property, these may be used instead of electric power. In other words, the change amount of the timing difference information 501, the phase difference information 601, or the amplitude difference information 701 may be measured as the certainty information, and the presence / absence information 6 may be created based on the change amount. However, as in Embodiments 1 and 2, it is possible to obtain the effect of reducing the circuit or program by measuring the power of the received signal and using it as certainty information.
[0203]
The burst pull-in method of the present invention inputs the received signal consisting of the preamble and the subsequent data area without being limited to such details, and completes the burst pull-in by the end of the preamble. In the burst demodulation method of demodulating the data area and outputting the original signal, a signal presence / absence determination step for inputting a digital signal generated based on the received signal and outputting presence / absence information, and a digital signal and presence / absence And a digital demodulation step of switching the burst pull-in method based on presence / absence information, performing burst pull-in and demodulation, and outputting an original signal.
[0204]
In the signal presence / absence determination step, the received signal is input, the probability that the modulated signal is included is measured, the certainty step that is output as certainty information, the certainty information is input, and the amount of increase is measured. An increase amount step for outputting as a certainty increase amount, and a presence / absence step for inputting the certainty increase amount, obtaining a degree of stability when the reception signal rises and stabilizing, and outputting as presence / absence information (Hereinafter referred to as feature 1).
[0205]
With this feature 1, it is possible to perform fine burst drawing. Specifically, in the first and second embodiments, a total of five stages of “signal absence”, “signal instability” and two stages of “signal stability” are provided, and the pulling accuracy is gradually emphasized. It was assumed that the fine pull-in operation was performed with fine pull-in adjustment. However, this may not be the case, for example, it may be divided into three stages of “signal absent”, “signal unstable” and “signal stable”, or “signal absent” (“signal absent” and “signal unstable”). ), “Signal present” (“signal stable”). By doing so, it can be realized by a circuit or program smaller than those of the first and second embodiments.
[0206]
Alternatively, the value of the presence / absence information 6 may be gradually reduced every step so that the accuracy of the pull-in is emphasized and the pull-in operation with fine pull-in adjustment may be performed. By doing so, the pull-in speed can be made faster than in the first and second embodiments.
[0207]
In the first and second embodiments, the configuration is such that the falling edge of the received signal is detected and “no signal” is present, but this need not be the case. Even if the mechanism is removed, if the power value 301 is greater than the delay power 302 by a or more (a is a constant), it is determined that the received signal is now rising, and the presence / absence information 6 is temporarily set to “signal unstable”. Has been. For this reason, the signal becomes unstable once when the preamble rises, so the same operation as in the first and second embodiments is performed. With this configuration, since there is no “signal absent” state, there are only two states, “No signal” (“Signal unstable”) and “Signal present” (“Signal stable”). Becomes easier.
[0208]
The burst pull-in method of the present invention is not limited to any details, and in addition to the above feature 1, the presence / absence information has at least two states of “no signal” and “signal present”, and the presence / absence step increases certainty. When the large amount is larger than a (a is a positive value), the presence / absence information is “no signal”, and when the certainty increase amount is smaller than a, the presence / absence information is “no signal”, and the presence / absence information is It is characterized by “with signal” (hereinafter referred to as feature 2). With this feature 2, it is possible to perform a more detailed burst pull than feature 1.
[0209]
In addition to the feature 2, the presence / absence information includes at least three states of “signal absence”, “signal instability”, and “signal stability”. The method includes a reduction amount step for inputting the certainty information, measuring the amount of reduction, and outputting it as the certainty reduction amount. In the presence / absence step, the amount of increase in certainty and the amount of decrease in certainty are input and the presence / absence information is output. When the amount of certainty increase is larger than a (a is a positive value), When the certainty decrease amount is larger than b (b is a positive value), the presence / absence information is “signal absent”, the certainty increase amount is smaller than a, and the certainty decrease amount is smaller than b. If the presence / absence information before one time point is “signal unstable”, the presence / absence information is “signal stable” (hereinafter referred to as feature 3). With this feature 3, it is possible to perform a more detailed burst pull than feature 2.
[0210]
In the burst pull-in method of the present invention, in addition to the feature 3, the “signal instability” of the presence / absence information has n state values of 1, 2,..., N (n is a natural number). Hereinafter, “signal instability” of the state value k is referred to as “signal instability k”.) In the presence / absence step, if the certainty increase amount is larger than a, the presence / absence information one time before is “signal instability”. The state value is incremented (hereinafter referred to as feature 4).
[0211]
In addition to the feature 3, the “signal stability” of the presence / absence information has m state values of 1, 2,..., M (m is a natural number, below). In the presence / absence step, when the certainty decrease amount is smaller than b and the certainty increase amount is smaller than a, the presence / absence information before one time point is “ If the signal is unstable, the presence / absence information is “signal present 1”, and if the presence / absence information one time before is “signal stable”, the state value is incremented (hereinafter referred to as feature 5). By these features 4 or 5, it is possible to perform a more detailed burst pull than feature 3.
[0212]
In the first and second embodiments, three examples of the clock pull-in operation, the clock reproduction, the carrier reproduction, and the gain adjustment are mentioned as the burst pull-in operation. For example, gain adjustment is unnecessary in digital modulation that includes information only in the phase component and does not include information in the amplitude component, such as PSK. In addition, carrier reproduction is unnecessary in digital modulation in which information is included only in the amplitude component and information is not included in the phase component as in ASK. In addition, in a modulation method in which information is put on and off of a signal, only carrier recovery may be performed without performing carrier recovery and gain adjustment.
[0213]
In some cases, a burst pull-in operation other than the above three may be required. For example, when there is a large error between the carrier frequency handled by the digital down converter 153 and the carrier frequency at the time of actual transmission, the operation of “frequency adjustment” is necessary.
[0214]
The burst pull-in method of the present invention is not limited to such details. In addition to the feature 1, the digital demodulation step outputs an original signal and timing information. It is created by sampling the received signal based on the information. In the digital demodulation step, a signal based on the digital signal (hereinafter referred to as a corrected digital signal) and presence / absence information are input, a burst pulling step for outputting a pulling execution variable, a digital signal is input, and the original signal is converted. And a demodulation step for outputting. In the burst pull-in step, a burst pull-in operation is performed by obtaining a pull-in execution variable so that the corrected digital signal is in an ideal state based on the characteristics of the preamble. It is characterized by adopting a different method depending on.
[0215]
As a specific configuration, the burst pulling step in the burst pulling method of the present invention includes a difference step for obtaining difference information based on the difference between the state of the digital signal and the ideal state, and an adjustment value from the difference information. An adjustment step; and a pull-in execution step for obtaining a new pull-in execution variable from the adjustment value and the previously determined pull-in execution variable. The adjustment step is characterized in that the difference value having the same value is increased when the presence / absence information indicates instability and is decreased when the presence / absence information indicates stability.
[0216]
【The invention's effect】
As described above, according to the burst demodulation method of the present invention, when data is decoded by inputting a reception signal composed of a preamble and the subsequent data area, the burst pull-in is performed before the end of the preamble. Is completed, the original signal can be output by demodulating the data area. At this time, there is provided a signal presence / absence determination step for outputting presence / absence information, and a digital demodulation step for switching the burst pull-in method based on the presence / absence information to perform burst pull-in and demodulation and outputting the original signal. In the signal presence / absence determination step, the probability of including the modulated signal by inputting the received signal is measured and output as certainty information. Also, certainty information is input, the amount of increase is measured and output as the amount of certainty increase. Further, the amount of increase in certainty is input, and the degree of stability when the received signal rises and stabilizes is obtained. By doing so, it is possible to determine that there is a burst at a constant timing of the preamble regardless of the difference in received power caused by the difference in the magnitude of transmission power or the amount of attenuation in the transmission path. In this way, it is possible to perform the finest burst pull-in, so that the burst pull-in can be completed with a short preamble.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an overall configuration of a burst demodulation device according to a first embodiment of the present invention.
FIG. 2 is a configuration diagram of a sampling unit in the first embodiment.
FIG. 3 is a configuration diagram of a signal presence / absence determination unit in the first embodiment.
4 is a configuration diagram of a digital demodulation unit in Embodiment 1. FIG.
FIG. 5 is a configuration diagram of a burst pull-in unit in the first embodiment.
6 is a configuration diagram of a clock recovery unit in Embodiment 1. FIG.
FIG. 7 is a configuration diagram of a carrier reproducing unit in the first embodiment.
FIG. 8 is a configuration diagram of a gain adjustment unit in the first embodiment.
FIG. 9 is a signal point arrangement diagram of a preamble.
FIG. 10 is an explanatory diagram of sample timing adjustment.
FIG. 11 is an explanatory diagram of amplitude adjustment.
FIG. 12 is an explanatory diagram of phase adjustment.
FIG. 13 is a signal point arrangement diagram of 16QAM.
FIG. 14 is an explanatory diagram of burst pulling in when receiving power is detected.
FIG. 15 is an explanatory diagram of burst pulling when a change in received power is detected.
FIG. 16 is a block diagram showing an overall configuration of a burst demodulation device according to Embodiment 2 of the present invention.
FIG. 17 is a program diagram of a burst demodulation method in the second embodiment.
18 is an overall configuration diagram of a burst mode demodulator described in Patent Document 1. FIG.
19 is a configuration diagram of a clock phase estimation circuit described in Patent Document 1. FIG.
FIG. 20 is an overall configuration diagram of a burst signal demodulator in a third conventional example.
FIG. 21 is a configuration diagram of a burst detection means described in Patent Document 3.
[Explanation of symbols]
1 Received signal
2 Digital I signal
3 Digital Q signal
4 Timing information
5 original signal
6 Presence / absence information
51 Sampling unit
52 Digital demodulator
53 Signal presence / absence judgment unit
101 AD clock
102 Digital received signal
103 Digital I signal
104 Digital Q signal
151 AD clock section
152 AD converter
153 Digital down converter
154,155 filters
201 Amplitude adjusted I signal
202 Q signal after amplitude adjustment
203 Phase-adjusted I signal (corrected digital I signal)
204 Phase-adjusted Q signal (corrected digital Q signal)
205 original signal points
207 Amplitude information
208 Phase information
251,252 Multiplier
253 Complex multiplier
254 Signal point determination unit
255 Parallel-serial converter
257 Burst lead-in part
301 Power value
302 Delay power
351 Power measurement unit
352 Delay part
353 judgment part
354 Maximum value storage
451 Clock recovery unit
452 Carrier playback unit
453 Gain adjuster
501 Timing difference information
502 Timing adjustment information
551 Timing difference information output unit
552 Timing adjustment value output section
553 Timing information output unit
601 Phase difference information
602 Phase adjustment information
651 Phase difference information output unit
652 Phase adjustment value output section
653 Phase information output unit
701 Amplitude difference information
702 Amplitude adjustment information
751 Amplitude difference information output unit
752 Amplitude adjustment value output section
753 Amplitude information output section

Claims (30)

A burst demodulation method in which a reception signal composed of a preamble and a data area following the preamble is input, the burst pull-in is completed by the end of the preamble, the original signal is output by demodulating the data area Because
A signal presence / absence determination step of inputting a digital reception signal obtained by digitally converting the reception signal and outputting presence / absence information indicating the certainty of whether or not the modulated signal is included in the digital reception signal;
A digital demodulation step of inputting the digital received signal and the presence / absence information, switching a burst pull-in method based on the presence / absence information, performing burst pull-in and demodulation, and outputting the original signal; And
The signal presence / absence determining step includes:
A certainty information output step of inputting the digital received signal, measuring the probability of including the modulated signal, and outputting the result as certainty information;
An increase amount output step of inputting the certainty information, measuring the increase amount, and outputting the result as the certainty increase amount;
A presence / absence information output step of inputting the certainty increase amount, obtaining a degree of stability when the digital reception signal rises and stabilizes, and outputting the degree as the presence / absence information. Burst demodulation method. The presence / absence information is information indicating at least two states of no signal and signal,
The presence / absence information output step includes:
When the certainty increase amount is larger than a (a is a positive value), the presence / absence information is not signaled, and when the certainty increase amount is smaller than a, if the presence / absence information one time before is not signaled, 2. The burst demodulation method according to claim 1, wherein presence / absence information is signaled. The presence / absence information is information indicating at least three states of signal absence, signal instability, and signal stability,
The signal presence / absence determining step includes:
The method further includes a reduction amount output step of inputting the certainty information, measuring the reduction amount, and outputting the result as the certainty reduction amount,
The presence / absence information output step includes:
The certainty increase amount and the certainty decrease amount are input and the presence / absence information is output, and when the certainty increase amount is larger than a (a is a positive value), the presence / absence information is signaled. When the certainty decrease amount is larger than b (b is a positive value), the presence / absence information is absent, and the certainty increase amount is smaller than a and the certainty decrease amount is smaller than b. 2. The burst demodulation method according to claim 1, wherein if the presence / absence information before one time point is unstable, the presence / absence information is stabilized. The signal instability of the presence / absence information has n state values of 1, 2,..., N (n is a natural number), and the signal instability of the state value k (k is a natural number less than n) is signal unstable. k and
The presence / absence information output step includes:
4. The burst demodulation method according to claim 3, wherein if the certainty increase amount is larger than a, the value of k is incremented if the presence / absence information one time before is unstable. The signal stability of the presence / absence information has m state values of 1, 2,..., M (m is a natural number), and the signal stability of the state value j (j is a natural number of m or less) is represented as signal presence j. ,
The presence / absence information output step includes:
When the certainty decrease amount is smaller than b and the certainty increase amount is smaller than a, if the presence / absence information before one time point is unstable, the presence / absence information is set to 1 with the presence / absence information. 4. The burst demodulation method according to claim 3, wherein if the signal is stable, the value of j is incremented. The correction digital signal is obtained by digitally converting the received signal and correcting it by burst pulling,
The digital demodulation step includes
A burst pulling step for inputting the correction digital signal and the presence / absence information and outputting a new pull execution variable;
An original signal output step of inputting the corrected digital signal and outputting the original signal; and
The burst pulling step includes:
Based on the characteristics of the preamble, a burst pull-in operation is performed by obtaining the pull-in execution variable so that the state of the correction digital signal becomes an ideal state, and a different method is used depending on the presence / absence information. 2. The burst demodulation method according to claim 1, wherein: The burst pulling step includes:
A difference information output step for obtaining difference information based on a difference between the state of the corrected digital signal and the ideal state;
An adjustment value output step for obtaining an adjustment value of the pull-in execution variable from the difference information;
A pull-in execution variable output step for obtaining a new pull-in execution variable from the adjustment value and the pull-in execution variable obtained last time;
The adjustment value output step includes:
7. The adjustment value for the difference information having the same value is increased when the presence / absence information indicates instability, and is decreased when the presence / absence information indicates stability. Burst demodulation method. The corrected digital signal has been subjected to a process of sampling the received signal based on at least timing information,
The burst pulling step includes:
A clock recovery step for inputting the corrected digital signal and the presence / absence information and outputting the timing information;
The clock recovery step includes
7. The burst demodulation method according to claim 6, wherein a method for obtaining the timing information is switched according to the presence / absence information. The corrected digital signal is generated by oversampling the received signal twice.
The clock recovery step includes
Three samples of the corrected digital signal are held, and the zero-cross method is used so that one of the first and third samples has an ideal signal amplitude of A, the other is -A, and the second sample is zero. Based on the timing information,
A timing difference information output step for obtaining timing difference information based on a difference between the state of the corrected digital signal obtained by the timing information and the ideal state;
A timing adjustment value output step for obtaining a timing adjustment value from the timing difference information;
A timing information output step for obtaining new timing information from the timing adjustment value and the timing information obtained last time,
The timing adjustment value output step includes:
The timing adjustment value with respect to the timing difference information having the same value is increased when the presence / absence information indicates instability, and is decreased when the presence / absence information indicates stability. 9. The burst demodulation method according to 8. The digital demodulation step includes
The digital reception signal and phase information that is one of the pull-in execution variables are input, the phase adjustment is performed by complex multiplication of the digital reception signal by the phase information, and a complex digital signal that is output as the corrected digital signal is output. A multiplication step,
The burst pulling step includes:
A carrier reproduction step of inputting the corrected digital signal and the presence / absence information and outputting the phase information;
The carrier reproduction step includes
7. The burst demodulation method according to claim 6, wherein a method for obtaining the phase information is switched according to the presence / absence information. The carrier reproduction step includes
The correction digital signal is input, and the phase information is obtained so that the phase on the IQ plane of the correction digital signal alternately becomes an ideal state of 0 and π for each symbol,
A phase difference information output step for obtaining phase difference information based on a difference between the state of the corrected digital signal and the ideal state;
A phase adjustment value output step for obtaining a phase adjustment value from the phase difference information;
A phase information output step for obtaining new phase information from the phase adjustment value and the phase information obtained last time,
The phase adjustment value output step includes:
The phase adjustment value for the phase difference information having the same value is increased when the presence / absence information indicates instability, and is decreased when the presence / absence information indicates stability. The burst demodulation method according to 10. The digital demodulation step includes
A multiplication step of inputting the digital reception signal and amplitude information which is one of the pull-in execution variables, performing amplitude adjustment by multiplying the digital reception signal by the amplitude information, and outputting as the corrected digital signal Further comprising
The burst pulling step includes:
A gain adjustment step of inputting the corrected digital signal and the presence / absence information and outputting the amplitude information;
The gain adjustment step includes
7. The burst demodulation method according to claim 6, wherein a method for obtaining the amplitude information is switched according to the presence / absence information. The gain adjustment step includes
The correction digital signal is input, and the amplitude information is obtained so that each of the amplitudes I and Q of the correction digital signal is in an ideal state in which a reference value is obtained,
An amplitude difference information output step for obtaining amplitude difference information based on a difference between the state of the corrected digital signal and the ideal state;
An amplitude adjustment value output step for obtaining an amplitude adjustment value from the amplitude difference information;
An amplitude information output step for obtaining new amplitude information from the amplitude adjustment value and the previously obtained amplitude information,
The amplitude adjustment value output step includes:
The amplitude adjustment value with respect to the amplitude difference information having the same value is increased when the presence / absence information indicates instability, and is decreased when the presence / absence information indicates stability. 12. The burst demodulation method according to 12. 2. The burst demodulation method according to claim 1, wherein in the certainty information output step, the certainty information is measured by measuring the power of the received signal. A program medium storing a program according to any one of the burst demodulation methods according to claim 1. A burst demodulator that inputs a received signal consisting of a preamble and a data area that follows the preamble, completes the burst pull-in by the end of the preamble, demodulates the data area, and outputs the original signal Because
A signal presence / absence determination unit that inputs a digital reception signal obtained by digitally converting the reception signal and outputs presence / absence information indicating whether or not the modulated signal is included in the digital reception signal;
A digital demodulator that inputs the digital received signal and the presence / absence information, switches a burst pull-in method based on the presence / absence information, performs burst pull-in and demodulation, and outputs the original signal; And
The signal presence determination unit
A certainty information output unit that inputs the digital received signal, measures the probability of including the modulated signal, and outputs the result as certainty information;
An increase amount output unit that inputs the certainty information, measures the increase amount, and outputs the result as the certainty increase amount;
A presence / absence information output unit that inputs the certainty increase amount, obtains a degree of stability when the digital reception signal rises and stabilizes, and outputs the degree as the presence / absence information; Burst demodulator. The presence / absence information is information indicating at least two states of no signal and signal,
The presence / absence information output unit includes:
When the certainty increase amount is larger than a (a is a positive value), the presence / absence information is not signaled, and when the certainty increase amount is smaller than a, if the presence / absence information one time before is not signaled, The burst demodulator according to claim 16, wherein presence / absence information is signaled. The presence / absence information is information indicating at least three states of signal absence, signal instability, and signal stability,
The signal presence determination unit
Further comprising a reduction amount output unit for inputting the certainty information, measuring the reduction amount, and outputting the result as the certainty reduction amount,
The presence / absence information output unit includes:
The certainty increase amount and the certainty decrease amount are input and the presence / absence information is output, and when the certainty increase amount is larger than a (a is a positive value), the presence / absence information is signaled. When the certainty decrease amount is larger than b (b is a positive value), the presence / absence information is absent, and the certainty increase amount is smaller than a and the certainty decrease amount is smaller than b. 17. The burst demodulator according to claim 16, wherein if the presence / absence information one time before is unstable, the presence / absence information is stabilized. The signal instability of the presence / absence information has n state values of 1, 2,..., N (n is a natural number), and the signal instability of the state value k (k is a natural number less than n) is signal unstable. k and
The presence / absence information output unit includes:
19. The burst demodulator according to claim 18, wherein if the certainty increase amount is larger than a, the value of k is incremented if the presence / absence information one time before is unstable. The signal stability of the presence / absence information has m state values of 1, 2,..., M (m is a natural number), and the signal stability of the state value j (j is a natural number of m or less) is represented as signal presence j. ,
The presence / absence information output unit includes:
When the certainty decrease amount is smaller than b and the certainty increase amount is smaller than a, if the presence / absence information one time before is signal unstable, the presence / absence information is set to 1 with the presence / absence information. 19. The burst demodulator according to claim 18, wherein if the signal is stable, the value of j is incremented. The correction digital signal is obtained by digitally converting the received signal and correcting it by burst pulling,
The digital demodulator
A burst pull-in unit that inputs the correction digital signal and the presence / absence information and outputs a new pull-in execution variable;
An original signal output unit that inputs the corrected digital signal and outputs the original signal;
The burst lead-in part is
Based on the characteristics of the preamble, a burst pull-in operation is performed by obtaining the pull-in execution variable so that the state of the correction digital signal becomes an ideal state, and a method of obtaining different depending on the presence / absence information is used. The burst demodulator according to claim 16, characterized in that The burst lead-in part is
A difference information output unit for obtaining difference information based on a difference between the state of the corrected digital signal and the ideal state;
An adjustment value output unit for obtaining an adjustment value of the pull-in execution variable from the difference information;
A pull-in execution variable output unit for obtaining a new pull-in execution variable from the adjustment value and the pull-in execution variable obtained last time;
The adjustment value output unit
22. The adjustment value for the difference information having the same value is increased when the presence / absence information indicates instability, and is decreased when the presence / absence information indicates stability. Burst demodulator. The corrected digital signal has been subjected to a process of sampling the received signal based on at least timing information,
The burst lead-in part is
A clock recovery unit that inputs the corrected digital signal and the presence / absence information and outputs the timing information,
The clock recovery unit
The burst demodulation apparatus according to claim 21, wherein the timing information is switched in accordance with the presence / absence information. The corrected digital signal is generated by oversampling the received signal twice.
The clock recovery unit
Three samples of the corrected digital signal are held, and the zero-cross method is used so that one of the first and third samples has an ideal signal amplitude of A, the other is -A, and the second sample is zero. Based on the timing information,
A timing difference information output unit for obtaining timing difference information based on a difference between the state of the corrected digital signal obtained by the timing information and the ideal state;
A timing adjustment value output unit for obtaining a timing adjustment value from the timing difference information;
A timing information output unit for obtaining new timing information from the timing adjustment value and the timing information obtained last time;
The timing adjustment value output unit includes:
The timing adjustment value with respect to the timing difference information having the same value is increased when the presence / absence information indicates instability, and is decreased when the presence / absence information indicates stability. The burst demodulator according to claim 23. The digital demodulator
The digital reception signal and phase information that is one of the pull-in execution variables are input, phase adjustment is performed by complex multiplication of the phase information to the digital reception signal, and a complex digital signal that is output as the corrected digital signal is output. A multiplier,
The burst lead-in part is
A carrier reproducing unit that inputs the corrected digital signal and the presence / absence information and outputs the phase information;
The carrier reproduction unit
The burst demodulator according to claim 21, wherein a method of obtaining the phase information is switched according to the presence / absence information. The carrier reproduction unit
The correction digital signal is input, and the phase information is obtained so that the phase on the IQ plane of the correction digital signal alternately becomes an ideal state of 0 and π for each symbol,
A phase difference information output unit for obtaining phase difference information based on a difference between the state of the corrected digital signal and the ideal state;
A phase adjustment value output unit for obtaining a phase adjustment value from the phase difference information;
A phase information output unit for obtaining new phase information from the phase adjustment value and the phase information obtained last time;
The phase adjustment value output unit is
The phase adjustment value for the phase difference information having the same value is increased when the presence / absence information indicates instability, and is decreased when the presence / absence information indicates stability. 25. A burst demodulation device according to 25. The digital demodulator
A multiplier that inputs the digital received signal and amplitude information that is one of the pull-in execution variables, performs amplitude adjustment by multiplying the digital received signal by the amplitude information, and outputs the corrected digital signal Further comprising
The burst lead-in part is
The correction digital signal and the presence / absence information are input, and a gain adjustment unit that outputs the amplitude information is provided.
The gain adjusting unit is
The burst demodulator according to claim 21, wherein the method for obtaining the amplitude information is switched according to the presence / absence information. The gain adjusting unit is
The correction digital signal is input, and the amplitude information is obtained so that each of the amplitudes I and Q of the correction digital signal is in an ideal state in which a reference value is obtained,
An amplitude difference information output unit for obtaining amplitude difference information based on a difference between the state of the corrected digital signal and the ideal state;
An amplitude adjustment value output unit for obtaining an amplitude adjustment value from the amplitude difference information;
An amplitude information output unit for obtaining new amplitude information from the amplitude adjustment value and the previously obtained amplitude information,
The amplitude adjustment value output unit includes:
The amplitude adjustment value with respect to the amplitude difference information having the same value is increased when the presence / absence information indicates instability, and is decreased when the presence / absence information indicates stability. 27. A burst demodulator according to 27. 17. The burst demodulator according to claim 16, wherein the certainty information output unit measures the power of the received signal to obtain the certainty information. A signal is transmitted to the burst demodulation device using the burst demodulation method according to any one of claims 1 to 14, or to the burst demodulation device according to any one of claims 16 to 29. A burst transmission device,
A burst signal consisting of a preamble followed by a data area is transmitted,
The preamble is less than 30 symbols, and the time width of one symbol is the same as the time width of one symbol in the data area.

Images