JP5745293B2 - Receiving method and receiving apparatus for spread spectrum communication system - Google Patents

Description

  The present invention relates to a receiving method and receiving apparatus for a spread spectrum communication system, and more particularly to a receiving method and receiving apparatus for a spread spectrum communication system suitable for long-distance wireless communication.

  In a sensor network that is expected to be put into practical use with the arrival of the ubiquitous information society, there is a demand to increase the communication distance even though the communication speed may be slow. For example, warehouse inventory management, farm management, and wildlife monitoring using wireless communication.

  On the other hand, the communication distance of the current RFID (Radio Frequency IDentification), which is regarded as promising as a communication device mounted on a sensor, is only tens of meters.

  Therefore, if the communication distance can be expanded while taking advantage of RFID, the realization of a ubiquitous information society will be further promoted.

  According to Shannon's theorem, the theoretical limit of communication capacity C is defined by the following equation.

  Here, W is the bandwidth of the communication path, S is the signal power, and N is the noise power. Assuming that the received power (signal power S) is attenuated in proportion to the communication distance D to the α power, the following equation can be derived.

  Here, a is a constant. Therefore, it can be seen that the communication distance D can be increased by reducing the communication capacity C (lowering the communication speed).

  However, in actual communication, the communication distance is limited by performance degradation due to frequency shift and time shift. Hereinafter, performance degradation due to frequency shift and time shift will be described.

(1) Performance degradation due to frequency shift FIG. 1 is a system configuration diagram of general spread spectrum communication. The spread spectrum communication system includes a transmission device 1 and a reception device 2.

  The transmission apparatus 1 multiplies the spread signal 5 by which the transmission signal 3 is spread by the spread code 4, the carrier frequency generation means 6, the signal generated by the carrier frequency generation means 6 and the output signal of the spread means 5. The multiplier 7 for converting the frequency of the output signal, and the transmitting antenna 8 for transmitting the output signal of the multiplier 7 as a transmission signal wirelessly.

  The receiving device 2 multiplies the receiving antenna 9 that receives the transmission signal from the transmitting device 1, the carrier frequency generating means 10, the signal generated by the carrier frequency generating means 10 and the output signal of the receiving antenna 9. The multiplier 11 for converting the frequency of the output signal and the despreading means 13 for despreading the spectrum of the output signal of the multiplier 11 with the spread code 12 are configured. Then, the output of the despreading means 13 becomes the received signal 14.

  Applying Shannon's deformation theorem shown in Eq. 2 to spread spectrum communication with bandwidth W, it can be seen that the communication distance D can be expanded by narrowing the band, that is, by increasing the spreading sequence length (decreasing the communication speed). .

  However, in an actual communication system, there is a slight carrier frequency shift between the transmission device 1 and the reception device 2, and this causes the phase of the reception signal to rotate during the despreading period. As a result, a correct processing gain cannot be obtained, and the communication distance D is limited.

  The spread spectrum normalization gain Gp is given by the following equation, where θ is the phase rotation that occurs during the despreading period.

40kHz carrier frequency 2, the chip rate 1kc / s, the transmitting apparatus 1 and receiving apparatus 2 frequency stability of the case of a 10ppm respectively, showing the relationship of the spreading sequence length L and normalization processing gain G _p. In this figure, it can be seen that as the spreading sequence length L increases, the normalization processing gain G _p drops sharply from 1. FIG. 3 shows the relationship between the spreading sequence length L and the processing gain G _p (not normalized). In this figure, it can be seen that as the spreading sequence length L increases, the processing gain G _p peaks and the spreading ratio is limited. A dotted line extending vertically from the horizontal axis in the figure indicates a position where the limitation of the diffusion ratio starts. From these figures, it can be seen that if the spreading sequence length L is increased, a sufficient processing gain G _p cannot be obtained.

(2) Performance degradation due to time shift In addition to the carrier frequency, a slight data clock frequency shift occurs between the transmitter 1 and the receiver 2. As a result, a chip shift of the spreading sequence occurs during the despreading period, and similarly, a correct processing gain cannot be obtained, and the communication distance D is limited.

In FIG. 4, when the carrier frequency is 40 kHz, the chip speed is 1 kc / s, and the frequency stability of the carrier frequency of the transmission device 1 and the reception device 2 is 10 ppm, respectively, the reverse occurs between the transmission device 1 and the reception device 2. The relationship between the spreading sequence length L and the processing gain due to the effect of chip shift occurring during the spreading period is shown. In this figure, when spreading sequence length L is large, it can be seen that the normalization processing gain G _p is depressed rapidly from 1.

  In actual communication, both the frequency shift and the time shift are affected at the same time, and the communication distance D is limited by the performance deterioration of the frequency shift and the time shift.

  In order to solve such a problem, the conventional technique detects a toggle point candidate existing in the received signal carrier by a correlation operation between the expected signal held in advance and the received signal carrier on the receiving side. A method of demodulating a received signal by multiplying the received signal by a spread code shifted based on the detection result is disclosed (Patent Document 1).

  Also disclosed are a method for demodulating a signal with a frequency shift by delay detection (Patent Document 2), a carrier frequency offset correction method (Patent Document 3), and the like.

Japanese Patent No. 3639839 Japanese Patent Laid-Open No. 9-93217 JP2007-202088

  However, in the above Patent Documents 1 to 3, the spread spectrum sequence length is set in a range that is not affected by the carrier frequency deviation of the transmitter / receiver, and the carrier frequency of the transmitter is detected from the received signal having a good S / N. The method to do is adopted. When it is necessary to use a long spread spectrum sequence length, for example, a high-accuracy oscillator such as an oscillator with a temperature compensation function is used.

  Therefore, the spread spectrum sequence length is set in a range where the influence of the carrier frequency deviation between the transmitter and receiver is large, and the transmission signal is narrowed, that is, the communication speed is reduced, so that transmission over a long distance without increasing the transmission power is performed. It was not a thing.

  For example, in FIG. 5, regarding the relationship between the signal S and the noise N transmitted in the spread spectrum communication system, if there is a partial sequence transmission signal S and noise N divided into n, the partial sequence transmission signal S is added. When the amplitude is increased by n times, the signal power is n2 times and the noise power is n times. Here, if the output of the transmission signal S is increased, the S / N is improved, but the upper limit of the high output radio wave is determined by the regulations of the Radio Law.

  FIG. 6 shows the relationship between the signal S of a partial sequence divided into n received by the spread spectrum communication system and the noise N. On the receiving side, the carrier frequency shift (transmission of each partial sequence between the transmitting side and the transmitting side). Therefore, even if the partial sequence received signal S is added, the S / N is poor, and the receiving side needs to demodulate under a condition where the S / N is bad.

  Accordingly, an object of the present invention is to provide a receiving method and a receiving apparatus for a spread spectrum communication system that can eliminate the limitation of narrowing the signal band and perform wireless long-distance communication.

The invention according to claim 1 of the present invention frequency-converts a received signal with a carrier frequency signal, divides a spread sequence of length L into partial sequences of length ΔL for the frequency-converted received signal, A despreading vector is calculated for each partial sequence, the phase of the despreading vector calculated for each partial sequence is rotated by a fixed angle, positive and negative, and branched into two branches until the end of the spreading sequence. A combined despread vector obtained by generating a binary tree by continuing to branch into two branches, and combining the despread vectors calculated for each partial series for a plurality of paths from the beginning to the end of the binary tree Each of the plurality of paths is detected as the despread data to be demodulated, and the path having the maximum norm of the combined despread vector is detected .
The carrier frequency is F [Hz], the frequency stability of the carrier frequency on the transmission side and A _t [ppm], the frequency stability of the carrier frequency on the receiving side is A _r [ppm], the chip rate R _C [cps], the length of the spreading sequence is L [chip], and the phase rotation allowed in the despread vector calculation process is α [rad],
A frequency pitch Δf for dividing the spread sequence at equal intervals is expressed by the following equation:

  As a result, the spread sequence of the received signal length L is divided into n by the following equation:

  The length of the partial sequence ΔL is not more than the value shown in the following equation.

  It is characterized by that.

In the invention according to claim 2 of the present invention, as the phase of the despreading vector is rotated by a certain angle positively or negatively for each partial series, the time axis change due to the accumulation of the phase rotation of the despreading vector is: When the minimum slide interval is reached when calculating the despreading vector, the partial sequence input before despreading is shifted in consideration of the time lag Δt defined by the following equation :

  It is characterized by that.

The invention according to claim 3 of the present invention divides a spread sequence of length L into partial sequences of length ΔL for carrier frequency generating means for generating a carrier frequency and a received signal frequency-converted by the carrier frequency. a spreading sequence dividing means, the in spreading sequence division means each partial sequence obtained by dividing a matched filter is a despreading means for calculating the despreading vector by despreading the received signal, the phase of the previous Kigyaku spreading vector Binary tree generating means for rotating the image to a predetermined angle by rotating the image to positive and negative, branching to two branches, and branching to the two branches until the end of the spreading sequence, and a plurality of binary trees from the beginning to the end For each path, a norm of a combined despreading vector obtained by combining the despread vectors calculated for each partial series is calculated, and among the plurality of paths, the nodal of the combined despread vector is calculated. With a maximum value detecting means which beam is detected as the despread data to be demodulated paths having the maximum on the basis of the despread data of the maximum value detecting means, and demodulating means for demodulating the received signal, and
The spreading sequence of length L of the received signal, the carrier frequency is F [Hz], the frequency stability of the transmitter and A _t [ppm], the frequency stability of the receiver and A _r [ppm], chips When the speed is R _C [cps], the length of the spreading sequence is L [chip], and the phase rotation α [rad] allowed in the despreading vector calculation process, the spreading sequence is equally spaced The frequency pitch Δf to be divided into

  As a result, the spread sequence of the received signal length L is divided into n by the following equation:

  The spreading sequence dividing means is configured such that the length ΔL of the partial sequence is equal to or less than the value shown in the following equation.

The invention according to claim 4 of the present invention involves pre-Symbol the phase of the despread vector for each partial sequence to cause a predetermined angular rotation to positive and negative, the time axis change by accumulation of the phase rotation of the despread vector A time shift compensation means for shifting the partial sequence input to the matched filter in consideration of a time shift Δt defined by the following equation when the minimum slide interval of the matched filter is reached: To do.

According to the first aspect of the present invention, a result obtained by detecting, as despread data to be demodulated, a path having the maximum norm of the combined despread vector among a plurality of paths from the beginning to the end of the binary tree. Therefore, the phase shift of the despread vector calculated for each partial series, that is, the signal with the smallest frequency deviation is detected, and the received signal with the smallest carrier frequency deviation can be demodulated. Wireless long-distance communication becomes possible without the limitation of narrowing the bandwidth.

According to claim 2 of the present invention, as the phase of the despreading vector is rotated positive and negative by a certain angle for each partial series, the time axis change due to the accumulation of the phase rotation of the despreading vector is calculated as the despreading vector. When the minimum slide interval is reached, the time shift can be compensated by taking into account the time shift Δt and shifting the partial sequence to be input before despreading, eliminating the limitation of signal narrowing, Wireless long-distance communication is possible.

According to claim 3 of the present invention, the spread sequence of the received signal is divided to calculate a despread vector for each partial sequence, and the phase of the despread vector is rotated by a certain angle positively or negatively and branched into two branches. And a binary tree generating means for continuing the two branches until the end of the spreading sequence, and the despreading vector calculated for each partial sequence for a plurality of paths from the beginning to the end of the binary tree. A maximum value detecting unit that calculates a norm of a combined despread vector and detects a path having a maximum norm of the combined despread vector among the plurality of paths as despread data to be demodulated; As a result, it is possible to demodulate the received signal with the carrier frequency deviation minimized, thereby eliminating the limitation of narrowing the signal band and enabling wireless long-distance communication.

According to claim 4 of the present invention, as the phase of the despreading vector is rotated positive and negative by a certain angle for each partial series, the time axis change due to the accumulation of the phase rotation of the despreading vector is minimized. By providing a time shift compensation means that shifts the input signal of the matched filter by the time shift Δt when the interval is reached, it is possible to compensate for the time shift, eliminating the limitation of signal narrowing, and wireless Long distance communication is possible.

It is a system block diagram of the conventional spread spectrum communication. FIG. 6 is a diagram illustrating a relationship between a spreading sequence length L and a normalization processing gain Gp. It is a figure which shows the relationship between spreading sequence length L and processing gain Gp. It is a figure which shows the relationship between the chip | tip shift | offset | difference which arises in the despreading period between transmitter / receivers, and a processing gain. It is a figure which shows the relationship between the transmission signal and noise of a spread spectrum communication system. It is a figure which shows the relationship between the received signal and noise of a spread spectrum communication system. It is a schematic block diagram of the frequency-axis approach of a comparative example. It is a schematic block diagram of the time-axis approach in one Example of this invention. It is a conceptual diagram which shows the division | segmentation of a spread spectrum series same as the above. It is a figure which shows a synthetic | combination de-spreading vector when there is no phase rotation for every partial series. It is a figure which shows a synthetic | combination de-spreading vector in case there exists a phase rotation for every partial series. It is a figure which shows the synthetic | combination de-spreading vector of m partial series same as the above. It is a block diagram of the binary tree of the Example of this invention. It is a conceptual diagram explaining search operation | movement by binary tree production | generation by a time-axis approach same as the above. FIG. 4 is a conceptual diagram illustrating an operation for detecting, as despread data, a path having the maximum norm of a combined despread vector from a search result by binary tree generation. It is a figure which shows the calculation result of the frequency-axis approach characteristic of a comparative example. It is a figure which shows the calculation result of the time-axis approach characteristic of the Example of this invention. It is a figure which shows the simulation result of the frequency-axis approach characteristic of a comparative example. It is a figure which shows the simulation result of the time-axis approach characteristic of the Example of this invention. It is a figure which shows the S / N characteristic at the time of applying the frequency-axis approach of a comparative example and the time-axis approach of the Example of this invention in a wired experiment ((alpha) = (pi) / 2). It is a figure which shows the S / N characteristic at the time of applying the frequency-axis approach of a comparative example and the time-axis approach of the Example of this invention in a wired experiment ((alpha) = (pi) / 4). It is a figure which shows the comparison of the computational complexity of the frequency-axis approach of a comparative example, and the time-axis approach of the Example of this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Although the present invention compensates for the carrier frequency deviation of the receiving apparatus with respect to the transmitting apparatus by the time axis approach, for the sake of easy understanding, the carrier frequency deviation compensation method by the frequency axis approach which is the prior art is used. First, a comparative example will be described.

1. Operation Principle of Frequency Axis and Time Axis Approach (1) Frequency Axis Approach First, as a comparative example, a frequency axis approach for compensating for a carrier frequency deviation as seen from the frequency axis will be described.

FIG. 7 is a schematic configuration diagram of the receiving device 20 that realizes the frequency axis approach. Specifically, the receiving antenna 21 that receives a signal transmitted wirelessly, and the frequency obtained by dividing the deviation frequency from the upper limit f _H to the lower limit f _L of the carrier frequency F on the receiving side into n frequency pitch Δf intervals. In order to frequency-convert the received signal by multiplying the corresponding n carrier frequency generating means 22 and the signal generated by the n carrier frequency generating means 21 and the received signal received by the receiving antenna 21 Multiplier 23, matched filter 24, which is a despreading means for despreading each output signal of multiplier 23, and each despread signal are compared, and the largest signal with a small carrier frequency deviation is selected as received data And a demodulator 26 for demodulating the selected signal.

  A frequency difference between a transmission device (not shown) and a reception device 20 varies due to a manufacturing environment of an oscillator (not shown) incorporated in the carrier frequency generation means 2 and a usage environment such as temperature. However, it is assumed that the despreading period is sufficiently short, and here there is no frequency fluctuation within the despreading period.

In the frequency axis approach, carrier frequency deviation is compensated from the frequency axis. Now, the carrier frequency F [Hz], the frequency stability of the carrier frequency of the transmitter and A _t [ppm], the frequency stability of the carrier frequency of the receiving device A _r [ppm], the chip rate R _C [cps ], The spreading sequence length is L [chip], and the phase rotation allowed in the despreading process is α [rad], the upper limit f _H and the lower limit f _L of the carrier frequency F are expressed by the following equations.

The receiving apparatus 20 shown in FIG. 7 divides the shift frequency from the upper limit frequency f _H to the lower limit frequency f _L into frequency pitch Δf intervals, and performs n carrier frequency demodulation and despreading of the received signal at each divided frequency. The generation means 22, the multiplier 23, and the matched filter 24 are used, and the output value from each matched filter 24 is compared by the comparator 25 to detect the maximum value as received data.

  The normalization processing gain Gps of the frequency axis approach is given by substituting α for θ in Equation 3 when the phase rotates α in the despreading process.

  Next, the calculation amount of the frequency axis approach is obtained.

  The spreading sequence length L is obtained by using the frequency pitch Δf.

  Given in. The frequency pitch Δf is now

It becomes. The shift frequency of the band 2 (A _t + A _r ) F is divided into n by the frequency pitch Δf.

Therefore, when using a matched filter 24 to despreading calculation amount N _S of the frequency axis approach is given by the following equation.

  The computational complexity of the frequency axis approach increases in proportion to L to the third power.

(2) Time Axis Approach Next, as an embodiment of the present invention, a time axis approach for compensating for a carrier frequency deviation as seen from the time axis will be described.

  FIG. 8 is a schematic configuration diagram of a receiving apparatus 2 that realizes the receiving method (time-axis approach) of the present invention. Specifically, a receiving antenna 31 for receiving a signal transmitted wirelessly, a carrier frequency generating means 32 for generating a signal of a predetermined carrier frequency on the receiving side, a received signal and a carrier frequency generating means received A multiplier 33 that multiplies the output signal of 32 and converts the frequency of the received signal, a matched filter 34 that is a despreading means that despreads the output of the multiplier 33, a transmitter (not shown), and a receiver 2 A binary tree generating means 35 for searching a carrier frequency shift range, a searching means 37 having a maximum value detecting means 36, a demodulating means 38 for demodulating a received signal based on a search result of the searching means 37, It is configured with.

  The operation procedure of the time axis approach performed by the receiving apparatus 2 is as follows.

(I) Dividing the spreading sequence into partial sequences The spreading sequence with length L is divided into partial sequences with length ΔL in which the phase rotation is α [rad]. Here, α is a phase rotation that allows the degradation of the processing gain G _p . The normalization processing gain G _p given by Equation 3 hardly deteriorates when the phase rotation θ generated in the despreading process is small. For example, G _p when θ is π / 4 is about 0.974. The time axis approach takes advantage of this feature. FIG. 9 shows spreading sequence division, which is performed by spreading sequence division means (not shown) provided in the preceding stage or subsequent stage of the multiplier 33.

(II) Despreading of partial sequences The reception antenna 31 despreads the received signal with I and Q orthogonal axes for each partial sequence divided by the spreading sequence dividing means. The I-axis here means the in-phase component of the waveform by quadrature modulation, and the Q-axis means the quadrature component. The despreading result by the matched filter 34 is treated as (despreading vector).

For example, FIG. 10 shows a composition in which m despread vectors ΔL1, ΔL2,... ΔLm calculated for each partial series have no phase rotation, and the despread vectors ΔL1, ΔL2,. A despreading vector is shown. Such an ideal case, the normalization processing gain G _p is the loss no L next to the I-axis and coaxially. On the other hand, if the despread vectors ΔL1, ΔL2,... ΔLm have a phase rotation of θ by θ as shown in FIG. 11, the phase shift accumulates for m pieces as shown in FIG. The vector becomes smaller. For example, when m is 20 and θ is π / 12, the normalization processing gain G _p is 0.19L. 10 to 12 are diagrams for explaining the despreading vector, and the present invention is not applied thereto.

(III) Configuration of Binary Tree The despread vector of the partial sequence obtained by the matched filter 34 is branched into two branches whose phases are rotated by ± α by the next binary tree generating means 35. The binary tree generating means 35 continues this operation until the end of the spreading sequence to complete the binary tree.

  FIG. 13 shows the configuration of the binary tree. Since FIG. 13 shows the phase rotation operation, the phase shift accompanying the carrier frequency shift in actual operation is not included here. Therefore, for example, when the + α phase is rotated from ΔL1 (ΔL2) and then the −α phase is further rotated (ΔL3), the accumulation of phase rotation is zero. In addition, at the tip of the despreading vector of the partial series of ΔL3, the two branches intersect at a point where the phase rotation becomes zero, and at the tip of the despreading vector of the partial series of ΔL4, the phase rotation is + α, − Two branches intersect at point α. Therefore, in the search by the actual binary tree generating means 35, it is possible to calculate as one branch by omitting the overlapping calculation after the intersection.

(IV) Detection of Maximum Value of Vector Sum The maximum value detecting means 36 obtains the norm of the combined despreading vector for each path with respect to the path consisting of n partial series from the initial state to the final branch of the binary tree. Thereafter, the path having the maximum norm is detected as despread data to be demodulated and sent to the demodulator 38. FIG. 14 is a conceptual diagram for explaining the search operation by the binary tree generating means 35 based on the time axis approach. This figure shows a state in which a binary tree search is performed with a phase shift for each partial series. Here, comparing FIG. 13 with FIG. 14, in FIG. 13, six despread vectors of the partial sequence of ΔL4 are described, whereas in FIG. 14, eight are described. However, also in FIG. 14 with a phase shift, since two despread vectors out of the eight are vectors having the same magnitude and direction as the other two despread vectors, the calculation is omitted. Calculation is for 6 lines. This can be explained in the same way as omitting the redundant calculation after the intersection in FIG. 13, but in more detail, in each partial series ΔL, the despread vectors having the same rotation angle of the accumulated phase are used. Means that the calculation only needs to be performed once.

FIG. 15 is a conceptual diagram showing an operation in which the maximum value detecting means 36 detects a path in which the norm of the despreading vector obtained by synthesizing the despreading vector for each partial series is maximum. As shown in this figure, if the path with the maximum norm of the combined despreading vector is selected and its normalization gain is G _p , the phase shift of the despreading vector for each partial sequence will be minimized. Therefore, the carrier frequency shift is minimized.

The normalization processing gain G _pt of the time axis approach is obtained when the phase is alternately rotated by + α and −α for each partial series, and is given by the following equation as obtained in Equation 3.

For example, when α is π / 4, G _pt is about 0.9L.

  Note that the maximum value detection means 36 stores or reads data such as a despread vector for each partial series and a composite despread vector for each path in the memory 40 which is a storage means (not shown) for each path. The norm of the combined despread vector is obtained, or the path having the maximum norm is detected as despread data to be demodulated.

  Next, the calculation amount of the time axis approach performed by the receiving device 2 of FIG. 8 is obtained.

The length ΔL of each partial sequence is obtained by dividing the spread sequence length L into n pieces defined by Equation 9, and is expressed by the following equation. The above (1) as in the case of the frequency axis approach, the frequency stability of the carrier frequency of the transmitting apparatus and A _t [ppm], the frequency stability of the carrier frequency in the receiving apparatus A _r [ppm] To do.

The computation amount _NT of the time axis approach is obtained from the matched filter computation of each partial correlation, the configuration of the binary tree, and the detection of the maximum value of the vector norm. Here, the sum of the calculation amounts of the matched filters for each partial correlation is equal to the calculation amount of the matched filter 34 having the spreading sequence length L, and the remaining calculations are sufficiently small. Therefore, the computation amount _NT of the time axis approach can be approximated by the following equation.

The computation amount _NT of the time axis approach is the square of the spreading sequence length L, and the signal processing load on the receiving device 2 side serving as the management center is small compared to the frequency axis approach proportional to the cube of the spreading sequence length L. That is, by adopting the configuration of FIG. 8 instead of the configuration of FIG. 7, despread data (received data) that minimizes the deviation of the carrier frequency after the reception antenna 31 receives the received signal is used. The processing speed until output can be reduced from the proportionality of the spreading sequence length L to the square.

  Next, the compensation for the time lag described above will be discussed. As means for compensating for the time lag, time lag compensation means 29 and 39 are provided in the preceding stage of the matched filter 24 shown in FIG. 7 and the preceding stage of the matched filter 34 shown in FIG.

  In the transmission device 1 and the reception device 2, the carrier frequency and the data clock frequency may be generated by separate oscillators. Here, the case where the carrier frequency generation means 22 and 32 generate the same oscillator is considered.

At this time, if the difference between both data clock frequencies is Δf, the time axis of the spreading sequence between the transmission apparatus 1 and the reception apparatus 2 is such that the transmission carrier frequency is f _S and the reception carrier frequency f _R. Scales to (f _R −Δf) / f _R times. Therefore, in the receiving apparatus 2 that realizes the frequency axis approach shown in FIG. 7, the time shift compensation means 29 expands and contracts the time axis according to the frequency shift, and then despreads to compensate for the frequency shift and the time shift. It becomes possible.

  On the other hand, in the receiving apparatus 2 that implements the time axis approach shown in FIG. 8, the time shift Δt associated with the phase rotation α for each partial sequence is expressed by the following equation.

  Therefore, in the time axis approach, when the time axis change due to the accumulation of phase rotation reaches the minimum slide interval of the matched filter 34, the input to the partial sequence matched filter 34 is shifted by the time shift Δt of the above equation (11). It is possible to compensate for frequency shift and time shift.

2. Performance comparison of frequency axis and time axis approaches (1) Frequency axis and time axis approach characteristics Carrier frequency is 40kHz, chip speed is 1kc / s, carrier frequency frequency stability of transmitter 1 and receiver 2 is 10ppm each FIG. 16 shows the calculation results of the frequency axis approach characteristics in the case of three cases of α = π / 8, π / 4, and π / 2. The calculation result of the time axis approach characteristic is also shown in FIG. FIG. 18 shows the simulation result of the frequency axis approach characteristic, and FIG. 19 shows the simulation result of the time axis approach characteristic. In the frequency axis approach, the receiving apparatus 2 shown in FIG. 7 is used, and in the time axis approach, the receiving apparatus 2 shown in FIG. 8 is used.

  From FIG. 16 to FIG. 19, it can be seen that both the frequency axis approach and the time axis approach have no performance deterioration when the spreading sequence length is increased, and the carrier frequency shift and time shift can be compensated. Further, the performance deterioration amount can be adjusted by the allowable phase rotation α. Increasing α increases the allowable error and increases the deterioration amount. However, if α is decreased, the deterioration amount decreases and the normalization processing gain approaches 1.

  Since the comparative example and the embodiment of the present invention can be applied not only to wireless but also to wired, simulation results when applied to wired are shown in FIGS.

20 and 21 show the results of simulation experiments when α is π / 2 and π / 4 and the S / N is good (60 dB) and bad (-30 dB), and the frequency axis approach (comparative example). , The time axis approach (an embodiment of the present invention), and the characteristics of theoretical calculation values are shown. In either case, when the spreading sequence length L is 10 ³ or more, the result is the same as the theoretical calculation value. In the case where the S / N is bad (−30 dB), if the spreading sequence length L is small, there is a deviation from the theoretical calculation value, which is considered to be due to the noise generation state in the experiment.

(2) Comparison of calculation amount FIG. 22 shows calculation amounts of the frequency axis approach and the time axis approach shown in Equations 17 and 20, respectively.

  When the spreading sequence length L is increased, the frequency axis approach formula is proportional to the cube of L, whereas the time axis approach is L square. Further, the amount of computation of the time axis approach does not depend on the allowable phase rotation α.

  Therefore, it can be seen that in long-distance communication with a long spreading sequence length, the amount of computation required for the time axis approach is one-several digits of the frequency axis approach, and the signal processing load on the management center side can be reduced.

(3) Summary The relationship between communication distance and communication speed was derived from Shannon's theorem, and it was shown that the communication distance can be expanded by lowering the communication speed. However, when this modified Shannon's theorem is applied to spread spectrum communication, it has been found that the frequency shift and the time shift generated between the transmitter and the receiver become problems.

  Therefore, the long-distance wireless communication using spread spectrum is based on the frequency axis approach and the time axis approach. The compensation method of frequency shift and time shift was examined. It has been found that these compensation methods have almost no performance degradation even when the spreading sequence is lengthened, and the degradation amount can be adjusted.

  In addition, the amount of computation required for deviation compensation was found to be proportional to the cube of the spreading sequence length in the frequency axis approach, while the square of the spreading sequence length is sufficient in the time axis approach. Considering the significant improvement in CPU performance due to technological innovation, the time axis approach seems to be practical enough. In addition, the receiving method of the present invention is advantageous because the wireless device mounted on the sensor can be used as it is.

  As described above, in the receiving method of the present embodiment, the received signal is frequency-converted with a carrier frequency signal, and for the received signal that has been frequency-converted, the spread sequence of length L is divided into partial sequences of length ΔL, A despreading vector is calculated for each partial sequence, the phase of the despreading vector calculated for each partial sequence is rotated by a fixed angle, positive and negative, and branched into two branches. The bifurcation tree is generated by branching to the branches of the book, and the norm of the combined despreading vector obtained by synthesizing the despreading vectors calculated for each partial series is obtained for a plurality of paths from the beginning to the end of the binary tree. Each is calculated, and a path having the maximum norm of the combined despreading vector among a plurality of paths is detected as despread data to be demodulated.

Further, the carrier frequency is F [Hz], the frequency stability of the transmitting side and A _t [ppm], the frequency stability of the receiving side is A _r [ppm], the chip rate and R _C [cps] When the length of the spreading sequence is L [chip] and the phase rotation allowed in the despreading vector calculation process is α [rad], the frequency pitch Δf for dividing the spreading sequence at equal intervals is As a result, the spread sequence having the length L of the received signal is divided into n by the above equation 5, and the length ΔL of the partial sequence is divided as shown in the above equation 6.

  In this case, among the plurality of paths from the beginning to the end of the binary tree, the path having the maximum norm of the combined despreading vector is detected as the despread data to be demodulated. Since the signal with the smallest phase shift of the calculated despreading vector, that is, the frequency shift is detected, and the received signal can be demodulated with the carrier frequency shift minimized, the restriction on the narrowing of the signal is eliminated. Wireless long distance communication becomes possible.

Further, in the receiving method of the present embodiment, the carrier frequency is F [Hz], the frequency stability of the transmitting side and A _t [ppm], the frequency stability of the receiving side is A _r [ppm], chips When the speed is R _C [cps], the spreading sequence length is L [chip], and the phase rotation allowed in the despreading process is α [rad], the phase of the despreading vector for each subsequence is When the time axis change due to the accumulation of the phase rotation of the despreading vector reaches the minimum slide interval when calculating the despreading vector as the positive and negative angles are rotated by a certain angle, the time shift Δt defined by the above equation 21 is set. Considering this, the partial series to be input is shifted before despreading.

  In this case, as the phase of the despreading vector is rotated by a certain angle positively or negatively for each partial series, the time axis change due to the accumulation of the phase rotation of the despreading vector reaches the minimum slide interval when calculating the despreading vector. When the time difference Δt is taken into consideration, the partial sequence input before despreading is shifted to compensate for the time shift, eliminating the limitation of narrowing the signal band and enabling wireless long-distance communication. Become.

  Further, in the receiving apparatus 2 of the present embodiment, the carrier frequency generating means 32 that generates the carrier frequency, the matched filter 34 that is a despreading means that despreads the received signal frequency-converted by the carrier frequency, and the spread of the received signal Divide the sequence and calculate the despreading vector for each subsequence, rotate the phase of the despreading vector to a certain angle, branch to two branches, and generate a binary tree that continues to branch until the end of the spread sequence The norm of the synthesized despread vector obtained by synthesizing the despread vector calculated for each partial series is calculated for each of the means 35 and the plurality of paths from the beginning to the end of the binary tree. Based on the despread data of the maximum value detection means 36 and the maximum value detection means 36 for detecting the path having the maximum despread vector norm as the despread data to be demodulated, the received signal is recovered. And a demodulating means 38 for adjusting.

Further, the spreading sequence of length L of the received signal, the carrier frequency is F [Hz], the frequency stability of the transmitter 1 and _{A t [ppm], A r} [ppm frequency stability of the receiver 2 ], The chip speed is R _C [cps], the spreading sequence length is L [chip], and the phase rotation allowed in the despreading process is α [rad]. The frequency pitch Δf is expressed by the above equation 4, whereby the spread sequence having the length L of the received signal is divided into n by the above equation 5, and the length ΔL of the partial sequence is the length expressed by the above equation 6. A spreading sequence dividing means is provided.

  In this case, the spread sequence of the received signal is divided to calculate a despread vector for each partial sequence, the phase of the despread vector is rotated by a certain angle positively or negatively, branched into two branches, and the last of the spread sequence Binary tree generation means 35 that continues the two branches until and a combined despread vector obtained by synthesizing the despread vectors calculated for each partial series for a plurality of paths from the beginning to the end of the binary tree And a maximum value detecting means 36 for detecting, as despread data to be demodulated, a path having the maximum norm of the combined despreading vector among the plurality of paths. Since it is possible to demodulate the received signal with the frequency deviation minimized, there is no restriction on narrowing the signal band and wireless long-distance communication is possible.

The receiving apparatus 2 of this embodiment, the carrier frequency is F [Hz], the frequency stability of the transmitter 1 and A _t [ppm], the frequency stability of the receiver 2 and A _r [ppm], When the chip speed is R _C [cps], the spreading sequence length is L [chip], and the phase rotation allowed in the despreading process is α [rad], the phase of the despreading vector is set for each subsequence. When the time axis change due to the accumulation of the phase rotation of the despreading vector reaches the minimum slide interval of the matched filter 34 as the angle is rotated positively and negatively by a certain angle, the time shift Δt defined by the above equation 7 is set. In consideration of this, a time shift compensation means 39 for shifting the partial sequence input to the matched filter 34 is provided.

  In this case, as the phase of the despreading vector is rotated positive and negative by a certain angle for each partial series, the time axis change due to the accumulation of the derotating vector phase rotation reaches the minimum slide interval of the matched filter 34. By providing the time shift compensation means 39 for shifting the input signal of the matched filter 34 by the time shift Δt, it is possible to compensate for the time shift, eliminate the limitation of signal narrowing, and perform wireless long distance communication. It becomes possible.

  In addition, this invention is not limited to the said Example, It can change in the range which does not deviate from the meaning of this invention. For example, in the search by binary tree generation in the above embodiment, despreading vector calculation is performed on the branches obtained by branching all the partial sequences. That is, all branches are calculated to the last depth. However, in the middle of the generation of the binary tree, the calculation may be reduced for a branch having a clearly large phase shift to reduce the calculation amount.

2 Receiver
32 Frequency variation means
34 matched filters
35 Binary tree generation means
36 Maximum value detection means
37 Search means
38 Demodulation means
39 Time shift compensation

Claims (4)

Frequency conversion of received signal with carrier frequency signal,
Dividing the spread sequence of length L into partial sequences of length ΔL for the frequency-converted received signal;
Calculate the despreading vector for each partial series,
The phase of the despreading vector calculated for each partial series is rotated by a certain angle, positive and negative, and branched into two branches,
Continue branching to the two branches until the end of the spreading sequence to generate a binary tree;
For a plurality of paths from the beginning to the end of the binary tree, each calculates a norm of a combined despread vector obtained by combining the despread vectors calculated for each partial series,
Detecting a path having a maximum norm of the combined despreading vector among the plurality of paths as despread data to be demodulated ;
The carrier frequency is F [Hz], the frequency stability of the transmitting side and A _t [ppm], the frequency stability of the receiving side is A _r [ppm], the chip rate and R _C [cps], When the length of the spreading sequence is L [chip] and the phase rotation allowed in the despreading vector calculation process is α [rad],
A frequency pitch Δf for dividing the spread sequence at equal intervals is expressed by the following equation:

As a result, the spread sequence of the received signal length L is divided into n by the following equation:

A receiving method of a spread spectrum communication system, wherein the length ΔL of the partial sequence is equal to or less than a value represented by the following equation .

With prior Symbol the phase of the despread vector for each partial sequence to cause a predetermined angular rotation to positive and negative, the time axis change by accumulation of the phase rotation of the despread vector is the minimum slide distance upon calculation of the despread vector When you reach
Taking into account the time shift Δt which is defined by the following equation, reception method according to claim 1 spread-spectrum communication system, wherein the shifting the partial sequence to be entered before the despreading.

Carrier frequency generating means for generating a carrier frequency;
Spreading sequence dividing means for dividing a spread sequence of length L into partial sequences of length ΔL for a received signal frequency-converted by the carrier frequency ;
A matched filter that is a despreading unit that despreads the received signal and calculates a despread vector for each partial sequence divided by the spreading sequence dividing unit ;
By a predetermined angle rotates the phase sign of prior Kigyaku diffusion vector, it branched into branch 2, and a binary tree generating means to continue the branch to branch of the two until the end of the spreading sequence,
For each of a plurality of paths from the beginning to the end of the binary tree, a norm of a combined despread vector obtained by combining the despread vectors calculated for each partial series is calculated, and the combined among the plurality of paths A maximum value detecting means for detecting a path having the maximum despread vector norm as despread data to be demodulated;
A demodulating means for demodulating a received signal based on the despread data of the maximum value detecting means ,
The carrier frequency is F [Hz], the frequency stability of the transmitter and A _t [ppm], the frequency stability of the receiver and A _r [ppm], the chip rate and R _C [cps], the spreading sequence And L [chip], and the phase rotation α [rad] allowed in the despread vector calculation process,
A frequency pitch Δf for dividing the spread sequence at equal intervals is expressed by the following equation:

As a result, the spread sequence of the received signal length L is divided into n by the following equation:

A receiving apparatus for a spread spectrum communication system, characterized in that the spreading sequence dividing means is configured such that the length ΔL of the partial sequence is equal to or less than a value represented by the following equation .

With prior Symbol the phase of the despread vector for each partial sequence to cause a predetermined angular rotation to positive and negative, the time axis change by accumulation of the phase rotation of the despread vector reaches a minimum slide distance of the matched filter 4. The spread spectrum communication system according to claim 3, further comprising time shift compensation means for shifting the partial sequence input to the matched filter in consideration of a time shift Δt defined by the following equation: Receiver device.

Images