JP6906782B2 - Contention-based communication system - Google Patents

Description

The present invention relates to a contention-based communication system in which a plurality of terminals can simultaneously transmit data in the same radio resource.

The 5th generation mobile communication system (5G) is being energetically researched and developed in Japan for practical use in 2020. 5G is required to be a wireless system that not only realizes large-capacity and high-speed communication but also supports the social development of IoT (Internet of Things). 5G needs to accommodate a huge number of IoT terminals such as sensors in a base station, and is also required to support applications such as automatic driving systems where delay is almost unacceptable.

LTE (Long Term Evolution), which is the current mobile phone communication standard, does not fully consider the development of IoT, and regarding the number of terminals that can be connected at the same time and the allowable delay time when IoT becomes widespread. It is difficult to fully meet the requirements of 5G. In LTE, when a terminal (UE: User Equipment) transmits data to a base station (eNB: evolved NodeB), the eNB adopts a method of exclusively allocating resources to each UE and the number of simultaneous connections. Is limited (only one UE can be connected with one wireless resource per antenna). In addition, a negotiation time for resource allocation is required before the UE actually transmits data, and a certain delay is unavoidable (example: 10 ms or less as a target value in IMT-Advanced).

The present invention aims to increase the number of simultaneously accommodated terminals to 5 to 10 times or more the current number and to realize a delay of 5 ms or less. Toward this realization, the present application proposes a contention-based communication system in which a plurality of terminals simultaneously transmit data in the same wireless resource while being based on LTE. With this communication method, it is possible to increase the number of simultaneous connections of wireless communication terminals (hereinafter referred to as terminals) and reduce the delay.

The above communication method is realized by a frame configuration including a preamble unit and a data unit. The preamble section estimates the propagation path characteristics of the base station and each terminal, and the data section transmits the payload. Even if a collision occurs in signals transmitted from a plurality of terminals at the same time, the data unit can be separated / decoded by using the propagation path estimated value obtained from the preamble unit. Further, in the design of the preamble portion, signals from a plurality of terminals are separated so that an estimated value of the propagation path of each terminal can be obtained.

First, the upstream data transmission procedure in LTE will be described.
(1) Random access procedure The random access procedure in LTE is shown in FIG. The details of each step in FIG. 1 are shown below.

<Step 1>: Random Access Preamble
The terminal to be accessed transmits a random access preamble (RAP) to the eNB. The RAP is designed so that even if a plurality of UEs transmit at the same time, they can be separated in the eNB.
<Step 2>: Random Access Response
Upon receiving the RAP, the eNB sends a random access response to the UE. The random access response includes a timing advance command (Non-Patent Document 1) for adjusting the transmission timing of the UE, a signal instructing the radio resource transmitted by the UE in step 3, and the like.
<Step 3>: Scheduled Transmission
The UE uses the allocated radio resource to send the terminal identifier and resource allocation request.
<Step 4>: Contention Resolution
The eNB notifies the UE of the terminal identifier that permits the use of the resource, and allocates the resource. If the UE that tried the random access cannot recognize the terminal identifier of its own station, it assumes that the random access has failed and starts over from step 1.
When the terminal identifier of the own station is notified, data transmission is performed in the data transmission stage.

(2) Random access preamble (RAP)
FIG. 2 shows a configuration example of RAP in LTE. In the time axis direction, the RAP is composed of a 100 μs cyclic prefix (CP), a 100 μs guard time (GT: Guard Time), and an 800 μs preamplifier series. The frame time of RAP is 1 ms, which is the same as one subframe time of LTE. In the frequency axis direction, 864 subcarriers (of which 25 subcarriers are guard subcarriers) are arranged in the band of 1.08 MHz corresponding to the 6 resource blocks of LTE. Table 1 shows the parameters of RAP.

FIG. 3 shows the signal processing of the RAP. The preamble series is generated by embedding the Zadoff-Chu series (Non-Patent Document 3) in a subcarrier (excluding the guard subcarrier).
In the transmitting means 200, the Zadofu-chu sequence X _u (p) inserted into the subcarrier is expressed by the following equation, which is generated by the Zadofu-chu sequence generator 1.

Here, p is the index of the subcarrier, u is the series number of the Zadofu-chu series, and _Nzc (= 839) is the series length. u is defined by eNB. The Zadofu-chu series is used for the preamble series because it has the characteristic that the power is constant in the frequency domain and the time domain and has an ideal cyclic autocorrelation.
As is well known, the Zadofu-chu series described above is defined as one of the constant amplitude zero autocorrelation (CAZAC) series. The CAZAC series has a constant amplitude on the time axis, and the autocorrelation function is a delta function. The number 1 is _{a definition when N zc} is an odd number, and when the N zc is an even number, p (p + 1) of the number 1 is replaced ^{by p 2.} In the present invention, since _Nzc is mainly set as a prime number, the number 1 is used in the following description.

Next, the generated Zadofu-chu series is given a phase rotation by the phase shifter 3 in order to change the peak position of the correlation value in the time axis signal. By shifting the peak position, the number of UEs that can be detected simultaneously in the eNB can be increased. X _{u and v} (p) obtained by giving phase rotation to X u (p) are shown in the following equation.

Here, C _v = vN _CS indicates the shift amount of the position where the peak appears, and N _CS indicates the interval (excluding the propagation delay difference) between the peak positions when two or more UEs transmit. Next, since X _{u and v} are transmitted as time-axis signals, IDFT (Inverse Discrete Fourier Transform) is applied by the inverse discrete Fourier transform 4 to obtain the following equation.

Here, n is the index of the time signal. After CP is added to these x _{u and v} by the cyclic prefix adder 5, the signal is transmitted from the antenna.
The eNB receiving means 210 receives a signal in which noise is added to _{x u and v to which CP is added by passing through a propagation path. The signals ru and v} in which the CP is deleted from the received signal by the cyclic prefix remover 7 are represented as follows.

Here, x _{u and v} are cyclic signals, h _v is an impulse response (IR), and w is noise. Then _{r u, v} is converted signals _{R u} in the frequency _domain, the _v in the discrete Fourier transformer 8 by a DFT (Discrete Fourier Transform), the following manner.

Here, H and W represent discrete Fourier transform signals of h and w, respectively. Next, Ru _{and v are obtained by multiplying the complex conjugate conj (X u} ) of the Zadofu-chu series Xu from the Zadofu-chu series generator 9 by the multiplier 10 to obtain the following equation.

Next, when H'v is _subjected to the inverse discrete Fourier transform again with the inverse discrete Fourier transformer 11, the following equation is obtained. w'is the inverse discrete Fourier transform of W'.

Here, is a matter relating to the present invention, the eNB as shown in equation (7), because the IR that C _v for the time shift of a terminal is obtained by the detection, the UE requesting a random access I understand the existence. Further, since the UE can change the position where the IR appears by v, the eNB can grasp the random access request of each UE unless a plurality of UEs use the same v. The above is the process of step 1, and after that, the process proceeds to step 2 in which the eNB transmits a random access response.

Takeshi Hattori, Masanobu Fujioka, Tomoo Morohashi (translated), All 4G LTE / LTE-Advanced, pp.393-pp405, Maruzen Publishing, Tokyo, 2015. 3GPP TS 36.300: LTE; Evolved Universal Terrestrial Radio Access (E-UTRA) and Evolved Universal Terrestrial Radio Access Network (E-UTRAN); Overall description; Stage 2. D.C.Chu, Polyphase, Polyphase codes with good periodic correlation properties, IEEE T. Inform. Theory 18 (4), pp531-532, Jul. 1972. Axnas, Johan, et al. "Successive interference cancellation techniques for LTE downlink." 2011 IEEE 22nd International Symposium on Personal, Indoor and Mobile Radio Communications. Pp.1793-1797. 2011. Divsalar, Dariush, Marvin K. Simon, and Dan Raphaeli. "Improved parallel interference cancellation for CDMA." IEEE Transactions on Communications 46.2, pp.258-268, 1998. Takizawa, Moriyama, Odo, Tezuka, Murakami, Ishizu, Kojima, "High-efficiency communication method for mobile phone networks that accommodate a large number of devices-Basic study on interference suppression / removal technology-," Shingaku Giho, RCS-2016, Kanagawa ， Oct.2016. Z. Nadir & M. Idrees Ahmad, “Path loss Determination Using Okumura-Hata Model and Cubic Regression for Missing Data for Oman”, Proceeding of IMECS, Vol. 2, 2010. S. Sarooshyari & N. Madaya, “Introduction to mobile radio propagation and characterization of frequency bands” wireless comm. Technologies, IEEE, 16: 332: 559, 1996. A. Neskovic, N. Neskovic & G. Paunovic, “Modern approaches in modeling of mobile radio systems propagation environment”, IEEE Communications Surveys, 2000. IEEE 802.16.3c-01 / 29r4, Channel Models for Fixed Wireless Applications V. Erceg, et al. "An empirically based path loss model for wireless channels in suburban environments." IEEE Journal on selected areas in communications Vol.17, No7, pp.1205-1211, 1999. Rec.ITU-R BT.1368-11, “Planning criteria, including protection ratios, for digital terrestrial television services in the VHF / UHF bands,” 2014.

Even if a collision occurs in signals transmitted from a plurality of terminals at the same time, the data unit can be separated / decoded by using the propagation path estimated value obtained from the preamble unit.

The contention type communication system of the present invention is a contention type communication system that transmits a packet signal having a frame configuration including a preamble unit and a data unit from a plurality of wireless terminals and receives the packet signal at a common base station.
The preamble section includes a signal for estimating propagation path characteristics, and the predetermined series signal is phase-shifted by the amount assigned to each wireless terminal, and the data section provides terminal identification information and a payload. Including
Each of the plurality of wireless terminals transmits after adding a cyclic prefix and a guard time to the preamble unit and the data unit.
The base station that has received the packet signal removes the cyclic prefix of the preamble portion, performs Fourier conversion to convert the time axis signal into a frequency axis signal, and is a signal corresponding to the complex conjugate of the predetermined series signal. Is multiplied by, and inverse Fourier conversion is performed to convert the frequency axis signal into a time axis signal, and each of the impulse response signals, which is the time axis signal obtained by this conversion, is distributed to the slots of the wireless terminals by a time filter and distributed. Each transmission line characteristic is estimated from each of the impulse response signals, and the data unit is separated and decoded using the estimated transmission line characteristic.

Further, the contention type communication system of the present invention is a contention type communication system in which a packet signal having a frame configuration including a preamble unit and a data unit is transmitted from a plurality of wireless terminals and received by a common base station. ,
The above wireless terminal
A first signal generator that generates a predetermined series signal,
A cyclic expansion unit that cyclically expands the output of the first signal generation unit, and a cyclic expansion unit.
The phase shift unit that shifts the output of the cyclic extension unit by the amount assigned to each wireless terminal, and the phase shift unit.
The first inverse Fourier transform unit for converting the output of the phase shift unit into a time axis signal, and
A cyclic prefix section for providing a cyclic prefix region in the output of the inverse Fourier transform section, and a cyclic prefix section.
A transmission means for transmitting the output of the cyclic prefix portion is provided.
The above base station
A receiving means for receiving a radio signal transmitted from the above transmitting means, and a receiving means.
A cyclic prefix remover for removing the cyclic prefix area from the received signal,
A first Fourier transform unit for converting the output signal of the cyclic prefix removal unit into a frequency axis signal, and
A second signal generator that generates a signal corresponding to the complex conjugate of the predetermined sequence signal, and
A multiplication unit that multiplies the output of the Fourier transform unit and the output of the second signal generation unit,
A frequency filter unit that multiplies the output of the multiplication unit by a predetermined window function that suppresses side lobes.
A second inverse Fourier transform unit that converts the output of the frequency filter unit into a time axis signal,
A time filter that classifies the signal from the second inverse Fourier transform unit into impulse response signals with a plurality of delay times, and
A second Fourier transformer that converts each of the above classified impulse response signals into each of the frequency axis signals, and
Each of the time filter units that multiply the correction window function that corrects each of the outputs of the second Fourier transform, and
Each of the third inverse Fourier transform units that convert each output of the time filter unit into a time axis signal, and
A decoding unit that estimates each transmission line characteristic from each of the impulse response signals converted to the time axis and uses each of the estimated transmission line characteristics to separate and decode each of the data units.
It is characterized by having.

The window function is characterized by being a Kaiser window.

The correction window function is characterized in that each of the outputs of the second Fourier transform is multiplied by a window function having a characteristic corresponding to the reciprocal of the window function.

The decoding unit utilizes sequential interference elimination or parallel interference elimination.

The sequence in the predetermined sequence signal is a Zadofuchu sequence, and the length of the sequence is a prime number.

According to the present invention, it is possible to realize not only an increase in the number of simultaneous connections of terminals but also a reduction in delay at the time of connection.

It is a figure which shows the random access procedure in LTE. It is a figure which shows the structure of the random access preamble (RAP) in LTE. It is a block diagram which shows the signal processing of a random access preamble (RAP) in LTE. It is a figure which shows the data transmission from the terminal of this invention to a base station. It is a figure which shows the frame structure example used in this invention. It is a block diagram which shows the signal processing of the RAP of this invention. It is a figure which shows the separation of the impulse response in this invention. It is a figure which shows the preamble success probability by a computer simulation. It is a figure which shows the relationship between the distance to a base station by a computer simulation, and SNR ((signal-noise ratio). It is a figure which shows the simulation result of PER (Packet Error Rate) with respect to SNR ((signal-noise ratio) using the parameters of Table 2. It is a figure which shows the simulation result of PER (Packet Error Rate) with respect to SNR ((signal-noise ratio) using the parameters of Table 3. A frequency filter 16 (Kaiser filter) is used as a window function in front of the inverse discrete Fourier transformer 11 to suppress the amplitude of the side lobe, and an inverse frequency filter 17 (inverse Kaiser filter) is used after the time filter 15 to filter the frequency. It is a block diagram which shows the structure which compensates the distortion generated in the impulse response caused by 16. It is a block diagram which shows the configuration example for generating the frame which consists of a preamble and data.

Hereinafter, the present invention will be described in detail with reference to a block diagram. Regarding the reference numerals in the block diagram, the same reference numerals shall be used for blocks having the same function unless there are special circumstances.

The random access procedure in LTE consists of four steps as shown in FIG. 1, and it takes time to allocate resources. On the other hand, in the present invention, for example, the preamble of FIG. 4 (b) and the data of FIG. 4 (c) are transmitted at once without resource allocation, that is, at the stage of step 1. Propose a method.

The communication procedure of the present invention is roughly as follows.
First, the preamble unit and the data unit are transmitted in the same subframe. Therefore, no time is required for resource allocation. The preamble unit is used to estimate the IR between each terminal and the base station. In addition, the estimated IR is used to separate and demodulate the data unit. Decoding of the data unit uses sequential interference cancellation (SIC: Successive Interference Cancellation) (Non-Patent Document 4) or parallel interference cancellation (PIC: Parallel Interference Cancellation) (Non-Patent Document 5). Details of the signal processing of the data unit are described in Document (Non-Patent Document 6).
By adopting such a subframe configuration and communication procedure, it is possible to eliminate the negotiation time and increase the number of simultaneously connected terminals by separating and decoding the colliding signals.

FIG. 5 shows a frame configuration of the data transmission method of this embodiment. As shown in FIG. 13, this frame can be generated by alternately switching between the preamble signal generated by the preamble unit and the data signal generated by the data unit at predetermined time intervals by the signal switching means 16. .. In the signal switching means 26, preamble generation is performed with a configuration similar to the configuration of FIG. Further, in the processing in the data unit, the scrambler 21 disperses the digital signal, the cyclic redundant error detection encoder 22 adds an error detection code, the encoder 23 for example turbo encodes, and the modulator for example QPSK. Modulate and insert, for example, R, CP and GT in Table 2 with an R, CP and GT inserter to generate an oversampled signal.

The frame of this embodiment is composed of two preamble parts and two data parts, for a total of four symbols. As for the parameters, two types were examined as shown in Tables 2 and 3, respectively. Considering the affinity with LTE, the subframe configuration uses the LTE RAP mechanism shown in FIG. In this example, the subcarrier spacing was widened from 1.35 kHz of LTE RAP to 5 kHz or 4.8 kHz to shorten one symbol length.
When a plurality of preambles are used in one frame, for example, the preambles under the conditions of Tables 2 and 3 can be used in the frame. In this case, it is desirable that the length of the Zadofu-chu series is a prime number as close as possible.

The preamble unit is used to extract the IR of each terminal. In addition, this IR is used to separate and demodulate the signal of the data unit where the collision has occurred. The block diagram of FIG. 6 shows the signal flow of transmission / reception of the preamble portion of this embodiment.
For the preamble, first, _{a Zadofu-chu sequence having a sequence length N'ZC} ( _N'ZC is a prime number) is generated by the Zadofu-chu sequence generator 1. Next, the Zadofu - carried out in Chu sequence a cyclic extension to match the number of subcarriers N _ZC present (Cyclic Extended) cyclic dilator 2. The cyclic extension is expressed by the following equation.

The processing after the cyclic expander 2 is the same as the procedure from the number (2) to the number (7) described above. That is, the Zadofu-chu series gives phase rotation with the phase shifter 3 in order to change the peak position of the correlation value in the time axis signal. Next, in order to transmit as a time axis signal, IDFT (Inverse Discrete Fourier Transform) is performed by the Inverse Discrete Fourier Transformer 4, which is the first inverse Fourier transform unit. Then, after adding a CP by the cyclic prefix adder 5, the signal is transmitted to the communication channel 6.

The transmitted signal has CP removed from the received signal by the cyclic prefix remover 7, and then the signal in the frequency domain is subjected to DFT (discrete Fourier transform) by the discrete Fourier transform 8 which is the first Fourier transform unit. Is converted to. _{Here, the complex conjugate conj (X u} ) of the Zadofu-chu series Xu from the Zadofu-chu series generator 9 is multiplied by the multiplier 10 and inverse-discrete by the inverse discrete Fourier transformer 11 which is the second inverse Fourier transform unit. Fourier transform.

Thus, after the inverse discrete Fourier transform by the inverse discrete Fourier transformer 11 of the receiving means 110 of the base station, the output signal h _'v is a time filter 12 as shown in FIG. 6 and 7 M = floor ( It _{is divided into NZC} / _NCS ) subslots, and the impulse response (IR) in each subslot is extracted. _(N ZC _{/ N CS} when the table 3 is not an integer, indicating the number of points (= 9) assigned point by point in the second half of the nine sub-slot. Figure 1 points the time the 7 α = 0.926μs remainder) .. If the IRs reproduced by the IDFT are within the time filter delimiters, the propagation path can be estimated with high accuracy, and a maximum of M IRs can be obtained with one preamble. Of the two preamble units, when each terminal uses only one of the preambles, 2M IRs can be obtained.
In each subslot, the estimated IR is used to separate and demodulate the data unit. Decoding of the data unit uses sequential interference cancellation (SIC: Successive Interference Cancellation) (Non-Patent Document 4) or parallel interference cancellation (PIC: Parallel Interference Cancellation) (Non-Patent Document 5). Details of the signal processing of the data unit are described in Document (Non-Patent Document 6).

The delay due to the cell radius covered by the base station is dealt with as follows. Since it is also contemplated that transmitting the frame without receiving timing advance command from the base station (non-patent document 1) In the present invention, CP to absorb transmission timing errors is provided with GT and N _CS. CP from Table 2 and Table 3, since the GT and _N shortest time in _CS are respectively 24.07μs and 16.67Myuesu, terminal - the distance between the base station until 3.6km, also for delays 2. It can absorb the delay caused by the round trip of radio waves between points of 5 km.
Further, in the actual propagation environment, it is necessary to consider the delay due to the multipath of about 5 μs in addition to the transmission timing error. Therefore, the distance between the terminal and the base station where the IR actually falls within the time filter delimiter is expected to be about 3 km according to the parameters in Table 2 and about 2 km in Table 3.

In addition, the collision of the data unit can be dealt with as follows. The data unit is used to transmit control information (including terminal identification information, etc.) and payload required for communication. In this embodiment, QPSK and a turbo code having a coding rate of 1/3 are assumed as the modulation method. However, in Table 2, 12 parity bits are deleted (punctured), and the coding rate is 292/864 (= 0.338 ≈ 1/3). The amount of data that can be transmitted at one time is 36 bytes (Table 2) and 37 bytes (Table 3). For the separation / demodulation of the collision signals in the data unit, SIC (sequential interference removal) or PIC (parallel interference removal) is applied using the IR estimated in the preamble unit described above (Non-Patent Document 6).

FIG. 12 shows a configuration example in which distortion in the impulse response is suppressed as compared with the configuration of FIG. This suppresses the amplitude of the side lobe by using a frequency filter 16 (Kaiser filter) as a window function in front of the inverse discrete Fourier transformer 11 which is the second inverse Fourier transform unit in the receiving means 120, and suppresses the amplitude of the side lobe of the time filter 15. Later, the window function 27 and the inverse frequency filter 17 (inverse Kaiser filter) are used to correct the distortion generated in the impulse response caused by the frequency filter 16. The inverse frequency filter 17 is converted into a frequency axis by the discrete Fourier transform 18 which is the second Fourier transform unit, the frequency band is selected by the frequency filter 19 having the inverse coefficient of the frequency filter 16, and the third inverse is used. It is converted into a time axis signal by the inverse discrete Fourier transformer 20 which is a Fourier transform unit.

The probability of successful transmission of the preamble by simulation will be described.
The base station obtains IRs of a plurality of terminals by using the preamble unit, but when two or more terminals select the same v in the equation (2), the plurality of IRs are shown in the same section of the time filter in FIG. Therefore, accurate IR cannot be obtained at the base station. Therefore, under the condition that v is randomly selected, the collision probability at which IRs of a plurality of terminals appear in the same time filter section was obtained by simulation. Since there are two preamble units in one frame, in the simulation, each terminal randomly assigns v1 and v2 to each of the two preambles and transmits them. In this paper, the occurrence of a collision is defined as the case where both v1 and v2 match v1 and v2 of other terminals, respectively.

^{The number of terminals is from the target of 1 million terminals per 1 km 2} in the 5th generation mobile communication system (5G) to a maximum of 10 million terminals, each terminal transmits at an average interval of 600 seconds, and the transmission timing follows the Poisson distribution. did. In addition, the base station is set to stand by on one channel with a band of 1.08 MHz (6 resource blocks), and the terminal is set to have a transmission opportunity once every 1 ms. FIG. 8 shows the probability that the preambles did not collide and the preambles were successfully transmitted. From the figure, it can be seen that when the number of terminals is 1 million, the collision probability of the preamble can be achieved at 5% or less. In the comparison between Tables 2 and 3, since M = 8 in Table 2 and M = 12 in Table 3, the parameters in Table 2 can cover the cell radius, but the collision probability can be reduced.

In this simulation, v was randomly selected, but it is worth considering an effective selection method for v. For example, it is assumed that the group A is divided into two groups, a group A consisting of terminals near the center of the cell and terminals at the end of the cell, and a group B consisting of terminals existing in the middle of the cell, and the v that can be selected by each group is different. Group A is a group in which a group of terminals with weak signals reaching the base station and a group of strong terminals coexist, and when a weak signal collides with a strong signal, a method is devised so that at least the IR of a strong signal can be obtained. Can be considered.

The link budget and packet error rate were also estimated as shown below.
First, for the link budget, we estimated how strong the random access preamble emitted from the terminal would reach the base station. The carrier frequency was set to 2.5 GHz as an example in this study. In this paper, the extended Hata model (Non-Patent Documents 7, 8 and 9) and the Erceg model (Non-Patent Documents 10 and 11) examined in 802.16 are used as the path loss model. Sub Urban was adopted for the extended Hata model, and Category C, which is a flat land, was adopted for the Erceg model. In addition, each short interval variation is not taken into consideration. Other parameters are shown in Table 4. FIG. 9 shows the relationship between the distance to the base station and the SNR.

Regarding the packet error probability, the packet error rate was obtained using computer simulation when the obtained IR and data signals did not collide (however, the characteristics when there is a collision in the data signal are described here. No. See Non-Patent Document 4). The error correction code uses a turbo code having a coding rate of 292/864 and 300/900, and the interleave uses the LTE standard to perform repeated decoding eight times. The propagation path model is a Typical Urban (6-pass model) (Non-Patent Document 12), a SUI3 model, and a SUI4 model (Non-Patent Document 10). The antenna is omnidirectional and the terminal is assumed to be in a quasi-stationary state. Equalization is performed in the frequency domain, and only phase compensation is performed. The simulation results are shown in FIGS. 10 (parameters in Table 2) and FIG. 11 (parameters in Table 3). In the comparison between FIGS. 10 and 11, the performance of FIG. 11 is slightly better than that of FIG. 10 because the coding rate is lower and the number of bits is larger. The characteristics of SUI3 in both FIGS. 10 and 11 are inferior to those of the other channel models because the fluctuation of fading in the frequency axis direction is small. Since the SUI 3 is a path model in which a delay wave having a large delay time does not exist, it is considered that a large frequency diversity effect cannot be obtained in a band of 1.08 MHz. In this example, in the case of Typical Urban and SUI4, an SNR of 15 dB is required to realize a packet error rate of 0.1%.

In order to further improve the characteristics, it is considered necessary to apply a turbo equalizer that can update the propagation path estimation value.

1 Zadofu-chu series generator 2 Cyclic expander 3 Phase shifter 4 Inverse discrete Fourier transform 5 Cyclic prefix adder 6 Communication channel 7 Cyclic prefix remover 8 Discrete Fourier transform 9 Zadofu-chu series generator 10 multiplier 11 inverse discrete Fourier transformer 12-time filter _{13 1, 13 2, ···,} 13 M-1 subslot 14 detector 15 time filter 16 frequency filter 17 inverse frequency filter 18 discrete Fourier transformer 19 frequency filter 20 opposite Discrete Fourier Transform 21 Scrambler 22 Patrol Redundant Error Detection Encoder 23 Encoder 24 Modulator 25 R, CP and GT Insert 26 Signal Switching Means 27 Window Function 100 Transmitting Means 110, 120 Receiving Means 200 Transmitting Means 210 Receiving means

Claims (6)

It is a contention type communication system that transmits a packet signal having a frame configuration including a preamble unit and a data unit from a plurality of wireless terminals and receives it at a common base station.
The preamble section includes a signal for estimating propagation path characteristics, and the predetermined series signal is phase-shifted by the amount assigned to each wireless terminal, and the data section provides terminal identification information and a payload. Including
Each of the plurality of wireless terminals transmits after adding a cyclic prefix and a guard time to the preamble unit and the data unit.
The base station that has received the packet signal removes the cyclic prefix of the preamble portion, performs Fourier conversion to convert the time axis signal into a frequency axis signal, and is a signal corresponding to the complex conjugate of the predetermined series signal. Is multiplied by, and inverse Fourier conversion is performed to convert the frequency axis signal into a time axis signal, and each of the impulse response signals, which is the time axis signal obtained by this conversion, is distributed to the slots of the wireless terminals by a time filter and distributed. Each transmission line characteristic is estimated from each of the above impulse response signals, and the data unit is separated and decoded using the estimated transmission line characteristic.
A contention type communication system characterized by this. It is a contention type communication system that transmits a packet signal having a frame configuration including a preamble unit and a data unit from a plurality of wireless terminals and receives it at a common base station.
The above wireless terminal
A first signal generator that generates a predetermined series signal,
A cyclic expansion unit that cyclically expands the output of the first signal generation unit, and a cyclic expansion unit.
The phase shift unit that shifts the output of the cyclic extension unit by the amount assigned to each wireless terminal, and the phase shift unit.
The first inverse Fourier transform unit for converting the output of the phase shift unit into a time axis signal, and
A cyclic prefix section for providing a cyclic prefix region in the output of the inverse Fourier transform section, and a cyclic prefix section.
A transmission means for transmitting the output of the cyclic prefix portion is provided.
The above base station
A receiving means for receiving a radio signal transmitted from the above transmitting means, and a receiving means.
A cyclic prefix remover for removing the cyclic prefix area from the received signal,
A first Fourier transform unit for converting the output signal of the cyclic prefix removal unit into a frequency axis signal, and
A second signal generator that generates a signal corresponding to the complex conjugate of the predetermined sequence signal, and
A multiplication unit that multiplies the output of the Fourier transform unit and the output of the second signal generation unit,
A frequency filter unit that multiplies the output of the multiplication unit by a predetermined window function that suppresses side lobes.
A second inverse Fourier transform unit that converts the output of the frequency filter unit into a time axis signal,
A time filter that classifies the signal from the second inverse Fourier transform unit into impulse response signals with a plurality of delay times, and
A second Fourier transformer that converts each of the above classified impulse response signals into each of the frequency axis signals, and
Each of the time filter units that multiply the correction window function that corrects each of the outputs of the second Fourier transform, and
Each of the third inverse Fourier transform units that convert each output of the time filter unit into a time axis signal, and
A decoding unit that estimates each transmission line characteristic from each of the impulse response signals converted to the time axis and uses each of the estimated transmission line characteristics to separate and decode each of the data units.
A contention type communication system characterized by being equipped with. The contention type communication system according to claim 2, wherein the window function is a Kaiser window. The correction window function according to claim 2 or 3, wherein each of the outputs of the second Fourier transform is multiplied by a window function having a characteristic corresponding to the reciprocal of the window function. Contention type communication system. The contention-type communication system according to any one of claims 2 to 4, wherein the decoding unit uses sequential interference elimination or parallel interference elimination. The above sequence is in a predetermined sequence signal, Zadofu - a Chu sequence, the communication system of the contention system according to claim 1, any one of 5, wherein the length of the sequence is a prime number.

Images