WO2022247082A1

WO2022247082A1 - Transmission method in which pseudo-random sequence and sparse concatenated code are superimposed in bit field

Info

Publication number: WO2022247082A1
Application number: PCT/CN2021/121021
Authority: WO
Inventors: 陈为刚; 何亚龙; 韩昌彩
Original assignee: 天津大学
Priority date: 2021-05-28
Filing date: 2021-09-27
Publication date: 2022-12-01
Also published as: CN113300993A; CN113300993B

Abstract

Disclosed in the present invention is a transmission method in which a pseudo-random sequence and a sparse concatenated code are superimposed in a bit field. The method comprises: (1) at a sending end, firstly, performing channel coding on an information sequence to be transmitted, according to a sparse coding scheme, converting each channel coded symbol into a sparse sequence having a small proportion of code elements "1", then, performing bit-by-bit exclusive OR superimposition on the sparse sequence and a pseudo-random sequence, so as to obtain a sequence to be transmitted, and finally, modulating and sending same; and (2) at a receiving end, firstly, performing a sliding correlation decision on a received signal and a local pseudo-random sequence, so as to realize frame synchronization and transmission symbol synchronization, further removing the pseudo-random sequence from the signal, then, calculating, in a demodulator, hard decision information and soft decision information of each bit, sending same to a decoder for decoding and calculating hard output information and soft output information of each sparse coded codeword, according to a mapping rule the same as that at the sending end, obtaining hard decision information and soft decision information of the channel coded symbol corresponding to each coded codeword, and obtaining an original information sequence by means of channel decoding.

Description

Transmission method of superimposed pseudo-random sequence and sparse concatenated coding in bit field

technical field

The invention relates to the field of digital communication, in particular to a transmission method of superimposing pseudo-random sequences and sparse concatenated coding in bit domains.

Background technique

When data information is transmitted in the communication system, a frame is generally composed of several symbols as the basic unit of data transmission. The position of each digital time slot can be identified according to the frame positioning signal. Therefore, in the digital communication network, frame Synchronization is an indispensable prerequisite for subsequent data processing. However, with the development of modern error correction coding and decoding technology, the frame synchronization locking and tracking threshold of the communication system is facing higher requirements. The conventional frame synchronization method has a high probability of false locking and cannot achieve stable tracking synchronization. The frame synchronization detection performance can be effectively improved by increasing the length of the frame synchronization word, but this will bring additional spectrum overhead and destroy the original frame structure of the system. Especially when the receiver works under low signal-to-noise ratio, such as burst communication, the traditional synchronous detection method is difficult to implement, and the accuracy will be greatly limited.

Short burst signal duration is short, difficult to capture and detect, and has strong anti-interference and anti-capture capabilities. This excellent characteristic makes short burst signals widely used in communication systems that require strong concealment, such as: military Communication and the emergency life-saving communication that appeared in recent years. Under low signal-to-noise ratio, short-burst communication generally requires less information to be transmitted, and uses an extremely low bit rate coding scheme and configures an extremely low information transmission rate to achieve stronger anti-interference capabilities. The combination of extremely low rate and extremely low bit rate can achieve reliable transmission under extremely low signal-to-noise ratio, but at the same time, it also makes the duration of the data segment longer. If the frame header is also longer, the duration of the entire burst is longer. It greatly reduces its concealment, so it is difficult to achieve high concealment and high reliability of burst communication at the same time. Therefore, it is very important to study the signal acquisition and synchronization method of burst communication and improve the synchronization detection performance of the receiver in the low SNR environment. Necessary for reliable communication at low signal-to-noise ratios. In order to achieve frame synchronization, there are usually two types of methods, one is to insert some special code groups in the transmission information stream as the head and tail marks of each frame, and the receiving end can realize frame synchronization according to the positions of these special code groups; Self-synchronization is realized by using the different characteristics of the code groups themselves, without the need for external special code groups.

The method of inserting a special code group to achieve frame synchronization mainly includes the coherent insertion method and the interval insertion method. The coherent insertion method is to insert the frame synchronization code group at the beginning of each frame. In this frame synchronization method, it is The transmitted data is compiled into frames, each frame contains one or more code groups, and a special character is added to the head of the frame to indicate the beginning of a frame. The receiving end searches the received bit stream, and once this special character is detected, the starting position of the frame is obtained, and the code groups in the frame are divided accordingly. However, in some cases, the frame synchronization code group is not inserted in the information code stream, but a frame synchronization code is inserted every certain number of information symbols, and the receiving end can search and detect the received symbols. Determine the position of the frame synchronization code. The search and detection of the frame synchronization position can generally use the external synchronization method or the self-synchronization method. Specifically, after the receiving end receives the data from the sending end, it performs a sliding correlation between the local sequence and the receiving sequence. If it is at the boundary position of the synchronous capture frame , the correlation function of the two sequences will have a sharp peak, and the synchronization position information can be obtained by using the correlation peak. The external synchronization method often inserts a pilot sequence with good autocorrelation characteristics before each frame, that is, frame synchronization codes, such as Barker codes, m sequences, and so on. Then, sliding detection is performed on the received signal at the receiving end, and the position of the frame header is calculated according to the correlation characteristics of the frame synchronization code, so as to realize frame synchronization. Since the magnitude of the received signal may be smaller than the noise magnitude under low SNR, the frame synchronization code is generally detected by correlation accumulation and threshold judgment. The autocorrelation of the frame synchronization code is characterized by a large difference between the main lobe and the side lobe. The sharper the peak of the main lobe, the higher the accuracy of the decision. This difference can be improved by increasing the length of the frame synchronization code. When the frame synchronization The longer the code, that is, the longer the pilot sequence, the higher the effective gain of correlation accumulation will be.

In addition, autocorrelation or differential coherent detectors are often used in OFDM systems, and detection methods based on the correlation between the inserted pilot sequence and the received signal are also widely used, and the performance of this method is better than that of non- Coherent detection and autocorrelation detectors, but cross-correlation detectors consume a lot of resources. The research results show that under the same signal-to-noise ratio, the optimal criterion algorithm has nearly 3dB gain than the simple correlation criterion algorithm. Some researchers proposed a parallel search scheme based on matched filters. This scheme uses matched filters to calculate the correlation value of the synchronization code at each sampling time. If the correlation value at a certain time is greater than the decision threshold, it can be judged as was successfully captured, and the offset phase at which this correlation peak is located is the estimated value of frame sync detection. The traditional window search peak detection algorithm also uses a threshold judgment method. If the correlation peak value exceeds the threshold, it is judged that the frame header capture is successful. However, the performance of this detection method is very sensitive to the threshold setting. In order to avoid the dependence of the peak detector on the preset threshold, Someone proposed a peak detector without setting a threshold, limiting the peak search to a finite window as a segment, and then searching for the maximum position in the segment as an alternative, and then judging the maximum position in the next window Whether it matches this position, if it matches, the frame header search is successful and frame synchronization is completed, but this algorithm has a large amount of calculation and needs to rely on pilots, which greatly increases the complexity. In order to further reduce the complexity of the algorithm, based on the idea of continuous search algorithm, someone proposed the sliding correlation detection window method, which extracts a sample point from multiple sample points, and performs correlation analysis between the extracted sample point data and the local codeword , which proves the feasibility of this method to achieve signal capture, and can save search time to a certain extent.

In order to reduce the chip resources required for hardware implementation, a simplified method of using the local sequence and the hard decision results of the received signal to perform cross-correlation operations for detection is proposed in the prior art, but it is not as good as the correlation detection method based on soft decision information. performance, and in a low signal-to-noise ratio environment, when the signal is severely damaged by noise, this type of data-assisted detector needs longer frame header data to maintain the accuracy of signal capture and synchronization detection, resulting in a decrease in data transmission efficiency The decrease reduces the concealment and anti-interception of bursts, and also brings greater bandwidth requirements and reduces spectrum utilization.

Contents of the invention

The present invention provides a transmission method and device for superimposing pseudo-random sequences and sparse concatenated coding in the bit field. The receiver of the present invention can also realize accurate signal capture and synchronization under low signal-to-noise ratio, so as to complete effective signal transmission. Receive; the sending end performs sparse coding processing on the information sequence to be transmitted, so that after the pseudo-random sequence is superimposed on it, the frame synchronization accuracy under a certain signal-to-noise ratio can still be guaranteed, and compared with the traditional coding transmission method, it can Reliable transmission of information is achieved under low signal-to-noise ratio, see the following description for details:

A transmission method for superimposing pseudo-random sequences and sparse concatenated coding in bit fields, said method comprising the following steps:

(1) At the sending end, channel coding is first performed on the information sequence to be transmitted, and each channel coded symbol is converted into a sparse sequence with a small proportion of symbol "1" according to the sparse coding scheme, and then the sparse sequence and the pseudo-random sequence are one by one The sequence to be transmitted is obtained by bit XOR superposition, and finally modulated and sent;

(2) At the receiving end, firstly, the received signal and the local pseudo-random sequence are subjected to sliding correlation judgment to realize frame synchronization and transmission symbol synchronization, further remove the pseudo-random sequence in the signal, and then calculate each bit in the demodulator The hard decision and soft decision information are sent to the decoder for decoding and the hard output and soft output information of each sparsely coded codeword are calculated, and finally the channel coding corresponding to each codeword is obtained according to the same mapping rules as the sending end Symbol hard-decision information and soft-decision information are channel-decoded to obtain the original information sequence.

The beneficial effects of the technical solution provided by the invention are:

1. The concatenated coding transmission method of superimposing pseudo-random sequences proposed by the present invention can realize two functions of frame synchronization and information transmission simultaneously with one code stream and carrier;

2. The present invention performs sparse coding processing on the information sequence to be transmitted at the sending end, so that after the pseudo-random sequence is superimposed on it, the accuracy of frame synchronization under a certain signal-to-noise ratio can still be guaranteed;

3. Compared with the traditional coded transmission method, the present invention can realize reliable transmission of information at a lower signal-to-noise ratio.

Description of drawings

Fig. 1 is the system block diagram that the scheme of the present invention realizes;

Fig. 2 is a schematic diagram of the structure of the sequence to be transmitted after superimposing the pseudo-random sequence in the present invention;

Fig. 3 is the implementation flowchart of the sending end based on the low-heavy sequence mapping of the present invention;

Fig. 4 is the schematic representation of the waveform of the sequence set V ₄ corresponding to 2 coded bits w fixed to 1 in the present invention;

Fig. 5 is the schematic representation of the waveform of the sequence set V ₈ corresponding to 3 code bits w fixed to 1 in the present invention;

FIG. 6 is a schematic representation of a waveform of a sequence set constructed with w fixed as 1 in the present invention;

Fig. 7 is a schematic representation of a waveform of a sequence set with w not fixed structure in the present invention;

Fig. 8 is the implementation flow chart of the sending end based on the low repetition code word mapping of the present invention;

Fig. 9 is a flow chart of decoding at the receiving end based on low-heavy sequence mapping in the present invention;

Fig. 10 the present invention is based on the decoding flow chart of the receiving end of low repetition codeword mapping;

FIG. 11 is a simulation diagram of frame synchronization detection performance in Embodiment 1 of the present invention;

Fig. 12 is the specific embodiment 1 of the present invention under AWGN channel, RS code and LDPC code are respectively used as the decoding bit error rate performance figure of outer code;

Fig. 13 is the frame synchronization detection error rate curve under the AWGN channel of Embodiment 2 of the present invention;

Fig. 14 is the frame synchronization detection error rate curve under the AWGN channel of the specific embodiment 3 of the present invention;

Fig. 15 is a performance diagram of the decoding bit error rate of the

specific embodiments

2 and 3 of the present invention under the AWGN channel.

Table 1 is the corresponding relationship between the coding symbol composed of 3 bits and the sequence set _V8 in the specific embodiment of the present invention, and 8 groups of binary sparse sequences;

Table 2 is a schematic diagram of the corresponding relationship between each group of binary sparse sequences and coded symbols in the sequence set V ₆₄ with a length of 8 chips corresponding to 6 coded bits, which is constructed w indefinitely in the present invention.

Detailed ways

In order to make the purpose, technical solution and advantages of the present invention clearer, the implementation manners of the present invention will be further described in detail below.

Aiming at the difficulties of short-burst communication transmission in low signal-to-noise ratio environment, in order to ensure the accuracy of burst detection and the reliability of data transmission, the embodiment of the present invention proposes a bit-field superimposed pseudo-random sequence and sparse coding Sequential low signal-to-noise ratio communication methods.

At the sending end, channel coding is first performed on the information sequence to be transmitted, and then each channel coding symbol is converted into a sparse sequence with a small proportion of symbol "1" according to a specially designed sparse coding scheme, and then the sparse sequence is combined with the pseudo-random sequence The sequence to be transmitted is obtained by bit-by-bit XOR superposition, and finally modulated and sent; at the receiving end, firstly, the received signal and the local pseudo-random sequence are subjected to sliding correlation judgment to achieve frame synchronization and transmission symbol synchronization, and further remove the pseudo-random sequence in the signal , and then calculate the hard decision and soft decision information of each bit in the demodulator, and send it to the decoder for decoding and calculate the hard output and soft output information of each sparse codeword, and finally according to the same According to the mapping rule, the hard-decision information and soft-decision information of the channel coding symbols corresponding to each codeword are obtained, and the original information sequence is obtained through channel decoding.

The method proposed by the invention can ensure high accuracy of frame synchronization and realize reliable transmission of information under low signal-to-noise ratio, and can realize interference-free separation of transmission data under certain synchronization code related performance loss. In addition, the method proposed by the present invention can flexibly select the correlation length according to channel conditions, the maximum length is the same as the frame length, and has good channel adaptability, thereby realizing the trade-off between complexity and performance in the sliding correlation detection algorithm.

Facing the field of digital communication, the present invention proposes a cascaded coding transmission method of superimposing pseudo-random sequences, selects pseudo-random sequences with good correlation as the carrier of information transmission, and still has good autocorrelation characteristics after being superimposed with sparse coding sequences, And there is no need to allocate time slots for the training sequence, which effectively improves the frequency band utilization. At the same time, using the correlation of the pseudo-random sequence, the sliding correlation acquisition method is used for signal acquisition, and the frame synchronization and transmission can be accurately realized without loss of transmission resources. Symbol synchronization, combined with concatenated codes to achieve reliable transmission of information under low signal-to-noise ratio, compared with traditional coding transmission methods, the present invention can use one code stream to simultaneously realize two functions of frame synchronization and information transmission. The communication solution in the present invention will be described in detail below in conjunction with the accompanying drawings.

The system realization block diagram of the scheme of the present invention is as shown in Figure 1, and the scheme of the present invention should comprise the following steps:

(1) At the sending end, channel coding is first performed on the information sequence to be transmitted, and then each channel coding symbol is converted into a sparse sequence with a small proportion of symbol "1" according to a specially designed sparse coding scheme, and then the sparse sequence is combined with The pseudo-random sequence is XOR-superposed bit by bit to obtain the sequence to be transmitted, as shown in Figure 2, and finally modulated and sent;

(2) At the receiving end, firstly, the received signal and the local pseudo-random sequence are subjected to sliding correlation judgment to realize frame synchronization and transmission symbol synchronization, further remove the pseudo-random sequence in the signal, and then calculate each bit in the demodulator The hard decision and soft decision information are sent to the decoder to decode and calculate the hard output and soft output information of each sparsely coded codeword. Finally, the channel corresponding to each codeword is obtained according to the same mapping rules as the sending end. Encode symbol hard-decision information and soft-decision information, and obtain the original information sequence through channel decoding.

Among them, the local pseudo-random sequence is specifically the same pseudo-random sequence m as that of the sender, and the receiver can select the sliding window length of the correlator according to the channel conditions, where the maximum length is the same as the frame length, and correspondingly, the correlation operation is performed within the sliding window The local sequence of is a binary sequence of the same length intercepted from the pseudo-random sequence m from its head to the rear.

Wherein, the above step (1) is specifically:

(1.1) After K information bits are channel-coded, an N-bit-long encoded codeword c is obtained. Each p-bit encoded codeword for c is a group, and the i-th group of bit streams can be expressed as [c _ip ,c _ip+1 , …,c _ip+p-1 ] to form the corresponding encoding symbols, denoted as e _i ,

Where i∈{0,1,…,N/P-1}, c _ip+1 is the second chip of the i-th bit stream, c _ip+j is the j-th chip of the i-th bit stream , N is the bit length of the codeword c, and p is the number of symbols in each group of bit streams.

(1.2) Construct a mapping sequence set V for realizing the sparseness of channel coding codewords, and design a special sparse coding scheme according to the mapping sequence set V, and obtain a binary sparse sequence s after sparse coding;

(1.3) Perform chip-by-chip XOR of the binary sparse sequence s and the pseudo-random sequence m to obtain the sequence t to be transmitted,

Modulate and send.

Wherein, the above-mentioned step (2) is specifically:

(2.1) The receiving end receives the transmission signal, first realizes signal capture through the cross-correlation detector according to the known pseudo-random sequence, and completes frame synchronization and transmission symbol synchronization;

(2.2) Remove the pseudo-random sequence in the received signal, and flip the sign of the received signal according to the pulse position of the locally known pseudo-random sequence, thereby recovering the superimposed sparse sequence and sending it to the demodulator;

(2.3) Calculate the hard decision and soft decision information of each bit in the demodulator, and then obtain the hard information and soft information of the channel coding symbols corresponding to each sparse sequence according to the corresponding mapping rules in the sparse coding scheme at the sending end, and then pass Channel decoding obtains the original information sequence.

Among them, the generation of the binary sparse sequence s in the above step (1.2) specifically has two sparse coding implementation methods, which are:

The first one: use low-heavy sequence mapping to implement sparse coding, as shown in Figure 3. First, the weight w of each sequence in the constructed sequence set may not be completely equal. On the premise of ensuring the sparsity of the sequence, w can be arbitrarily smaller than the sequence The value of half the length is shown in Figures 4, 5, 6, 7, 8 and Table 1. Then, the mapping relationship between the channel coding symbols and the sparse sequence set is established to obtain the sparse coding codeword sequence.

Among them, the sequence in which the density of the symbol "1" in the binary sequence is less than 1/2 is called a low-heavy sequence; specifically, the implementation steps are as follows:

(3.1) The sequence set includes 2 ^p groups of different binary sparse sequences, expressed as

The length of each group of sequences is l chips, and a group of binary sparse sequences whose position number is a can be expressed as

l≤2 ^p , α=0,1,...,2 ^p -1;

(3.2) Select the binary sparse sequence corresponding to the position number in the sequence set V and the encoding symbol value as the transmission symbol, that is, when the encoding symbol is e _i , select the binary sparse sequence whose position number is e _i in the sequence set V

As the corresponding transmission symbol, sparse coding is completed;

(3.3) Design the corresponding decoding algorithm for the low-weight sequence mapping scheme to realize sparse coding.

The second method: use low-repetition codeword mapping to realize sparse coding, as shown in Figure 8, first, use the information bits of some low-repetition codewords of the error-correcting code to construct a sequence set, which can make each group of sequences in the sequence set undergo error-correcting coding The obtained codewords are all low-weight codewords, and then the mapping relationship between the channel coding symbols and the information bits of the low-weight codewords is established, and finally the low-heavy error correction coding is performed on the mapped sequence to obtain the sparse code sequence of words.

Wherein, the code word whose symbol "1" density is less than 1/2 in the binary code word sequence is called low repetition code word.

Specifically, the implementation steps are as follows:

(4.1) Taking the Golay(23,12) code as an example, the mapping sequence set V is composed of the 12-bit information bits of the encoded codewords with

code weights

0, 7 and 8 in the Golay(23,12) code, where

Represents a set of binary sequences whose position number is a, α=0,1,...,N/P-1;

(4.2) According to the mapping rules, select the binary sequence corresponding to the position number and the encoding symbol value in the sequence set V as the transmission sequence, that is, when the encoding symbol is c _i , select the binary sequence whose position number is c _i in the sequence set V

As the corresponding transmission sequence, complete the mapping;

(4.3) Golay encoding is performed on the mapped transmission sequence, that is, every 12 bits is a binary sparse sequence s encoded as a 23-bit long binary sparse sequence s with a weight of 0, 7 or 8;

(4.4) Design a corresponding decoding algorithm for the low-repetition codeword mapping scheme to realize sparse coding.

Wherein, the above-mentioned step (2.1) is specifically:

(5.1) Send the received signal y and the local pseudo-random sequence m into the cross-correlation detector for sliding correlation, and the output R(k, g) is the cross-correlation value, which can be expressed as:

Among them, N _k is the length of the sliding window of the correlator, y(n+g) is the sequence after the received signal is shifted by g bits, m(n) is the pseudo-random sequence in the sliding window, k represents the sequence number output by the correlator, n is the serial number of the digital sample.

(5.2) Detect the correlation results at any time, search for the correlation peak position, and the judgment criterion is:

in,

Indicates the searched correlation peak position, representing the arrival of the data frame.

(5.3) After successfully completing the code capture, determine the estimated value of frame synchronization according to the position of the correlation peak, find the starting position of one frame of the received signal, establish a pulse sequence consistent with the start and end times of the sending end, and realize frame synchronization;

(5.4) Divide the received signal into a group of 1 chip as the received symbol to realize the synchronization of the transmitted symbols, and send them to the demodulator to calculate the soft information of each symbol in turn.

Referring to Fig. 9, in the above step (3.3), the specific steps of the decoding algorithm designed for the low-heavy sequence mapping to realize the sparse coding scheme are as follows:

(6.1) The received signal of every l chip forms a symbol and is sent to the demodulator, where the nth group of received signal is expressed as

(6.2) In the demodulator, each symbol uses the l-dimensional Gaussian distribution density function to calculate the soft information according to the sequence set V, expressed as:

in,

is the symbol likelihood information of the received symbol corresponding to the position number a of v _a in the sequence set V, σ ² is the noise variance,

is the jth chip of the nth group of received signals,

is the j-th chip of the sequence numbered a in the sequence set V.

(6.3) Normalize the obtained symbol likelihood information as the initial probability without decoding and directly send it to the multi-ary decoder for decoding, or convert it into bit soft information and send it to binary decoding The device decodes, and the initial symbol probability of the nth group of information bits corresponding to the symbol a is expressed as

The calculation method is:

Referring to Fig. 10, in the above-mentioned step (4.4), the specific steps of the decoding algorithm for the scheme design of realizing sparse coding for the low-repetition codeword mapping are:

(7.1) Calculate the logarithmic likelihood information r _i of each bit in the demodulator and obtain the corresponding hard decision result u _i , where,

(7.2) According to the obtained bit log likelihood information and hard decision results, Golay decoding is performed with the correlation difference as the decoding metric, where the correlation difference

The codeword with the smallest correlation difference is the Golay decoding result;

Among them, ri _· s _i represents the product of the i-th chip of the two sequences, _ri is the i-th chip of the received sequence, and s _i is the i-th chip of the sparse sequence in the mapping table.

(7.3) After the Golay decoding is completed, the hard decision and soft decision information of the channel coding symbols corresponding to each codeword are obtained according to the same mapping rules as the sending end, and are output to the channel decoder;

(7.4) According to the obtained hard decision and soft decision information of the channel coding symbols, complete channel decoding and restore transmission information.

A low signal-to-noise ratio communication method designed by the present invention, in which a pseudo-random sequence and a sparse coded sequence are superimposed in a bit field, can realize two functions of frame synchronization and information transmission at the same time with one code stream and a carrier, without allocating additional time slots for pilots . Three specific embodiments are given below, all under the AWGN channel, adopting the BPSK modulation method to illustrate the feasibility of the method of the present invention, and to further understand the purpose, characteristics and advantages of the invention.

Example 1

The channel coding used in this embodiment is the RS(511, k) code defined on the GF(512) domain, the code rate is k/511, the length of the information sequence is 511 symbols, and each symbol has 9 bits. The pseudo-random sequence adopts an m-sequence with a series number of 14 and a feedback coefficient of 42103 (octal).

As shown in the system model in Figure 1, the specific process of the scheme is as follows:

(1) The random information source takes every 9 bits of information as a group, obtains an information sequence with a symbol length of k, and sends it to the RS encoder for encoding to obtain a 512-ary RS code word e=[e ₀ , with a symbol length of 511 e ₁ ,...,e ₅₁₀ ];

(2) Construct a mapping table, select the information bits corresponding to the codewords with weights of 7 or 8 in 511 Golay (23,12) codes, and add all 0 codewords to form a mapping table, corresponding to GF each element of (512);

(3) Select the binary sequence corresponding to the position number in the mapping sequence set and the encoding symbol value as the transmission symbol, that is, if the encoding symbol is e _i , then select the binary sparse sequence in the sequence set V

As a transmission symbol, the symbol mapping is completed;

(4) A sequence with a bit length of 4599 is mapped and Golay(23,12) encoded to obtain 511 groups of transmission symbols with a length of 23 chips, and the sparse sequence s composed of these transmission symbols is processed into m with a length of 11753 After the sequence m is XORed chip by chip, the sequence x to be transmitted is obtained;

Among them, the part of the m-sequence exceeding 11753 is truncated.

Wherein, the sequence to be transmitted is actually a pseudo-random sequence with 33% chip inversion, and the overall transmission code rate is 9k/(511×23).

(5) carry out BPSK modulation to the sequence x to be transmitted and send it sequentially;

(6) After the signal passes through the AWGN channel, the signal arriving at the receiving end is y, and then y and the local sequence m are sent to the sliding correlation detector for continuous correlation processing, and the output R(k,g) is the cross-correlation value , see formula (2) for the calculation method;

(7) Detect the correlation results at any time, and search for the correlation peak position, the judgment criterion is shown in formula (3);

(8) After successfully completing the code capture, determine the estimated frame synchronization value according to the correlation peak position, find the starting position of one frame of the received signal, set up a pulse sequence consistent with the start and end times of the sending end, and realize frame synchronization;

(9) Divide the received signal into a group of 23 chips as the received symbols to realize the synchronization of the transmitted symbols, and send the symbols to the demodulator in turn to calculate the hard decision information and soft decision information of each bit. Calculate the logarithmic likelihood information r _i of each bit in the demodulator and obtain the corresponding hard decision result u _i , see formula (6) for the hard decision method;

(10) Utilize the known pseudo-random sequence to process the hard decision and soft decision information obtained by the demodulator, and remove the superimposed pseudo-random sequence;

(11) According to the obtained bit log likelihood information and hard decision results, Golay decoding is performed with the correlation difference as the decoding metric, where the correlation difference

The codeword with the smallest correlation difference is the result of Golay decoding, and the hard decision is directly made;

(12) Obtain the RS coding symbols corresponding to the Golay decoding hard decision results according to the same mapping rules as the sending end, and output them to the RS decoder;

(13) The RS decoder adopts a hard-decision decoding algorithm to complete RS decoding according to the obtained hard-decision information of the channel coded symbols to obtain the original information sequence.

First, statistics are made on the error rate of frame synchronization detection in this embodiment. As shown in Figure 11, the performance of frame synchronization detection increases with the increase of the length of the sliding correlation window. When the correlation length is 11753, the m-sequence has the optimal frame synchronization Detection performance, when the signal-to-noise ratio is -22dB, the error rate reaches 10 ^-4 , when the correlation length is only 1023, the frame synchronization detection error rate reaches 10 ^-4 when the signal-to-noise ratio is -5dB, which verifies the proposed The scheme has excellent frame synchronization detection performance under low signal-to-noise ratio.

Then, the error performance of this embodiment is simulated, as shown in Figure 12, in this embodiment, the RS code is used as the outer code, and when the overall code rate of the concatenated code is 0.192, when the signal-to-noise ratio is -3.2dB, the system's translation The code error bit error rate (BER) can reach 10 ^-5 . In this scheme, the NB-LDPC code with a code rate of 1/2 is used as the outer code. When the overall code rate of the concatenated code is 0.196, and the signal-to-noise ratio is -6.2dB, the decoding bit error rate (BER) of the system can be 10 ^-5 ; with the NB-LDPC code of 1/3 code rate as the outer code, when the overall code rate of the concatenated code is 0.131, and the signal-to-noise ratio is -7.8dB, the system's decoding bit error rate (BER ) can reach 10 ^-5 .

Example 2

The channel coding adopted in this embodiment is an LDPC code defined on the GF(8) domain, the adopted parity check matrix H has a code length of 384 symbols, a code rate of 1/2, and an information sequence length of 192 symbols, corresponding to, The code length is 1152 bits, and the information sequence is 576 bits long. The 4 non-zero elements in each row of the parity check matrix are randomly selected from the non-zero element set {1,2,...,7} corresponding to the finite field. The pseudo-random sequence adopts an m-sequence with a series number of 12 and a feedback coefficient of 10123 (octal).

The sequence set used for sparse coding contains 8 groups of binary sparse sequences with a length of 8 chips, expressed as:

V ₈ ={v ₀ , v ₁ ,...,v ₇ }

The sparse sequence whose position number is a in the sequence set is expressed as

where α∈{0,1,...,7}. Figure 5 shows the waveform representation of the sequence set _V8 , and Table 1 shows the correspondence between subsequences and coding symbols in the sequence set _V8 .

Table 1

(1) A random information source generates a 576-bit information sequence, and each 3-bit information is used as a group to obtain an information sequence f=[f ₀ ,f ₁ ,…,f ₁₉₁ ] with a symbol length of 192 and send it to the LDPC encoder , encode according to its check matrix H to obtain an octal LDPC codeword e=[e ₀ , e ₁ ,..., e ₃₈₃ ] whose symbol length is 384;

(2) Select the binary sparse sequence corresponding to the position number and coded symbol value in the sequence set V ₈ as the transmission symbol, that is, if the coded symbol is e _i , then select the binary sparse sequence in the sequence set V

As a transmission symbol, sparse coding is completed;

(3) The codeword with a symbol length of 384 is sparsely orthogonally encoded to obtain 384 groups of transmission symbols with a length of 8 chips, and the sparse sequence composed of these transmission symbols

XOR with the pseudo-random sequence m with a length of 3072 chip by chip to obtain the sequence t to be transmitted;

Wherein, the sequence to be transmitted is actually a pseudo-random sequence with 12.5% chip inversion, and the overall transmission code rate is 3/16.

(4) carry out BPSK modulation to the sequence t to be transmitted and send it sequentially;

(5) For the actual execution steps of signal capture and synchronization, refer to steps (6)-(8) in Embodiment 1;

(6) Divide the received signal, the received signal of every 8 chips

Form a symbol and send it to the demodulator;

(7) In the demodulator, each group of received signals uses an ₈ -dimensional Gaussian distribution density function to calculate the symbol likelihood information of the corresponding data according to the sequence set V8, and the calculation method is shown in formula (4);

(8) After the symbol likelihood information is normalized, it is directly sent to the octal LDPC decoder for decoding as the initial probability without decoding. The initial symbol probability of the nth group of information bits corresponding to the symbol a is expressed as

See formula (5) for the calculation method;

In this embodiment, the Fast Fourier Transform-Confidence Propagation (FFT-BP) algorithm on the GF(8) domain is used for decoding. Statistics, and compared with the frame synchronization detection error rate of the frame header in the traditional frame structure, where the frame header is also composed of m-sequence. As shown in Figure 14, when the correlator sliding window length is 3072, the frame synchronization detection performance of this embodiment is optimal, that is, when the signal-to-noise ratio is -17dB, the detection error rate reaches 10 ^-5 , which is close to the length Frame synchronization detection performance for 2047 frame headers. When the correlation lengths are 1023 and 2047 respectively, the local sequences in the sliding window are the first 1023 chips and the first 2047 chips of the pseudo-random sequence, and the detection error rates are -12dB and -15dB respectively It reaches 10 ^-5 , which is close to the frame synchronization detection performance of the frame header composed of m-sequences with lengths of 511 and 1023 respectively.

Then, the system error performance of this embodiment under the AWGN channel is simulated. As shown in Figure 16, when the signal-to-noise ratio is -0.8dB in this embodiment, the decoding bit error rate (BER) can reach 10 ^-6 . , there is a performance gain of about 1.3dB. .

Example 3

At the sending end, a group of 6-bit codewords is used as a group for sparse coding, and the sequence set V ₆₄ used is shown in Table 2, expressed as V ₆₄ =[v ₀ ,v ₁ ,…,v ₆₃ ], including 64 groups of length It is a binary sparse sequence of 8 chips, wherein there are 8 sequences with weight w=1, and 56 sequences with weight w=3, and the sparse sequence whose position number is a in the sequence set is expressed as

where α∈{0,1,...,63}. The specific implementation process of the embodiment is as follows.

Table 2

The channel coding adopted in this embodiment is an LDPC code defined on the GF(64) domain, the adopted parity check matrix H has a code length of 384 symbols, a code rate of 1/2, and an information sequence length of 192 symbols, corresponding to, The code length is 2304 bits, and the information sequence is 1152 bits long. The 4 non-zero elements of each row of the parity check matrix are randomly selected from the set of non-zero elements of the corresponding finite field. The pseudo-random sequence used is the same as that of the specific embodiment 1, and the implementation process of the scheme is similar to that of the specific embodiment 2.

In this embodiment, the FFT-BP algorithm on the GF(64) domain is used for decoding. First, the frame synchronization detection error rate of this embodiment is counted under the AWGN channel, and compared with the frame header in the traditional frame structure Frame sync detection error rates are compared. As shown in Figure 13, when the correlator sliding window length is 3072, this embodiment achieves the optimal frame synchronization detection performance, that is, when the signal-to-noise ratio is -9dB, the error rate reaches 10 ^-5 , which is close to the length of 255 Frame synchronization detection performance of frame headers composed of m-sequences. When the correlation length is 2047, the detection error rate reaches 10 ^-5 when the signal-to-noise ratio is -dB, which is better than the frame synchronization detection performance of the frame header whose length is 127.

Then, the system error performance of this embodiment under the AWGN channel is simulated. As shown in FIG. 15 , in this embodiment, when the SNR is -2.2dB, the BER can reach 10 ^-6 . Compared with the transmission scheme using BPSK combined with the same LDPC code, there is a performance gain of about 1dB.

It can be seen from the simulation results of the embodiment that the scheme of the present invention can ensure the accuracy of burst capture and frame synchronization detection under low signal-to-noise ratio, and can adapt to different channel conditions by selecting different correlation lengths. At the same time, compared with the traditional transmission scheme, it also has a certain error performance gain, which is suitable for short burst communication.

In the embodiments of the present invention, unless otherwise specified, the models of the devices are not limited, as long as they can complete the above functions.

Those skilled in the art can understand that the accompanying drawing is only a schematic diagram of a preferred embodiment, and the serial numbers of the above-mentioned embodiments of the present invention are for description only, and do not represent the advantages and disadvantages of the embodiments.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within range.

Claims

A kind of transmission method of superimposed pseudo-random sequence and sparse concatenated coding in bit field, it is characterized in that, described method comprises the following steps:

(1) At the sending end, channel coding is first performed on the information sequence to be transmitted, and each channel coded symbol is converted into a sparse sequence with a small proportion of symbol "1" according to the sparse coding scheme, and then the sparse sequence and the pseudo-random sequence are one by one The sequence to be transmitted is obtained by bit XOR superposition, and finally modulated and sent;

(2) At the receiving end, firstly, the received signal and the local pseudo-random sequence are subjected to sliding correlation judgment to realize frame synchronization and transmission symbol synchronization, further remove the pseudo-random sequence in the signal, and then calculate each bit in the demodulator The hard decision and soft decision information are sent to the decoder for decoding and the hard output and soft output information of each sparsely coded codeword are calculated, and finally the channel coding corresponding to each codeword is obtained according to the same mapping rules as the sending end Symbol hard-decision information and soft-decision information are channel-decoded to obtain the original information sequence.
A kind of transmission method of superimposed pseudo-random sequence and sparse concatenated coding in bit field according to claim 1, it is characterized in that, described step (1) is specifically:

(1.1) K information bits are channel-coded to obtain an N-bit codeword c, c is a group of codewords of p bits each, and the i-th group can be expressed as [c ip , c ip+1 ,..., c ip+p-1 ] to form the corresponding encoding symbols, expressed as e i ,

(1.2) Construct the mapping sequence set V used to realize the sparseness of channel coding code words, and design a special sparse coding scheme according to the mapping sequence set, and obtain the binary sparse sequence s after sparse coding;

(1.3) Perform chip-by-chip XOR of the binary sparse sequence s and the pseudo-random sequence m to obtain the sequence t to be transmitted,
Modulate and send.
A kind of transmission method of superimposed pseudo-random sequence and sparse concatenated coding in bit field according to claim 1, it is characterized in that, described step (2) is specifically:

(2.1) The receiving end receives the transmission signal, realizes signal capture through a cross-correlation detector according to a known pseudo-random sequence, and completes frame synchronization and transmission symbol synchronization;

(2.2) Remove the pseudo-random sequence in the received signal, flip the received signal according to the pulse position of the locally known pseudo-random sequence, recover the superimposed sparse sequence, and send it to the demodulator;

(2.3) Calculate the hard decision and soft decision information of each bit in the demodulator, and then obtain the hard information and soft information of the channel coding symbols corresponding to each sparse sequence according to the corresponding mapping rules in the sparse coding scheme at the sending end, and then pass Channel decoding obtains the original information sequence.
The transmission method of a kind of bit field superposition pseudo-random sequence and sparse concatenated coding according to claim 2, it is characterized in that, described step (1.2) is specifically:

(1.2.1) Use low-heavy sequence mapping to realize sparse coding. First, the weight w of each group of sequences in the constructed sequence set is not completely equal. Under the premise of ensuring sequence sparsity, w is any value less than half the sequence length, and then , establish the mapping relationship between the channel coding symbols and the sparse sequence set, and obtain the sparse coding codeword sequence;

(1.2.2) Use the low-repetition codeword mapping to realize sparse coding. First, use the information bits of some low-repetition codewords of the error-correcting code to construct a sequence set, so that the codewords obtained after error-correcting coding for each sequence in the sequence set are The low-weight codeword, and then establish the mapping relationship between the channel coding symbols and the information bits of the low-weight codeword, and finally perform low-heavy error correction coding on the mapped sequence to obtain a sparsely coded codeword sequence.
According to claim 1, a kind of transmission method of superimposing pseudo-random sequence and sparse concatenated coding in bit field, it is characterized in that, at the receiving end, the sliding correlation judgment is carried out between the received signal and the local pseudo-random sequence, so as to realize frame synchronization and Transmission symbol synchronization, the specific steps are:

(3.1) Send the received signal y and the local pseudo-random sequence m to the cross-correlation detector for sliding correlation, and the output R(k,g) is expressed as:

Among them, N k is the length of the sliding window of the correlator, y(n+g) is the sequence after the received signal is shifted by g bits, m(n) is the pseudo-random sequence in the sliding window, k represents the sequence number output by the correlator, n is the serial number of the digital sample;

(3.2) Detect the correlation results at any time, search for the correlation peak position, and the judgment criterion is:

in,
Indicates the searched correlation peak position, representing the arrival of the data frame;

(3.3) After successfully completing the code capture, determine the estimated frame synchronization value according to the correlation peak position, find the starting position of a frame of the received signal, establish a pulse sequence consistent with the start and end times of the sending end, and realize frame synchronization;

(3.4) Divide the received signal into a group of 1 chip as the received symbol to realize the synchronization of the transmitted symbols, and send them to the demodulator to calculate the soft information of each symbol in turn.
The transmission method of a kind of bit field superposition pseudo-random sequence and sparse concatenated coding according to claim 4, it is characterized in that, described step (1.2.1) is specifically:

The sequence set consists of 2 p sets of distinct binary sparse sequences denoted as
The length of each group of sequences is l chips, and a group of binary sparse sequences whose position number is a can be expressed as

α=0,1,..., 2p -1;

Select the binary sparse sequence corresponding to the position number in the sequence set V and the encoding symbol value as the transmission symbol, that is, when the encoding symbol is e i , select the binary sparse sequence whose position number is e i in the sequence set V
As the corresponding transmission symbol, sparse coding is completed;

A corresponding decoding algorithm is designed for the sparse coding scheme of low-weight sequence mapping.
A kind of transmission method of superimposed pseudo-random sequence and sparse concatenated coding in bit field according to claim 5, it is characterized in that, described step (1.2.2) is specifically:

Select the binary sequence corresponding to the position number and the coded symbol value in the sequence set V as the transmission sequence according to the mapping rule;

Golay encoding is performed on the mapped transmission sequence, and a corresponding decoding algorithm is designed for the sparse coding scheme realized by low-repetition codeword mapping.
According to claim 6, a bit-field superimposed pseudo-random sequence and sparse concatenated coding transmission method is characterized in that, the corresponding decoding algorithm is designed for the scheme of low-heavy sequence mapping to realize sparse coding, and the specific steps are as follows :

(6.1) The received signal of each 1 chip forms a symbol, which is sent to the demodulator, and each symbol calculates the soft information by using the 1-dimensional Gaussian distribution density function according to the sequence set V, expressed as:

in,
The received symbol corresponds to the symbol likelihood information of the position number a of v a in the sequence set V;

(6.2) Normalize the obtained symbol likelihood information as the initial probability without decoding and directly send it to the multi-ary decoder for decoding, or convert it into bit soft information and send it to binary decoding The device decodes, and the initial symbol probability of the nth group of information bits corresponding to the symbol a is expressed as
The calculation method is:
According to claim 7, a bit-field superimposed pseudo-random sequence and sparse concatenated coding transmission method, characterized in that, the corresponding decoding algorithm is designed for the scheme of implementing sparse coding for low-heavy code word mapping, and the specific steps are as follows :

(7.1) Calculate the logarithmic likelihood information r i of each bit in the demodulator and obtain the corresponding hard decision result u i ;

(7.2) According to the obtained bit logarithmic likelihood information and hard decision results, Golay decoding is carried out with the correlation difference as the decoding measure, and the codeword with the smallest correlation difference is the Golay decoding result;

(7.3) After the Golay decoding is completed, the hard decision and soft decision information of the channel coding symbols corresponding to each codeword are obtained according to the same mapping rules as the sending end, and are output to the channel decoder;

(7.4) Complete channel decoding according to the obtained hard decision and soft decision information of channel coding symbols, and obtain the original information sequence.