4096 QAM (Quadrature Amplitude Modulation) probability shaping coding modulation method used under Rayleigh fading channel

Abstract

In order to improve high-reliability data transmission performance in a high spectral efficiency scene, the invention provides a 4096QAM probability shaping coding modulation method used under a Rayleigh fading channel, according to the method, through average mutual information AMI theoretical analysis and frame error rate simulation verification, when the spectral efficiency is 9.60 bit / s / Hz, compared with a traditional equal-probability rule 4096QAM coding modulation scheme, the probability shaping coding modulation method has the advantages that the probability shaping coding modulation efficiency is greatly improved, and the probability shaping coding modulation efficiency is greatly improved. A 4096 QAM probability shaping coding modulation scheme under a Rayleigh fading channel is provided, so that a simulation gain of 1.35 dB can be obtained; when the spectral efficiency is 10.00 bit / s / Hz, compared with a traditional equal-probability rule 4096 QAM coding modulation scheme, the provided 4096 QAM probability shaping coding modulation scheme under the Rayleigh fading channel can obtain the simulation gain of 11.30 dB, and compared with the traditional equal-probability rule 4096 QAM coding modulation scheme, the reliable transmission performance of the system can be improved.

Description

Technical Field

[0001] The invention belongs to the field of wireless communication, and in particular relates to a 4096QAM probability shaping coding modulation method for a Rayleigh fading channel. Background Art

[0002] The sixth generation of mobile communications (6G) needs to meet the requirements of extremely high reliability, higher spectrum efficiency, ultra-low latency, etc. At present, the fifth generation of mobile communications (5G) communication system adopts a quadrature amplitude modulation (QAM) constellation with equal probability and uniform distribution. For high spectrum efficiency data transmission, the current 5G system adopts 256QAM constellation modulation and 1024QAM constellation modulation.

[0003] In an additive white Gaussian noise (AWGN) channel, under high spectral efficiency, the use of an equal-probability regular QAM constellation will result in a shaping loss of 1.53 dB. In order to obtain shaping gain, geometric shaping (GS) and probabilistic shaping (PS) can be used. Geometric shaping obtains shaping gain by satisfying equal-probability and unequal-interval distribution of constellation points, and probabilistic shaping obtains shaping gain by satisfying unequal-probability and equal-interval distribution of constellation points. The optimization of geometric shaping is related to the signal-to-noise ratio, and its hardware implementation is complex; probabilistic shaping generates an amplitude sequence with unequal-probability distribution by using a constant composition distribution matching (CCDM). Compared with geometric shaping, probabilistic shaping keeps the position of the constellation points unchanged, can support different low-density parity code (LDPC) code rates at the same time, and has the advantages of flexible code rate matching and easy hardware implementation.

[0004] In Rayleigh fading channels, an existing rotated QAM constellation modulation probability shaping scheme optimizes the constellation rotation angle and constellation probability distribution through average mutual information (AMI) theoretical analysis. However, for the traditional regular QAM coding modulation scheme and the PS QAM coding modulation scheme, when the same code rate channel coding is used, the spectral efficiency of the two schemes is not equal. The PS scheme will cause code rate loss, that is, when the same LDPC code rate is used, the spectral efficiency of the PS QAM coding modulation scheme is less than that of the traditional regular QAM coding modulation scheme. Summary of the invention

[0005] In order to further improve the spectrum efficiency, the present invention proposes a 4096QAM probability shaping coding modulation method under Rayleigh fading channel, which can improve the reliable transmission performance of the system compared with the traditional equal-probability rule 4096QAM coding modulation scheme.

[0006] To achieve the above object, the technical solution of the present invention is a 4096QAM probability shaping coding modulation method under Rayleigh fading channel, 1) At the transmitter, the k-bit source sequence is uniformly distributed After distribution matching (DM), we can obtain the unequal probability amplitude sequence. , Among them, the unequal amplitude sequence satisfies the probability distribution , the code rate of the distribution matching is ; Amplitude sequence pass Mapped to a binary bit sequence , is the amplitude mapping function; 2) In the 4096QAM probability shaped coding modulation system, for Constellation modulation, LDPC code rate Need to meet ;when , equal probability source bit sequence The length of is 0; when , equal probability source bit sequence The length is , then the bit sequence needs to be and Combine to get the sequence , convert the sequence As the LDPC coded information bit, LDPC encoding is performed; 3) After LDPC encoding, the system bit symbol sequence is obtained and check digit symbol sequence ; 4) When the LDPC code rate meets When the system bit symbol sequence The bit sequence is used for the amplitude bit and sign bit of constellation modulation, where the amplitude bit sequence The length is bit; the systematic bit symbol sequence The remaining length is The bit sequence and check bit sequence Combined as the symbol bit sequence of constellation modulation , the length of the sign bit sequence is Bit; 5) Set the sign bit sequence and amplitude bit sequence Perform constellation modulation; 6) The complex signal after constellation modulation is ,in, represents transpose, and Represents complex signals The in-phase and orthogonal components of and They are represented by a 64ASK constellation symbol respectively. The total system bit rate is, (1) For 4096QAM modulation, the system spectrum effect is ; 7) Channel State Information (CSI) at the receiving end. After passing through the Rayleigh fading channel, the expression of the received signal is: (2) in, is the channel fading coefficient, which satisfies the mean of 0 and the variance of 1. N represents a complex Gaussian noise sequence with a mean of 0 and a variance of , after phase equalization, the received signal is (3) in, represents the pseudo-inverse operation of the matrix, represents the norm operation, N is circularly symmetric, sequence sum N The sequences satisfy the same distribution, i.e. The mean of the sequence is 0 and the variance is ; The signal sequence Decomposed into two parallel channels, the expression is as follows: (4) in, represents a diagonal matrix, represents an independent random variable with Rayleigh distribution, and its probability density function is expressed as ; and are the in-phase component and orthogonal component of the random variable Z respectively; and The mean and variance are both 0. After the maximum a posteriori (MAP) probability detection, the bit-level log-likelihood ratio information of the LDPC code is obtained. For the two-dimensional PS QAM constellation modulation, the Gaussian conditional probability used is the complex Gaussian conditional probability, which is expressed as (5) in, is the complex Gaussian conditional probability, and They are the in-phase component and quadrature component of the complex signal respectively. is a random variable The determined value of 8) Perform LDPC demodulation and decoding; 9) After LDPC decoding, the binary bit sequence is obtained ; The bit sequence To split the , where the bit sequence After the inverse amplitude mapping function Matching the inverse distribution After that, we get the estimated sequence of part of the transmitted sequence , completing the entire decoding process.

[0007] The 4096QAM modulation adopts two-dimensional probability shaping (PS) and Gray mapping coding.

[0008] The two-dimensional probability shaping (PS) is to divide a 4096QAM modulation symbol into two 64 amplitude shift keying (ASK) modulation symbols. For 64ASK constellation modulation, one amplitude is mapped to 5 bits, that is, the amplitude bit is 5 bits, and the remaining 1 bit is the sign bit. The sign bit length of the entire constellation modulation is n bits, and the amplitude bit length is 5n bits.

[0009] The beneficial effect of the coding modulation method of the present invention is to improve the high-reliability data transmission performance in a high-spectral-efficiency scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 This is the system block diagram of the 4096QAM probability shaped coding modulation scheme under Rayleigh fading channel.

[0011] Figure 2 The frame error rate performance of 4096QAM is 1.

[0012] Figure 3 The frame error rate performance for 4096QAM is 2. DETAILED DESCRIPTION

[0013] like Figure 1 As shown, a 4096QAM probability shaped coding modulation method under Rayleigh fading channel, 1) At the transmitter, the k-bit source sequence is uniformly distributed After distribution matching (DM), we can obtain the unequal probability amplitude sequence. , Among them, the unequal amplitude sequence satisfies the probability distribution , the code rate of the distribution matching is ; Amplitude sequence pass Mapped to a binary bit sequence , is the amplitude mapping function; 4096QAM modulation uses two-dimensional probability shaping (PS) and Gray mapping coding; two-dimensional probability shaping (PS) is to divide a 4096QAM modulation symbol into two 64 amplitude shift keying (ASK) modulation symbols. For 64ASK constellation modulation, one amplitude is mapped to 5 bits, that is, the amplitude bit is 5 bits, and the remaining 1 bit is the sign bit. The symbol bit length of the entire constellation modulation is n bits, and the amplitude bit length is 5n bits; 2) In the 4096QAM probability shaped coding modulation system, for Constellation modulation, LDPC code rate Need to meet ;when , equal probability source bit sequence The length of is 0; when , equal probability source bit sequence The length is , then the bit sequence needs to be and Combine to get the sequence , convert the sequence As the LDPC coded information bit, LDPC encoding is performed; 3) After LDPC encoding, the system bit symbol sequence is obtained and check digit symbol sequence ; 4) When the LDPC code rate meets When the system bit symbol sequence The bit sequence is used for the amplitude bit and sign bit of constellation modulation, where the amplitude bit sequence The length is bit; the systematic bit symbol sequence The remaining length is The bit sequence and check bit sequence Combined as the symbol bit sequence of constellation modulation , the length of the sign bit sequence is Bit; 5) Set the sign bit sequence and amplitude bit sequence Perform constellation modulation; 6) The complex signal after constellation modulation is ,in, represents transpose, and Represents complex signals The in-phase and orthogonal components of and They are represented by a 64ASK constellation symbol respectively. The total system bit rate is, (1) For 4096QAM modulation, the system spectrum effect is ; 7) Channel State Information (CSI) at the receiving end. After passing through the Rayleigh fading channel, the expression of the received signal is: (2) in, is the channel fading coefficient, which satisfies the mean of 0 and the variance of 1. N represents a complex Gaussian noise sequence with a mean of 0 and a variance of , after phase equalization, the received signal is (3) in, represents the pseudo-inverse operation of the matrix, represents the norm operation, N is circularly symmetric, sequence sum N The sequences satisfy the same distribution, i.e. The mean of the sequence is 0 and the variance is ; The signal sequence Decomposed into two parallel channels, the expression is as follows: (4) in, represents a diagonal matrix, represents an independent random variable with Rayleigh distribution, and its probability density function is expressed as ; and are the in-phase component and orthogonal component of the random variable Z respectively; and The mean and variance are both 0. After the maximum a posteriori (MAP) probability detection, the bit-level log-likelihood ratio information of the LDPC code is obtained. For the two-dimensional PS QAM constellation modulation, the Gaussian conditional probability used is the complex Gaussian conditional probability, which is expressed as (5) in, is the complex Gaussian conditional probability, and They are the in-phase component and quadrature component of the complex signal respectively. is a random variable The determined value of 8) Perform LDPC demodulation and decoding; For each received QAM signal , the binary demodulation bit log-likelihood ratio information is, , in, , , For binary LDPC codes, the bit log-likelihood ratio information Can be directly used for binary LDPC decoding; 9) After LDPC decoding, the binary bit sequence is obtained ; The bit sequence To split the , where the bit sequence After the inverse amplitude mapping function Matching the inverse distribution After that, we get the estimated sequence of part of the transmitted sequence , completing the entire decoding process.

[0014] Conduct simulation analysis. The simulation compares the frame error rate of the 4096QAM PS coded modulation system and the 4096QAM iso-uniform regular coded modulation system under Rayleigh fading channels, where the channel coding uses a binary LDPC code with a regularity of 3 and a code length of 12000 bits. The probability distribution of the PS amplitude satisfies the MB distribution.

[0015] Figure 2The figure shows the frame error rate performance when the total code rate is 4 / 5, that is, the spectrum efficiency is 9.60 bit / s / Hz. Among them, the LDPC code rate of the 4096QAM uniform coding modulation system is 4 / 5, and the LDPC code rate of the 4096QAM PS coding modulation system is 5 / 6. When the frame error rate FER=10 -3 When 4096QAM modulation is used and the spectrum efficiency is 9.60 bits / s / Hz, the proposed 4096QAM PS coded modulation system has a performance gain of 1.35 dB compared with the traditional 4096QAM uniform regular coded modulation system.

[0016] Figure 3 The figure shows the frame error rate performance when the total code rate is 5 / 6, that is, the spectral efficiency is 10.00 bit / s / Hz. Among them, the LDPC code rate of the 4096QAM uniform regular coding modulation system is 5 / 6, and the LDPC code rate of the 4096QAM PS coding modulation system is 7 / 8. When the frame error rate FER=10 -3 When 4096QAM modulation is used and the spectrum efficiency is 10.00 bits / s / Hz, the proposed 4096QAM PS coded modulation system has a performance gain of 1.30 dB compared with the traditional 4096QAM uniform regular coded modulation system.

[0017] Therefore, when the spectral efficiency is 9.60 bit / s / Hz, the proposed 4096QAM probability shaped coded modulation scheme under Rayleigh fading channel can obtain a simulation gain of 1.35 dB compared with the traditional equal-probability regular 4096QAM coded modulation scheme. When the spectral efficiency is 10.00 bit / s / Hz, the proposed 4096QAM probability shaped coded modulation scheme under Rayleigh fading channel can obtain a simulation gain of 1.30 dB compared with the traditional equal-probability regular 4096QAM coded modulation scheme.

Claims

1. A 4096QAM probability shaped coding modulation method under Rayleigh fading channel, characterized in that: 1) At the transmitter, the k-bit source sequence is uniformly distributed After distribution matching (DM), we can obtain the unequal probability amplitude sequence. , Among them, the unequal amplitude sequence satisfies the probability distribution , the code rate of the distribution matching is ; Amplitude sequence pass Mapped to a binary bit sequence , is the amplitude mapping function; 2) In the 4096QAM probability shaped coding modulation system, for Constellation modulation, LDPC code rate Need to meet ;when , equal probability source bit sequence The length of is 0; when , equal probability source bit sequence The length is , then the bit sequence needs to be and Combine to get the sequence , convert the sequence As the LDPC coded information bit, LDPC encoding is performed; 3) After LDPC encoding, the system bit symbol sequence is obtained and check digit symbol sequence ; 4) When the LDPC code rate meets When the system bit symbol sequence The bit sequence is used for the amplitude bit and sign bit of constellation modulation, where the amplitude bit sequence The length is bit; the systematic bit symbol sequence The remaining length is The bit sequence and check bit sequence Combined as the symbol bit sequence of constellation modulation , the length of the sign bit sequence is Bit; 5) Set the sign bit sequence and amplitude bit sequence Perform constellation modulation; 6) The complex signal after constellation modulation is ,in, represents transpose, and Represents complex signals The in-phase and orthogonal components of and They are represented by a 64ASK constellation symbol respectively. The total system bit rate is, (1) For 4096QAM modulation, the system spectrum effect is ; 7) Channel State Information (CSI) at the receiving end. After passing through the Rayleigh fading channel, the expression of the received signal is: (2) in, is the channel fading coefficient, which satisfies the mean of 0 and the variance of 1. N represents a complex Gaussian noise sequence with a mean of 0 and a variance of , after phase equalization, the received signal is (3) in, represents the pseudo-inverse operation of the matrix, represents the norm operation, N is circularly symmetric, sequence sum N The sequences satisfy the same distribution, i.e. The mean of the sequence is 0 and the variance is ; The signal sequence Decomposed into two parallel channels, the expression is as follows: (4) in, represents a diagonal matrix, represents an independent random variable with Rayleigh distribution, and its probability density function is expressed as ; and are the in-phase component and orthogonal component of the random variable Z respectively; and The mean and variance are both 0. After the maximum a posteriori (MAP) probability detection, the bit-level log-likelihood ratio information of the LDPC code is obtained. For the two-dimensional PS QAM constellation modulation, the Gaussian conditional probability used is the complex Gaussian conditional probability, which is expressed as (5) in, is the complex Gaussian conditional probability, and They are the in-phase component and quadrature component of the complex signal respectively. is a random variable The determined value of 8) Perform LDPC demodulation and decoding; 9) After LDPC decoding, the binary bit sequence is obtained ; The bit sequence To split the , where the bit sequence After the inverse amplitude mapping function Matching the inverse distribution After that, the estimated sequence of the partial transmission sequence is obtained , completing the entire decoding process.

2. The 4096QAM probability shaped coding modulation method under Rayleigh fading channel according to claim 1, characterized in that: The 4096QAM modulation system adopts two-dimensional probability shaping (PS) and Gray mapping coding.

3. The 4096QAM probability shaped coding modulation method under Rayleigh fading channel according to claim 2, characterized in that: The two-dimensional probability shaping (PS) is to divide a 4096QAM modulation symbol into two 64 amplitude shift keying (ASK) modulation symbols. For 64ASK constellation modulation, one amplitude is mapped to 5 bits, that is, the amplitude bit is 5 bits, and the remaining 1 bit is the sign bit. The sign bit length of the entire constellation modulation is n bits, and the amplitude bit length is 5n bits.

Images