A polar code transmission method based on transmission channel quantization level optimization

Abstract

The present invention relates to a polarization code transmission method based on transmission channel quantization level optimization, which solves the defect that different antennas will produce different precisions through different ADC links based on mobile communication MIMO system compared with the prior art. The present invention includes the following steps: the transmitting end screens the channel and performs coded transmission of polarization code; the receiving end constructs the transmission channel quantization parameter based on the minimum distortion quantization algorithm; the receiving end quantizes according to the transmission channel quantization level and performs BP decoding. The present invention optimizes quantizers of different precisions under polarization code BP decoding, improves the performance of polarization code decoder in a system with different quantization precisions, and improves the transmission efficiency of polarization code.

Description

Technical Field

[0001] The present invention relates to the field of 5G communication technology, and specifically to a polarization code transmission method based on transmission channel quantization level optimization. Background Art

[0002] From the first generation (1G) mobile communication system to the fifth generation (5G) mobile communication system, mobile communication has developed into a technology that has a comprehensive impact on people's daily lives. Although the fourth generation (4G) mobile communication system can already provide high-speed and high-quality network conditions, people's demand for higher communication technology is increasingly insatiable, so 5G came into being. Currently, 5G includes three main application scenarios: enhanced mobile broadband (eMBB), massive machine type communication (mMTC) and ultra-reliable low latency (URLLC). Polar code was first proposed in 2008. It has the advantages of efficient coding and low complexity decoding. Therefore, it was selected as the coding scheme for short code length control channels in 5G, attracting the research of many scholars at home and abroad.

[0003] Another technology that plays a vital role in 5G systems is massive multiple-input multiple-output (MIMO). MIMO has been widely used due to its larger bandwidth and higher communication capacity. However, in massive MIMO systems, different antennas transmit to analog-to-digital converters (ADCs) with different precision. In order to obtain high-quality communication, high-precision ADCs are generally used, but the power consumption of ADCs increases exponentially with the quantization level. The higher the precision, the greater the power consumption. In order to effectively reduce the power consumption of ADCs, low-precision ADCs are proposed. However, low-precision quantization will cause more serious nonlinear distortion, which is a great challenge to the accuracy of information transmission.

[0004] In order to solve the problem that different antennas will have different accuracies when passing through different ADC chains, it is proposed to optimize the corresponding relationship between the quantization level and the received symbols in the codeword.

[0005] Therefore, in the polar code information sequence After coding, modulation and transmission through the additive white Gaussian noise channel (AWGN), the channel output is obtained. How to Placing different quantization levels at different locations becomes a technical issue that needs to be explored and solved. Summary of the invention

[0006] The purpose of the present invention is to solve the defect in the prior art that different antennas will produce different accuracies through different ADC links based on a mobile communication MIMO system, and to provide a polarization code transmission method based on transmission channel quantization level optimization to solve the above problem.

[0007] In order to achieve the above object, the technical solution of the present invention is as follows:

[0008] A polar code transmission method based on transmission channel quantization level optimization comprises the following steps:

[0009] The transmitter screens the channel and performs polarization code encoding and transmission: The transmitter calculates the channel parameter vector P e , according to the channel parameter vector P e The information bit channel and the frozen bit channel are selected based on the value of ; the information bit channels constitute an information bit channel set I, and the frozen bit channels constitute a frozen bit channel set F; Q CRC check bits are added to the information bits, the information sequence is encoded and modulated by polar code, and then sent through the AWGN channel;

[0010] The receiving end constructs the transmission channel quantization parameters based on the minimum distortion quantization algorithm: In order to minimize the quantization error, the minimum distortion quantization algorithm is obtained by taking the derivative of the distortion degree D; the quantization level is set to L, generally L is an integer power of 2, and the transmission channel quantization level optimization rule is constructed according to the quantization level L and the minimum distortion quantization algorithm. The rule is determined by the endpoints x of the input range. j And output information y j Composition, x j is L-1, y j There are L; the optimization rule stipulates that if the input information is in x j and x j+1 interval, the output is the same value y j ;

[0011] The receiving end quantizes the polar code according to the quantization level of the transmission channel and performs BP decoding: The receiving end receives the polar code sent by the transmitting end, quantizes the polar code according to the quantization level of the transmission channel, and then performs BP decoding.

[0012] The transmitting end screening the channel and performing coded transmission of the polarization code comprises the following steps:

[0013] For the channel that transmits information, the channel parameter vector P is calculated using the Tal-Vardy construction method. e ;

[0014] According to the channel parameter vector P e The value of Filter out the information bit channel and the frozen bit channel, the steps are as follows:

[0015] The channel parameter vector P e The values ​​of are arranged in descending order to obtain the sorted channel parameter vector P temp , and calculate P temp Each element in P e The index value in gets the index vector P idx ;

[0016] According to the index vector P idx The values ​​of the first K elements in are arranged in ascending order to obtain the information bit channel set I, where K is the number of information bits; the values ​​of the remaining NK elements are arranged in ascending order to obtain the frozen bit channel set F;

[0017] Q CRC check codes are added to the end of the information bit, and then the information is transmitted, polar code encoding is completed, and it is sent after BPSK modulation and AWGN channel.

[0018] The receiving end constructs the transmission channel quantization parameter based on the minimum distortion quantization algorithm, including the following steps:

[0019] According to the distortion degree D of the quantization process, the minimum distortion quantization algorithm is derived, and the steps are as follows:

[0020] The probability density function of the input sequence of the quantizer is p(x). The minimum distortion quantization means that the quantization error ε=xy is the minimum. x is the input signal amplitude, y is the quantized output, and f(xy) is the error function. Then the distortion is

[0021] Taking the derivative of D with respect to x and y, we get the necessary condition for minimizing the distortion, namely

[0022] x j =(y j +y j-1 ) / 2or y j =2x j -y j-1 ,j=2,…,N

[0023]

[0024] x j Yes j and j-1 The midpoint between j is p(x) at x j and x j+1 The centroid of the area between;

[0025] Construct the optimization rule of transmission channel quantization level: according to different signal-to-noise ratios SNR and quantization levels L, use the minimum distortion quantization algorithm to iteratively calculate the quantization parameter x j and j , the steps are as follows:

[0026] Assume that the probability density function p(x) is an even function. First, find the quantization parameter on the right side of the X-axis origin. The quantization parameter on the left side of the X-axis origin is obtained symmetrically through the quantization parameter on the right side of the X-axis origin. Find the overall quantization parameter. First, let x1 be 0 and y1 be 0.01. At this time, y1 is an assumed value. If y is judged later, L If it is not a suitable centroid, then y1 must be reselected;

[0027] By formula Since the area of ​​p(x) between x1 and y1 is equal to the area of ​​p(x) between y1 and x2, x2 is calculated; then, by using the formula y j =2x j -y j-1 (1), thus we get y2;

[0028] The minimum distortion quantization algorithm is used to iteratively calculate each quantization parameter x to the right of the X-axis origin. j and j Then, we can use formula (2) to determine y L Is it x L and ∞, that is, calculate x L and L The area between p(x) and y L Is the difference s between the area of ​​p(x) and ∞ no more than 0.001? If s is no more than 0.001, then we assume that y L is x L The centroid of the area between and ∞;

[0029] If s exceeds 0.001, reselect y1. When s is greater than 0, subtract y1 from 0.001. When s is less than 0, add y1 to 0.001. Repeat the iteration until y is determined to be L is x L The centroid of the area between and ∞;

[0030] The obtained x j and j The value of is symmetric about the y-axis, and the overall quantization parameters x and y are obtained.

[0031] The receiving end quantizes according to the transmission channel quantization level and performs BP decoding, comprising the following steps:

[0032] Receive the polarization code sent by the transmitter;

[0033] Quantizing the polar code according to the transmission channel quantization level optimization rule;

[0034] Perform BP decoding:

[0035] The received N codewords are sent to the optimized quantizer, and the corresponding quantization level L is selected for the channel position of the N codewords; after setting the quantization level L, each codeword is compared with the elements in the quantization parameter x matrix; when they are in the same interval, the elements in the corresponding quantization parameter y matrix are output to complete the quantization of the channel output; after obtaining the quantization value, BP decoding is performed, and the decoding is completed when the set maximum number of iterations is reached or the early termination condition is met.

[0036] The quantizing process of the polar code according to the transmission channel quantization level optimization rule comprises the following steps:

[0037] Optimize the quantizer of the transmission channel: divide the N transmission channels into two parts, the first case is: one part is the information bit channel selected according to the screening channel, and the other part is the frozen bit channel; the second case is: one part is the information bit transposed channel, and the other part is the frozen bit transposed channel; the third case is: one part is the first N-(K+Q) channels equal to the number of frozen bit channels, and the other part is the last K+Q channels equal to the number of information bit channels; set corresponding quantization levels for the two parts of the transmission channel respectively;

[0038] According to the optimized quantizer, the codewords of different channels are compared with the quantization parameters x of different precisions, and the corresponding quantization output y is output to complete the quantization of the polarization code.

[0039] Beneficial Effects

[0040] The polar code transmission method based on transmission channel quantization level optimization of the present invention performs optimization of quantizers with different precisions under polar code BP decoding compared with the prior art, improves the performance of the polar code decoder in a system with different quantization precisions, and improves the transmission efficiency of the polar code.

[0041] The present invention uses a minimum distortion quantization algorithm and sets a probability density function p(x) as the sum of two normal distributions with average values ​​of positive and negative 1, so as to match the polar code coding situation; after the information sequence is encoded, BPSK modulated and passes through an AWGN channel, it is transmitted to an optimized quantizer to quantize different channels; after obtaining the quantized value, BP decoding is performed, and the decoding is completed when the set maximum number of iterations is reached or the early termination condition is met.

[0042] The simulation results of the present invention show that, using the method described in the present invention, in an AWGN channel, under the conditions of code length N=256 and code rate R=0.5, for a quantizer with a combination of quantization levels 2 and 16, the first N-(K+Q) channels equal to the number of frozen bit channels select a quantization level of 2, and the last K+Q channels equal to the number of information bit channels select a quantization level of 16, which has the best decoding performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a method sequence diagram of the present invention;

[0044] Figure 2 This is a system model diagram of the optimization method of different precision quantizers under polar code BP decoding according to the present invention;

[0045] Figure 3 The error performance comparison diagram of BP decoding after constructing a quantizer with overall quantization levels of 2, 4, 8, 16, and 32 for N codewords output by the channel in an AWGN channel;

[0046] Figure 4 To optimize the quantizer with quantization levels of 2 and 16 at different positions of the output channel in the AWGN channel, the error performance comparison chart of BP decoding is performed;

[0047] Figure 5 To optimize the quantizer with quantization levels of 2 and 32 at different positions of the output channel in the AWGN channel, the error performance comparison of BP decoding is performed. DETAILED DESCRIPTION

[0048] In order to have a further understanding and recognition of the structural features and the effects achieved by the present invention, a preferred embodiment and accompanying drawings are used for detailed description as follows:

[0049] The present invention first calculates the channel parameter Pe by using the Tal-Vardy algorithm, and then selects the information bit and the frozen bit according to the channel parameter; Q CRC check bits are added to the information bit, the information sequence is encoded and modulated by polarization code, and then sent through the AWGN channel; quantizers with different precisions are constructed, and the channel output is quantized with different precisions; finally, the BP decoding algorithm is used for decoding to obtain the error performance.

[0050] like Figure 1 As shown, the polar code transmission method based on transmission channel quantization level optimization according to the present invention comprises the following steps:

[0051] In the first step, the transmitter screens the channel and performs polarization code encoding and transmission. The transmitter calculates the channel parameter vector P e , according to the channel parameter vector P e The value of is used to select the information bit channel and the frozen bit channel; the information bit channels constitute the information bit channel set I, and the frozen bit channels constitute the frozen bit channel set F; Q CRC check bits are added to the information bits, the information sequence is encoded and modulated by polar code, and then sent through the AWGN channel. The specific steps are as follows:

[0052] (1) For the channel that transmits information, the channel parameter vector P is calculated using the Tal-Vardy construction method. e .

[0053] (2) Based on the channel parameter vector P e The value of Filter out the information bit channel and the frozen bit channel, the steps are as follows:

[0054] A1) The channel parameter vector P e The values ​​of are arranged in descending order to obtain the sorted channel parameter vector P temp , and calculate P temp Each element in P e The index value in gets the index vector P idx ;

[0055] A2) According to the index vector P idx The values ​​of the first K elements in are arranged in ascending order to obtain the information bit channel set I, where K is the number of information bits; the values ​​of the remaining NK elements are arranged in ascending order to obtain the frozen bit channel set F.

[0056] (3) Q CRC check codes are added to the end of the information bits, and then the information is transmitted, polar code encoding is completed, and the information is sent after BPSK modulation and AWGN channel.

[0057] In the second step, the receiving end constructs the transmission channel quantization parameters based on the minimum distortion quantization algorithm. In order to minimize the quantization error, the minimum distortion quantization algorithm is obtained by taking the derivative of the distortion degree D; the quantization level is set to L, generally L is an integer power of 2, and the transmission channel quantization level optimization rule is constructed based on the quantization level L and the minimum distortion quantization algorithm. The rule is determined by the endpoints x of the input range. j And output information y j Composition, x j is L-1, y j There are L; the optimization rule stipulates that if the input information is in x j and x j+1 interval, the output is the same value y j .

[0058] Because the equation about the distortion D is unsolvable without numerical values, it is impossible to calculate the minimization of quantization distortion. Therefore, in order to calculate the input information with a known normal distribution amplitude and expected square error, the probability density function p(x) is set to be the sum of two normal distributions with average values ​​of positive and negative 1 respectively, and the quantization parameters are obtained using an iterative algorithm to achieve the purpose of minimum distortion quantization.

[0059] The receiving end constructs the transmission channel quantization parameter based on the minimum distortion quantization algorithm, including the following steps:

[0060] (1) Based on the distortion degree D of the quantization process, the minimum distortion quantization algorithm is derived. The steps are as follows:

[0061] B1) The probability density function of the input sequence of the quantizer is p(x). The minimum distortion quantization means that the quantization error ε=xy is the minimum. x is the input signal amplitude, y is the quantized output, and f(xy) is the error function. Then the distortion is

[0062] B2) Taking the derivative of D with respect to x and y, we obtain the necessary condition for minimizing distortion, namely

[0063] x j =(y j +y j-1 ) / 2or y j =2x j -y j-1 ,j=2,…,N

[0064]

[0065] x j Yes j and j-1 The midpoint between j is p(x) at x j and x j+1 The centroid of the area between.

[0066] (2) Constructing the optimization rule of the transmission channel quantization level: According to different signal-to-noise ratios (SNR) and quantization levels (L), the minimum distortion quantization algorithm is used to iteratively calculate the quantization parameter x j and j , the steps are as follows:

[0067] C1) Assume that the probability density function p(x) is an even function. First, find the quantization parameter on the right side of the X-axis origin. The quantization parameter on the left side of the X-axis origin is obtained symmetrically through the quantization parameter on the right side of the X-axis origin. Find the overall quantization parameter. First, let x1 be 0 and y1 be 0.01. At this time, y1 is an assumed value. If y is judged later, L If it is not a suitable centroid, then y1 must be reselected;

[0068] C2) Through the formula Since the area of ​​p(x) between x1 and y1 is equal to the area of ​​p(x) between y1 and x2, x2 is calculated; then, by using the formula y j =2x j -y j-1 (1), thus we get y2;

[0069] C3) Use the minimum distortion quantization algorithm to iteratively calculate each quantization parameter x to the right of the X-axis origin j and j Then, we can use formula (2) to determine y L Is it x L and ∞, that is, calculate x L and L The area between p(x) and y L Is the difference s between the area of ​​p(x) and ∞ no more than 0.001? If s is no more than 0.001, then we assume that y L is x L The centroid of the area between and ∞;

[0070] C4) If s exceeds 0.001, reselect y1. When s is greater than 0, subtract y1 from 0.001; when s is less than 0, add y1 to 0.001 and iterate again until y is determined to be L is x L The centroid of the area between and ∞;

[0071] C5) The obtained x j and j The value of is symmetric about the y-axis, and the overall quantization parameters x and y are obtained.

[0072] The third step is that the receiving end quantizes the polar code according to the transmission channel quantization level and performs BP decoding: the receiving end receives the polar code sent by the transmitting end, quantizes it according to the transmission channel quantization level, and then performs BP decoding. Figure 2 As shown, here, the channel output value is output through the AWGN channel and sent to the optimized quantizers with different precisions, and the output values ​​of different channels are quantized with different precisions; the obtained quantized output value is BP decoded, and the decoding is completed when the set maximum number of iterations is reached or the early termination condition is met.

[0073] Due to the different precision requirements of ADC in actual applications, it is difficult to meet the requirements of transmission quality whether it is high-precision or low-precision overall quantization of channel output. Therefore, based on the transmission channel quantization level optimization rule, different channels are quantized at different quantization levels, which improves the performance of the polar code decoder without affecting power consumption and causing nonlinear distortion.

[0074] The specific steps are as follows:

[0075] (1) Receive the polarization code sent by the transmitter.

[0076] (2) The polar code is quantized according to the transmission channel quantization level optimization rule.

[0077] D1) Optimizing the quantizer of the transmission channel: N transmission channels are divided into two parts. In the first case, one part is the information bit channel selected according to the screening channel, and the other part is the frozen bit channel; in the second case, one part is the information bit transposed channel, and the other part is the frozen bit transposed channel; in the third case, one part is the first N-(K+Q) channels equal to the number of frozen bit channels, and the other part is the last K+Q channels equal to the number of information bit channels; corresponding quantization levels are set for the two parts of the transmission channel respectively;

[0078] D2) According to the optimized quantizer, the code words of different channels are compared with the quantization parameters x of different precisions, and the corresponding quantization output y is output to complete the quantization of the polar code.

[0079] (3) Perform BP decoding:

[0080] The received N codewords are sent to the optimized quantizer, and the corresponding quantization level L is selected for the channel position of the N codewords; after setting the quantization level L, each codeword is compared with the elements in the quantization parameter x matrix; when they are in the same interval, the elements in the corresponding quantization parameter y matrix are output to complete the quantization of the channel output; after obtaining the quantization value, BP decoding is performed, and the decoding is completed when the set maximum number of iterations is reached or the early termination condition is met.

[0081] like Figure 3 As shown, Figure 3 The error performance comparison chart of BP decoding is shown in the figure after constructing a quantizer with overall quantization levels of 2, 4, 8, 16, and 32 for the N codewords output by the channel in the AWGN channel. Figure 3 It can be seen from the figure that as the quantization level increases, the bit error performance of the quantization algorithm of the present invention gradually approaches the bit error performance of the unquantized one, and the higher the quantization level, the smaller the quantization error.

[0082] Figure 4 To optimize the quantizer with quantization levels of 2 and 16 at different positions of the output channel in the AWGN channel, the error performance comparison of BP decoding is performed. Figure 4 It can be seen that the best bit error performance is achieved when the quantization level is 2 for the first N-(K+Q) channels of the output channel and 16 for the last K+Q channels. Similarly, Figure 5 is a combination of quantization levels 2 and 32, from Figure 5 It can be seen that the best bit error performance is achieved when the quantization level is 2 for the first N-(K+Q) channels of the output channel and the quantization level is 32 for the last K+Q channels.

[0083] The present invention solves the problem that different antennas will have different accuracies when passing through different ADC links, proposes to optimize the corresponding relationship between the quantization level and the received symbol in the codeword, improves the performance of the decoder in a system with different quantization accuracies, optimizes quantizers with different accuracies and performs BP decoding. The present invention improves the performance of the polarization code decoder in situations where different quantization accuracies exist.

[0084] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A polar code transmission method based on transmission channel quantization level optimization, characterized in that: The following steps are involved: 11) The transmitter selects the channel and performs polarization code encoding transmission: The transmitter calculates the channel parameter vector P e , according to the channel parameter vector P e The value of is used to select information bit channels and frozen bit channels; the information bit channels constitute an information bit channel set I, and the frozen bit channels constitute a frozen bit channel set F; Q CRC check bits are added to the information bits, the information sequence is encoded and modulated by polarization codes, and then sent through the AWGN channel; the transmitting end selects channels and performs polarization code encoding and transmission, including the following steps: 111) For the channel that transmits information, the channel parameter vector P is calculated using the Tal-Vardy construction method e ; 112) According to the channel parameter vector P e The value of Filter out the information bit channel and the frozen bit channel, the steps are as follows: 1121) The channel parameter vector P e The values ​​of are arranged in descending order to obtain the sorted channel parameter vector P temp , and calculate P temp Each element in P e The index value in gets the index vector P idx ; 1122) According to the index vector P idx The values ​​of the first K elements in are arranged in ascending order to obtain the information bit channel set I, where K is the number of information bits; the values ​​of the remaining NK elements are arranged in ascending order to obtain the frozen bit channel set F; 113) Q CRC check codes are added to the end of the information bit, and then the information is transmitted, polar code encoding is completed, and the information is sent after BPSK modulation and AWGN channel; 12) The receiving end constructs the transmission channel quantization parameters based on the minimum distortion quantization algorithm: In order to minimize the quantization error, the distortion degree D is derived to obtain the minimum distortion quantization algorithm; the quantization level is set to L, generally L is an integer power of 2, and the transmission channel quantization level optimization rule is constructed based on the quantization level L and the minimum distortion quantization algorithm. The rule is determined by the endpoints x of the input range. j And output information y j Composition, x j is L-1, y j There are L; the optimization rule stipulates that if the input information is in x j and x j+1 interval, the output is the same value y j ; The receiving end constructs the transmission channel quantization parameter based on the minimum distortion quantization algorithm, including the following steps: 121) According to the distortion degree D of the quantization process, the minimum distortion quantization algorithm is derived, and the steps are as follows: 1211) The probability density function of the input sequence of the quantizer is p(x), the minimum distortion quantization means the quantization error ε=xy is the minimum, x is the input signal amplitude, y is the quantized output, f(xy) is the error function, then the distortion is 312) Taking the derivative of D with respect to x and y, we can obtain the necessary condition for minimizing the distortion, namely xj=(yj+yj-1) / 2hour y j =2x j -y j-1 ,j=2,L,N x j Yes j and j-1 The midpoint between j is p(x) at x j and x j+1 The centroid of the area between; 122) Construct the optimization rule of transmission channel quantization level: according to different signal-to-noise ratios SNR and quantization levels L, use the minimum distortion quantization algorithm to iteratively calculate the quantization parameter x j and j , the steps are as follows: 1221) Assume that the probability density function p(x) is an even function. First, find the quantization parameter on the right side of the X-axis origin. The quantization parameter on the left side of the X-axis origin is obtained symmetrically through the quantization parameter on the right side of the X-axis origin. Find the overall quantization parameter. First, let x1 be 0 and y1 be 0.

01. At this time, y1 is an assumed value. If y is judged later, L If it is not a suitable centroid, then y1 must be reselected; 1222) through the formula Since the area of ​​p(x) between x1 and y1 is equal to the area of ​​p(x) between y1 and x2, x2 is calculated; then, by using the formula y j =2x j -y j-1 (1), thus we get y2; 1223) Use the minimum distortion quantization algorithm to iteratively calculate each quantization parameter x to the right of the X-axis origin j and j Then, we can use formula (2) to determine y L Is it x L and ∞, that is, calculate x L and L The area between p(x) and y L Is the difference s between the area of ​​p(x) and ∞ no more than 0.001? If s is no more than 0.001, then we assume that y L is x L The centroid of the area between and ∞; 1224) If s exceeds 0.001, reselect y1. When s is greater than 0, subtract y1 from 0.001; when s is less than 0, add y1 to 0.001, and iterate again until y is determined. L is x L The centroid of the area between and ∞; 1225) The obtained x j and j The value of is symmetric about the y-axis, and the overall quantization parameters x and y are obtained; 13) The receiving end quantizes according to the quantization level of the transmission channel and performs BP decoding: The receiving end receives the polar code sent by the transmitting end, quantizes according to the quantization level of the transmission channel, and then performs BP decoding.

2. The polar code transmission method based on transmission channel quantization level optimization according to claim 1, characterized in that: The receiving end quantizes according to the transmission channel quantization level and performs BP decoding, comprising the following steps: 21) receiving the polarization code sent by the transmitter; 22) quantizing the polar code according to the transmission channel quantization level optimization rule; 23) Perform BP decoding: The received N codewords are sent to the optimized quantizer, and the corresponding quantization level L is selected for the channel position of the N codewords; after setting the quantization level L, each codeword is compared with the elements in the quantization parameter x matrix; when they are in the same interval, the elements in the corresponding quantization parameter y matrix are output to complete the quantization of the channel output; after obtaining the quantization value, BP decoding is performed, and the decoding is completed when the set maximum number of iterations is reached or the early termination condition is met.

3. The polar code transmission method based on transmission channel quantization level optimization according to claim 2, characterized in that: The quantizing process of the polar code according to the transmission channel quantization level optimization rule comprises the following steps: 31) Optimizing the quantizer of the transmission channel: N transmission channels are divided into two parts. In the first case, one part is the information bit channel selected according to the screening channel, and the other part is the frozen bit channel; in the second case, one part is the information bit transposed channel, and the other part is the frozen bit transposed channel; in the third case, one part is the first N-(K+Q) channels equal to the number of frozen bit channels, and the other part is the last K+Q channels equal to the number of information bit channels; corresponding quantization levels are set for the two parts of the transmission channel respectively; 32) According to the optimized quantizer, the code words of different channels are compared with the quantization parameters x of different precisions, and the corresponding quantization output y is output to complete the quantization of the polarization code.

Images