CN115987303A - Generation polynomial acquisition method and system based on differential evolution algorithm - Google Patents
Generation polynomial acquisition method and system based on differential evolution algorithm Download PDFInfo
- Publication number
- CN115987303A CN115987303A CN202211637119.0A CN202211637119A CN115987303A CN 115987303 A CN115987303 A CN 115987303A CN 202211637119 A CN202211637119 A CN 202211637119A CN 115987303 A CN115987303 A CN 115987303A
- Authority
- CN
- China
- Prior art keywords
- generator polynomial
- value
- feasible
- code
- false
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Error Detection And Correction (AREA)
Abstract
The invention discloses a generating polynomial obtaining method and a generating polynomial obtaining system based on a differential evolution algorithm, wherein the method comprises the following steps: generating a plurality of initial information bits to construct an initial information bit set; setting initial parameters of a differential evolution algorithm, wherein the initial parameters comprise population quantity, differential evolution factors, crossing factors and termination conditions; constructing a feasible first generator polynomial set and a feasible second generator polynomial set according to the population number, wherein the feasible first generator polynomial set comprises a plurality of feasible first generator polynomials, and the feasible second generator polynomial set comprises a plurality of feasible second generator polynomials; constructing a fitness function of the generator polynomial according to the initial information bit set, the feasible first generator polynomial set and the feasible second generator polynomial set; and adopting a differential evolution algorithm to carry out optimization solution on the fitness function so as to obtain the optimal first generator polynomial set and the optimal second generator polynomial set.
Description
Technical Field
The invention relates to the technical field of rail transit, in particular to a generating polynomial acquiring method and system based on a differential evolution algorithm.
Background
In a rail transit signal system, a computer interlocking system is one of core signal systems for ensuring train running safety, and is used for establishing and controlling an interlocking relation among a station signal machine, a point switch, a rail circuit and other equipment so as to prevent various potential hazards of trains in railway transportation operation, such as train derailment, train workshop collision and the like. In view of the safety-critical properties of computer interlocking systems, a high degree of reliability of their interlocking functions, in particular safety-related functions, is required; meanwhile, the design requirements on computer interlocking system safety software in the standard TB/T3027-2015 technical computer interlocking condition for railway stations are as follows: "the Hamming code distance of the information code of the variables related to the driving safety and the different values of the same variable should not be less than 4", the purpose of the requirement is that when the computer interlocking system is in random failure, the values of the variables will not jump to other allowable states to prevent the wrong output, thereby avoiding the driving accident.
According to the above requirements, information codes with a code distance of less than 4 should be removed from all available code groups, which allows the number of available code groups used to represent variables in the interlocking system to be compressed. However, as the size of the interlocking station is larger and larger now, more code blocks are required to represent different devices in the interlocking system data, which leads to a contradiction between the number of compressed available code blocks and the increasing demand of available code blocks for interlocking stations. In addition, currently, available code groups configured for different interlocking stations are often generated by using the same set of commonly used generator polynomials, so that certain misuse risks may exist in interlocking system data between different stations. Generally, a selected generator polynomial is utilized to carry out polynomial division on values in a coding space to obtain a coding code group, and then code words with code distances not meeting requirements in the coding space are removed according to requirements, and finally an available code group is obtained; it can be seen that both the number and value of available codes are directly related to the selected generator polynomial. Therefore, how to determine the appropriate generator polynomial has become the key to solve the above contradiction and reduce the above misuse risk.
Disclosure of Invention
The invention aims to provide a generator polynomial acquisition method and a generator polynomial acquisition system based on a differential evolution algorithm, which can effectively expand a selection pool of generator polynomials, realize the configuration of different station differentiated code blocks, and simultaneously enable the pairwise code distance of the configured code blocks to be as large as possible.
In order to achieve the purpose, the invention is realized by the following technical scheme:
a generator polynomial obtaining method based on differential evolution algorithm is used for double-channel information coding, each channel outputs binary code words consisting of information bits and check bits, and the check bits are obtained through the information bits and generator polynomials corresponding to the channels, and the method comprises the following steps:
generating a plurality of initial information bits to construct an initial information bit set;
setting initial parameters of a differential evolution algorithm, wherein the initial parameters comprise population quantity, differential evolution factors, crossing factors and termination conditions;
constructing a feasible first generator polynomial set and a feasible second generator polynomial set according to the population number, wherein the feasible first generator polynomial set comprises a feasible plurality of first generator polynomials, and the feasible second generator polynomial set comprises a feasible plurality of second generator polynomials;
constructing a fitness function of a generator polynomial according to the initial information bit set, the feasible first generator polynomial set and the feasible second generator polynomial set; and
and carrying out optimization solution on the fitness function by adopting the differential evolution algorithm so as to obtain the optimal first generator polynomial set and the optimal second generator polynomial set.
Optionally, the initial information bit is a binary number of K bits, and the first generator polynomial and the second generator polynomial are both binary numbers of R +1 bits; wherein R and K are both positive integers, and the last bit of the first generator polynomial and the second generator polynomial are both 1.
Optionally, the step of generating a plurality of initial information bits to construct an initial information bit set includes:
sequentially evaluating K binary numbers of the initial information bits in a first value-evaluating interval to obtain 2 K -2 of said initial information bits; and the first value interval is (000 …, 111 …), wherein the number of 0 in the left end point 000 … of the first value interval is K, and the number of 1 in the right end point 111 … is K.
Optionally, the step of constructing a feasible first generator polynomial set and a feasible second generator polynomial set according to the population number includes:
respectively carrying out random value taking on front R-bit binary numbers of the first generator polynomial and the second generator polynomial in a second value-taking interval according to the population quantity to obtain a plurality of feasible first generator polynomials and a plurality of feasible second generator polynomials; and the second value interval is (000 …, 111 …), wherein the number of 0 in the left end point 000 … of the second value interval is R, and the number of 1 in the right end point 111 … is R.
Optionally, the step of constructing a fitness function of a generator polynomial according to the initial information bit set, the feasible first generator polynomial set, and the feasible second generator polynomial set includes:
selecting an initial information bit with the last bit value of 1 from the initial information bit set as a true value information bit to obtain a true value information bit set;
taking a reverse code for each true value information bit in the true value information bit set to obtain a corresponding false value information bit;
performing modulo-2 division on each of the true information bits and the false information bits by using any feasible first generator polynomial to obtain a first parity bit corresponding to each of the true information bits and a first false check bit corresponding to each of the false information bits;
performing modulo-2 division on each of the true information bits and the false information bits by using any feasible second generator polynomial to obtain a second true check bit corresponding to each of the true information bits and a second false check bit corresponding to each of the false information bits;
splicing each truth value information bit with the corresponding first truth value check bit to obtain a first truth value binary code word; splicing each false value information bit and the corresponding first false value check bit to obtain a first false value binary code word;
splicing each true value information bit with the corresponding second true value check bit to obtain a second true value binary code word; splicing each false value information bit and the corresponding second false value check bit to obtain a second false value binary code word;
forming a pair of truth code groups from the first and second true binary codewords having the same true information bits and a pair of false code groups from the first and second false binary codewords having the same false information bits;
calculating code distances of the first true value binary codeword and the second true value binary codeword in each of the true value code groups, and calculating code distances of the first false value binary codeword and the second false value binary codeword in each of the false value code groups;
and acquiring a corrected total code distance corresponding to the feasible first generator polynomial and the feasible second generator polynomial according to the code distances of all the true value code groups and the code distances of all the false value code groups, wherein the corrected total code distance is used as a fitness index of the fitness function.
Optionally, the step of obtaining a corrected total code distance corresponding to the feasible first generator polynomial and the feasible second generator polynomial according to the code distances of all the code groups includes:
summing the code distances of all the true value code groups and all the false value code groups to obtain an uncorrected total code distance;
calculating a penalty value according to the minimum value of the preset code distance; and
and correcting the uncorrected total code distance according to the punishment value to obtain the corrected total code distance.
Optionally, the penalty value is calculated by using the following formula:
P＝a 0 ×10 n+1 +a 1 ×10 n +a 2 ×10 n-1 +a 3 ×10 n-2 +…+a n ×10 1
wherein P represents a penalty value; n represents the minimum value of the preset code distance; a is 0 、a 1 、a 2 、a 3 、…、a n Respectively representing the times of code distances of 0, 1, 2, 3, …, n in all the true value code groups and all the false value code groups.
Optionally, the corrected total code distance is calculated by using the following formula:
D Z ＝D C -P
wherein D is Z Representing the corrected total code distance; d C Representing the uncorrected total code distance.
Optionally, the step of performing an optimization solution on the fitness function by using the differential evolution algorithm to obtain the optimal first generator polynomial set and the optimal second generator polynomial set includes:
forming an initial population by using a feasible plurality of first generator polynomials and a feasible plurality of second generator polynomials, wherein any one of the first generator polynomials and any one of the second generator polynomials form an individual;
calculating a corrected total code distance corresponding to each individual of the initial population;
performing variation and cross operation on each individual of the initial population according to the differential evolution factor and the cross factor to obtain a progeny population consisting of new individuals;
calculating the corrected total code distance corresponding to each individual of the filial generation population;
comparing the corrected total code distances of any two individuals in the initial population and the offspring population, and reserving the individual with the larger corrected total code distance to enter the next offspring population until the termination condition is reached; and
and taking the individuals in the corresponding population when the termination condition is reached as an optimal first generator polynomial and an optimal second generator polynomial to obtain an optimal first generator polynomial set and an optimal second generator polynomial set.
Optionally, the termination condition is a maximum number of iterations.
Optionally, the number of the first generator polynomial and the second generator polynomial in each population is the same as the number of the populations.
Optionally, the set of true information bits includes 2 K-1 -1 true information bits; and the number of the false value information bits is the same as the number of the true value information bits.
On the other hand, the invention also provides a generating polynomial acquiring system based on the differential evolution algorithm, which adopts the generating polynomial acquiring method based on the differential evolution algorithm to acquire the optimal first generating polynomial set and the optimal second generating polynomial set.
Compared with the prior art, the invention has at least one of the following advantages:
according to the generating polynomial acquiring method and system based on the differential evolution algorithm, an initial information bit set can be constructed by generating initial information bits, and a feasible first generating polynomial set and a feasible second generating polynomial set can be constructed according to set initial parameters of the differential evolution algorithm; the fitness function of the generator polynomial can be constructed according to the initial information bit set, the feasible first generator polynomial set and the feasible second generator polynomial set, the fitness function is optimized and solved by adopting a differential evolution algorithm, the optimal first generator polynomial set and the optimal second generator polynomial set can be obtained, so that the selection pool of the generator polynomial is effectively expanded, the differentiated code block configuration of different stations is realized, the misuse of data is fundamentally avoided, and the pairwise code distance of the configured code blocks is as large as possible.
Compared with the traversing method, the code distance calculation amount or calculation cost of the corresponding code group can be greatly reduced by only comparing the first true value binary code word and the second true value binary code word with the same true value information bit and the first false value binary code word and the second false value binary code word with the same false value information bit, and the time required by calculating the optimal solution can be reduced in geometric multiples by adopting a differential evolution algorithm for optimal solution, so that the method has operability.
The invention can improve the number of available code groups in the coding space and the code distance between the code groups so as to adapt to the variable representation requirement of the interlocking station with higher and higher complexity and improve the safety.
The initial parameters of the differential evolution algorithm can be adjusted, and the initial parameter parameters can be adjusted to meet different requirements on code distances in different use scenes.
Compared with the fixed generator polynomial method widely used by the current system, the method has the advantages that the first generator polynomial and the second generator polynomial are optimized and solved by using the differential evolution algorithm, the selection pool of the generator polynomial can be effectively expanded, different generator polynomials are selected for different interlocking stations to configure the code blocks, and therefore misuse of configuration data is avoided.
Drawings
Fig. 1 is a flowchart of a generator polynomial obtaining method based on a differential evolution algorithm according to an embodiment of the present invention;
fig. 2 is a flowchart for calculating a corrected total code distance in a generator polynomial obtaining method based on a differential evolution algorithm according to an embodiment of the present invention;
fig. 3 is a flowchart of a differential evolution algorithm in a generator polynomial obtaining method based on the differential evolution algorithm according to an embodiment of the present invention.
Detailed Description
The following describes the generator polynomial obtaining method and system based on the differential evolution algorithm in detail with reference to the accompanying drawings and the detailed description. The advantages and features of the present invention will become more apparent from the following description. It is to be noted that the drawings are in a very simplified form and are all used in a non-precise scale for the purpose of facilitating and distinctly aiding in the description of the embodiments of the present invention. To make the objects, features and advantages of the present invention comprehensible, reference is made to the accompanying drawings. It should be understood that the structures, ratios, sizes, and the like shown in the drawings and described in the specification are only used for matching with the disclosure of the specification, so as to be understood and read by those skilled in the art, and are not used to limit the implementation conditions of the present invention, so that the present invention has no technical significance, and any structural modification, ratio relationship change or size adjustment should still fall within the scope of the present invention without affecting the efficacy and the achievable purpose of the present invention.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
With reference to fig. 1 to 3, this embodiment provides a generator polynomial obtaining method based on a differential evolution algorithm, which is used for two-channel information encoding, where each channel outputs a binary codeword composed of an information bit and a check bit, and the check bit is obtained through the information bit and a generator polynomial corresponding to the channel; the generator polynomial obtaining method comprises the following steps: step S110, generating a plurality of initial information bits to construct an initial information bit set; step S120, setting initial parameters of a differential evolution algorithm, wherein the initial parameters comprise population quantity, differential evolution factors, cross factors and termination conditions; step S130, constructing a feasible first generator polynomial set and a feasible second generator polynomial set according to the population quantity, wherein the feasible first generator polynomial set comprises a plurality of feasible first generator polynomials, and the feasible second generator polynomial set comprises a plurality of feasible second generator polynomials; step S140, constructing a fitness function of a generator polynomial according to the initial information bit set, the feasible first generator polynomial set and the feasible second generator polynomial set; and step S150, optimizing and solving the fitness function by adopting the differential evolution algorithm to obtain the optimal first generator polynomial set and the optimal second generator polynomial set.
Referring to fig. 1, the initial information bits are K bits, and the first generator polynomial and the second generator polynomial are R +1 bits; wherein R and K are both positive integers, the binary number is a number consisting of 0 and 1, and the last bit of the first generator polynomial and the second generator polynomial are both 1.
It is understood that the step S110 includes: sequentially evaluating K binary numbers of the initial information bits in a first value-evaluating interval to obtain 2 K -2 bits of said initial information; and the first value interval is (000 …, 111 …), wherein the number of 0's in the left end point 000 … of the first value interval is K, and the number of 1's in the right end point 111 … of the first value interval is K。
The step S130 includes: respectively carrying out random value taking on the front R-bit binary numbers of the first generator polynomial and the second generator polynomial in a second value-taking interval according to the population quantity to obtain a plurality of feasible first generator polynomials and a plurality of feasible second generator polynomials; and the second value interval is (000 …, 111 …), wherein the number of 0 in the left end point 000 … of the second value interval is R, and the number of 1 in the right end point 111 … of the second value interval is R.
Specifically, in this embodiment, the two-channel information encoding is a NISAL encoding technique, where the two channels may be denoted as a first channel and a second channel, the first generator polynomial is used for encoding information of the first channel, and the second generator polynomial is used for encoding information of the second channel. More specifically, the initial information bit may be a binary number of 16 bits, and the first generator polynomial and the second generator polynomial may be binary numbers of 17 bits, that is, both the values of K and R are 16, and the first value range and the second value range are (0000000000000000, 1111111111111111111), which may be represented by (0x10000, 0x1ffff) in hexadecimal notation. And then sequentially dereferencing the 16-bit binary number of the initial information bit in the first dereferencing interval to obtain the initial information bit set, where the initial information bit set includes 65534 initial information bits, and as can be seen from the first dereferencing interval, 0x0000 and 0xFFFF are not used as the initial information bits. After the first 16-bit binary numbers of the first generator polynomial and the second generator polynomial are subjected to random value taking in the second value taking interval, a feasible first generator polynomial set and a feasible second generator polynomial set can be obtained, and the number of the first generator polynomial in the feasible first generator polynomial set and the number of the second generator polynomial in the feasible second generator polynomial set are both the same as the population number. Preferably, any feasible first generator polynomial and any feasible second generator polynomial are not the same, but the invention is not limited thereto.
Referring to fig. 1 and fig. 2, the step S140 includes: step S1401, selecting an initial information bit with a last value of 1 from the initial information bit set as a true value information bit to obtain a true value information bit set; step S1402, taking an inverse code for each true value information bit in the true value information bit set to obtain a corresponding false value information bit, and obtaining a false value information bit set; step S1403, performing modulo-2 division on each of the true information bits and the false information bits by using any feasible first generator polynomial to obtain a first true check bit corresponding to each of the true information bits and a first false check bit corresponding to each of the false information bits; step S1404, performing modulo-2 division on each of the true information bits and the false information bits by using any feasible second generator polynomial to obtain a second true check bit corresponding to each of the true information bits and a second false check bit corresponding to each of the false information bits; step S1405, concatenating each true value information bit with the corresponding first true value check bit to obtain a first true value binary codeword; splicing each false value information bit and the corresponding first false value check bit to obtain a first false value binary code word; step S1406, concatenating each true value information bit with the corresponding second true value check bit to obtain a second true value binary codeword; splicing each false value information bit and the corresponding second false value check bit to obtain a second false value binary code word; step S1407, forming a pair of truth value code groups by the first truth value binary code word and the second truth value binary code word having the same truth value information bits, and forming a pair of false value code groups by the first false value binary code word and the second false value binary code word having the same false value information bits; step S1408, calculating code distances of the first true value binary codeword and the second true value binary codeword in each of the true value code groups, and calculating code distances of the first false value binary codeword and the second false value binary codeword in each of the false value code groups; and step S1409, obtaining a corrected total code distance corresponding to the feasible first generator polynomial and the feasible second generator polynomial according to the code distances of all the true value code groups and the code distances of all the false value code groups, where the corrected total code distance is used as the fitness index of the fitness function.
It is understood that the step S1409 includes: summing the code distances of all the true value code groups and all the false value code groups to obtain an uncorrected total code distance; calculating a penalty value according to the minimum value of the preset code distance; and correcting the uncorrected total code distance according to the punishment value to obtain the corrected total code distance.
The penalty value is calculated using the following formula:
P＝a 0 ×10 n+1 +a 1 ×10 n +a 2 ×10 n-1 +a 3 ×10 n-2 +…+a n ×10 1 (1)
wherein P represents a penalty value; n represents the minimum value of the preset code distance, and the default value of n can be 4 according to the interlocking technical condition; a is a 0 、a 1 、a 2 、a 3 、…、a n Respectively representing the number of code distances of 0, 1, 2, 3, …, n in all the true value code groups and all the false value code groups.
The corrected total code distance is calculated by adopting the following formula:
D Z ＝D C -P(2)
wherein D is Z Representing the corrected total code distance; d C Representing the uncorrected total code distance.
Specifically, in this embodiment, in the step S1401, for each of the initial information bits, a value of a last code bit, i.e. a rightmost code bit, of the initial information bit may define that the initial information bit represents a true value or a false value, and the true value and the false value are in an inverse code relationship with each other. More specifically, if the last bit of the initial information bit has a value of 1, the initial information bit is the true information bit. In step S1402, the true value information bit pair can be obtained by taking the inverse code of each code bit in the true value information bitThe corresponding false value information bits. Therefore, each of the true value information bits and each of the false value information bits are also binary numbers of K bits; the number of the false information bits is the same as the number of the true information bits, and the number of the true information bits is half of the number of the initial information bits, i.e., the set of true information bits includes 2 K-1 A true value information bit, said set of false value information bits comprising 2 K-1 -1 false value information bit, but the invention is not limited thereto.
Specifically, in the present embodiment, since the feasible first generator polynomial and the feasible second generator polynomial are both R + 1-bit binary numbers, the first true value check bit, the first false value check bit, the second true value check bit and the second false value check bit obtained in the steps S1403 and S1404 are all R-bit binary numbers, and the first true value binary code word, the first false value binary code word, the second true value binary code word and the second false value binary code word obtained in the steps S1405 and S1406 are all R + K-bit binary numbers. More specifically, when both R and K take on values of 16, the first true value binary codeword is a 32-bit binary codeword consisting of 16 true value information bits and 16 first true value check bits, and the structure is shown in table 1; similarly, the second true binary codeword is a 32-bit binary codeword consisting of 16 bits of the true information bits and 16 bits of the second true check bits, the first false binary codeword is a 32-bit binary codeword consisting of 16 bits of the false information bits and 16 bits of the first false check bits, the second false binary codeword is a 32-bit binary codeword consisting of 16 bits of the false information bits and 16 bits of the second false check bits, and the first true binary codeword, the first false information bit, the second true binary codeword and the second false binary codeword can also be represented by hexadecimal, but the invention is not limited thereto.
TABLE 1 first truth value binary code word structure table
Specifically, in this embodiment, in the step S1407, the first true value binary codeword and the second true value binary codeword having the same true value information bits may be defined as a pair of the true value code groups; similarly, the first and second dummy binary codewords having the same dummy information bits can be defined as a pair of the dummy code groups, and when the value of K is 16, the number of the true information bits and the number of the dummy information bits are 32767, that is, 2 K-1 -1, when the number of the first true binary codeword, the second true binary codeword, the first false binary codeword and the second false binary codeword are all 32767, the number of the true code group and the false code group are also 32767 pairs, i.e. 2 K-1 -1 pairs and the detailed data table of the truth code group and the false value code group is shown in table 2, wherein each pair of the first truth binary codeword and the second truth binary codeword in the truth code group and each pair of the first false value binary codeword and the second false value binary codeword in the false value code group adopt hexadecimal representation.
TABLE 2 data table of true code blocks and false code blocks
Specifically, in this embodiment, in step S1408, performing two-by-two xor on the first true value binary codeword and the second true value binary codeword in each pair of the true value code groups, shifting the obtained calculation result to the right bit by bit according to the length of the result codeword, and adding 1 to the result codeword to obtain the code distance of each pair of the true value code groups, that is, the number of different codes at the corresponding position between the first true value binary codeword and the second true value binary codeword in each pair of the true value code groups; similarly, the code distance of each pair of the dummy value code groups can be obtained by comparing the first dummy value binary code word and the second dummy value binary code word in each pair of the dummy value code groups, but the invention is not limited thereto.
Specifically, in this embodiment, in consideration of the compliance requirement and the related specification requirement of the technical condition, a penalty function may be introduced in the step S1409, that is, the penalty value is calculated according to the preset minimum code distance value, so as to correct the calculated uncorrected total code distance, and further obtain the corrected total code distance. Preferably, the preset code distance minimum value may be 4, that is, n =4; a can be obtained by checking the number of code distances 0, 1, 2, 3 and 4 in all the true value code groups and all the false value code groups 0 、a 1 、a 2 、a 3 And a 4 The penalty value is substituted into the formula (1), and then the penalty value is substituted into the formula (2), so that the corrected total code distance can be obtained, wherein the corrected total code distance is used as a fitness index of the fitness function, and the larger the corrected total code distance is, the better the corrected total code distance is. More specifically, the fitness function may be expressed as: d Z = max f (first generator polynomial, second generator polynomial) where f () is fitness function, D Z The indication fitness index is the corrected total code distance, but the invention is not limited thereto.
Referring to fig. 1 and fig. 3, the step S150 includes: step S1501, forming an initial population by using a feasible plurality of first generator polynomials and a feasible plurality of second generator polynomials, wherein any one of the first generator polynomials and any one of the second generator polynomials form an individual; step S1502, calculating a corrected total code distance corresponding to each individual of the initial population; step S1503, performing variation and cross operation on each individual of the initial population according to the differential evolution factor and the cross factor to obtain a progeny population consisting of new individuals; step S1504, calculating a corrected total code distance corresponding to each individual of the offspring population; step S1505, comparing the corrected total code distance of any two individuals in the initial population and the offspring population, and reserving the individual with larger corrected total code distance to enter the next offspring population until the termination condition is reached; and step S1506, using the first generator polynomial and the second generator polynomial, which are the individuals in the population corresponding to the termination condition, as the optimal first generator polynomial and second generator polynomial, so as to obtain the optimal first generator polynomial set and the optimal second generator polynomial set.
Specifically, in this embodiment, the differential evolution algorithm is a random search algorithm based on population data iteration, and has a simple structure and a fast convergence rate, and the differential evolution algorithm is adopted to solve the optimization problem, so that the solution efficiency can be effectively improved. More specifically, in step S1501, the first generator polynomial and the second generator polynomial in the feasible first generator polynomial set and the feasible second generator polynomial set constructed according to the population quantity may constitute the initial population. In the step S1502, the corrected total code distance of each individual in the initial population is calculated according to the steps S1402 to S1408, but the invention is not limited thereto.
Specifically, in this embodiment, in the step S1503, the offspring population may be generated through a mutation mechanism and a crossover mechanism, and the mutation formula is:
r’＝r 1 +F*(r 2 -r 3 )(3)
wherein r' is an individual in the progeny population, r 1 、r 2 、r 3 Are different individuals in the randomly selected initial population, and F is the differential evolution factor, which is generally a constant. The cross factor then represents r 1 、r 2 、r 3 Probability of being selected into a progeny population, but the invention is not so limited.
Specifically, in this embodiment, in step S1504, the corrected total code distance of each individual in the child population is also calculated according to steps S1402 to S1408. In step S1505, comparing the corrected total code distance between any two individuals in the initial population and the offspring population, and keeping the individual with the larger corrected total code distance, so as to enter the next offspring population; meanwhile, whether the termination condition is met or not needs to be judged, if the termination condition is not met, the steps S1503-S1504 are repeated to carry out iterative computation; if the termination condition is reached, stopping the calculation and outputting an optimization result, that is, outputting the individuals (that is, the optimal first generator polynomial and the optimal second generator polynomial) in the population corresponding to the termination condition, and the optimal first generator polynomials and the optimal second generator polynomials can correspond to the optimal first generator polynomial set and the optimal second generator polynomial set, thereby effectively expanding the selection pool of generator polynomials. Optionally, the termination condition is a maximum number of iterations; the number of the first generator polynomial and the second generator polynomial in each population is the same as the number of the populations; preferably, the number of the populations is 300. In addition, it should be noted that the optimization result output in this embodiment is directly related to the selection of the preset parameter, and if a larger population size and a larger number of iterations are selected, the performance of the optimized first generator polynomial and the optimized second generator polynomial may be further improved, but the present invention is not limited thereto.
In addition, in this embodiment, compared to the traversal method, by comparing only the first true value binary codeword and the second true value binary codeword with the same true value information bits and the first false value binary codeword and the second false value binary codeword with the same false value information bits, the code distance calculation amount or calculation cost of the corresponding code group can be greatly reduced, and the time required for calculating the optimal solution can be reduced by geometric multiples by using the differential evolution algorithm for the optimal solution. If the method is a traversal method, firstly, pairwise combinations of all feasible first generator polynomials and second generator polynomials in the second value range (0x10000, 0x1FFFF) need to be calculated, and 65534 × (65534-1)/2 different combinations exist; then, after any two of the first generator polynomial and the second generator polynomial are combined, the code distances of all code groups need to be calculated, and it needs to calculate (2 x 65534-1) x (2 x 65534)/2 times, and it needs about 9h for one calculation using a common personal computer (i 5 processor, main frequency 2.1 GHz). Therefore, the traversal calculation of the above problem is completed under the existing conditions, and about (65534 × 65533/2) × 9h is needed for obtaining the optimal solution finally, about 2.2 × e6 years, and the method has no actual operability.
On the other hand, the embodiment also provides a generator polynomial acquiring system based on a differential evolution algorithm, which acquires an optimal first generator polynomial set and an optimal second generator polynomial set by using the generator polynomial acquiring method based on the differential evolution algorithm.
In summary, the present embodiment provides a method and a system for acquiring a generator polynomial based on a differential evolution algorithm, where an initial information bit set can be constructed by generating an initial information bit, and a feasible first generator polynomial set and a feasible second generator polynomial set can be constructed according to set initial parameters of the differential evolution algorithm; the fitness function of the generator polynomial can be constructed according to the initial information bit set, the feasible first generator polynomial set and the feasible second generator polynomial set, the fitness function is optimized and solved by adopting a differential evolution algorithm, the optimal first generator polynomial set and the optimal second generator polynomial set can be obtained, the selection pool of the generator polynomial is effectively expanded, the differentiated code block configuration of different stations is realized, the misuse of data is fundamentally avoided, and the pairwise code distances of the configured code blocks are as large as possible. Compared with the traversal method, the code distance calculation amount or calculation cost of the corresponding code group can be greatly reduced by only comparing the first true value binary code word and the second true value binary code word with the same true value information bit and the first false value binary code word and the second false value binary code word with the same false value information bit, and meanwhile, the time required by calculating the optimal solution can be reduced in a geometric multiple manner by adopting a differential evolution algorithm to perform optimal solution, so that the method has operability; in addition, the number of available code groups in the coding space and the code distance between the code groups can be increased, so that the variable representation requirement of the interlocking station with higher and higher complexity is met, and the safety is improved. Meanwhile, the initial parameters of the differential evolution algorithm can be adjusted, and different requirements on code distances under different use scenes can be met by adjusting the initial parameter parameters.
While the present invention has been described in detail with reference to the preferred embodiments, it should be understood that the above description should not be taken as limiting the invention. Various modifications and alterations to this invention will become apparent to those skilled in the art upon reading the foregoing description. Accordingly, the scope of the invention should be limited only by the attached claims.
Claims (13)
1. A generator polynomial obtaining method based on differential evolution algorithm is used for double-channel information coding, each channel outputs binary code words consisting of information bits and check bits, and the check bits are obtained through the information bits and generator polynomials corresponding to the channels, and the generator polynomial obtaining method is characterized by comprising the following steps:
generating a plurality of initial information bits to construct an initial information bit set;
setting initial parameters of a differential evolution algorithm, wherein the initial parameters comprise population quantity, differential evolution factors, crossing factors and termination conditions;
constructing a feasible first generator polynomial set and a feasible second generator polynomial set according to the population number, wherein the feasible first generator polynomial set comprises a feasible plurality of first generator polynomials, and the feasible second generator polynomial set comprises a feasible plurality of second generator polynomials;
constructing a fitness function of a generator polynomial according to the initial information bit set, the feasible first generator polynomial set and the feasible second generator polynomial set; and
and adopting the differential evolution algorithm to carry out optimization solution on the fitness function so as to obtain the optimal first generator polynomial set and the optimal second generator polynomial set.
2. The differential evolution algorithm-based generator polynomial acquisition method according to claim 1, wherein the initial information bit is a binary number of K bits, and both the first generator polynomial and the second generator polynomial are binary numbers of R +1 bits; wherein R and K are both positive integers, and the last bit of the first generator polynomial and the second generator polynomial are both 1.
3. The method for obtaining generator polynomial based on differential evolution algorithm as claimed in claim 2, wherein the step of generating a plurality of initial information bits to construct the initial information bit set comprises:
sequentially evaluating K binary numbers of the initial information bits in a first value-evaluating interval to obtain 2 K -2 of said initial information bits; and the first value interval is (000 …, 111 …), wherein the number of 0 in the left end point 000 … of the first value interval is K, and the number of 1 in the right end point 111 … is K.
4. The method for generating polynomial expressions to obtain claim 3 based on differential evolution algorithm, wherein the step of constructing a feasible first set of generating polynomials and a feasible second set of generating polynomials according to the population number comprises:
respectively carrying out random value taking on the front R-bit binary numbers of the first generator polynomial and the second generator polynomial in a second value-taking interval according to the population quantity to obtain a plurality of feasible first generator polynomials and a plurality of feasible second generator polynomials; and the second value interval is (000 …, 111 …), wherein the number of 0 in the left end point 000 … of the second value interval is R, and the number of 1 in the right end point 111 … is R.
5. The differential evolution algorithm-based generator polynomial obtaining method according to claim 4, wherein the step of constructing a fitness function of a generator polynomial from the initial set of information bits, the feasible first set of generator polynomials and the feasible second set of generator polynomials comprises:
selecting the initial information bit with the last bit value of 1 from the initial information bit set as a true value information bit to obtain a true value information bit set;
taking a reverse code for each true value information bit in the true value information bit set to obtain a corresponding false value information bit;
performing modulo-2 division on each of the true information bits and the false information bits by using any feasible first generator polynomial to obtain a first true check bit corresponding to each of the true information bits and a first false check bit corresponding to each of the false information bits;
performing modulo-2 division on each of the true information bits and the false information bits by using any feasible second generator polynomial to obtain a second true check bit corresponding to each of the true information bits and a second false check bit corresponding to each of the false information bits;
splicing each truth value information bit with the corresponding first truth value check bit to obtain a first truth value binary code word; splicing each false value information bit and the corresponding first false value check bit to obtain a first false value binary code word;
splicing each true value information bit with the corresponding second true value check bit to obtain a second true value binary code word; splicing each false value information bit and the corresponding second false value check bit to obtain a second false value binary code word;
forming a pair of truth code groups from the first and second true binary codewords having the same true information bits and a pair of false code groups from the first and second false binary codewords having the same false information bits;
calculating code distances of the first true value binary codeword and the second true value binary codeword in each of the true value code groups, and calculating code distances of the first false value binary codeword and the second false value binary codeword in each of the false value code groups;
and acquiring a corrected total code distance corresponding to the feasible first generator polynomial and the feasible second generator polynomial according to the code distances of all the true value code groups and the code distances of all the false value code groups, wherein the corrected total code distance is used as a fitness index of the fitness function.
6. The method of claim 5, wherein the step of obtaining a modified total code distance corresponding to the feasible first generator polynomial and the feasible second generator polynomial according to the code distances of all the code groups comprises:
summing the code distances of all the true value code groups and all the false value code groups to obtain an uncorrected total code distance;
calculating a penalty value according to the minimum value of the preset code distance; and
and correcting the uncorrected total code distance according to the punishment value to obtain the corrected total code distance.
7. The differential evolution algorithm-based generator polynomial acquisition method according to claim 6, characterized in that the penalty value is calculated using the following formula:
P＝a 0 ×10 n+1 +a 1 ×10 n +a 2 ×10 n-1 +a 3 ×10 n-2 +…+a n ×10 1
wherein P represents a penalty value; n represents the minimum value of the preset code distance; a is 0 、a 1 、a 2 、a 3 、…、a n Respectively representing the times of code distances of 0, 1, 2, 3, …, n in all the true value code groups and all the false value code groups.
8. The method of claim 7, wherein the modified total code distance is calculated by the following formula:
D Z ＝D C -P
wherein D is Z Representing the corrected total code distance; d C Representing the uncorrected total code distance.
9. The method for obtaining polynomials of claim 5, characterized in that the step of using the differential evolution algorithm to optimize the fitness function to obtain the optimal first set of polynomials and the optimal second set of polynomials comprises:
forming an initial population by using a feasible plurality of first generator polynomials and a feasible plurality of second generator polynomials, wherein any one of the first generator polynomials and any one of the second generator polynomials form an individual;
calculating a corrected total code distance corresponding to each individual of the initial population;
performing variation and cross operation on each individual of the initial population according to the differential evolution factor and the cross factor to obtain a progeny population consisting of new individuals;
calculating a corrected total code distance corresponding to each individual of the filial generation population;
comparing the corrected total code distances of any two individuals in the initial population and the offspring population, and reserving the individual with the larger corrected total code distance to enter the next offspring population until the termination condition is reached; and
and taking the individuals in the corresponding population when the termination condition is reached as an optimal first generator polynomial and an optimal second generator polynomial to obtain an optimal first generator polynomial set and an optimal second generator polynomial set.
10. The differential evolution algorithm-based generator polynomial acquisition method of claim 9, characterized in that the termination condition is a maximum number of iterations.
11. The method of claim 9, wherein the number of the first generator polynomial and the second generator polynomial in each population is the same as the number of the populations.
12. The differential evolution algorithm-based generator polynomial acquisition method of claim 5, characterized in that the set of true value information bits includes 2 K-1 -1 true information bits; and the number of the false value information bits is the same as the number of the true value information bits.
13. A system for acquiring generator polynomials based on a differential evolution algorithm, characterized in that the optimal first generator polynomial set and the optimal second generator polynomial set are acquired by using the method for acquiring generator polynomials based on a differential evolution algorithm as claimed in any of claims 1 to 12.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211637119.0A CN115987303A (en) | 2022-12-16 | 2022-12-16 | Generation polynomial acquisition method and system based on differential evolution algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211637119.0A CN115987303A (en) | 2022-12-16 | 2022-12-16 | Generation polynomial acquisition method and system based on differential evolution algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115987303A true CN115987303A (en) | 2023-04-18 |
Family
ID=85973297
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211637119.0A Pending CN115987303A (en) | 2022-12-16 | 2022-12-16 | Generation polynomial acquisition method and system based on differential evolution algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115987303A (en) |
-
2022
- 2022-12-16 CN CN202211637119.0A patent/CN115987303A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9287896B2 (en) | Signal segmentation method and CRC attachment method for reducing undetected error | |
| CN106888026B (en) | Segmented polarization code coding and decoding method and system based on LSC-CRC (least significant likelihood-Cyclic redundancy check) decoding | |
| EP2482464A1 (en) | Encoding apparatus, decoding apparatus, encoding method, decoding method, and communication system | |
| CN101814922B (en) | Multi-bit error correcting method and device based on BCH (Broadcast Channel) code and memory system | |
| US7392461B2 (en) | Decoding for algebraic geometric code associated with a fiber product | |
| CN102412847A (en) | Method and apparatus for decoding low density parity check code using united node processing | |
| US20080320370A1 (en) | CRC generator polynomial select method, CRC coding method and CRC coding circuit | |
| CN1756090B (en) | Channel encoding apparatus and method | |
| CN107239362B (en) | Parallel CRC (Cyclic redundancy check) code calculation method and system | |
| CN101764620B (en) | Apparatus and method for decoding using channel code | |
| US11184034B2 (en) | Method and device for decoding staircase code, and storage medium | |
| CN107306140A (en) | A kind of GEL codeword structures coding and method, device and the relevant device of decoding | |
| CN115987303A (en) | Generation polynomial acquisition method and system based on differential evolution algorithm | |
| KR20080030329A (en) | Method for reduced complexity encoder generating low density parity check codes | |
| CN107404320A (en) | For recombinate the ldpc decoding apparatus and correlation technique of decoding | |
| EP2195928B1 (en) | Encoding method and device for tail-biting trellis ldpc codes | |
| CN110519012B (en) | Method and device for coding and decoding polarization code | |
| Forney | Use of a sequential decoder to analyze convolutional code structure (Corresp.) | |
| JP4379329B2 (en) | CRC generator polynomial selection method, CRC encoding method, and CRC encoding circuit | |
| CN101471672B (en) | Method for generating check matrix and corresponding encoding method and encoder | |
| Fitriani et al. | Single-bit parity detection and correction using hamming code 7-bit model | |
| Tjalkens | On the principal state method for run-length limited sequences | |
| WO2019137231A1 (en) | Decoding method and device | |
| CN110324048A (en) | The coding method of RA-LDPC-CC and encoder in a kind of communication modulation systems | |
| US3439334A (en) | Processing signal information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40084336 Country of ref document: HK |