We study the detailed malicious code propagating process in scale-free networks with link weights that denotes traffic between two nodes. It is found that the propagating velocity reaches a peak rapidly then decays in...We study the detailed malicious code propagating process in scale-free networks with link weights that denotes traffic between two nodes. It is found that the propagating velocity reaches a peak rapidly then decays in a power-law form, which is different from the well-known result in unweighted network case. Simulation results show that the nodes with larger strength are preferential to be infected, but the hierarchical dynamics are not clearly found. The simulation results also show that larger dispersion of weight of networks leads to slower propagating, which indicates that malicious code propagates more quickly in unweighted scale-free networks than in weighted scale-free networks under the same condition. These results show that not only the topology of networks but also the link weights affect the malicious propagating process.展开更多
A revised weight-coded evolutionary algorithm (RWCEA) is proposed for solving multidimensional knapsack problems. This RWCEA uses a new decoding method and incorporates a heuristic method in initialization. Computatio...A revised weight-coded evolutionary algorithm (RWCEA) is proposed for solving multidimensional knapsack problems. This RWCEA uses a new decoding method and incorporates a heuristic method in initialization. Computational results show that the RWCEA performs better than a weight-coded evolutionary algorithm pro-posed by Raidl (1999) and to some existing benchmarks, it can yield better results than the ones reported in the OR-library.展开更多
Constant weight code is an important error-correcting control code in communications. Basic structure of constant weight codes for some arriving at Johnson bound, A（n, 2u, w）, is presented. Some correlative property...Constant weight code is an important error-correcting control code in communications. Basic structure of constant weight codes for some arriving at Johnson bound, A（n, 2u, w）, is presented. Some correlative propertys of the codes, the solution of arriving at Johnson bound, and the results on the couple constant code and some constant weight codes are discussed. The conclusion is verified through four examples.展开更多
Let F_qbe afinite field with q=pmelements,where pis an odd prime and mis apositive integer.Here,let D_0={(x_1,x_2)∈F_q^2\{(0,0)}:Tr(x_1^(pk1+1)+x_2^(pk2+1))=c},where c∈F_q,Tr is the trace function fromFF_qtoFpand m/...Let F_qbe afinite field with q=pmelements,where pis an odd prime and mis apositive integer.Here,let D_0={(x_1,x_2)∈F_q^2\{(0,0)}:Tr(x_1^(pk1+1)+x_2^(pk2+1))=c},where c∈F_q,Tr is the trace function fromFF_qtoFpand m/(m,k_1)is odd,m/(m,k_2)is even.Define ap-ary linear code C_D =c(a_1,a_2):(a_1,a_2)∈F_q^2},where c(a_1,a_2)=(Tr(a_1x_1+a_2x_2))_((x1,x2)∈D).At most three-weight distributions of two classes of linear codes are settled.展开更多
In this paper we consider the problem of finding bounds on the size of lexicographic constant-weight equidistant codes over the alphabet of three, four and five elements with 2 ≤ w 【n ≤ 10. Computer search of lexic...In this paper we consider the problem of finding bounds on the size of lexicographic constant-weight equidistant codes over the alphabet of three, four and five elements with 2 ≤ w 【n ≤ 10. Computer search of lexicographic constant-weight equidistant codes is performed. Tables with bounds on the size of lexicographic constant-weight equidistant codes are presented.展开更多
Deep holes are very important in the decoding of generalized RS codes, and deep holes of RS codes have been widely studied, but there are few works on constructing general linear codes based on deep holes. Therefore, ...Deep holes are very important in the decoding of generalized RS codes, and deep holes of RS codes have been widely studied, but there are few works on constructing general linear codes based on deep holes. Therefore, we consider constructing binary linear codes by combining deep holes with binary BCH codes. In this article, we consider the 2-error-correcting binary primitive BCH codes and the extended codes to construct new binary linear codes by combining them with deep holes, respectively. Furthermore, three classes of binary linear codes are constructed, and then we determine the parameters and the weight distributions of these new binary linear codes.展开更多
A hybrid decoding algorithm is proposed for nonbinary low-density parity-check （LDPC） codes, which combines the weighted symbol-flipping （WSF） algorithm with the fast Fourier trans- form q-ary sum-product algorit...A hybrid decoding algorithm is proposed for nonbinary low-density parity-check （LDPC） codes, which combines the weighted symbol-flipping （WSF） algorithm with the fast Fourier trans- form q-ary sum-product algorithm （FFT-QSPA）. The flipped position and value are determined by the symbol flipping metric and the received bit values in the first stage WSF algorithm. If the low- eomplexity WSF algorithm is failed, the second stage FFT-QSPA is activated as a switching strategy. Simulation results show that the proposed hybrid algorithm greatly reduces the computational complexity with the performance close to that of FFT-QSPA.展开更多
The paper introduces the state reduction algorithm and accelerated state reduction algorithm are used to compute the distance weight enumerator(transfer function)T[x,y] of convolutional codes.Next use computer simulat...The paper introduces the state reduction algorithm and accelerated state reduction algorithm are used to compute the distance weight enumerator(transfer function)T[x,y] of convolutional codes.Next use computer simulation to compare upper bound on the bit error probability on an additive white Gaussian noise(AWGN) for maximum free distance(MFD) codes of previously found and optimum distance spectrum(ODS) codes with rate 1/4,overall constraint length are 5 and 7,respectively.Finally,a method of how to search for good convolutional codes is given.展开更多
The paper introduces the state reduction algorithm and accelerated state reduction algorithm are used to compute the distance weight enumerator(transfer function) T[x,y] of convolutional codes.Next use computer simula...The paper introduces the state reduction algorithm and accelerated state reduction algorithm are used to compute the distance weight enumerator(transfer function) T[x,y] of convolutional codes.Next use computer simulation to compare upper bound on the bit error probability on an additive white Gaussian noise (AWGN) for maximum free distance (MFD) codes of previously found and optimum distance spectrum (ODS) codes with rate 1/4,overall constraint length are 5 and 7,respectively. Finally,a method of how to search for good convolutional codes is given.展开更多
基于数据驱动的单元测试代码自动化生成技术存在覆盖率低和可读性差的问题,难以应对日益增长的测试需求。大语言模型(LLM)在代码生成任务中显示了极大的潜力,然而由于代码数据的功能风格和编码风格的差异,LLM面临灾难性遗忘和资源受限这...基于数据驱动的单元测试代码自动化生成技术存在覆盖率低和可读性差的问题,难以应对日益增长的测试需求。大语言模型(LLM)在代码生成任务中显示了极大的潜力,然而由于代码数据的功能风格和编码风格的差异,LLM面临灾难性遗忘和资源受限这2个挑战。为了解决这些问题,提出将编码风格和功能风格同步迁移微调的思想,并开发一种高效的LLM微调训练方法用于单元测试用例生成。首先,利用广泛使用的指令数据集对LLM进行指令对齐,并按任务类型对指令集分类;同时,提取并存储具有任务特征的权重增量;其次,设计一个自适应风格提取模块,该模块包含抗噪声干扰学习和编码风格回溯学习,以应对不同的代码编写风格;最后,在目标域分别对功能风格增量和编码风格增量进行联合训练,以实现在目标域低资源情况下的高效适配和微调。在SF110 Corpus of Classes数据集上的测试用例生成实验结果表明,所提方法的结果均优于对比方法,与主流代码生成LLM Codex、Code Llama和DeepSeek-Coder相比,所提方法的编译率分别提高0.8%、43.5%和33.8%、分支覆盖率分别提高3.1%、1.0%和17.2%;行覆盖率分别提高4.1%、6.5%和15.5%,验证了所提方法在代码生成任务上的优越性。展开更多
Let p =ef +1 be an odd prime with positive integers e and f. In this paper, we calculate the values of Gauss periods of order e =3, 4, 6 over a finite field GF(q), where q is a prime with q≠p. As applications, severa...Let p =ef +1 be an odd prime with positive integers e and f. In this paper, we calculate the values of Gauss periods of order e =3, 4, 6 over a finite field GF(q), where q is a prime with q≠p. As applications, several cyclotomic sequences of order e =3, 4, 6 are employed to construct a number of classes of cyclic codes over GF(q) with prime length. Under certain conditions, the linear complexity and reciprocal minimal polynomials of cyclotomic sequences are calculated, and the lower bounds on the minimum distances of these cyclic codes are obtained.展开更多
A novel Code-Matched interleaver is proposed which decreases the number of the low weight codewords to improve the performance of the Turbo code. The modified design can adapt more kinds of Turbo codes determined by t...A novel Code-Matched interleaver is proposed which decreases the number of the low weight codewords to improve the performance of the Turbo code. The modified design can adapt more kinds of Turbo codes determined by the generator matrix, while it doesn＇t decrease the bit error rate performance of Turbo codes at moderate to high signal to noise ratio. At the same time, in Rayleigh fading channel, the new Code-Matched interleaver can also debase the error floor.展开更多
基金Supported by the National Natural Science Foundation of China (90204012, 60573036) and the Natural Science Foundation of Hebei Province (F2006000177)
文摘We study the detailed malicious code propagating process in scale-free networks with link weights that denotes traffic between two nodes. It is found that the propagating velocity reaches a peak rapidly then decays in a power-law form, which is different from the well-known result in unweighted network case. Simulation results show that the nodes with larger strength are preferential to be infected, but the hierarchical dynamics are not clearly found. The simulation results also show that larger dispersion of weight of networks leads to slower propagating, which indicates that malicious code propagates more quickly in unweighted scale-free networks than in weighted scale-free networks under the same condition. These results show that not only the topology of networks but also the link weights affect the malicious propagating process.
文摘A revised weight-coded evolutionary algorithm (RWCEA) is proposed for solving multidimensional knapsack problems. This RWCEA uses a new decoding method and incorporates a heuristic method in initialization. Computational results show that the RWCEA performs better than a weight-coded evolutionary algorithm pro-posed by Raidl (1999) and to some existing benchmarks, it can yield better results than the ones reported in the OR-library.
文摘Constant weight code is an important error-correcting control code in communications. Basic structure of constant weight codes for some arriving at Johnson bound, A（n, 2u, w）, is presented. Some correlative propertys of the codes, the solution of arriving at Johnson bound, and the results on the couple constant code and some constant weight codes are discussed. The conclusion is verified through four examples.
文摘Let F_qbe afinite field with q=pmelements,where pis an odd prime and mis apositive integer.Here,let D_0={(x_1,x_2)∈F_q^2\{(0,0)}:Tr(x_1^(pk1+1)+x_2^(pk2+1))=c},where c∈F_q,Tr is the trace function fromFF_qtoFpand m/(m,k_1)is odd,m/(m,k_2)is even.Define ap-ary linear code C_D =c(a_1,a_2):(a_1,a_2)∈F_q^2},where c(a_1,a_2)=(Tr(a_1x_1+a_2x_2))_((x1,x2)∈D).At most three-weight distributions of two classes of linear codes are settled.
文摘In this paper we consider the problem of finding bounds on the size of lexicographic constant-weight equidistant codes over the alphabet of three, four and five elements with 2 ≤ w 【n ≤ 10. Computer search of lexicographic constant-weight equidistant codes is performed. Tables with bounds on the size of lexicographic constant-weight equidistant codes are presented.
文摘Deep holes are very important in the decoding of generalized RS codes, and deep holes of RS codes have been widely studied, but there are few works on constructing general linear codes based on deep holes. Therefore, we consider constructing binary linear codes by combining deep holes with binary BCH codes. In this article, we consider the 2-error-correcting binary primitive BCH codes and the extended codes to construct new binary linear codes by combining them with deep holes, respectively. Furthermore, three classes of binary linear codes are constructed, and then we determine the parameters and the weight distributions of these new binary linear codes.
基金Supported by the National High Technology Research and Development Programme of China(No.2009AAJ128,2009AAJ208,2010AA7010422)
文摘A hybrid decoding algorithm is proposed for nonbinary low-density parity-check （LDPC） codes, which combines the weighted symbol-flipping （WSF） algorithm with the fast Fourier trans- form q-ary sum-product algorithm （FFT-QSPA）. The flipped position and value are determined by the symbol flipping metric and the received bit values in the first stage WSF algorithm. If the low- eomplexity WSF algorithm is failed, the second stage FFT-QSPA is activated as a switching strategy. Simulation results show that the proposed hybrid algorithm greatly reduces the computational complexity with the performance close to that of FFT-QSPA.
文摘The paper introduces the state reduction algorithm and accelerated state reduction algorithm are used to compute the distance weight enumerator(transfer function)T[x,y] of convolutional codes.Next use computer simulation to compare upper bound on the bit error probability on an additive white Gaussian noise(AWGN) for maximum free distance(MFD) codes of previously found and optimum distance spectrum(ODS) codes with rate 1/4,overall constraint length are 5 and 7,respectively.Finally,a method of how to search for good convolutional codes is given.
文摘The paper introduces the state reduction algorithm and accelerated state reduction algorithm are used to compute the distance weight enumerator(transfer function) T[x,y] of convolutional codes.Next use computer simulation to compare upper bound on the bit error probability on an additive white Gaussian noise (AWGN) for maximum free distance (MFD) codes of previously found and optimum distance spectrum (ODS) codes with rate 1/4,overall constraint length are 5 and 7,respectively. Finally,a method of how to search for good convolutional codes is given.
文摘基于数据驱动的单元测试代码自动化生成技术存在覆盖率低和可读性差的问题,难以应对日益增长的测试需求。大语言模型(LLM)在代码生成任务中显示了极大的潜力,然而由于代码数据的功能风格和编码风格的差异,LLM面临灾难性遗忘和资源受限这2个挑战。为了解决这些问题,提出将编码风格和功能风格同步迁移微调的思想,并开发一种高效的LLM微调训练方法用于单元测试用例生成。首先,利用广泛使用的指令数据集对LLM进行指令对齐,并按任务类型对指令集分类;同时,提取并存储具有任务特征的权重增量;其次,设计一个自适应风格提取模块,该模块包含抗噪声干扰学习和编码风格回溯学习,以应对不同的代码编写风格;最后,在目标域分别对功能风格增量和编码风格增量进行联合训练,以实现在目标域低资源情况下的高效适配和微调。在SF110 Corpus of Classes数据集上的测试用例生成实验结果表明,所提方法的结果均优于对比方法,与主流代码生成LLM Codex、Code Llama和DeepSeek-Coder相比,所提方法的编译率分别提高0.8%、43.5%和33.8%、分支覆盖率分别提高3.1%、1.0%和17.2%;行覆盖率分别提高4.1%、6.5%和15.5%,验证了所提方法在代码生成任务上的优越性。
基金Supported by the National Natural Science Foundation(NNSF)of China(No.11171150)Foundation of Science and Technology on Information Assurance Laboratory(No.KJ-13-001)+1 种基金Funding of Jiangsu Innovation Program for Graduate Education(CXLX13-127,Fundamental Research Funds for the Central Universities)Funding for Outstanding Doctoral Dissertation in NUAA(BCXJ-13-17)
文摘Let p =ef +1 be an odd prime with positive integers e and f. In this paper, we calculate the values of Gauss periods of order e =3, 4, 6 over a finite field GF(q), where q is a prime with q≠p. As applications, several cyclotomic sequences of order e =3, 4, 6 are employed to construct a number of classes of cyclic codes over GF(q) with prime length. Under certain conditions, the linear complexity and reciprocal minimal polynomials of cyclotomic sequences are calculated, and the lower bounds on the minimum distances of these cyclic codes are obtained.
文摘A novel Code-Matched interleaver is proposed which decreases the number of the low weight codewords to improve the performance of the Turbo code. The modified design can adapt more kinds of Turbo codes determined by the generator matrix, while it doesn＇t decrease the bit error rate performance of Turbo codes at moderate to high signal to noise ratio. At the same time, in Rayleigh fading channel, the new Code-Matched interleaver can also debase the error floor.
