Current Issue

25 June 2023, Volume 44 Issue 6

Previous Issue    Next Issue

Papers

Research on new frameworks and key technologies for intelligent emergency command communication networks

Li WANG, Aiguo FEI, Ping ZHANG, Lianming XU

2023, 44(6):  1-11.  doi:10.11959/j.issn.1000-436x.2023112

Asbtract ( 342 )   HTML ( 100)   PDF (2103KB) ( 887 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The new generation of emergency command communication network is the basic means and important support for enhancing China’s emergency response capabilities such as national natural disasters and accidents, and is a key component of establishing a scientific emergency management technology system.Focusing on the requirements of communication, navigating, and sensing capabilities in “intelligent emergency”, a theoretical method and framework for intelligent emergency command and communication networks was proposed, which was introduced from three aspects in terms of network deployment, resource allocation, and assisted decision-making.The difficulties and technical ideas of multi-objective dynamic network deployment for communication, navigating, and sensing, multi-dimension efficient resource allocation for communication, computation, and caching, and multi-level decision-making and intelligent enhancement with cloud-edge-terminal collaboration were analyzed and discussed.It provides theoretical methods and key technical supports for building a new generation of emergency command and communication network in China.

Optimized design of sensing transmission and computing collaborative industrial Internet

Jingbo LI, Li MA, Yang LI, Yingxun FU, Dongchao MA

2023, 44(6):  12-22.  doi:10.11959/j.issn.1000-436x.2023118

Asbtract ( 233 )   HTML ( 51)   PDF (2488KB) ( 446 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In view of the new scenario and time-sensitive requirements of sensing-transmission end to computing end in the industrial Internet, and to solve the problem of end-to-end delay in applications that were not up to standard, a network optimized design scheme for sensing-transmission cooperative computing was proposed.First, the sensing-transmission computing cooperative network architecture based on cloud-edge-end was proposed to promote the deep integration of multidimensional systems and heterogeneous resources.Secondly, the network topology construction mode was optimized to reduce the average path length of the network, shorten the propagation distance, and offset the problem of signal decay due to the use of higher frequencies.Finally, the routing method was improved, and an integrated algorithm of topology optimization and routing strategy was designed to reduce the queuing delay in the network, control the end-to-end delay within a bounded range, and weaken the “long tail phenomenon”.The experimental results show that the scheme reduces the end-to-end delay and ensures the real-time and reliability of the network by optimizing the network topology and routing strategy and integrating the design.

Research on test strategy for randomness based on deep learning

Dongyu CHEN, Hua CHEN, Limin FAN, Yifang FU, Jian WANG

2023, 44(6):  23-33.  doi:10.11959/j.issn.1000-436x.2023111

Asbtract ( 199 )   HTML ( 40)   PDF (1104KB) ( 356 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In order to achieve better test performance, researches on the randomness test strategies based on deep learning were conducted, including the batch average strategy proposed by EUROCRYPT 2021 and the selection strategy for data unit size.By introducing the randomness statistical test model based on deep learning methods, the statistical distribution and test power expression of two test strategies were theoretically derived, and it was pointed out that: (i) the batch average strategy could amplify the prediction accuracy of the model, but it was prone to an increase in the probability of the second type of error in statistics, instead reducing the statistical test power; (ii) the smaller data units of the deep model generally obtained the more powerful statistical tests.Based on the above understanding, a new bit-level deep learning model was proposed for randomness statistical tests, which gained the advantage of prediction with 80 times fewer parameters and 50% samples, compared with the previous work on linear congruent generator (LCG) algorithm, and achieved significant prediction advantages with 10～20 times fewer parameters by extending the model to apply to 5～7 rounds of Speck, compared with the model proposed by Gohr.

Self-adaptive differential evolution algorithm based on population state information

Weijie MAI, Weili LIU, Jinghui ZHONG

2023, 44(6):  34-46.  doi:10.11959/j.issn.1000-436x.2023113

Asbtract ( 244 )   HTML ( 39)   PDF (1337KB) ( 304 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The local optimum and stagnation state information of the population seriously affects the performance of differential evolution (DE) algorithm.An advanced DE algorithm with population state processing measures was proposed to address the above two issues.When the population falled into the local optimum, the individuals in the population were learned randomly by LBFGS method to improve the global quality of the solution, and Gaussian mutation was employed to trigger new individuals to jump out of local optimum.As for the stagnation state, the covariance matrix of the population was applied to reorganize the target individuals based on the rotation of the spatial coordinates to suppress the stagnation state of the population and enhance the global search ability of the algorithm.In addition, a new selection strategy was designed, which built an external archive to store abandoned individuals after greedy selection.When the trial individual was inferior to the target individual, the algorithm no longer generated the next generation with greedy selection strategy, but made reasonable intelligent selection around the external archive to ensure that the algorithm converges to the global optimum.Compared with eight state-of-the-art DE algorithms on 29 benchmark functions, the experimental results show that the proposed algorithm has better performance in terms of the solution accuracy and convergence speed.

Research and implementation of reputation-based inter-domain routing selection mechanism

Shiqi ZHAO, Xiaohong HUANG, Zhigang ZHONG

2023, 44(6):  47-56.  doi:10.11959/j.issn.1000-436x.2023114

Asbtract ( 153 )   HTML ( 32)   PDF (952KB) ( 388 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

To solve the problem of lack of validation for exchanging messages in BGP, a inter-domain routing mechanism, which consisted of a reputation evaluation mechanism and a reputation-based BGP optimal routing algorithm, was proposed.The reputation evaluation mechanism used a distributed autonomous system (AS) alliance architecture, which divided node routing behavior in detail.The service domain and observation weight were used as indicators to quantify the impact of node behavior.By designing a feedback mechanism, the reputation value not only reflected the good and bad of nodes, but also reflected the node’s resistance to malicious attacks.The reputation-based BGP routing selection algorithm adds a “security” policy to the existing routing selection algorithm by filtering routes containing low-reputation nodes and selecting the best route among high reputation routes.The experimental results show that the proposed mechanism outperform most existing reputation mechanisms by avoiding routes with vulnerable nodes and restraining the propagation of illegal routes, thereby providing a more secure inter-domain routing environment.

Performance analysis of optical differential spatial modulation over atmospheric joint effect

Hui ZHAO, Jin LI, Weiwen MA, Wenchao DENG, Tianqi ZHANG, Yuanni LIU

2023, 44(6):  57-69.  doi:10.11959/j.issn.1000-436x.2023117

Asbtract ( 105 )   HTML ( 15)   PDF (1405KB) ( 388 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Traditional optical spatial modulation techniques all rely on the premise that the system receiving end can obtain accurate channel state information.The optical differential spatial modulation (ODSM) system effectively avoids complex channel estimation, but related research only analyzes the system performance under a single turbulent state, and ignores the effects of factors such as pointing errors and path loss in actual free space optical communication systems.Therefore, the Málaga turbulence channel, which could characterize all atmospheric turbulence states, was used to derive the upper bound expression and diversity order of the bit error rate of the ODSM system under the atmospheric joint effect such as turbulence, pointing error, and path loss.The analysis results show that compared with other optical spatial modulation schemes, the ODSM system avoids complex channel estimation, making the system no longer affected by channel estimation errors, thus possessing higher anti-interference and stability.The bit error rate performance of the ODSM system decreases as the intensity of turbulence and pointing error increase.By optimizing parameters such as the number of optical antennas, the number of photodetectors, and the modulation signal order, the bit error rate performance of the ODSM system can be further improved.

Design of knowledge enhanced semantic communication receiver

Rongpeng LI, Bingyan WANG, Honggang ZHANG, Zhifeng ZHAO

2023, 44(6):  70-76.  doi:10.11959/j.issn.1000-436x.2023106

Asbtract ( 168 )   HTML ( 25)   PDF (910KB) ( 420 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

To address the problem that existing semantic communication do not make sufficient use of prior knowledge and have limited decoding capability at the receiver side, a knowledge enhanced semantic communication framework was proposed, in which the receiver could more actively utilize the prior knowledge in the knowledge base for semantic reasoning and decoding, without extra modifications to the neural network structure of the transmitter.Specifically, a transformer-based knowledge extractor was designed to find relevant factual triples for the received noisy signal.Extensive simulation results on the WebNLG dataset demonstrate that the proposed framework has significantly improved performance on the basis of knowledge graph enhanced decoding.

Buffer-aided cooperative NOMA with power transfer

Long YANG, Li ZHAO, Yuchen ZHOU, Bingtao HE, Jian CHEN

2023, 44(6):  77-89.  doi:10.11959/j.issn.1000-436x.2023107

Asbtract ( 136 )   HTML ( 14)   PDF (1891KB) ( 330 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

To improve the throughput of multiuser simultaneous wireless information and power transfer systems, a buffer-aided cooperative non-orthogonal multiple access scheme with power transfer was designed to maximize the average throughput under the constraints of average/peak power, user’s rate, and the buffer stability.To reduce the optimization complexity, the Lyapunov’s method was introduced to convert the long-term average optimization problem into a series of time-discrete subproblems, and an adaptive transmission and resource allocation optimization algorithm was proposed, where the working mode, the user scheduling and the power allocation were dynamically optimized according to the time-varying channel/buffer state.Simulation results demonstrate that compared with the existing schemes, the proposed scheme can significantly enhance the average throughput whilst achieving the tradeoff between the delay and throughput.

Scheduling framework based on reinforcement learning in online-offline colocated cloud environment

Ling MA, Qiliang FAN, Ting XU, Guanchen GUO, Shenglin ZHANG, Yongqian SUN, Yuzhi ZHANG

2023, 44(6):  90-102.  doi:10.11959/j.issn.1000-436x.2023119

Asbtract ( 183 )   HTML ( 16)   PDF (3127KB) ( 334 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Some reinforcement learning-based scheduling algorithms for cloud computing platforms barely considered one scenario or ignored the resource constraints of jobs and treated all machines as the same type, which caused low resource utilization or insufficient scheduling efficiency.To address the scheduling problems in online-offline colocated cloud environment, a framework named JobFusion was proposed.Firstly, an efficient resource partitioning scheme was built in the cloud computing platform supporting virtualization technology by integrating the hierarchical clustering method with connectivity constraints.Secondly, a graph convolutional neural network was utilized to embed the attributes of elastic dimension with various constraints and the jobs with various numbers, to capture the critical path information of workflow.Finally, existing high-performance reinforcement learning methods were integrated for scheduling jobs.According to the results of evaluation experiments, JobFusion improves the resource utilization by 39.86% and reduces the average job completion time by up to 64.36% compared with baselines.

Ring-based efficient batch authentication and group key agreement protocol with anonymity in Internet of vehicles

Haibo ZHANG, Kai LAN, Zhou CHEN, Ruyan WANG, Can ZOU, Mingyue WANG

2023, 44(6):  103-116.  doi:10.11959/j.issn.1000-436x.2023055

Asbtract ( 205 )   HTML ( 28)   PDF (2714KB) ( 480 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Aiming at the problem that the current batch authentication and key agreement protocol were relied on semi-trusted road side unit (RSU) and were not suitable for key update in large-scale Internet of vehicles (IoV), a ring-based efficient batch authentication and group key agreement protocol with anonymity in IoV was proposed.The anonymity was ensured by the pseudonym mechanism.The authentication key pairs were constructed by the chaotic map, and the batch authentication for many vehicles was quickly completed by a small number of bilinear maps.The joining and leaving of vehicles in large-scale IoV scenario were fully considered, a ring session group was efficiently constructed by using the semi-group property of chaotic maps, and a group key establishment and update mechanism suitable for large-scale vehicles was designed.In addition, a pseudonym update mechanism and an anonymous tracing mechanism were designed to ensure a more secure session process.At the same time, the BAN logic model was used to prove the semantic security of the protocol.The security analysis and simulation results show that the proposed protocol has multiple security attributes and certain efficiency advantages.

Direct localization algorithm of the aerial target based on external radiation source

Nan XIA, Danyang GAO, Baohui XING, Yaning WANG

2023, 44(6):  117-124.  doi:10.11959/j.issn.1000-436x.2023120

Asbtract ( 194 )   HTML ( 27)   PDF (4300KB) ( 404 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In order to improve the accuracy of aerial target location and ensure its concealment, a direct location algo-rithm of the aerial target based on external radiation source was proposed.Fixed broadcast signals were used as transmit-ting sources, and the signals scattered by the aerial targets were received by multiple receivers.A direct relationship be-tween the synchronized multi-dimensional baseband signals and the position coordinates of the target was established, and then the spatial cross-spectrum function and joint matching function were constructed.Through iterative spectral peak search, the positioning results of the aerial target could be optimized, and the impact of direct signals and multipath signals on the ground could be effectively suppressed.Simulation results show that the performance of the proposed al-gorithm is better than that of the traditional two-step localization method and other direct localization methods in the case of low signal-to-noise ratio (SNR), and it can achieve high-precision localization of the three-dimensional mobile target.

Functional complementarity relationship enhanced cloud API recommendation method

Zhen CHEN, Wenhui CHEN, Xiaowei LIU, Dianlong YOU, Linlin LIU, Limin SHEN

2023, 44(6):  125-137.  doi:10.11959/j.issn.1000-436x.2023093

Asbtract ( 111 )   HTML ( 13)   PDF (2388KB) ( 398 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Current cloud application program interface (API) recommendation methods mainly use similarity calculation or historical calls of Mashup to generate recommendation results, while ignoring the beneficial functional complementarity (FC) between Mashup and cloud API.To address the above issue, a FC relationship enhanced cloud API recommendation approach was proposed.Firstly, label co-occurrence was applied to describe the FC relationship.Then, the FC score was calculated to describe the degree of FC between the cloud API and the Mashup, and the FC vector was learned to describe the potential FC relationship.Based on this, FC scores and FC vectors were embedded into the cloud API recommendation model, so that FC relationship played a key role in the cloud API recommendation process.Experiments were conducted on real-world cloud API datasets, and the AUC, F1 and HR@5 of the proposed approach improved by an average of 2.32%, 1.86% and 9.15%, respectively, in the sparse scenario.Finally, the proposed approach can improve the accuracy of cloud API recommendation results, while improving the recommendation performance of long-tail cloud API.

Byzantine-robust federated learning over Non-IID data

Xindi MA, Qinghua LI, Qi JIANG, Zhuo MA, Sheng GAO, Youliang TIAN, Jianfeng MA

2023, 44(6):  138-153.  doi:10.11959/j.issn.1000-436x.2023115

Asbtract ( 377 )   HTML ( 84)   PDF (2173KB) ( 741 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The malicious attacks of Byzantine nodes in federated learning was studied over the non-independent and identically distributed dataset , and a privacy protection robust gradient aggregation algorithm was proposed.A reference gradient was designed to identify “poor quality” shared gradients in model training, and the influence of heterogeneity data on Byzantine node recognition was reduced by reputation evaluation.Meanwhile, the combination of homomorphic encryption and random noise obfuscation technology was introduced to protect user privacy in the process of model training and Byzantine node recognition.Finally, through the evaluation over the real-world datasets, the simulation results show that the proposed algorithm can accurately and efficiently identify Byzantine attack nodes while protecting user privacy and has good convergence and robustness.

Adaptive random early detection algorithm based on network traffic level grade prediction

Debin WEI, Chengsheng PAN, Li YANG, Zuoren YAN

2023, 44(6):  154-166.  doi:10.11959/j.issn.1000-436x.2023092

Asbtract ( 123 )   HTML ( 21)   PDF (4673KB) ( 124 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In view of the problem that the calculation of average queue length and maximum packet drop probability in random early detection algorithm and its variants reflect the changes of network traffic slowly, an adaptive random early detection algorithm based on network traffic level grade prediction was proposed.Based on the statistical characteristics of self-similar network traffic, the transition probability table of network traffic level grade was established, and a grade prediction method of self-similar network traffic level with low complexity and high accuracy was proposed.Furthermore, the prediction results were applied to calculate the average queue length in equal interval and adjust the maximum packet drop probability.Under the condition of fixed and variable bottleneck link capacity, it is found that regardless of the degree of self-similarity of network traffic, the proposed algorithm can improve the throughput and packet loss rate, especially when the Hurst parameter is large and the traffic is light.

Construction method of type-Ⅱ even-length Z-complementary pair with large zero correlation zone width

Xiaoyu CHEN, Lianfeng SUN, Yihan ZHANG

2023, 44(6):  167-174.  doi:10.11959/j.issn.1000-436x.2023121

Asbtract ( 89 )   HTML ( 12)   PDF (875KB) ( 289 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

The construction method of Type-Ⅱ even-length Z-complementary pair (ZCP) was proposed by using the inter-leaving technique.The basic framework of interleaving structure with Golay complementary pair (GCP) and its mate as the basic sequences was proposed.Furthermore, the necessary conditions were given for the framework in order that the resulting ZCP could have large zero correlation zone (ZCZ) width.The main parameter form of the constructed ZCP was((2k+1)N,(2k+1)N-k)-ZCP, where N was the length of the chosen GCP.Although not all even lengths can be covered, the constructed ZCP has the ZCZ width that can achieve or approach the upper bound of Type-Ⅱ even-length binary ZCP, and has small peak-to-mean envelope power ratio whose value is less than or equal to 4.The resultant sequences can be applied to multi-carrier spread spectrum system to eliminate more asynchronous interference and reduce nonlinear distortion.

Quantum cryptanalysis of lightweight block cipher Piccolo

Xiaoni DU, Xiangyu WANG, Lifang LIANG, Kaibin LI

2023, 44(6):  175-182.  doi:10.11959/j.issn.1000-436x.2023109

Asbtract ( 135 )   HTML ( 16)   PDF (1283KB) ( 301 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

By taking the characteristics of the structure of Piccolo algorithm RP permutation into consideration, a 3-round quantum distinguisher was proposed.Based on Grover meets Simon algorithm, the 6-round of quantum key recovery attack was given.The results show that the key can be recovered 56 bit with the time complexity 2 ²⁸ and the occupation of 464 qubit.Moreover, if attack rounds r＞6，the time complexity is 2 ^28+16(r-6), which is $\frac{1}{2^{68}}$ of Grover quantum brute-force search.The time complexity of the proposed attack method is significantly reduced compared with Grover search and is also better than that of traditional cryptanalysis, which lays a foundation for the subsequent research on quantum attacks of lightweight block ciphers.

GenFedRL: a general federated reinforcement learning framework for deep reinforcement learning agents

Biao JIN, Yikang LI, Zhiqiang YAO, Yulin CHEN, Jinbo XIONG

2023, 44(6):  183-197.  doi:10.11959/j.issn.1000-436x.2023122

Asbtract ( 392 )   HTML ( 65)   PDF (1359KB) ( 415 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

To solve the problem that intelligent devices equipped with deep reinforcement learning agents lack effective security data sharing mechanisms in the intelligent Internet of things, a general federated reinforcement learning (GenFedRL) framework was proposed for deep reinforcement learning agents.The joint training through model-sharing technology was realized by GenFedRL without the need to share the local private data of deep reinforcement learning agents.Each agent device’s data and computing resources could be effectively used without disclosing the privacy of its private data.To cope with the complexity of the real communication environment and meet the need to accelerate the training speed, a model-sharing mechanism based on synchronization and parallel was designed for GenFedRL.Combined with the model structure characteristics of common deep reinforcement learning algorithms, general federated reinforcement learning algorithm suitable for single network structure and multi-network structure was designed based on the FedAvg algorithm, respectively.Then, the model sharing mechanism among agents with the same network structure was implemented to protect the private data of various agents better.Simulation experiments show that common deep reinforcement learning algorithms still perform well in GenFedRL even in the harsh communication environment where most data nodes cannot participate in training.

Comprehensive Review

Survey on community detection method based on random walk

Yang GAO, Hongli ZHANG

2023, 44(6):  198-210.  doi:10.11959/j.issn.1000-436x.2023108

Asbtract ( 270 )   HTML ( 42)   PDF (1285KB) ( 466 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Random walk techniques achieve high accuracy and efficiency in community detection.To summarize and analyze community detection methods based on random walk, the random walk technique was classified into personalized PageRank, heat kernel diffusion and other random walk methods, and community detection was classified into tasks of local community detection and global community structure identification.A detailed overview of different techniques based on random walk and their application to the tasks of community detection was provided, problems in existing methods were analyzed, and future research directions were pointed out.Finally, evaluation metrics of community detection accuracy for different community detection tasks were summarized in terms of similarity and structure respectively to facilitate research in community detection.

Correspondences

Research on geomagnetic indoor high-precision positioning algorithm based on generative model

Shuai MA, Ke PEI, Huayan QI, Hang LI, Wen CAO, Hongmei WANG, Hailiang XIONG, Shiyin LI

2023, 44(6):  211-222.  doi:10.11959/j.issn.1000-436x.2023104

Asbtract ( 148 )   HTML ( 23)   PDF (3116KB) ( 479 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

Aiming at the current bottleneck of constructing a fine geomagnetic fingerprint library that required a lot of labor costs, two generative models called the conditional variational autoencoder and the conditional confrontational generative network were proposed, which could collect a small number of data samples for a given location, and generate pseudo-label fingerprints.At the same time, in order to solve the problem of low positioning accuracy of single-point geomagnetic fingerprints, a geomagnetic sequence positioning algorithm based on attention mechanism of convolutional neural network-gated recurrent unit was designed, which could effectively use the spatial and temporal characteristics of fingerprints to achieve precise positioning.In addition, a real-time, portable mobile terminal data collection and positioning system was also designed and built.The actual test shows that the proposed model can effectively construct the available geomagnetic fingerprint database, and the average error of the proposed algorithm can reach 0.16 m.

SCLF algorithm for polar code based on GRU network assistance and path metric exchange

Jun LI, Lejia ZHOU, Zhengquan LI, Ru JI, Jintao ZHU, Xingxin LIU, Ziyi LIU

2023, 44(6):  223-237.  doi:10.11959/j.issn.1000-436x.2023110

Asbtract ( 99 )   HTML ( 15)   PDF (3938KB) ( 384 )   Knowledge map   

Figures and Tables | References | Related Articles | Metrics

In order to solve the problems of low transmission performance, inaccurate flip set construction and high complexity of existing polar codes successive cancellation list bit-flip (SCLF) algorithms, a SCLF algorithm for polar codes based on GRU network assistance and path metric exchange was proposed.First the decoding state was redivided, and a bit-flip strategy was proposed by combining the ideas of path metric exchange and reverse judgment.Secondly, a flip set construction method was proposed by training the GRU network to locate the first erroneous bit in the decoding.Finally, a multi-bit flipping rule was proposed by sorting the path metric for each low-order flip and fixing the low-order reliable bits before performing the high-order flip.Simulation results show that compared with the existing SCLF algorithms, the proposed algorithm improves the accuracy of identifying the first error bit by 18～24% at low signal noise ratio.Under single-bit and multi-bit flipping, the proposed algorithm has a performance gain of up to 0.3 dB and 0.2 dB, respectively, and the online decoding complexity is lower.