Aug 2023, Volume 24 Issue 8

  • Select all
  • Review
    Zhenxin MU, Jie PAN, Ziye ZHOU, Junzhi YU, Lu CAO
    2023, 24(8): 1093-1116.

    For complex functions to emerge in artificial systems, it is important to understand the intrinsic mechanisms of biological swarm behaviors in nature. In this paper, we present a comprehensive survey of pursuit-evasion, which is a critical problem in biological groups. First, we review the problem of pursuit-evasion from three different perspectives: game theory, control theory and artificial intelligence, and bio-inspired perspectives. Then we provide an overview of the research on pursuit-evasion problems in biological systems and artificial systems. We summarize predator pursuit behavior and prey evasion behavior as predator-prey behavior. Next, we analyze the application of pursuit-evasion in artificial systems from three perspectives, i.e., strong pursuer group vs. weak evader group, weak pursuer group vs. strong evader group, and equal-ability group. Finally, relevant prospects for future pursuit-evasion challenges are discussed. This survey provides new insights into the design of multi-agent and multi-robot systems to complete complex hunting tasks in uncertain dynamic scenarios.

  • Review
    Jie CHEN, Dandan WU, Ruiyun XIE
    2023, 24(8): 1117-1142.

    Three technical problems should be solved urgently in cyberspace security: the timeliness and accuracy of network attack detection, the credibility assessment and prediction of the security situation, and the effectiveness of security defense strategy optimization. Artificial intelligence (AI) algorithms have become the core means to increase the chance of security and improve the network attack and defense ability in the application of cyberspace security. Recently, the breakthrough and application of AI technology have provided a series of advanced approaches for further enhancing network defense ability. This work presents a comprehensive review of AI technology articles for cyberspace security applications, mainly from 2017 to 2022. The papers are selected from a variety of journals and conferences: 52.68% are from Elsevier, Springer, and IEEE journals and 25% are from international conferences. With a specific focus on the latest approaches in machine learning (ML), deep learning (DL), and some popular optimization algorithms, the characteristics of the algorithmic models, performance results, datasets, potential benefits, and limitations are analyzed, and some of the existing challenges are highlighted. This work is intended to provide technical guidance for researchers who would like to obtain the potential of AI technical methods for cyberspace security and to provide tips for the later resolution of specific cyberspace security issues, and a mastery of the current development trends of technology and application and hot issues in the field of network security. It also indicates certain existing challenges and gives directions for addressing them effectively.

  • Orginal Article
    Linna ZHOU, Zhigao LU, Weike YOU, Xiaofei FANG
    2023, 24(8): 1143-1155.

    In the field of reversible data hiding (RDH), designing a high-precision predictor to reduce the embedding distortion and developing an effective embedding strategy to minimize the distortion caused by embedding information are the two most critical aspects. In this paper, we propose a new RDH method, including a predictor based on a transformer and a novel embedding strategy with multiple embedding rules. In the predictor part, we first design a transformer-based predictor. Then, we propose an image division method to divide the image into four parts, which can use more pixels as context. Compared with other predictors, the transformer-based predictor can extend the range of pixels for prediction from neighboring pixels to global ones, making it more accurate in reducing the embedding distortion. In the embedding strategy part, we first propose a complexity measurement with pixels in the target blocks. Then, we develop an improved prediction error ordering rule. Finally, we provide an embedding strategy including multiple embedding rules for the first time. The proposed RDH method can effectively reduce the distortion and provide satisfactory results in improving the visual quality of data-hidden images, and experimental results show that the performance of our RDH method is leading the field.

  • Orginal Article
    Kaili QI, Minqing ZHANG, Fuqiang DI, Yongjun KONG
    2023, 24(8): 1156-1168.

    To improve the embedding capacity of reversible data hiding in encrypted images (RDH-EI), a new RDH-EI scheme is proposed based on adaptive quadtree partitioning and most significant bit (MSB) prediction. First, according to the smoothness of the image, the image is partitioned into blocks based on adaptive quadtree partitioning, and then blocks of different sizes are encrypted and scrambled at the block level to resist the analysis of the encrypted images. In the data embedding stage, the adaptive MSB prediction method proposed by Wang and He (2022) is improved by taking the upper-left pixel in the block as the target pixel, to predict other pixels to free up more embedding space. To the best of our knowledge, quadtree partitioning is first applied to RDH-EI. Simulation results show that the proposed method is reversible and separable, and that its average embedding capacity is improved. For gray images with a size of 512×512, the average embedding capacity is increased by 25565 bits. For all smooth images with improved embedding capacity, the average embedding capacity is increased by about 35530 bits.

  • Orginal Article
    Xiuli CHAI, Xiuhui CHEN, Yakun MA, Fang ZUO, Zhihua GAN, Yushu ZHANG
    2023, 24(8): 1169-1180.

    With the substantial increase in image transmission, the demand for image security is increasing. Noise-like images can be obtained by conventional encryption schemes, and although the security of the images can be guaranteed, the noise-like images cannot be directly previewed and retrieved. Based on the rank-then-encipher method, some researchers have designed a three-pixel exact thumbnail preserving encryption (TPE2) scheme, which can be applied to balance the security and availability of images, but this scheme has low encryption efficiency. In this paper, we introduce an efficient exact thumbnail preserving encryption scheme. First, blocking and bit-plane decomposition operations are performed on the plaintext image. The zigzag scrambling model is used to change the bit positions in the lower four bit planes. Subsequently, an operation is devised to permute the higher four bit planes, which is an extended application of the hidden Markov model. Finally, according to the difference in bit weights in each bit plane, a bit-level weighted diffusion rule is established to generate an encrypted image and still maintain the same sum of pixels within the block. Simulation results show that the proposed scheme improves the encryption efficiency and can guarantee the availability of images while protecting their privacy.

  • Orginal Article
    Fengda ZHANG, Kun KUANG, Long CHEN, Zhaoyang YOU, Tao SHEN, Jun XIAO, Yin ZHANG, Chao WU, Fei WU, Yueting ZHUANG, Xiaolin LI
    2023, 24(8): 1181-1193.

    To leverage the enormous amount of unlabeled data on distributed edge devices, we formulate a new problem in federated learning called federated unsupervised representation learning (FURL) to learn a common representation model without supervision while preserving data privacy. FURL poses two new challenges: (1) data distribution shift (non-independent and identically distributed, non-IID) among clients would make local models focus on different categories, leading to the inconsistency of representation spaces; (2) without unified information among the clients in FURL, the representations across clients would be misaligned. To address these challenges, we propose the federated contrastive averaging with dictionary and alignment (FedCA) algorithm. FedCA is composed of two key modules: a dictionary module to aggregate the representations of samples from each client which can be shared with all clients for consistency of representation space and an alignment module to align the representation of each client on a base model trained on public data. We adopt the contrastive approach for local model training. Through extensive experiments with three evaluation protocols in IID and non-IID settings, we demonstrate that FedCA outperforms all baselines with significant margins.

  • Orginal Article
    Liang WANG, Shunjiu HUANG, Lina ZUO, Jun LI, Wenyuan LIU
    2023, 24(8): 1194-1213.

    The problem of data right confirmation is a long-term bottleneck in data sharing. Existing methods for confirming data rights lack credibility owing to poor supervision, and work only with specific data types because of their technical limitations. The emergence of blockchain is followed by some new data-sharing models that may provide improved data security. However, few of these models perform well enough in confirming data rights because the data access could not be fully under the control of the blockchain facility. In view of this, we propose a rightconfirmable data-sharing model named RCDS that features symbol mapping coding (SMC) and blockchain. With SMC, each party encodes its digital identity into the byte sequence of the shared data by generating a unique symbol mapping table, whereby declaration of data rights can be content-independent for any type and any volume of data. With blockchain, all data-sharing participants jointly supervise the delivery and the access to shared data, so that granting of data rights can be openly verified. The evaluation results show that RCDS is effective and practical in data-sharing applications that are conscientious about data right confirmation.

  • Orginal Article
    Baoxiong XU, Jianxin YI, Feng CHENG, Ziping GONG, Xianrong WAN
    2023, 24(8): 1214-1230.

    In radar systems, target tracking errors are mainly from motion models and nonlinear measurements. When we evaluate a tracking algorithm, its tracking accuracy is the main criterion. To improve the tracking accuracy, in this paper we formulate the tracking problem into a regression model from measurements to target states. A tracking algorithm based on a modified deep feedforward neural network (MDFNN) is then proposed. In MDFNN, a filter layer is introduced to describe the temporal sequence relationship of the input measurement sequence, and the optimal measurement sequence size is analyzed. Simulations and field experimental data of the passive radar show that the accuracy of the proposed algorithm is better than those of extended Kalman filter (EKF), unscented Kalman filter (UKF), and recurrent neural network (RNN) based tracking methods under the considered scenarios.

  • Perspective
    Yingbo LI, Zhao LI, Yucong DUAN, Anamaria-Beatrice SPULBER
    2023, 24(8): 1231-1238.