
Current Calls for Papers
Learn More

Current Calls for Papers
Learn More

Special Issue: Recipients of CIE Science and Technology Awards
Learn More

Special Issue: Recipients of CIE Science and Technology Awards
Learn More
|
WELCOME TO |
| Acta Electronica Sinica |
|
Acta Electronica Sinica, launched in 1962, is the flagship academic journal of the Chinese Institute of Electronics (CIE). The journal publishes cutting-edge research findings in the fields of electronics, information science, and related disciplines.
|
| LEARN MORE |
Quick Entry
- Ex-Submission System
- Submission System
- Contact Us

2026Volume 54Issue 8
|
ACTA ELECTRONICA SINICA CURRENT ISSUE: |
Recent Issues
![]() |
![]() |
![]() |
Articles Online
MOREOn-device Agent Systems
LIU Sicong, WU Fengmin, ZHAO Wanbing, ZHENG Yawen, CAI Yongyan, LI Teng, LU Wanting, ZHOU Xingshe, LIU YunhaoCorrected ProofPages: 1-37(2026) DOI: 10.12263/DZXB.20261120Abstract:Recent advances in foundation models and embedded hardware are transforming on-device intelligence from fixed-task inference into on-device agent systems that continuously operate in open environments. Beyond autonomous perception, intent understanding, action execution, and feedback adaptation, these systems face multi-source runtime dynamics arising from changing task intents, environmental states, performance requirements, and resource availability, making fixed model configurations and execution strategies increasingly inadequate. Consequently, on-device agent systems are evolving from task-level autonomy toward system-level autonomy, requiring runtime-elastic adaptation capabilities across perception, reasoning, memory, decision-making, and action execution. This shift challenges conventional post-hoc optimization and calls for structured adaptation pathways that are explicit, computable, and reconfigurable. Neuro-symbolic methods provide a promising solution: neural networks handle open-world understanding, uncertainty, and complex semantics, while learning implicit patterns from multi-source data; symbolic mechanisms explicitly represent objects, relations, processes, and constraints, transforming implicit dependencies within agent systems into adjustable structures for runtime adaptation. By combining neural learning with symbolic reasoning, on-device agent systems can not only understand complex dynamic environments but also leverage structured information to rapidly identify changes, constrain decisions, and perform local adjustments, thereby enabling runtime-elastic adaptation under resource constraints. This work systematically summarizes neuro-symbolic approaches to on-device agent systems from six perspectives: structured cognition, hierarchical task planning, lightweight world models, elastic reasoning orchestration, feedback-driven self-correction, and cross-level computation-resource mapping. We further discuss open challenges including self-evolving neuro-symbolic mechanisms, compute-memory integration, etc. Overall, this work highlights how neuro-symbolic methods can support the evolution of on-device agents from task-level to system-level autonomy and enable runtime-elastic adaptation, providing design insights for next-generation on-device artificial intelligence (AI) systems.36
11
0
2026-11-20
TrustLink: Blockchain-Driven Cross-Attestation for Heterogeneous Trusted Hardware
HUANG Junqin, KONG Linghe, CHEN GuihaiCorrected ProofPages: 1-22(2026) DOI: 10.12263/DZXB.20260993Abstract:Heterogeneous trusted execution environments (TEEs) are built on different hardware architectures and employ separate trust chains as well as incompatible remote attestation protocols, making it difficult to establish unified trust in distributed scenarios such as cross-cloud collaboration and privacy-preserving computing. An attestation approach that relies on a single proxy requires all participating organizations to place their trust in the same third party, while directly moving the process of evidence verification onto a blockchain introduces additional challenges in upgrading verification logic, validating externally retrieved results, and protecting trusted application identities. To address these challenges, this paper proposes TrustLink, a blockchain-driven cross-attestation scheme for heterogeneous trusted hardware. TrustLink uses smart contracts as a decentralized trust mediator and relies on blockchain consensus to unify heterogeneous evidence verification and attestation-state management. A storage-computation decoupled architecture separates the proxy contract from verification contracts. The proxy contract is responsible for maintaining attestation requests, their associated results, and the mappings to the corresponding verification modules, while platform-specific modules check certificate chains, report signatures, and acceptance policies. Committee governance supports the registration of new TEE types, verification-module upgrades, and revocation of attestation results. Because TEEs rely on peripheral hosts to retrieve blockchain information, the scheme verifies block-header finality and Merkle paths inside the TEE and combines these checks with successful execution status, request identifiers, and session-binding information to determine whether an attestation result can be accepted. Each connection uses a fresh challenge and an ephemeral public key. The two directions of mutual attestation share a session context, and key confirmation is performed after both parties pass verification to establish a secure channel bound to the current attestation. A hash-masking method based on one-time public keys is also designed to reduce cross-session linkability caused by stable on-chain identity identifiers, provided that the original trusted application measurement is undisclosed to the adversary and its candidate distribution has sufficient min-entropy. A formal security model is established to analyze report unforgeability, on-chain transaction inclusion, replay resistance, identity unlinkability, and governance security. The FISCO BCOS prototype provides a unified interface for Intel SGX, Intel TDX, AMD SEV-SNP, and Arm CCA evidence formats, and signed test evidence is used to validate cross-component attestation and mutual attestation workflows. Experiments with fixed evidence samples show that the mean latency of same-chain read-only verification ranges from 2.135 to 12.323 ms, with P99 latency not exceeding 14.264 ms. Under a uniform four-TEE workload at a concurrency level of 32, the average throughput reaches 1 819.23 requests/s. With the default minimum block-sealing interval of 500 ms, the mean latency of the unidirectional cross-component workflow ranges from 1.019 to 1.026 s, while the mean latency of mutual attestation across the six heterogeneous TEE pairs ranges from 2.041 to 2.055 s. Attestation latency is primarily determined by transaction confirmation. Experiments on concurrency, batch processing, network delay, and failure recovery further show that TrustLink supports unified heterogeneous evidence processing with millisecond-level verification overhead and improves processing capacity through parallel verification and batched writes, and can serve as a scalable foundation for attestation in trusted collaboration involving multiple independent organizations.32
11
0
2026-11-20
Multi-Target Activity Sensing Based on Harmonic Modulation with Space-Time-Coding Metasurface Antenna
SUN Yimiao, YE Haoran, LI Huidong, XIE Yadong, CHENG Qiang, HE YuanCorrected ProofPages: 1-9(2026) DOI: 10.12263/DZXB.20260803Abstract:With the continued growth in demand for intelligent sensing, activity sensing has already been applied in a variety of scenarios. Wireless sensing technology has become one of the main technologies for achieving activity sensing because of its advantages, including contactless operation and privacy-friendly features. However, due to the limitations imposed by signal bandwidth and the capabilities of sensing devices, multi-target activity sensing currently remains an open problem. Wireless signals that have been reflected by different sensing targets are superimposed on one another when they arrive at the receiver. The signal bandwidth available to wireless sensing systems and the hardware capabilities of their sensing devices are often limited, making it difficult to directly distinguish and separate these mixed signals. Existing methods employ signal processing techniques at the receiver to decompose the mixed signals, or alternatively deploy smartphones in the vicinity of the sensing targets and use the near-field channel variations acquired by these smartphones to sense different targets. However, these methods typically require sufficient differences between the signals reflected by different targets, or their sensing performance is limited by the hardware capabilities of the sensing devices used in the system. In this paper, we propose RISecho, a multi-target sensing method that uses a metasurface antenna to sense the activities of multiple targets. RISecho uses the metasurface antenna to convert the incident wireless signals into multiple harmonic beams, whose frequencies are coupled to the directions in which they propagate. As the harmonic beams illuminate different sensing targets, they separate the sensing information associated with these targets by allowing the information from different targets to be carried by signals at different frequencies. Once the receiver has received these harmonic signals, it can directly separate them in the frequency domain and then extract the sensing information contained in the separated signals. To address the challenge of sensing targets whose directions are unknown, RISecho employs a target discovery algorithm to identify the directions in which targets are present. The metasurface antenna generates scanning beams periodically to scan different directions for sensing targets. The receiver extracts Doppler variations in each direction and uses these variations to determine whether a sensing target is present in the corresponding direction. Based on the receiver's decisions about whether targets are present in different directions, the metasurface antenna allocates harmonics to the corresponding sensing targets. To address the interference caused by sidelobes, RISecho optimizes the space-time modulation codes used by the metasurface antenna so as to reduce the gain of these sidelobes. The receiver then applies an interference suppression algorithm to further process the harmonic signals it receives. After correcting the Doppler shifts based on the frequency differences among the harmonics, RISecho uses a sensing neural network to identify the activity states of the sensing targets. Experimental results obtained across a range of settings show that RISecho can simultaneously sense four targets and achieve an activity recognition accuracy of 94.37%, representing a 13.48% improvement over the baseline method. When the receiver is located in a non-line-of-sight region relative to the metasurface antenna, RISecho achieves an activity recognition accuracy that is more than 30% higher than that of the baseline method. When multiple sensing targets are present simultaneously, the average target detection accuracy remains above 96%.79
24
0
2026-11-16
A Deep-Learning-Assisted Fast Inverse Design Method for Pancharatnam-Berry Phase Metasurface Unit Cells
LIANG Jingyi, CHEN Long, LI XinyuCorrected ProofPages: 1-16(2026) DOI: 10.12263/DZXB.20260817Abstract:Pancharatnam-Berry (PB) phase metasurfaces can introduce, through unit-cell rotation, a geometric phase equal to twice the rotation angle in the cross-circularly polarized channel. They provide continuous phase coverage and simple structures, making it convenient to construct phase gradients. However, the cross-circularly polarized amplitude, co-circularly polarized amplitude, and phase of practical PB unit cells are jointly affected by factors including geometric parameters, operating frequency, and inter-unit coupling. Traditional design methods that rely on empirical experience and point-by-point parameter sweeps using full-wave simulations have high computational costs and struggle to meet the need for rapid iteration. To address this problem, this paper proposes a deep-learning-based fast inverse design method for PB phase metasurface unit cells. First, truncated Fourier series representing the real and imaginary parts of a complex quantity are used to explicitly describe the periodic variation of geometric parameters with the target phase variable. A MultiLayer Perceptron (MLP) then predicts the corresponding Fourier coefficients from the target amplitude-type feature and operating frequency. This transforms point-by-point structural-parameter prediction into the learning and reconstruction of complete parameter curves, thereby eliminating the numerical discontinuities caused by phase-periodicity boundaries and rotational-angle equivalence. Second, a parameter-response dataset is constructed using full-wave simulations, and a differentiable forward surrogate model is established to represent circular-polarization scattering parameters by their real and imaginary parts. Continuous interpolation provides a continuous mapping from geometric parameters to electromagnetic responses, avoiding the effect of phase wrapping on modeling accuracy. On this basis, the inverse MLP is cascaded with the forward surrogate model to jointly constrain amplitude and phase errors in a unified complex-response space. A hard-sample weighting mechanism is introduced to improve training stability and the consistency of inverse-design results under multiple-solution conditions. The framework can be applied to both unit-cell design for arbitrary cross-polarized S-parameters and unit-cell design for arbitrary polarization conversion. The two tasks thus share the same parameter representation, forward validation, and loss-constraint mechanism, reducing the complexity of repeated modeling for different design objectives. Simulation and measurement results show that, in the design of a 4-bit cross-polarized phase unit-cell set at 7.2 GHz, the 16 unit cells’ SLR have an average amplitude of 0.987 8 and a minimum amplitude of 0.986 3, with an average phase error of 1.12° and a maximum phase error of 3.30°. A 50×50 focusing metasurface constructed from the designed unit cells forms a distinct focal spot in both simulations and measurements. For arbitrary polarization-conversion unit-cell design, when 7.0 GHz and 7.5 GHz are used as test frequencies, the target Polarization Conversion Ratio (PCR) is set to 0.5, and the target phase is swept, the PCR error is approximately 0.05 and the maximum phase-difference error does not exceed 3.73°. These results show that the proposed method can balance design accuracy and computational efficiency, providing a feasible approach to the rapid design of PB phase metasurface unit cells and related devices.65
22
0
2026-11-16
A Large-Scale Multi-Objective Optimization Algorithm Based on Dynamic
-Variational Autoencoder Dimensionality Reduction and Multi-Operator Collaborative Sampling
WANG Pengtao, WU Xiangjuan, JI Zhichao, HE ChanghuiCorrected ProofPages: 1-18(2026) DOI: 10.12263/DZXB.20260796Abstract:Large-scale multi-objective optimization problems (LSMOPs) usually contain a large number of decision variables, which makes evolutionary algorithms inefficient in high-dimensional search spaces and difficult to balance convergence and diversity. To address this issue, this paper proposes a large-scale multi-objective optimization algorithm based on dynamic -variational autoencoder ( -VAE) dimensionality reduction and multi-operator collaborative sampling. In the proposed algorithm, the population is first mapped into a low-dimensional latent space by an improved -VAE model, and a dynamic adaptive latent-space update strategy is designed to adjust the strength of the KL divergence constraint according to the evolutionary stage. This strategy controls the degree of decision-variable decoupling in the latent space and prevents excessive loss of reconstruction information from the initial population during the early stage of latent-space evolution, thereby reducing the risk of model collapse. Moreover, the model is trained periodically to adapt to the changes in the population distribution. To overcome the limitation that a single latent-space sampling strategy can hardly maintain both convergence and diversity during offspring generation, a latent-space multi-operator collaborative sampling strategy is proposed. This strategy combines three sampling operators, namely elite interpolation, global random sampling, and boundary exploration, to improve the quality of candidate solutions. Meanwhile, a dynamic adjustment strategy for operator combination based on population stagnation perception is designed. The population update situation is measured by the effective survival rate of the offspring after environmental selection. When the effective survival rate of the offspring in consecutive generations is lower than the set threshold, the combination weights are updated based on the comprehensive survival rate of each operator during the stagnation period, and a minimum weight is set to retain the participation opportunities of each operator, thereby dynamically allocating the resources for offspring generation and alleviating the stagnation of evolution. Finally, on the LSMOP1 to LSMOP9 test functions, 2-objective and 3-objective cases with 500, 1 000 and 2 000 decision variables are set, and compared with 5 representative algorithms. The results show that the proposed algorithm has stronger competitiveness compared to many advanced large-scale multi-objective optimization algorithms.112
30
0
2026-11-11
Prediction-Driven Dual-Time-Scale Joint Optimization of AFDM-ISAC Waveform and Beam
LI Jiawei, WANG Dawei, PANG Linna, WEI Zhongxiang, JIN Yi, YANG Weichao, LI LiCorrected ProofPages: 1-16(2026) DOI: 10.12263/DZXB.20260962Abstract:This paper addresses scenarios involving continuous sensing of highly mobile aerial targets and multi-user downlink communication by proposing a prediction-driven, dual-time-scale joint design scheme for affine frequency division multiplexing (AFDM)-based integrated sensing and communication (ISAC) waveforms and beams. The core motivation stems from a fundamental conflict: target mobility induces dynamic variations in delay and Doppler, yet updating AFDM parameters on a slot-by-slot basis to track these variations incurs prohibitive signaling and synchronization overhead. To resolve this tension, a dual-time-scale framework is established that couples superframe-level waveform selection with slot-level beam updating, striking an effective trade-off between sensing accuracy and communication efficiency. Within the proposed framework, at the start of each superframe, the base station employs an extended Kalman filter (EKF) to obtain the target posterior state and error covariance, performs multi-step prediction of the target state, and constructs a delay-Doppler envelope with confidence margin through error propagation. Based on this envelope and the delay-Doppler support of the communication link, a feasible AFDM codebook is selected and fixed throughout the superframe, while communication and sensing beams are updated on a slot-by-slot basis according to real-time channel conditions and target prior information. On the modeling front, an effective rate model is developed that accounts for the overheads of data, pilot, and guard regions, as well as chirp-cyclic prefixes. Meanwhile, this paper constructs a data-averaged Fisher information matrix, thereby deriving the joint range-radial velocity Cramér-Rao bound (CRB). Building on these models, a joint optimization problem is formulated to maximize the average effective rate subject to constraints on transmit power, user quality of service, and the sensing CRB. To tackle the resulting non-convex problem, a joint optimization algorithm is proposed that integrates predictive envelope screening with successive convex approximation (SCA). By constructing a quadratic fractional form and a positive semidefinite affine lower bound for the Fisher matrix, and invoking the Schur complement, the beamforming subproblem under a fixed codebook is converted into a standard convex cone optimization problem, enabling efficient solution. Simulation results demonstrate that the proposed scheme achieves stable convergence within a modest number of iterations, improving the average effective rate by 62.9% over the maximum ratio transmission (MRT) benchmark. Moreover, the EKF-based tracking algorithm attains an average root mean square error of 1.698 1 m for three-dimensional aerial target localization. These two sets of results collectively validate the effectiveness of the proposed dual-time-scale AFDM-ISAC waveform design scheme in fulfilling both sensing and communication tasks.96
44
0
2026-11-10
Featured Articles
MOREPAPERS
A Combined Retrieval Method by Fusing Image and Text Features AI Introduction
QIN Yu-shu, YANG Liang-huai, ZHU Yan-chao, GONG Wei-huaVol. 53, Issue 2, Pages: 558-567(2025) DOI: 10.12263/DZXB.20240679Abstract:With the explosive growth of image data in the field of e-commerce, target image retrieval has become a challenging work in information retrieval research. The existing traditional image retrieval models only rely on a single text description or similar image, which is difficult to accurately capture the user’s retrieval intention, resulting in unsatisfactory retrieval results. In order to solve this problem, this paper proposes a combined retrieval method that fuses image and text features. Swin Transformer (SwinT) is used to extract the multi-layer features of the reference image, and the image and text features are fused at multiple levels, so that the text features can modify the reference image features at multi-level and fine-grained, and get closer to the target image features. Then, the modified image features and the target image features are embedded in a space for similarity measurement, and the batch-based classification loss is used to optimize the retrieval performance. Experimental results on Fashion200k, MIT-States and CSS datasets show that the proposed method improves the performance by 5 percentage points on average compared with the existing mainstream methods.1456
559
2
2025-12-08
PAPERS
Feature Anomaly Detection and Pseudo-Label Regression for Adversarial Domain Adaptation
PAN Jie, LIU Bo, ZOU Xiao-yuVol. 53, Issue 1, Pages: 128-140(2025) DOI: 10.12263/DZXB.20240074Abstract:In unsupervised domain adaptation tasks, the source and target domains usually do not satisfy the independent and identical distribution assumption. In order to generate the usable labels for the target domain, classical domain adaptation methods select the category with the highest prediction probability of the classifier as the pseudo-label of the target sample. Thus, the pseudo-label inevitably contains certain noise information, which may cause negative transfer to the domain adaptation model. In addition, traditional adversarial domain adaptation methods usually consider the global distribution between domains and ignore the category information of samples. How to extract discriminative category-level features in domain adaptation tasks is also an important problem. Therefore, an unsupervised adversarial domain adaptation method is proposed using feature anomaly detection and pseudo-label regression. The target samples of the same class predicted by the classifier are formed into the category subdomain within the target domain. The Gaussian uniform mixture model is used to detect the subdomain samples with abnormal distance from the class mean. The posterior probability of the samples is calculated and the correctness of the sample pseudo-labels in the subdomain is measured, which is used as a loss factor to limit the influence of pseudo-labels on the model in training. Meanwhile, the pseudo-label regression function is used to reduce the difference between the predicted label and the high-confidence pseudo-label of the classifier. The category constraint of the unlabeled target domain is adopted to improve the distinguishability of feature categories. Experimental results show that the average recognition accuracy of the proposed method on datasets Office-31, Image-CLEF, and Office-Home are 90.2%, 89.6%, and 69.5%, respectively, which are all higher than the related popular algorithms.912
452
4
2025-12-08
Special Focus: Recipient of CIE Outstanding Doctoral Dissertation
Non-Line-of-Sight Human Vital-Sign Sensing Aided by Reconfigurable Intelligent Surfaces
LI Xin-yu, ZHANG Jing-yuan, CAI Zi-xuan, QIN Shi-long, MA Qian, YOU Jian-wei, CUI Tie-junVol. 53, Issue 1, Pages: 1-13(2025) DOI: 10.12263/DZXB.20240674Abstract:This paper proposes a contactless human vital-sign sensing method in non-line-of-sight (NLoS) scenarios, which integrates reconfigurable intelligent surfaces (RISs) into wireless human sensing, and achieves NLoS human vital-sign sensing by utilizing the flexible manipulation capability of RISs over electromagnetic waves. Firstly, through a visual-aided mechanism, this paper locates human targets using the deep learning model Yolo-v7. Then, the metasurface optimizes the coding matrix based on the estimated human location, thereby altering the propagation path of sensing signals to achieve human sensing in non-line-of-sight scenarios. Additionally, addressing the shortcomings of traditional variational mode decomposition (VMD) algorithms in human vital-sign sensing, we further propose an improved VMD algorithm to achieve precise estimation of breathing rate and heartbeat rate. Experimental results demonstrate that the visual-aided module can accurately locate the position of human chest, thus enabling precise beamforming controlled by the intelligent metasurface. The proposed filtering-based VMD algorithm can achieve precise estimation of human vital signs, with average estimation errors of 0.6 RPM for respiration and 5.3 BPM for heartbeat, respectively. Furthermore, further analysis also demonstrates the effectiveness and accuracy of the RIS-based sensing scheme for human vital-sign sensing in NLoS scenarios.982
546
1
2025-12-08
PAPERS
Prototype Design and Implementation for Large-Scale Antenna Based Wireless Powered Communication System
CHE Yue-ling, LI Liang-zhu, ZENG Zhi-hao, LUO Sheng, WU Kai-shunVol. 51, Issue 12, Pages: 3397-3409(2023) DOI: 10.12263/DZXB.20230054Abstract:The wireless powered communication is a key technology to support energy sustainable wireless communications in the future 6G era. For the first time, this paper proposes a complete prototype design for the large-scale antenna based wireless powered communication system, and realizes this prototype system, which consists of a base station with a 32-transmit-antenna array, a user equipment and a feedback link, on the software defined radio platforms. At the base station, by designing flexible frame structures, this paper originally proposes a dynamic wireless energy transmission scheme to properly balance the required channel estimation accuracy and the wireless energy transmission efficiency. At the user equipment, by applying the practical non-linear energy harvesting and energy consumption models, this paper proposes a novel information-energy double threshold based battery management scheme to meet the requirements of different information and energy transmission. Based on the prompt status feedback from the user equipment, this paper completes a close-loop controlled system framework for adaptive wireless information and energy transmissions. Extensive experiments for both indoor and outdoor scenarios are implemented to verify the good environment adaptation and transmission performance of the wireless powered communication prototype system.648
401
0
2025-12-11
PAPERS
Edge Caching Scheme Based on Value of Information Age in the Internet of Vehicles
LIN Zhi-jian, HOU Ying, CAO Xiao-xiao, WU Wen-hao, CHEN Xiao-pei, CHEN Ping-pingVol. 51, Issue 12, Pages: 3410-3421(2023) DOI: 10.12263/DZXB.20221409Abstract:With the rapid development of the fifth generation (5G) mobile communications technology, Internet of vehicles (IoV) has become one of the important scenarios for 5G applications. Due to the urgent requirements of these applications for timely content caching, a dynamic content caching-enabled edge caching scheme is investigated to ensure timeliness, which allows vehicles to obtain cache content with freshness constraints from an edge caching-enabled access point (CAP). Furthermore, based on the idea of collaborative cache, a random model of CAP cache refreshing and delivery is established by probability theory. Meanwhile, considering the metric of hit rate with age of information (AoI) as the quality of service (QoS) index of IoV cache, a system model of caching performance value is developed. To ensure the effectiveness of content caching, two types of vehicular content requests with different AoI requirements are considered in a hierarchical slicing framework. Finally, the optimization problem to maximize the content caching value of cooperative content placement is formulated, which is a multi-objective, multi-dimensional and multi-choice knapsack problem and is proved to be NP-hard. Therefore, the problem is divided into several independent optimization problems, and a maximum incremental algorithm based on the age value of the content is designed to solve it. Simulation results show that the CAP cooperative content placement strategy outperforms other traditional methods.515
388
4
2025-12-11
PAPERS
An Attack Method Against the Modular Reduction Within a RSA-CRT Implementation Based on Template Attack
MA Xiang-liang, WU Li-ji, WANG Hong, ZHANG Xiang-min, HUANG Ke-zhen, LIU Yu-lingVol. 52, Issue 3, Pages: 689-695(2024) DOI: 10.12263/DZXB.20220175Abstract:At present, there are few researches on profile attacks against RSA-CRT implementation. This paper takes modular reduction operation as the research object, and a template attack method against RSA-CRT implementation is proposed. The core of this method is to solve the difficulty to recover the RSA-CRT private key from the Hamming weight of the intermediate value of ciphertext modular reduction. The characteristic of this method is to build a model based on the Hamming weight of the intermediate value derived from modular reduction. The Hamming weight can be obtained by collecting the power traces of chosen ciphertext modular reduction for template matching, and the intermediate value is recovered from the Hamming weight variation, the private key of the RSA-CRT algorithm can be further inferred based on the intermediate value. In addition, the advantage of this method is that ideally, templates based on the intermediate Hamming weight model can be shared, and there is no limit on the number of bits of the intermediate value for modelling, which can be in byte size, 64 bit size, or even the bit size of p. In the actual environment, it can be selected according to the leaked information. Finally, in this paper, the lowest byte of the intermediate value is selected to model to verify the feasibility of this method, and the defense suggestions are also provided.477
362
0
2025-12-11
01Global Dependency Guided Feature Reconstruction for Image Classification
YUAN Heng, WU Jingrui, ZHANG Shengchong322|1643|0
02Radar-Infrared Multi-View Cooperative Intelligent TBD Technology for Maritime Weak Target Detection
SUN Dian-xing, HUANG Ya-sheng, PENG Rui-hui, TAN Shun-cheng, WANG Guo-hong531|1601|0
03Tautologies and Quasi-Tautologies in Fuzzy Modal Logic
HU Ming-di, WANG Guo-jun133|1335|6
04Classification of Distributed GEP-BP Based on Grid Service
DENG Song, WANG Ru-chuan140|1324|2
XU Dan, WANG Qiang, JIANG Wen-li, ZHOU Yi-yu123|1321|1
06Construction of Non-Aliasing Pyramidal Transform
FENG Peng, WEI Biao, PAN Ying-jun, MI De-ling118|1303|2
07Study on Thermal Placement Optimization of Stacked 3D-MCM Based on Thermal Superposition Model
LIANG Ying, HUANG Chun-yue, YAN De-jin, LI Tian-ming121|1303|3
LIU Jing, GAO Yong116|1297|1
YANG Hong-bo, YU Zhong-yuan120|1287|0
10The Authentication Technology of P2P Network Based on Distributed PKI
DAI Zhan-feng, Wen Qiao-yan, LI Xiao-biao144|1285|3
Virtual Issues


































| Multi-media Resources |
| Read about our articles with data, invited articles, videos ...... |
媒体矩阵
icon

icon


icon



icon


iconMore
icon


Author Center


Author Guide


Subscription












