Chinese Journal of Electronics最新文献_第5页

Nearly-Tight Bounds for Private Stochastic Convex Optimization and Sparse Learning Under Heavy Tails 重尾条件下私有随机凸优化与稀疏学习的近紧界

IF 4.5 4区计算机科学

Chinese Journal of Electronics Pub Date : 2026-07-01 Epub Date: 2026-09-09 DOI: 10.23919/cje.2025.00.225

Youming Tao;Di Wang;Xiuzhen Cheng

{"title":"Nearly-Tight Bounds for Private Stochastic Convex Optimization and Sparse Learning Under Heavy Tails","authors":"Youming Tao;Di Wang;Xiuzhen Cheng","doi":"10.23919/cje.2025.00.225","DOIUrl":"https://doi.org/10.23919/cje.2025.00.225","url":null,"abstract":"We study differentially private stochastic convex optimization (DP-SCO) under heavy-tailed data, moving beyond standard assumptions like Lipschitz losses or gradients with bounded second moments. We focus on the setting where each gradient coordinate has only a bounded <tex>$(1+v)$</tex>-th moment for some <tex>$vin(0,1]$</tex>, allowing much heavier tails. We first propose an optimal private and robust mean estimator for the one-dimensional case, with matching upper and lower bounds. We then extend this to <tex>$d$</tex> dimensions, introducing a new high-dimensional differential privacy (DP) mean estimator with strong accuracy guarantees. A key technical result is a new concentration inequality for distributions with bounded <tex>$(1+v)$</tex>-th moments, of independent interest. Leveraging these tools, we design DP-SCO algorithms for convex and strongly convex settings, achieving nearly-tight error bounds. To address high-dimensionality, we study DP-SCO with a sparsity constraint (DP sparse learning), obtaining error rates logarithmic in <tex>$d$</tex> and optimal up to a factor of <tex>$O(s^{ast})$</tex>, where <tex>$s^{ast}$</tex> is the sparsity level. Finally, we validate our results with numerical experiments.","PeriodicalId":50701,"journal":{"name":"Chinese Journal of Electronics","volume":"35 4","pages":"1728-1740"},"PeriodicalIF":4.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11685215","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148909466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-Resolution Multi-Scale Graph Convolutional Networks for Fast Steady-State Thermal Prediction in Multi-Chip Integrated Systems 用于多芯片集成系统快速稳态热预测的多分辨率多尺度图卷积网络

IF 4.5 4区计算机科学

Chinese Journal of Electronics Pub Date : 2026-07-01 Epub Date: 2026-09-09 DOI: 10.23919/cje.2025.00.346

Chengxin Zhang;Yujie Liu;Quan Chen

{"title":"Multi-Resolution Multi-Scale Graph Convolutional Networks for Fast Steady-State Thermal Prediction in Multi-Chip Integrated Systems","authors":"Chengxin Zhang;Yujie Liu;Quan Chen","doi":"10.23919/cje.2025.00.346","DOIUrl":"https://doi.org/10.23919/cje.2025.00.346","url":null,"abstract":"With the increasing adoption of heterogeneous multi-chip integrated systems in high-performance computing, thermal coupling and non-uniform temperature distributions have become critical concerns, demanding fast and accurate steady-state thermal prediction methods. While recent approaches employing neural networks-such as convolutional neural networks and graph convolutional networks (GCNs)-offer improved computational speed, they often suffer from limitations in prediction accuracy and structural adaptability. To address these challenges, this paper proposes a multi-scale GCN-based method for rapid thermal field prediction. The proposed approach constructs graph structures by treating each pixel in the input maps as a node and dynamically aggregates node features across adjacency relationships defined at multiple spatial scales (scale 1, 2, and 3). This multi-scale, dynamic aggregation mechanism enables capturing thermal propagation patterns effectively at varying spatial resolutions. The network consists of a multi-scale encoder and a GCN-based decoder, trained with Softplus activation and dropout regularization. A simple linear iterative clustering-augmented power/material-discontinuity adaptive partition is developed to enhance boundary fidelity with fewer nodes, at a modest runtime cost. Furthermore, using a dataset we developed comprising over 200 chiplet configurations with diverse and flexible power-material layouts, the proposed method achieves a temperature prediction mean error within 0.68 °C. This approach delivers both high accuracy and computational efficiency, offering a viable solution for thermal-aware modeling and rapid optimization in complex multi-chip systems.","PeriodicalId":50701,"journal":{"name":"Chinese Journal of Electronics","volume":"35 4","pages":"1589-1598"},"PeriodicalIF":4.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11685237","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148909444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An Overview of Chiplet and Integrated Chip Testing Technologies 小片与集成芯片测试技术综述

IF 4.5 4区计算机科学

Chinese Journal of Electronics Pub Date : 2026-07-01 Epub Date: 2026-09-09 DOI: 10.23919/cje.2025.00.160

Zhao Zhang;Pengrong Lin;Zixiong Song;Tiening Wang;Ruohan Zhang;Peng Gao;Yimao Cai;Yong Wang

{"title":"An Overview of Chiplet and Integrated Chip Testing Technologies","authors":"Zhao Zhang;Pengrong Lin;Zixiong Song;Tiening Wang;Ruohan Zhang;Peng Gao;Yimao Cai;Yong Wang","doi":"10.23919/cje.2025.00.160","DOIUrl":"https://doi.org/10.23919/cje.2025.00.160","url":null,"abstract":"Chiplet and integrated chip technologies, one of the mainstream methods to enhance chip performance in the post-Moore era, have garnered significant attention from academia and industry. Testing technology plays a crucial role in ensuring the reliability and functionality of the product, and is a key area that requires breakthroughs in the development of chiplets and integrated chips. This paper offers a comprehensive overview of chiplet and integrated chip testing technologies, analyzing the current research status and challenges encountered in this critical field. In this paper, we first explore the testing process of integrated chips, including design for test (DfT), Known Good Die testing, integrated module testing, final testing, and system-level testing. The current research trends in integrated chip testing are reviewed, focusing on system-level DfT techniques and inter-chiplet interconnection testing methods. This article also introduces the testing equipment and fixtures required for integrated chip testing, as well as electronic design automation tools. Finally, the challenges and future directions faced by integrated chip testing technology are explored.","PeriodicalId":50701,"journal":{"name":"Chinese Journal of Electronics","volume":"35 4","pages":"1607-1630"},"PeriodicalIF":4.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11685238","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148909446","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Novel Differentially Private Low Rank Weight Fusion Approach 一种新的差分私有低秩权融合方法

IF 4.5 4区计算机科学

Chinese Journal of Electronics Pub Date : 2026-07-01 Epub Date: 2026-09-09 DOI: 10.23919/cje.2024.00.123

Yanqing Yao;Yingdong Wang;Chao Huang;Wenqi Ji

{"title":"A Novel Differentially Private Low Rank Weight Fusion Approach","authors":"Yanqing Yao;Yingdong Wang;Chao Huang;Wenqi Ji","doi":"10.23919/cje.2024.00.123","DOIUrl":"https://doi.org/10.23919/cje.2024.00.123","url":null,"abstract":"Differential privacy is widely applied in deep learning to mitigate the risk of sensitive information leakage. However, additional privacy protection measures, such as per-example gradient clipping, introduces extra computational overhead and storage requirements, which impede the exploration of differentially private deep learning methods. To address this challenge, we propose a novel differentially private low-rank weight fusion approach. Our approach utilizes a model weight composed of a shared weight and low-rank weights. The rank of the low-rank weights determines the number of sub-networks involved in the fusion process. During the training phase, we fix the shared weight and solely train the low-rank weights. After each training batch, we add noise to safeguard the privacy of the low-rank weight fusion operation, thereby achieving privacy-preserving training. This approach significantly reduces storage overhead and eliminates the need for per-example gradient computations. Additionally, the low-rank weight can be applied as a plug-in, particularly in transfer learning scenarios. We evaluate the knowledge transfer capability of our method and conduct experiments in both self-learning and transfer learning scenarios. Our results demonstrate that our approach achieves comparable accuracy to the batchensemble method without introducing additional noise.","PeriodicalId":50701,"journal":{"name":"Chinese Journal of Electronics","volume":"35 4","pages":"1756-1764"},"PeriodicalIF":4.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11685249","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148909463","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Formation-Guided Multimodal Representation Learning for Group Action Quality Assessment 群体行为质量评价的形成导向多模态表征学习

IF 4.5 4区计算机科学

Chinese Journal of Electronics Pub Date : 2026-07-01 Epub Date: 2026-09-09 DOI: 10.23919/cje.2025.00.486

Zexing Du;Wei Bo;Yumei Zhang;Xiaojun Wu

引用次数: 0

A Multifunctional Amplitude-Phase Control MMIC Based on 0.18 µm GaAs E/D pHEMT Technology 基于0.18µm GaAs E/D pHEMT技术的多功能幅相控制MMIC

IF 4.5 4区计算机科学

Chinese Journal of Electronics Pub Date : 2026-07-01 Epub Date: 2026-09-09 DOI: 10.23919/cje.2025.00.335

Liulin Hu;Qian Lin;Shuyue Yang;Haifeng Wu;Ruilan Yang

{"title":"A Multifunctional Amplitude-Phase Control MMIC Based on 0.18 µm GaAs E/D pHEMT Technology","authors":"Liulin Hu;Qian Lin;Shuyue Yang;Haifeng Wu;Ruilan Yang","doi":"10.23919/cje.2025.00.335","DOIUrl":"https://doi.org/10.23919/cje.2025.00.335","url":null,"abstract":"This paper presents the design of an amplitude-phase control multifunctional chip to address the requirements of wideband communication systems operating in the 2 GHz-6 GHz range. Fabricated using Gallium Arsenide enhancement- and depletion-mode pseudomorphic high electron mobility transistor (GaAs E/D pHEMT) technology, the chip integrates several key components including three single-pole-double-throw (SPDT) switches, three drive amplifiers, a 6-bit digital phase shifter, a 6-bit digital attenuator, two equalizers and a 12-bit level shift circuit. Notably, the design incorporates a spiral quadrature hybrid and reflection-type phase shifter, which ensures superior phase-shifting performance. Both amplifiers utilize a parallel feedback architecture to achieve a positive gain slope across the operating frequency range, effectively mitigating low-frequency gain challenges. Test results demonstrate that the chip delivers a receiving/transmitting gain greater than 14 dB, with a positive gain slope exceeding 3 dB. The 6-bit phase shifter exhibits a root-mean-square (RMS) phase error of less than 3°, while the 6-bit attenuator has an RMS amplitude error below 0.5 dB. The entire circuit is integrated into a compact area of 5 mm × 5 mm, with a direct current power consumption of 0.65 W. This multifunctional chip enables high-precision amplitude and phase control of radio frequency signals, making it highly suitable for widespread deployment use in transmitter/receiver modules.","PeriodicalId":50701,"journal":{"name":"Chinese Journal of Electronics","volume":"35 4","pages":"1660-1668"},"PeriodicalIF":4.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11685231","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148909422","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Cross-Layer Error Modeling and Deployment of Compressed Sensing Object Detection on Compute-in-Memory Systems 内存计算系统中压缩感知目标检测的跨层误差建模与部署

IF 4.5 4区计算机科学

Chinese Journal of Electronics Pub Date : 2026-07-01 Epub Date: 2026-09-09 DOI: 10.23919/cje.2025.00.501

Saike Zhu;Xiaoqian Huang;Yong Gong;Shifan Gao;Xianggao Wang;Hua Chen;Cimang Lu;Yi Zhao

引用次数: 0

A Novel Automatic Pole Determination Method for Macromodeling Multi-Port Interconnect Systems 一种新的多端口互连系统宏建模自动极点确定方法

IF 4.5 4区计算机科学

Chinese Journal of Electronics Pub Date : 2026-07-01 Epub Date: 2026-09-09 DOI: 10.23919/cje.2025.00.280

Zixing Ye;Hongshun Yin;Wenyong Zhou;Dawei Wang;Wensheng Zhao

引用次数: 0

Dynamic Vehicle Detection by Pose Estimation and Model Matching Using LiDAR Data 基于姿态估计和模型匹配的激光雷达动态车辆检测

IF 4.5 4区计算机科学

Chinese Journal of Electronics Pub Date : 2026-07-01 Epub Date: 2026-09-09 DOI: 10.23919/cje.2025.00.149

Kaiqi Liu;Fuxiang Kang;Shida Nie;Yuanyuan Deng;Wei Li;Jianqiang Wang

{"title":"Dynamic Vehicle Detection by Pose Estimation and Model Matching Using LiDAR Data","authors":"Kaiqi Liu;Fuxiang Kang;Shida Nie;Yuanyuan Deng;Wei Li;Jianqiang Wang","doi":"10.23919/cje.2025.00.149","DOIUrl":"https://doi.org/10.23919/cje.2025.00.149","url":null,"abstract":"Dynamic vehicle detection is of great significance for perceiving the dynamic development trend of the environment and plays an important role in construction of static environment and accurate localization. To detect vehicles robustly and quickly in various scenarios without training a large amount of data, a vehicle detection approach based on a novel pose estimation method is proposed using LiDAR remotely-sensed data. The distribution characteristics of point clouds on a vehicle surface are analyzed and described by normal vector histogram features. Considering the characteristics of vehicle surfaces, a vehicle measurement model consisting of four rectangles is proposed. Combined with the Kullback-Leibler divergence measure, the fitting degree between the point cloud clusters and a measurement model can be obtained, and the position and attitude of a candidate vehicle can be determined. With the estimated pose information, vehicle motion characteristics are utilized to determine the candidate vehicles. Finally, experiments on karlsruhe institute of technology and toyota technological institute datasets are implemented to verify the superiority of the proposed detection approach. Particularly, in distant areas, the F<inf>1</inf> score is increased by 9.6% compared with the previous pose estimation based on the coherent point drift method.","PeriodicalId":50701,"journal":{"name":"Chinese Journal of Electronics","volume":"35 4","pages":"1421-1435"},"PeriodicalIF":4.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11685212","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148909457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Safety-Enhanced Reinforcement Learning Framework for Vision-Language Models in Autonomous Driving 基于安全增强的自动驾驶视觉语言模型强化学习框架

IF 4.5 4区计算机科学

Chinese Journal of Electronics Pub Date : 2026-07-01 Epub Date: 2026-09-09 DOI: 10.23919/cje.2025.00.214

Shiqi Gao;Qi Song;Tianxiang Gong;Runhua Xu;Haoyi Zhou;Chao Li;Jianxin Li

{"title":"A Safety-Enhanced Reinforcement Learning Framework for Vision-Language Models in Autonomous Driving","authors":"Shiqi Gao;Qi Song;Tianxiang Gong;Runhua Xu;Haoyi Zhou;Chao Li;Jianxin Li","doi":"10.23919/cje.2025.00.214","DOIUrl":"https://doi.org/10.23919/cje.2025.00.214","url":null,"abstract":"While vision-language models (VLMs) demonstrate remarkable capabilities in autonomous driving applications, they face significant challenges in safety-critical environments due to their insufficient risk awareness, susceptibility to visual distractions, and challenges in effectively balancing safety requirements with learning efficiency. To mitigate these challenges, we propose safe-GRPO, a safe-aware group relative policy optimization framework specifically designed for safe decision-making in VLM-based driving systems. Safe-GRPO introduces three key innovations: i) an uncertainty-aware, safety-sensitive reward mechanism that dynamically adjusts reward and penalty scales according to both risk severity and model uncertainty; ii) an uncertainty-guided noise-enhanced training method that adaptively improves robustness against visual interferences; and iii) a risk-aware dynamic clipping strategy that controls policy updates based on scenario-specific risks. Extensive experiments on an enhanced real-world scenarios with driving theory tests dataset (3926 risk-annotated scenarios) show that safe-GRPO significantly outperforms existing approaches in decision accuracy, robustness, and risk prediction accuracy-achieving a 75.14% safe decision rate in high-risk scenarios, demonstrating notable advancements towards reliable real-world autonomous driving applications.","PeriodicalId":50701,"journal":{"name":"Chinese Journal of Electronics","volume":"35 4","pages":"1493-1506"},"PeriodicalIF":4.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11684963","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148909461","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0