Article
Engineering, Electrical & Electronic
Junjie Wu, Wei Chen
Summary: This paper studies the delay-optimal scheduling policy for energy harvesting-aided communications. The delay-minimal scheduling is obtained through Markov chain modeling and linear programming. A value iteration algorithm is presented to reduce computational complexity and a low complexity asymptotically optimal policy is proposed. A unified framework is presented for energy harvesting-aided communications with finite-capacity batteries based on large deviation theory.
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Mohammad Hatami, Markus Leinonen, Zheng Chen, Nikolaos Pappas, Marian Codreanu
Summary: This study focuses on a resource-constrained IoT network, where cache-enabled edge nodes handle on-demand requests from multiple users. By deciding whether to send fresh status updates or retrieve recent measurements from the cache, the edge nodes aim to minimize the average age of information of the received measurements. The study proposes two algorithms, an iterative algorithm and a relax-then-truncate algorithm, both of which successfully reduce the average on-demand age of information.
IEEE TRANSACTIONS ON COMMUNICATIONS
(2022)
Article
Computer Science, Information Systems
Meng-Lin Ku, Ting-Jui Lin
Summary: In this article, neural network-based transmit power control prediction schemes are designed for solar-powered energy harvesting communications, achieving satisfactory performance in both single-user and multi-user scenarios.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Mathematics, Applied
Yunbo Song, Dan Ye
Summary: This article explores the energy allocation problem for denial-of-service (DoS) attacks on remote state estimation in cyber-physical systems (CPSs), where attackers can disrupt packet transmission from sensors to remote estimators by harvesting energy. By maximizing estimation error covariance, the attacker adjusts jamming power based on energy uncertainties and limited battery storage, with imperfect feedback transformed into a Markov decision process. Optimal strategies for DoS attack energy allocation with imperfect acknowledgments and energy harvesting constraints are derived and verified through numerical examples.
APPLIED MATHEMATICS AND COMPUTATION
(2021)
Article
Computer Science, Information Systems
Wannian An, Chen Dong, Xiaodong Xu, Chao Xu, Shujun Han, Lei Teng
Summary: This article discusses a cooperative communication network that uses energy-harvesting decode-and-forward relays with buffers that harvest energy from the ambience. An opportunistic routing protocol is proposed to enhance data delivery in the network by selecting transmission paths based on node transmission priority. An algorithm based on state transition matrix is proposed to obtain the probability distribution of candidate broadcast node set. Theoretical expressions for the energy distribution in energy buffers and closed-form expressions for network outage probability and throughput are derived.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Computer Science, Information Systems
Kechen Zheng, Jiahong Wang, Xiaoying Liu, Xin-Wei Yao, Yang Xu, Jia Liu
Summary: This study investigates backscatter-aided energy harvesting cognitive radio networks in a multichannel scenario. A novel hybrid communication scheme is proposed, which is flexible in adapting to changes in energy and channel availabilities. Simulation results show that the proposed scheme performs superior in terms of throughput.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Computer Science, Information Systems
Tran Manh Hoang, Le Thi Thanh Huyen, Xuan Nam Tran, Pham Thanh Hiep
Summary: This work considers the use of a base station mounted on an unmanned aerial vehicle (UAV) called an aerial BS (ABS) to assist users in a nonorthogonal multiple access (NOMA) scheme. The ABS acts as a relay station and harvests communication power from ambient radio-frequency (RF) sources. The proposed system overcomes connectivity issues and energy limitations, and its performance can be improved by adjusting the UAV's altitude.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Information Systems
Kunyi Chen, Fatma Benkhelifa, Hong Gao, Julie A. McCann, Jianzhong Li
Summary: This work considers the AoI minimization scheduling problem in multihop energy harvesting wireless sensor networks. An energy-adaptive distributed algorithm is proposed and theoretical bounds are derived. Experimental results show that the algorithm outperforms other schemes in all scenarios.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Dipak Kumar Sah, Abhishek Hazra, Ramesh Kumar, Tarachand Amgoth
Summary: In this article, a modified Pro-energy prediction technique is proposed to control unnecessary errors in solar-based harvesting systems related to sensing devices. The proposed method uses prior energy measurements to show future energy status in respective time slots. Experimental observations validate the superior performance of the modified Pro-energy prediction technique compared to existing EMWA, WCMA, and Pro-energy methods.
IEEE SENSORS JOURNAL
(2023)
Article
Computer Science, Information Systems
Xiao-Ren Xu, Yi-Han Xu, Wen Zhou, Arumugam Nallanathan
Summary: This study investigates the resource allocation problem in Energy Harvesting-supported Cognitive Industrial Machine-to-Machine (EH-CI-M2M) network with the use of Unmanned Aerial Vehicles (UAVs) communication. The objective is to maximize the average energy efficiency by considering EH time slot assignment, transmit power control, and bandwidth allocation, while considering Quality of Service (QoS) and available energy status. The problem is approached through non-linear fractional programming and variable relaxation, and an iterative algorithm is proposed for optimization. Extensive simulation results demonstrate the superiority of the proposed scheme in terms of energy efficiency.
IEEE WIRELESS COMMUNICATIONS LETTERS
(2023)
Article
Computer Science, Information Systems
Kisong Lee, Hyun-Ho Choi, Woongsup Lee, Victor C. M. Leung
Summary: In this article, the authors investigate wireless-powered two-way communication in the presence of imperfect channel state information. They propose resource allocation schemes using a gradient algorithm and deep learning, and verify their performance through simulations.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Energy & Fuels
Ifrah Tahir, Ali Nasir, Abdullah Algethami
Summary: This paper focuses on developing a customized energy management system for a bank branch based on the structure and nature of activities in the building. A Markov Decision Process (MDP) model is proposed and solved using stochastic dynamic programming to generate an optimal control policy. The unique feature of the proposed model is its specificity for a bank building. Practical implementation and a case study are discussed to demonstrate the effectiveness of the model in achieving energy savings without compromising comfort.
Article
Engineering, Electrical & Electronic
Bao Zhao, Jiahua Wang, Wei-Hsin Liao, Junrui Liang
Summary: Piezoelectric transducers offer bidirectional energy conversion between mechanical and electrical forms, which can be utilized in various applications. This article introduces a bidirectional energy conversion circuit solution for time-sharing energy harvesting and vibration exciting purposes, achieved by developing a new control strategy on the synchronized triple bias-flip (S3BF) piezoelectric interface circuit.
IEEE TRANSACTIONS ON POWER ELECTRONICS
(2021)
Article
Computer Science, Information Systems
Xi Wei, Qi Zhu
Summary: Energy harvesting technology can solve the problem of users' energy consumption in disaster areas. This paper proposes a relay-based on simultaneous wireless information and power transfer (SWIPT) energy harvesting and information transmission time allocation optimization algorithm. Simulation results show that the proposed algorithm can provide energy for trapped users and improve data transmission rate.
Article
Engineering, Electrical & Electronic
Shengda Tang, Liansheng Tan, Tao Liu
Summary: A novel stochastic model, the extended Markov fluid flow model, is proposed for EH-WCS with reliable energy backup (REB), providing stationary distributions of key performance metrics and numerical analysis to study the effects of system parameters on performance. Both theoretical insights and numerical analyses are important for the design of EH-WCSs.
INTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS
(2021)
Article
Telecommunications
Haitao Zhao, Xiaoqing Li, Huiling Cheng, Jun Zhang, Qin Wang, Hongbo Zhu
Summary: This paper proposes an algorithm based on a deep convolutional neural network and a random forest to predict accident risks, and simulation results show that it outperforms existing algorithms in terms of AUC and accuracy.
CHINA COMMUNICATIONS
(2022)
Article
Materials Science, Textiles
Jun Zhang, Newman M. L. Lau, Yue Sun, Joanne Yip, Kit-Lun Yick, Winnie Yu, Jianming Chen
Summary: This study presents a nonlinear finite element model to simulate elderly breast deformation during arm abduction. The model has been validated and combined with questionnaire analysis, providing a basis for designing sports bras that are ergonomic.
TEXTILE RESEARCH JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Yifan Ma, Yifei Shen, Xianghao Yu, Jun Zhang, S. H. Song, Khaled B. Letaief
Summary: In this paper, a scalable and generalizable neural calibration framework for future wireless system design is proposed. The framework adopts a neural network to calibrate the input of conventional model-based algorithms and utilizes the permutation equivariance property of the topological structure to develop a generalizable neural network architecture. Simulation results show that the proposed approach enjoys significantly improved scalability and generalization compared with existing learning-based methods.
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Lumin Liu, Jun Zhang, Shenghui Song, Khaled B. Letaief
Summary: Federated learning (FL) is a powerful distributed machine learning framework that aggregates models trained by different clients without accessing their private data. Hierarchical FL, with a client-edge-cloud aggregation hierarchy, achieves high communication efficiency by leveraging the cloud server's access to multiple clients' data and the edge servers' proximity to the clients. Neural network quantization reduces communication overhead during model uploading.
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Jiawei Shao, Yuyi Mao, Jun Zhang
Summary: This paper investigates task-oriented communication for multi-device cooperative edge inference by proposing a learning-based communication scheme. The scheme optimizes local feature extraction and distributed feature encoding to reduce communication overhead and latency. The proposed method achieves a better rate-relevance tradeoff than existing methods for multi-view image classification and multi-view object recognition tasks.
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Wenwen Zhang, Chengdong Dong, Jun Zhang, Hangguan Shan, Eryun Liu
Summary: One-Shot Object Detection (OSOD) aims to detect novel object categories without fine-tuning by using query features to guide the detection process. Existing attention-based models face challenges in comprehensively utilizing context, handling cross-scale interactions, and addressing spatial misalignment issues. To address these problems, this paper proposes an adaptive context-and-scale-aware feature aggregation module (ACS) and employs a spatial transformer network (STN) to align features. Experimental results demonstrate the effectiveness of the proposed approach, outperforming baseline methods and achieving state-of-the-art results.
Article
Environmental Sciences
Jun Zhang, Ruixin Liang, Newman Lau, Qiwen Lei, Joanne Yip
Summary: This study developed a model order reduction approach to predict the real-time strain of senior women's breast skin during movement. By using motion capture experiments, eight body variables of the seniors were extracted and backpropagation artificial neural networks were built to predict the strain of the breast skin. After optimization, the R-value for the neural network model reached 0.99, showing acceptable accuracy. The computer-aided system of this study is validated as a robust simulation approach for conducting biomechanical analyses and predicting breast deformation.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2023)
Article
Engineering, Electrical & Electronic
Yifei Shen, Jun Zhang, S. H. Song, Khaled B. Letaief
Summary: Deep learning-based approaches have shown promise in wireless communications, but early attempts with neural network architectures from other domains performed poorly in large scale and unseen network settings. Graph neural networks (GNNs) have been adopted to exploit the domain knowledge in wireless communications and achieve near-optimal performance in large-scale networks with good generalization. However, the theoretical foundations and design guidelines are still unclear, hindering practical implementations.
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS
(2023)
Article
Computer Science, Information Systems
Yuchang Sun, Jiawei Shao, Yuyi Mao, Jessie Hui Wang, Jun Zhang
Summary: In this paper, we investigate a novel FEEL framework called semi-decentralized federated edge learning (SD-FEEL), where multiple edge servers collectively coordinate a large number of client nodes. By exploiting low-latency communication among edge servers, SD-FEEL incorporates more training data while achieving lower latency compared to conventional federated learning.
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2023)
Article
Computer Science, Information Systems
Xinyu Bian, Yuyi Mao, Jun Zhang
Summary: In the mMTC scenario, grant-free random access has emerged as a promising mechanism, but its potential has not been fully exploited. This article develops advanced receivers to improve the massive access performance by jointly designing activity detection, channel estimation, and data decoding. A turbo structure is adopted to tackle the algorithmic and computational challenges. Simulation results show that the turbo receiver effectively reduces errors and supports more active users compared to separate designs. Additionally, a low-cost SI-aided receiver outperforms conventional methods with lower complexity.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Zijian Li, Jiawei Shao, Yuyi Mao, Jessie Hui Wang, Jun Zhang
Summary: Federated learning (FL) is a popular privacy-preserving collaborative learning paradigm. However, it faces the challenge of non-independent and identically distributed (non-IID) data among clients. In this chapter, a novel framework called Synthetic Data Aided Federated Learning (SDA-FL) is proposed to address this non-IID challenge by sharing synthetic data. The framework includes local generative adversarial networks (GANs) for generating differentially private synthetic data and an iterative pseudo labeling mechanism for generating confident pseudo labels.
TRUSTWORTHY FEDERATED LEARNING, FL 2022
(2023)
Article
Computer Science, Information Systems
Long Chen, Jigang Wu, Jun Zhang, Hong-Ning Dai, Xin Long, Mianyang Yao
Summary: This study proposes a dependency-aware offloading scheme in MEC with edge-cloud cooperation under task dependency constraints. By dividing the offloading problem into two sub-problems and designing greedy algorithms, it achieves near-optimal performance and improved application finishing time. The simulation results and real-world experiments validate the effectiveness of the proposed algorithm.
IEEE TRANSACTIONS ON CLOUD COMPUTING
(2022)
Proceedings Paper
Telecommunications
Mianyang Yao, Long Chen, Jun Zhang, Jiale Huang, Jigang Wu
Summary: This paper investigates the impact of loading cost and dynamic user requests on caching and request routing in edge computing. It formulates a system throughput maximization problem and proposes a randomized rounding-based online algorithm to solve it. Experimental results demonstrate the effectiveness of the algorithm, achieving more than 42.7% throughput gain compared to baseline algorithms.
IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC 2022)
(2022)
Proceedings Paper
Telecommunications
Yuchang Sun, Jiawei Shao, Yuyi Mao, Jun Zhang
Summary: In this study, a novel semi-decentralized FEEL (SD-FEEL) architecture is investigated, where multiple edge servers collaborate to incorporate more data from edge devices in training. An asynchronous training algorithm is proposed to overcome the heterogeneity issue and a staleness-aware aggregation scheme is designed. Simulation results demonstrate the effectiveness of the proposed algorithm in achieving faster convergence and better learning performance.
IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC 2022)
(2022)
Proceedings Paper
Computer Science, Information Systems
Yuchang Sun, Jiawei Shao, Yuyi Mao, Jessie Hui Wang, Jun Zhang
Summary: In this paper, a novel framework called semi-decentralized federated edge learning (SD-FEEL) is proposed to improve the learning performance of federated edge learning (FEEL) by allowing model aggregation across different edge clusters. The training algorithm for SD-FEEL is presented and its convergence on non-independent and identically distributed (non-IID) data is proved. Experimental results confirm the effectiveness of SD-FEEL in achieving faster convergence compared to traditional federated learning architectures.
2022 IEEE WIRELESS COMMUNICATIONS AND NETWORKING CONFERENCE (WCNC)
(2022)
