期刊
PHYSICAL COMMUNICATION
卷 47, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.phycom.2021.101381
关键词
UAV; Federated learning; Latency; Energy consumption; Mobile edge computing
资金
- International Science and Technology Cooperation Projects of Guangdong Province [2020A0505100060]
- research program of Guangzhou University [YK2020008/YJ2021003]
- Natural Science Foundation of Guangdong Province [2021A1515011392]
This paper investigates the optimization of federated edge learning in UAV-enabled IoT networks, considering the limited batteries of devices. The study proposes a strategy based on deep deterministic policy gradient to jointly allocate computational resources and wireless bandwidth, aiming to reduce system cost and improve FEEL performance effectively. Simulation results demonstrate that the proposed strategy outperforms conventional methods by enabling devices to complete all rounds of FEEL with limited batteries.
In this paper, we investigate how to optimize the federated edge learning (FEEL) in unmanned aerial vehicle (UAV)-enabled Internet of Things (IoT) for B5G/6G networks. Federated learning is an effective framework to train a shared model between decentralized edge devices and servers without exchanging raw data, which helps protecting data privacy. In UAV-enabled IoT networks, latency and energy consumption are two important metrics limiting the performance of FEEL. Although most of the existing works have studied how to reduce latency and improve energy efficiency, only few of them have investigated the impact of the devices' limited batteries on FEEL. Motivated by this, we study the battery-constrained FEEL, where the UAVs can adjust their operating CPU-frequencies to prolong battery life and avoid dropping from federated learning training untimely. We optimize the system by jointly allocating the computational resources and wireless bandwidth in time-varying environments based on a deep deterministic policy gradient (DDPG) based strategy, where a linear combination of latency and energy consumption is used to evaluate the system cost. In this end, simulation results are demonstrated to show that the proposed strategy outperforms the conventional ones. In particular, it enables all the devices to complete all rounds of FEEL with limited batteries, and reduce the system cost effectively in the meantime. (C) 2021 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
Article
Engineering, Electrical & Electronic
Attention driven CWT-deep learning approach for discrimination of Radar PRI modulation
Purabi Sharma, Kandarpa Kumar Sarma
Summary: Electronic Warfare (EW) is becoming increasingly important with the proliferation of radio frequency (RF) systems and radar applications. This paper proposes an automatic approach based on continuous wavelet transform and a combination of convolutional neural network, multi-head self-attention mechanism, and long short-term memory for recognizing different types of pulse repetition interval modulation. The proposed method enhances performance and achieves robustness in noise-filled and imperfect channel knowledge environment.
PHYSICAL COMMUNICATION (2024)
Article
Engineering, Electrical & Electronic
Multipaths' statistics for scatterers with inverted elliptic-parabolic spatial density around the mobile
Chidera Linda Anioke, Chibuzo Joseph Nnonyelu
Summary: In this paper, the joint and marginal probability densities of multipaths' angles-of-arrival (AOA) and times-of-arrival (TOA) at the cellular base station are derived. A novel assumption is made where scatterers are assumed to be located in an elliptical region instead of a circular region to better model the elliptical footprint around the mobile station. The proposed model shows a better fit to empirical AOA data compared to existing models.
PHYSICAL COMMUNICATION (2024)
Article
Engineering, Electrical & Electronic
Intelligent UAV planning for task-offloading with limited buffer and multiple computing servers
Xuefeng Chen, Rui Ma
Summary: This paper proposes a UAV-enabled MEC network to optimize transmission delay, computation delay, and system energy consumption, and achieves the optimal solution through an intelligent optimization algorithm based on Deep Dueling Double Q-Network and Twin Delayed Deep Deterministic Policy Gradient.
PHYSICAL COMMUNICATION (2024)
Article
Engineering, Electrical & Electronic
One-bit spectrum sensing using Gustafson-Kessel fuzzy clustering for cognitive radio network
Saikat Majumder
Summary: This article addresses the problem of one-bit spectrum sensing in cognitive radio networks and proposes a GKFCM clustering-based algorithm for detecting primary user signals in white and correlated noise. By identifying conserved features in the quantization process and clustering decision vectors, the proposed algorithm improves detection performance.
PHYSICAL COMMUNICATION (2024)
Article
Engineering, Electrical & Electronic
A New Efficient Multi-Channel Fast NLMS (MC-FNLMS) Adaptive Algorithm for Audio Teleconferencing systems
Mohamed Zerouali, Mohamed Djendi
Summary: This paper proposes a new multichannel fast normalized least mean square (MC-FNLMS) algorithm for audio teleconferencing systems. The algorithm improves multichannel adaptive noise reduction by introducing a first order prediction process on the input signals. Compared with existing algorithms, the proposed algorithm demonstrates faster convergence speed, better noise reduction performance, and lower computational complexity.
PHYSICAL COMMUNICATION (2024)