Distributed Convex Optimization "Over-the-Air" in Dynamic Environments

被引:3
作者
Agrawal, Navneet [1 ]
Cavalcante, Renato Luis Garrido [2 ]
Yukawa, Masahiro [3 ]
Stanczak, Slawomir [2 ]
机构
[1] Tech Univ Berlin, Network Informat Theory Grp, D-10587 Berlin, Germany
[2] Heinrich Hertz Inst Nachrichtentech Berlin GmbH, Fraunhofer Inst Telecommun, D-10587 Berlin, Germany
[3] Keio Univ, Dept Elect & Elect Engn, Yokohama 2238522, Japan
来源
IEEE TRANSACTIONS ON SIGNAL AND INFORMATION PROCESSING OVER NETWORKS | 2024年 / 10卷
关键词
Adaptive algorithms; decentralized; distributed optimization; over-the-air function computation; superiorization; time-varying network; PROJECTED SUBGRADIENT METHOD; CONSENSUS ALGORITHMS; NETWORK; COMPUTATION; SET;
D O I
10.1109/TSIPN.2024.3423668
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
This paper presents a decentralized algorithm for solving distributed convex optimization problems in dynamic networks with time-varying objectives. The unique feature of the algorithm lies in its ability to accommodate a wide range of communication systems, including previously unsupported ones, by abstractly modeling the information exchange in the network. Specifically, it supports a novel communication protocol based on the "over-the-air" function computation (OTA-C) technology, that is designed for an efficient and truly decentralized implementation of the consensus step of the algorithm. Unlike existing OTA-C protocols, the proposed protocol does not require the knowledge of network graph structure or channel state information, making it particularly suitable for decentralized implementation over ultra-dense wireless networks with time-varying topologies and fading channels. Furthermore, the proposed algorithm synergizes with the "superiorization" methodology, allowing the development of new distributed algorithms with enhanced performance for the intended applications. The theoretical analysis establishes sufficient conditions for almost sure convergence of the algorithm to a common time-invariant solution for all agents, assuming such a solution exists. Our algorithm is applied to a real-world distributed random field estimation problem, showcasing its efficacy in terms of convergence speed, scalability, and spectral efficiency. Furthermore, we present a superiorized version of our algorithm that achieves faster convergence with significantly reduced energy consumption compared to the unsuperiorized algorithm.
引用
收藏
页码:610 / 625
页数:16
相关论文
共 52 条
[1]  
Abari O, 2016, Arxiv, DOI arXiv:1612.02307
[2]  
Agrawal N., 2023, P IEEE INT C AC SPEE, P1
[3]   A Scalable Max-Consensus Protocol For Noisy Ultra-Dense Networks [J].
Agrawal, Navneet ;
Frey, Matthias ;
Stanczak, Slawomir .
2019 IEEE 20TH INTERNATIONAL WORKSHOP ON SIGNAL PROCESSING ADVANCES IN WIRELESS COMMUNICATIONS (SPAWC 2019), 2019,
[4]   Blind Federated Edge Learning [J].
Amiri, Mohammad Mohammadi ;
Duman, Tolga M. ;
Gunduz, Deniz ;
Kulkarni, Sanjeev R. ;
Poor, H. Vincent .
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2021, 20 (08) :5129-5143
[5]   Federated Learning Over Wireless Fading Channels [J].
Amiri, Mohammad Mohammadi ;
Gunduz, Deniz .
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2020, 19 (05) :3546-3557
[6]  
[Anonymous], 2008, ADV NEURAL INFORM PR, DOI DOI 10.5555/2981562.2981710
[7]  
Bapat R.B., 1997, Encyclopedia of mathematics and its applications
[8]  
Bauschke HH, 2011, CMS BOOKS MATH, P1, DOI 10.1007/978-1-4419-9467-7
[9]  
Bjelakovic I, 2019, ANN ALLERTON CONF, P1146, DOI [10.1109/allerton.2019.8919875, 10.1109/ALLERTON.2019.8919875]
[10]   Energy-based sensor network source localization via projection onto convex sets [J].
Blatt, Doron ;
Hero, Alfred O., III .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2006, 54 (09) :3614-3619