Secure multiple adaptive kernel diffusion LMS algorithm for distributed estimation over sensor networks
Authors:
Open Access
- 20 September 2024
- journal article
- research article
- Published by Institution of Engineering and Technology (IET) in IET Wireless Sensor Systems
- Vol. 14 (6) , 477-483
Abstract
This paper introduces a kernel‐based approach to enhance the security of distributed estimation in the presence of adversary links. Adversary links often degrade distributed recovery algorithm performance in distributed estimation. The authors propose secure distributed estimation algorithms employing an adaptive kernel and adaptive combination coefficients derived from it. The authors’ method includes a multiple kernel approach with varied widths and a heuristic formula for combination coefficients, improving performance in the presence of adversary links. Additionally, the approach is extended to single exponential kernels with fixed and adaptive widths, treating them as special cases. The multiple kernel method is used because it provides more degrees of freedom compared to a single kernel, leading to better results. Simulation results show that the proposed multiple kernel approach achieves performance close to the diffusion least mean square algorithm in the absence of attacks. The adaptive nature of the kernel and coefficients enhances algorithm robustness, making it promising for secure distributed estimation in the presence of adversary links.Keywords
This publication has 22 references indexed in Scilit:
- A Novel Family of Robust Incremental Adaptive Algorithms for Distributed Estimation Based on Bregman DivergenceIEEE Sensors Letters, 2023
- Secure distributed estimation under Byzantine attack and manipulation attackEngineering Applications of Artificial Intelligence, 2022
- An adaptive unscented Kalman filter approach to secure state estimation for wireless sensor networksAsian Journal of Control, 2022
- Correction-based diffusion LMS algorithms for secure distributed estimation under attacksDigital Signal Processing, 2020
- A Secure Distributed Information Sharing Algorithm Based on Attack Detection in Multi-Task NetworksIEEE Transactions on Circuits and Systems I: Regular Papers, 2020
- Weighted diffusion continuous mixed p-norm algorithm for distributed estimation in non-uniform noise environmentSignal Processing, 2019
- Resilient Distributed Estimation: Sensor AttacksIEEE Transactions on Automatic Control, 2018
- Maximum Total Correntropy Diffusion Adaptation Over Networks With Noisy LinksIEEE Transactions on Circuits and Systems II: Express Briefs, 2018
- Secure Distributed Estimation Over Wireless Sensor Networks Under AttacksIEEE Transactions on Aerospace and Electronic Systems, 2018
- Distributed Spectrum Sensing for Cognitive Radio Sensor Networks Using Diffusion AdaptationIEEE Sensors Letters, 2017