کنفرانس مهندسی برق ایران

صفحه اصلی / سی و سومین کنفرانس بین المللی مهندسی برق

Optimal Receiver Placement in Distributed Passive Sensor Networks: A DRL Approach

نویسندگان :

Hossein Nikaein¹ Mohammad Reza Jabbari² Maedeh Kadkhodaie Elyaderani³ Saeed Gazor⁴

1- Isfahan University of Technology 2- Isfahan University of Technology 3- Isfahan University of Technology 4- Queen's University

کلمات کلیدی :

Deep Reinforcement Learning (DRL)،Distributed Passive Sensor Network (DPSN)،Receiver Placement

چکیده :

This paper addresses the optimal placement of Receivers (Rxs) in Distributed Passive Sensor Networks (DPSNs) to enhance the received Signal-to-Noise Ratio (SNR) in critical sub-regions within the surveillance area. We aim to maximize the average weighted SNR while ensuring a high minimum SNR across the entire area. We formulate a bi-objective optimization problem, which is solved numerically using Multi-Objective Genetic Algorithms (MOGA). Additionally, we develop a Deep Reinforcement Learning (DRL) framework using Proximal Policy Optimization (PPO) that learns optimal Rx placement strategies by balancing the dual objectives. Numerical simulations demonstrate that our approach effectively determines optimal Rx placements, enhancing SNR in high-priority areas while maintaining robust coverage and a uniform SNR distribution across the surveillance area.