Avoiding Jammers: A Reinforcement Learning Approach

TL;DR

This paper explores reinforcement learning strategies for cognitive radar to avoid jamming, analyzing performance under uncertainty with novel network modifications, and demonstrating improved results through simulations.

Contribution

It introduces a new approach to anti-jamming using RL with a softmax operator, replacing the traditional target network, and provides analysis under a POMDP model.

Findings

Softmax operator enhances RL performance over traditional target networks.

Deep Q-network and LSTM strategies effectively analyze anti-jamming performance.

Explicit jammer dynamics uncertainty expression improves understanding of radar robustness.

Abstract

This paper investigates the anti-jamming performance of a cognitive radar under a partially observable Markov decision process (POMDP) model. First, we obtain an explicit expression for uncertainty of jammer dynamics, which paves the way for illuminating the performance metric of probability of being jammed for the radar beyond a conventional signal-to-noise ratio ($\mathsf{SNR}$) based analysis. Considering two frequency hopping strategies developed in the framework of reinforcement learning (RL), this performance metric is analyzed with deep Q-network (DQN) and long short term memory (LSTM) networks under various uncertainty values. Finally, the requirement of the target network in the RL algorithm for both network architectures is replaced with a softmax operator. Simulation results show that this operator improves upon the performance of the traditional target network.