Abstract:In contemporary warfare, radar countermeasures have become multifunctional and intelligent,rendering the conventional jamming method and platform unsuitable for the modern radar countermeasures battlefield due to their limited efficiency. Reinforcement learning has been proven to be a practical solution for cognitive jamming decision-making in the cognitive electronic warfare. In this paper, we proposed a radar-jamming decision-making algorithm based on an improved Q-Learning algorithm. This improved Q-Learning algorithm ameliorated the problem of overestimating the Q-value that exists in the Q-Learning algorithm by introducing a second Q-table. At the same time, we performed a comprehensive design and implementation based on the classical Q-Learning algorithm, deploying it to a Field Programmable Gate Array (FPGA) hardware. We decomposed the implementation of the reinforcement learning algorithm into individual steps and described each step using a hardware description language. Then, the reinforcement learning algorithm can be computed on FPGA by linking the logic modules with valid signals. Experiments show that the proposed Q-Learning algorithm obtains considerable improvement in performance over the classical Q-Learning algorithm. Additionally, they confirm that the FPGA hardware can achieve great efficiency improvement on the radar-jamming decision-making algorithm implementation.

What problem does this paper attempt to address?

The paper attempts to address the issue that in modern warfare, traditional radar jamming methods are no longer suitable for complex radar countermeasure battlefields due to limited efficiency. With the development of electronic warfare, radar jamming has become multifunctional and intelligent, putting traditional jamming algorithms at a disadvantage. To this end, the authors propose a radar jamming decision-making method based on an improved Q-Learning algorithm and deploy it on an FPGA hardware platform to enhance decision-making speed and performance. Specifically, the paper conducts research and improvements in the following areas: 1. **Improved Q-Learning Algorithm**: The traditional Q-Learning algorithm has the problem of overestimating action values (Q-values), leading to unstable training results. The paper introduces a second Q-table, namely QA and QB tables, to reduce the overestimation issue. During each update, the maximum Q-value from the QB table is used to update the QA table, thereby improving the stability and performance of the algorithm. 2. **FPGA Hardware Implementation**: To efficiently implement intelligent radar jamming decisions, the paper designs an FPGA-based hardware platform. The FPGA platform combines the advantages of Application-Specific Integrated Circuits (ASIC) and Central Processing Units (CPU), allowing for reprogramming and having a large amount of parallel distributed storage resources, thus enhancing computational performance. 3. **System Architecture and Module Design**: The paper details how to decompose the reinforcement learning algorithm into multiple modules and design and implement them using Hardware Description Language (HDL). These modules include state transition modules, Q-table update modules, etc., achieving efficient algorithm operation on the FPGA through effective signal linking. Through these improvements and implementations, the paper aims to enhance the efficiency and accuracy of radar jamming decisions, enabling better responses to complex electromagnetic environments in modern warfare. Experimental results show that the improved Q-Learning algorithm significantly outperforms the traditional Q-Learning algorithm in terms of performance, and the FPGA hardware platform implementation further enhances the execution speed of the algorithm.

Radar-Jamming Decision-Making Based on Improved Q-Learning and FPGA Hardware Implementation

A Cognitive Electronic Jamming Decision-Making Method Based on Q-Learning and Ant Colony Fusion Algorithm

An Intelligent Anti-jamming Decision-making Method Based on Deep Reinforcement Learning for Cognitive Radar

Reinforcement Learning-Based Joint Adaptive Frequency Hopping and Pulse-Width Allocation for Radar anti-Jamming

Design of Anti-Jamming Waveforms for Cognitive Radar Based on Deep Reinforcement Learning

Reinforcement Learning based Anti-jamming Frequency Hopping Strategies Design for Cognitive Radar

Jamming Strategy Optimization through Dual Q-Learning Model against Adaptive Radar

GA-Dueling DQN Jamming Decision-Making Method for Intra-Pulse Frequency Agile Radar

Intelligent Decision Method of Slope Perturbing Based on Q-Learning for Anti-Deception Jamming

FPGA Implementation of High-Speed Data Acquisition System for High-Resolution Millimeter Wave Radar.

Radar Anti-Jamming Decision-Making Method Based on DDPG-MADDPG Algorithm

A Fast Learning Method for Optimal Jamming to Radar in Real-Time Environment

Adaptation of Frequency Hopping Interval for Radar Anti-Jamming Based on Reinforcement Learning

Design of anti‐jamming decision‐making for cognitive radar

Frequency Agile Anti-Interference Technology Based on Reinforcement Learning Using Long Short-Term Memory and Multi-Layer Historical Information Observation

Radar Jamming Decision-Making in Cognitive Electronic Warfare: A Review

A Cognitive Jamming Decision-making Method for Multi-functional Radar Based on Threat Assessment

Multi-Agent Reinforcement Learning for Anti-jamming Game of Frequency-Agile Radar

Airborne Radar Anti-Jamming Waveform Design Based on Deep Reinforcement Learning

Multifunctional Radar Cognitive Jamming Decision Based on Dueling Double Deep Q-Network

A Dynamic Game Strategy for Radar Screening Pulsewidth Allocation Against Jamming Using Reinforcement Learning