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

In order to improve the efficiency and adaptability of cognitive radar jamming decision-making, a fusion algorithm (Ant-QL) based on ant colony Q-Learning is proposed in this paper. The does not rely priori information enhances through real-time interactions between jammer target radar. At same time, it can be applied single multiple countermeasure scenarios with high effects. First, traditional DQN algorithms are discussed, decision-making model built for simulation verification each algorithm. Then, an improved address shortcomings both algorithms. By introducing pheromone mechanism using ε-greedy balance contradictory relationship exploration exploitation, greatly avoids falling into local optimum, thus accelerating convergence speed good stability robustness process. better adapt cluster environment future battlefields, extended cooperative decision-making. We map intelligent searching optimal path, jammers interact other obtain information. During process confrontation, method improves reduces need hardware power resources jammer. Assuming that number three, experimental results Ant-QL by 85.4%, 80.56% 72% compared Q-Learning, algorithms, respectively. process, very stable efficient, complexity low. After converge, average response times four 6.99 × 10−4 s, 2.234 10−3 2.21 s 1.7 show also have more advantages terms time after convergence.