A PSO-Neural Network Hybrid Algorithm for Optimal Jamming Resource Allocation

An improved particle swarm optimization algorithm integrated with neural network (PSONN) is proposed for cooperative jamming resource allocation in complex electronic warfare scenarios. A jamming resource allocation model is first designed to maximize effectiveness across time, frequency, space, energy, and jamming style, and then solved via PSONN. The hybrid algorithm leverages particle swarm optimization (PSO) to rapidly generate historical performances as inputs to the neural network. Based on fitness improvements, the network then adaptively adjusts the PSO's acceleration coefficients and updates its internal weights through backpropagation. This co-optimization process ultimately derives an optimal jamming resource allocation scheme. Simulation results show that PSONN reduces mean squared error by 93 % compared to IPSO, with only a 7.2 % increase in iterations. It also outperforms PSO in both convergence precision (MSE reduced by 83.3 %) and iteration efficiency (16.3% faster)