Network Anomaly Detection Using Wavelet Neural Networks Trained with Quantum-Behaved Particle Swarm Optimization

For network anomaly detection to improve detection rates for anomalous states and reduce false positives for normal states, this paper proposes a novel method that employs Quantum-behaved Particle Swarm Optimization (QPSO) to train Wavelet Neural Networks (WNN). The parameter set of the WNN is treated as a particle in the optimization algorithm, searching the global space for the parameter vector with the optimal fitness value. Key implementation involves encoding WNN parameters as particle positions and updating them using quantum behavior principles for enhanced convergence.