Generalized autoregressive spectrum estimation algorithm based on particles sward optimization

In this paper, an intelligent optimal algorithm of generalized AR spectrum estimation (GAR) based on Particles Sward Optimum (PSO) is presented. After the model of GAR spectrum estimation suitable for numerical optimum is established using the error estimator in Burg algorithm, GAR algorithm is developed based on PSO. The results of its numerical simulation results are given and discussed. In contrast to Burg algorithm, GAR shows its better performance of spectrum estimation in the gain of ratio of signal to noise (SNR), frequency shift, spectrum splitting and frequency resolution at the cost of computational quantity.