Least Mean Square Adaptive Filter Detection Based on Wavelet Transform in Spectrum Monitoring

The difficulty of spectrum detection and signal source perceivement is growing because of the increase in electromagnetic interference. Due to the complexity of the electromagnetic environment and the diversity of system components, micro or even weak signals sometimes submerge in noise and interference, difficult to detect. Aiming at a problem that the signal is not easy to detect, a signal detection algorithm based on electromagnetic spectrum monitoring combined with adaptive filtering method is proposed. In terms of theoretical verification, simulation analysis using analog simulation platform, and the algorithm performs performance testing. The main research contents of this paper are as follows: LMS algorithm simulation experiments are performed on the sinusoidal cycle signal. In response to the limitations of the fixed step LMS algorithm on convergence speed, the wavelet transform is used to perform frequency band separation, and the wavelet transform method is combined with the LMS algorithm to perform signal detection simulation. The simulation results show that the proposition combined with the wavelet transform and the LMS algorithm can detect useful signals. The convergence speed of the adaptive filtering algorithm is effectively improved compared to the LMS algorithm before the improvement. And the scheme can reduce steady state disorders, while effective elimination of interference on the spectrum, enhanced noise elimination capability.