Intelligent Adaptive Filtering Algorithm for Electromagnetic-Radiation Field Testing

To reduce the effect of background noise on the electromagnetic-radiation field testing of largeand medium-sized equipment, this work combines the multiscale wavelet transform, particle swarm parameter optimization, and adaptive filtering algorithm improvement to obtain an intelligent adaptive filtering algorithm that can be applied to electromagnetic-radiation field tests. After these improvements, the new algorithm overcomes the problem of traditional algorithm in which the latter cannot ensure both convergence speed and steady-state error that satisfy the real-time requirements of field tests. The multiscale wavelet analysis segments and reconstructs the electromagnetic radiation and background noise of equipment within the frequency domain, and decomposes the broadband signal into a narrowband signal to overcome the difficulty in configuring broadband-signal filtering parameters. The particle swarm parameter optimization uses error E[e² (n)] as a fitness function to optimize the adaptive algorithm parameters, simplifying the operations performed by engineering personnel. The algorithm is simulated, and the actual engineering experimental data are verified in the Laboratory Virtual Instrument Engineering Workbench software platform, which prove that the proposed algorithm is more suitable to practical engineering application than the traditional adaptive filtering algorithm.