A Novel Wavelet Denoising Pre-processing Algorithm for TDOA Localization

摘要

In order to reduce the distance measurement errors during TDOA localization, a novel pre-processing algorithm based on wavelet denoising is proposed. The wavelet pre-processing algorithm is able to flexibly select the number of wavelet decomposition levels based on the energy coefficient, then different thresholds are set for positive and negative wavelet coefficients which do not require noise estimation. Next, a new threshold function is designed to deal with wavelet coefficients which can incorporate the advantages of soft thresholding and hard thresholding, and finally the signal is reconstruct. Simulation results show that the wavelet preprocessing algorithm reduces distance measurement errors and improves the positioning accuracy under different environments.