Fast Resampling and Denoising of Spread Spectrum Signal for Real-Time Wireless Location

Narrowband noise is the most effective interference that can make measurement of spread spectrum signal undetected for wireless location. However, current methods for narrowband noise suppression are either very time-consuming or adding distortion to the signal received. In this paper, a fast resampling method and an adaptive Gaussian wavelet filter with optimal time-frequency localisation and variable bandwidth is proposed to suppress single and multiple narrowband noise that interferes with spread spectrum signals in the frequency domain. The filtering and resampling is computationally efficient such that it can be used for real-time implementation. A method by conducting magnitude spectral analysis on received signals in a detection channel is employed to determine which frequency contents in the spectral analysis output contain the interfering narrowband signals. Experimental work has demonstrated that the proposed resampling and wavelet filter reduces narrowband noise efficiently and improves signal to noise ratio (SNR) for peak detection leading to higher accuracy of timing measurement for wireless positioning.