A frequency-domain detection and estimation scheme for single-tone interference suppression

Abstract

For wideband communication systems, the ability against relatively narrowband interference, which is commonly modeled as single-tone interference, is of great importance. Without proper mitigation, i.e., interference suppression, the system performance can degrade significantly as the interference power increases. To cope with this problem, this paper develops an efficient frequency-domain single-tone interference detection and estimation scheme for interference suppression. After transforming the received signal to frequency domain via fast Fourier transform, and with possible power accumulation of consecutive fast Fourier transform windows for successful detection of relatively weak interference (e.g., at a signal-to-interference ratio around 24dB), we derive the probability density function of the received power within each tone and develop a threshold-based detection scheme, where the theoretical probabilities of false alarm and missed detection given a chosen threshold are also derived and confirmed by simulations. Based on the theoretical analyses, we further propose a high-accuracy and low-complexity scheme that estimates the frequency of single-tone interference.