A Low Complexity Piecewise Suboptimal Detector for Signals in Alpha-Stable Interference

The design of near-optimal detectors for binary signals in α-stable interference using two receive antennas is considered. The optimal detector requires more complex computation, whereas the linear (Gaussian) detector is practically simple but suffers from performance degradation with α-stable noise. In this paper, based on the optimal decision regions analysis, we propose to approximate the optimal decision boundaries by using linear segments which results in a novel piecewise detector. The proposed detector has much less complexity over the optimal detector. Also, the analytical evaluation shows near-optimal performance of the piecewise detector. Moreover, simulation results show that the proposed detector has robust performance for different values of α.