A Modified Energy Detector for Random Signals in Gaussian Noise

Abstract

This letter proposes a new modified energy detector for random signals in Gaussian noise. This new detector, whose decision variable is the weighted sum of the signal amplitude and its square, has a simple structure similar to the conventional energy detector. We propose a parameter optimization method to find the proper weighting factor by maximizing the error exponent of the misdetection probability based on large deviation theory. We also apply this method to optimize the power operation $p$ of another modified energy detector recently proposed by Chen. Numerical results show that, with the parameters obtained by the proposed method, both modified detectors achieve performance almost as well as the optimal detector.