A Signal Detection Method Using Goodness of Fit Test in Laplacian Noise

Signal detection is a challenging problem for many systems especially in the presence of non-Gaussian noise. However, there are few detectors can work well in this case. Motivated by this, we propose a signal detection algorithm based on goodness of fit (GoF) test in Laplacian noise. Processed data rather than original observations were used in the new GoF test. The new GoF test makes full use of the stochastic properties of Laplacian noise and is evaluated by Monte Carlo simulations including detection performance and the influence of noise uncertainty. It is shown that the proposed signal detection method outperforms the energy detection and other GoF tests in Laplacian noise.