In many scenarios, the undesirable disturbance at a communication receiver has to be modeled as the sum of Gaussian noise and impulsive interference for which closed form probability density function (pdf) generally does not exist. This makes optimum receiver implementation difficult. In this paper, motivated by the heavy tail nature of the resultant pdf, an alternate parametric solution is proposed. After estimation of the appropriate parameters, the alternate solution may then be written in closed form, leading to a relatively simple architecture. Simulation results show that the proposed receiver performance, in terms of bit error rate (BER), is almost indistinguishable from the optimum receiver while outperforming other sub-optimum receivers. Noise parameter estimation using empirical characteristic functions is also discussed.
(Publisher abstract provided.)