A Simple Expression for Bit Error Probability of Convolutional Codes under Class-A Interference

In order to establish emission limits to protect radio systems, the effect of the non-Gaussian properties of interference on the performance of coded radio transmission systems should be taken into account. In this paper, a simple closed-form expression for the bit error probability (BEP) of convolutional codes is developed for coherent BPSK systems assuming that the interfering signal is expressed by Middleton’s class A formula, which is a commonly used model of non-Gaussian interference. It is shown that the sum of identically distributed independent class A random variables becomes a new class A variable. Using this result, a simple expression for BEP is derived for Viterbi decoding with an unquantized soft decision. The validity of the expression is demonstrated by numerically simulating the BEP of a communication link. It is found that BEP after decoding under impulsive interference approaches that under Gaussian noise as the free distance of the code increases.