Theoretical Estimates of Burst Error Probability for Convolutional Codes

In this paper, we consider burst distribution at the output of Viterbi decoder for the binary symmetric channel. We suggest a new approach for the burst error probability evaluation based on active distances for the convolutional code. Using distance properties we also define a spectrum of active distances to analyze code performance. We obtain the expressions for lower and upper bounds on the burst error probability. These expressions are based on active distances and on their distance spectrum. In this paper, we consider convolutional codes with recursive encoder and rate one half.