A Parallel Algorithm for Computing the Weight Spectrum of Binary Linear Codes

We present a parallel SIMD (Single Instruction Multiple Data) algorithm for computing the weight spectrum of binary linear [n,k] codes using GPU (Graphics Processing Units) and CUDA (Compute Unified Device Architecture) programming platform. This algorithm is based on characteristic vector and Walsh-Hadamard transform and does not depend on the length of the code. We compare it with standard consecutive algorithms and give some experimental results.