CRC Aided List Decoding of Convolutional Codes

Sammanfattning: Designing for the best combination of channel codes and the decoding algorithm has always been a popular topic in channel coding. As a decoding algorithm for convolutional codes, Viterbi algorithm provides the most reliable performance with high complexity. List Viterbi algorithm(LVA) is an alternative solution to provide a good tradeoff between performance and complexity, and is widely used for high memory convolutional codes. In this thesis, a coding scheme for combining LVA and CRC will be introduced, and the implementation of CRC aided list decoding will be explained. This thesis compares the complexity and performance of this proposed decoding algorithm with Viterbi algorithm and list decoding. Simulations show that there are several parameters that will influence the performance, and that the proposed CRC aided list decoding can beat a traditional list decoding. Furthermore, the thesis investigates the performance of the decoder with different implementations under this coding scheme, and analyses the error probability with the help of the comparison.