A Decoding Algorithm for Spinal Codes over Fading Channel

Abstract

The quality of wireless channels between mobile vehicles can vary greatly, resulting in significant differences in the quality of received symbols. Additionally, Spinal codes decode sequentially, which means that a single poor quality symbol can cause the entire message to fail to decode. To address this issue, a dynamic stack decoding algorithm is proposed in this paper. By leveraging equalization results at the receiver, the decoder can make approximate assessments of symbol quality, enabling targeted adjustments of the search range for each layer. This approach reduces decoding redundancy by narrowing the search range for layers with better received symbol quality, while expanding the search range for layers with poorer quality, making it easier to capture the correct path. Simulation results validate the effectiveness of the proposed algorithm in improving rate performance and concurrently reducing computation cost.