Algorithms and Implementations for Practical and Energy-effecient Polar Decoders

Download or read book Algorithms and Implementations for Practical and Energy-effecient Polar Decoders written by Furkan Ercan and published by . This book was released on 2021 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: "Polar codes are a class of error-correcting codes that can provably achieve the channel capacity and have simple encoding and decoding mechanisms. Due to their attractive properties, the interest in polar codes has been increasing rapidly in recent years and they have been adopted for use in the $5^{\text{th}}$ generation (5G) wireless systems standard. Specifically, they have been chosen as the coding scheme for the control channel of enhanced mobile broadband (eMBB) use case, and they are being considered for other use cases within 5G. Successive cancellation (SC) decoding is the primary decoding algorithm of polar codes and has low implementation complexity. The two main problems of SC decoding is its mediocre error-correction performance at practical codeword lengths and its long latency due to its sequential nature. To overcome the latency problem, fast decoding techniques have been introduced to speed up the decoding process by an order of magnitude. Secondly, several SC-based decoding algorithms have been proposed to improve the decoding performance, such as SC-List (SCL) and SC-Flip (SCF) decoding. SCL decoding uses parallel SC decoders to improve error-correction performance and therefore suffers from high implementation complexity. On the other hand, the SCF decoding algorithm uses multiple iterations of SC decoding to improve error-correction performance and maintains a similar implementation complexity to that of SC decoding. Therefore, SCF is a promising low-complexity alternative to SCL decoding.This thesis covers several improvements for SC and SCF-based polar decoders. First, we describe how to utilize the hardware resources of fast SC decoding more efficiently and show how to improve the throughput. Second, we propose a partitioned decoding scheme for the SCF algorithm that is able to improve the error-correction performance and reduce the average number of iterations. Third, we describe how to implement energy-efficient polar decoders using fast SC and fast SCF algorithms. We propose the first fast SCF decoder in hardware and show that an energy-efficient approach with improved throughput is possible. Then, we describe the Thresholded SCF (TSCF) algorithm, which has improved error-correction performance and less computational complexity than the conventional SCF algorithm. We implement fast decoding techniques to create the Fast-TSCF decoder that is able to outperform decoders of similar performance in terms of throughput and area efficiency. Finally, we describe many simplifications and optimizations for the Dynamic SCF (DSCF) decoding algorithm, which is known for its significantly improved error-correction performance but has impractical computations. We replace its transcendental computations with simple approximations, introduce fast decoding techniques, reduce its computational complexity by using a theoretical framework, and demonstrate with hardware implementation. The proposed practical DSCF implementation is able to match the error-correction performance and throughput of SCL-based decoders with large list sizes and stands as a low-complexity alternative"--