Latency Reduced LTE-A Turbo-Code Decoding with Iteration Balancing on Transport Block Level
Conference: SCC 2017 - 11th International ITG Conference on Systems, Communications and Coding
02/06/2017 - 02/09/2017 at Hamburg, Germany
Proceedings: ITG-Fb. 268: SCC 2017
Pages: 6Language: englishTyp: PDFPersonal VDE Members are entitled to a 10% discount on this title
Weithoffer, Stefan; Wehn, Norbert (Microelectronic Systems Design Research Group, University of Kaiserslautern, 67663 Kaiserslautern, Germany)
The requirements of todays standards for wireless mobile applications require sophisticated Turbo-Code decoder architectures that support throughputs of 1-2 Gbps at low latencies. With the next generation of standards, however, the throughput requirements will be in the order of tens of Gbps, which is not achieved by further parallelization on Turbo-Code decoder level without sacrificing flexibility. We propose to leverage state-of-the-art iteration control and the grouping of code blocks into transport blocks as it is used in LTE-A to increase the throughput of Turbo-Code decoders by an iteration balancing across multiple code blocks and evaluate our approach for single and parallel Turbo-Code decoder instances. We show, that by using iteration balancing, either throughput gains of up to 50% are possible without a degradation in decoding performance or the decoding performance can be improved by up to 0:3dB for the same throughput.