Efficient soft-output detectors: Multi-core and GPU implementations in MIMOPack library

Error control coding ensures the desired quality of service for a given data rate and is necessary to improve re-laibility of Multiple-Input Multiple-Output (MIMO) communication systems. Therefore, a good combination of detection MIMO schemes and coding schemes has drawn attention in recent years. The most promising coding schemes are Bit-Interleaved Coded Modulation (BICM). At the transmitter the information bits are encoded using an error-correction code. The soft demodulator provides the reliability information in form of real valued log-likehood ratios (LLR). These values are used by the channel decoder to make final decisions on the transmitted coded bits. Nevertheless, these sophisticated techniques produce a significant increase in the computational cost and require large computational power. This paper presents a set of Soft-Output detectors implemented in CUDA and OpenMP, which allows to considerably decrease the computational time required for the data detection stage in MIMO systems. These detectors will be included in the future MIMOPack library, a High Performance Computing (HPC) library for MIMO Communication Systems. Experimental results confirm that these implementations allow to accelerate the data detection stage for different constellation sizes and number of antennas.