Low Complexity, Pairwise Layered Tabu Search for Large Scale MIMO Detection

Abstract

This paper presents a low complexity pairwise layered tabu search (PLTS) based detection algorithm for a large-scale multiple-input multipleoutput (MIMO) system. The proposed algorithm can compute two layers simultaneously and reduce the effective number of tabu searches. A metric update strategy is developed to reuse the computations from past visited layers. Also, a precomputation technique is adapted to reduce the redundancy in computation within tabu search iterations. Complexity analysis shows that the upper bound of initialization complexity in the proposed algorithm reduces from O(N 4 t ) to O(N 3 t ). The detection performance of the proposed detector is almost the same as the conventional complex version of LTS for 64QAM and 16QAM modulations. However, the proposed detector outperforms the conventional system for 4QAM modulation, especially in 16 × 16 and 8 × 8 MIMO. Simulation results show that the per cent of complexity reduction in the proposed method is approximately 75% for 64 × 64, 64QAM and 85% for 64 × 64 16QAM systems to achieve a BER of 10 −3 . Moreover, we have proposed a layer-dependent iteration number that can further reduce the upper bound of complexity with minor degradation in detection performance.

Performance analysis and comparison of approximate detectors with linear detectors in massive MIMO systems under imperfect channel state information

“…7 A plethora of M-MIMO detection techniques are covered in the literature, including linear, non-linear, local search, box detection, belief propagation, machine learning, and sparsity based detectors. [8][9][10][11][12][13][14][15][16] Optimal detectors like Maximum Aposteriori (MAP) and Maximal Likelihood (ML) detectors suffer from exponential computational complexity due to large antenna configurations and modulation sizes and, hence, are not feasible. 17 The belief propagation (BP) algorithm, a tree-based algorithm, can also give performance close to that of ML with low channel correlation.…”