Improved MIMO SIC Detection Exploiting ML Criterion

Abstract

In this paper, we propose an improved MIMO successive interference cancellation (SIC) detector taking maximumlikelihood (ML) criterion into account. Applying the ML criterion to multiple candidates obtained from all possible orderings of streams in the SIC, the proposed method can provide improved performance with complexity comparable to conventional ordered SIC for moderate number of spatial streams.Index Terms-MIMO spatial multiplexing, maximum likelihood (ML) criterion, successive interference cancellation (SIC)

Lattice-Reduction Aided Optimum and Sub-Optimum SIC Detection for MIMO Systems

“…All-ordering SIC (AOSIC) Compared to SIC which immediately detects symbols one by one by removing interference successively, ML-based detection using tree-search in a specific criterion has pros and cons in terms of complexity and performance. By constructively combining these detection schemes well, a detector with affordable complexity and near-ML performance can be considered in small-scale MIMO, which is called AOSIC [7].…”

“…Furthermore, there has been researched to find an optimum detection ordering of SIC [6]. Actually, optimum SIC can be implemented by way of finding an optimum order which is the candidate having the minimum Euclidean distance among all the candidates of detection ordering, which is called allordering SIC (AOSIC) [7]. However, an optimum order cannot perfectly prevent the error propagation because of undesirable channel environment such as correlation between different spatial paths and large channel gain applied to interference.…”

A low‐complexity iterative MIMO detection and decoding scheme using dimension reduction

“…That is, we cannot predetermine which is the best ordering in each instant, that is, time, frequency and spatial conditions of the channel. Based on this observation, we try all kinds of stream ordering and choose the best, similar to , which is employed for ML detection, as where …”

“…That is, we cannot predetermine which is the best ordering in each instant, that is, time, frequency and spatial conditions of the channel. Based on this observation, we try all kinds of stream ordering and choose the best, similar to [18] (22), the best ordering can be easily found using the MAP criterion based on Equation (22). Although different ordering schemes may be employed including ones based on pseudo-inverse, log-likelihood calculation, auto-correlation matrix [31], or multiple branches [32], we consider all the available ordering sets in an exhaustive manner, so-called all ordering, to perform the best SIC hard detection.…”

Iterative MIMO Reception Based on Low Complexity Soft Detection

“…이를 해결하기 위해서 ML 기 법을 변경한 근사적 최적 (sub-optimal) 기법들이 제시 되었으나 [5～6] 여전히 실제 사용하기에는 복잡도가 크거 나 성능이 만족스럽지 못하다. 이에 따라 본 논문에서 는 ML 검출을 위해 기존에 고안한 모든 순서 순차적 간섭제거 기법 (AOSIC) [7] 및 차원감소 소프트 검출기 법 (DRSD) [8] …”