Efficient Implementation of the MIMO Sphere Detector: Architecture and Complexity Analysis

Chang, Ronald Y.; Lin, Sian-Jheng; Chung, Wei-Ho

doi:10.1109/tvt.2012.2204075

Cited by 2 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Local search methods, while having the benefit of a tunable complexity with detection performance-speed tradeoff, are prone to the local minima problem if not given an extensive search. Tree search algorithms are also used in detectors [14]- [18], where the MIMO detection problem is formulated as a decision tree and a symbol is recovered at each layer of the tree. In [19], a statistical approach with the Monte Carlo tree search (MCTS) algorithm was proposed with hardware acceleration to recover the transmitted symbols at large MIMO setups.…”

Section: A Related Workmentioning

confidence: 99%

“…In addition, when the number of receive antennas is greater than the number of transmit antennas, as in the asymmetric 8 × 16 channel here, the channel matrix H tends to be more well-conditioned [2]. This is a favorable condition for traditional detectors such as the sphere decoder (producing extensive tree pruning and therefore higher detection speed and lower complexity) [18] and MMSE (producing generally satisfactory solutions) [4], and is similarly so for the tree search-based DRL-MCTS algorithm.…”

Section: B Drl-mcts Vs Mcts Comparisonmentioning

confidence: 99%

See 1 more Smart Citation

Deep Reinforcement Learning Aided Monte Carlo Tree Search for MIMO Detection

Mo,

Chang,

Kan

2021

Preprint

Self Cite

View full text Add to dashboard Cite

This paper proposes a novel multiple-input multipleoutput (MIMO) symbol detector that incorporates a deep reinforcement learning (DRL) agent into the Monte Carlo tree search (MCTS) detection algorithm. We first describe how the MCTS algorithm, used in many decision-making problems, is applied to the MIMO detection problem. Then, we introduce a self-designed deep reinforcement learning agent, consisting of a policy value network and a state value network, which is trained to detect MIMO symbols. The outputs of the trained networks are adopted into a modified MCTS detection algorithm to provide useful node statistics and facilitate enhanced tree search process. The resulted scheme, termed the DRL-MCTS detector, demonstrates significant improvements over the original MCTS detection algorithm and exhibits favorable performance compared to other existing linear and DNN-based detection methods under varying channel conditions.

show abstract

Section: A Related Workmentioning

confidence: 99%

Section: B Drl-mcts Vs Mcts Comparisonmentioning

confidence: 99%

Deep Reinforcement Learning Aided Monte Carlo Tree Search for MIMO Detection

Mo,

Chang,

Kan

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Among them, Sphere decoding is a very powerful algorithm to find the optimal solution in many scenarios [4][5][6][7][8][9][10][11][12]. Many works have been done to promote this algorithm [13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29]. To utilize the benefit of parallel calculation in signal process is a good method to promote the performance [30][31].…”

Section: Introductionmentioning

confidence: 99%