High-throughput lattice reduction for large-scale MIMO systems based on Seysen's algorithm
Farhana Sheikh, Alexios Balatsoukas‐Stimming, Chia‐Hsiang Chen · 2016
Lattice reduction aided multiple-input and multiple output (MIMO) detection has attracted significant attention recently due to its low complexity and excellent error rate performance. The most popular lattice reduction algorithms are the Lenstra-Lenstra-Lovász algorithm and Seysen's algorithm, although the former has received much more attention than the latter. In this work, we present a simplification to Seysen's lattice reduction algorithm, which reduces the periteration computational complexity from quadratic to linear with a small degradation in the quality of the resulting reduced lattice. Moreover, we present an efficient VLSI architecture which demonstrates the advantages of the proposed algorithm and can achieve a throughput of up to 91 Mmatrices/s at an operating frequency of 1 GHz.