MIMO (multiple input and multiple output) detection method based on lattice reduction

Abstract

The invention discloses a MIMO (multiple input and multiple output) detection method based on lattice reduction, which comprises the steps of 1, carrying out QR decomposition on a channel matrix of a MIMO system; 2, carrying out a QRCLLL lattice reduction operation on a result acquired by QR decomposition; 3, calculating boundary information expanded by tree search according to a result of the step 2; 4, carrying out selective fast enumeration expansion in an LR region according to boundary restriction; 5, converting a vector set acquired by completion of tree search into a constellation point region; and 6, acquiring an optimal vector from the acquired vectors. According to the invention, the performance difference between an LR based detection algorithm and the optimal MIMO detection is enabled to be further reduced under the condition that the number of expanded nodes of tree search is very small relative to that of traditional tree search, and the operation complexity of a detector is reduced.

Description

technical field [0001] The invention relates to the technical field of MIMO (Multiple Input and Muttiple Output) transmission, in particular to a lattice-based MIMO detection method. Background technique [0002] The MIMO system can provide great gains in data rate and signal reliability, making it the choice of many of today's most advanced wireless standards, such as wireless local area networks, next-generation wireless cellular networks, and more. However, with the development of wireless communication, the requirement of communication for MIMO technology is getting higher and higher, which makes it very meaningful to study high-performance MIMO system, and how to design MIMO detector becomes the focus. On the one hand, MIMO detectors need to separate and process multiple spatially multiplexed data received by multiple antennas at the same time, which makes the complexity of the detection method very high; on the other hand, while new communication standards propose high...

AI Technical Summary

Problems solved by technology

However, the so-called "very close" is only for 2×2 systems, and for higher-dimensional systems (such as 4×4 modes), the performance gap is still large, making detection with better performance urgently needed.

At present, in the existing low-complexity methods, the LR-assisted detection that has been proposed so far only directly combines LR and linear detection; the linked list generation method based on continuous interference elimination, but because the generation of the linked list has certain blindness, While increasing the complexity, the performance of the detection cannot be guaranteed

Images