Sphere-decoding detection method with fixed complexity

Abstract

The invention discloses a sphere-decoding detection method with fixed complexity, comprising the following steps: carrying out channel extension on channel matrixes of acquired received signals based on minimum mean squared error (MMSE) criterion to obtain extended channel matrixes; carrying out sequenced QR decomposition on the extended channel matrixes to obtain a group of displacement series as well as unitary matrixes and ordered upper triangular matrixes of elements on diagonal lines; multiplying the conjugate transpose of the unitary matrixes by the received signals to obtain received vectors; setting a parameter P and a parameter Mk (k is equal to 1, 2...P), starting from a last layer, and detecting the processed received signals layer by layer from back to front so as to generate a plurality of survivor paths in accordance with the formula in the specification; and outputting a path with the minimum Euclidean distance as a detection result from the plurality of survivor paths after all layers are detected. By using the sphere-decoding detection method, the complexity of the sphere-decoding detection can be reduced effectively at the same time of ensuring the system performances under the condition of high-order modulation (HOM) and high-dimension antenna configuration.

Description

Sphere decoding detection method with fixed complexity technical field The invention relates to the technical field of MIMO (Multiple Input and Multiple Output) in a wireless communication system, in particular to a fixed-complexity ball decoding detection method. Background technique MIMO technology has been widely used in recent years. The MIMO system can provide space diversity gain and space multiplexing gain, effectively increasing system capacity and improving link reliability. The capacity of the MIMO system increases linearly with the number of transmitting antennas, but due to the increase of antenna dimensions, the interference between transmitted data also increases. An effective receiver detection algorithm is crucial to guarantee system performance. Currently receiving algorithms include nonlinear detection and linear detection. The best detection algorithm for non-linear detection is maximum likelihood receiver detection (ML), which is especially beneficia...

AI Technical Summary

Problems solved by technology

However, when the modulation order of the system is high and the number of antennas is large, since the fixed-complexity sphere decoding algorithm retains all the constellation points similar to ML detection in the first few layers of detection, the whole detection algorithm still has relatively low High complexity increases the difficulty of implementation in practical systems

Images