A List Sphere Detector based on Dijkstra's Algorithm for MIMO-OFDM Systems

Abstract

A list sphere detector (LSD) based on Dijkstra's algorithm, namely increasing radius (IR) - LSD, is introduced for multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) systems. The complexity and the performance of the IR-LSD is compared to a breadth first and depth first search based LSDs. The required list sizes for considered LSDs are determined for a 4times4 system with quadrature amplitude modulations (QAMs). The complexity of the algorithms is relative to the number of visited nodes in the search tree structure. Thus, the algorithm complexities are compared by studying and comparing the number of visited nodes per received symbol vector by the algorithm via computer simulations. The IR-LSD is found to visit the least amount of nodes of the LSDs in the search tree in each studied case, and found the least complex especially with higher order constellations.