High-rate uncoded space-time labelling diversity with low-complexity detection

Abstract

Uncoded space-time labelling diversity (USTLD) is a recent scheme that improved the error performance compared to conventional multiple-input, multiple-output systems. Thus far, USTLD has suffered from limited achievable data rates, as the original model uses only two transmit antennas. This motivates for the work in this paper, where the USTLD model is extended to allow for any desired number of transmit antennas. An analytical bound for the average bit error probability of this high-rate USTLD (HR-USTLD) system is derived. This expression is verified using the results of Monte Carlo simulations, which show a tight fit in the high signal-to-noise ratio region. The increased data rates associated with larger transmit antenna arrays in HRUSTLD systems come at the cost of increased detection complexity. Therefore, this paper studies the application of low-complexity detection algorithms based on the popular QR decomposition technique and proposes a new algorithm specifically designed for HR-USTLD systems. Analysis of this algorithm in terms of accuracy and computational complexity is also provided and benchmarked against maximum-likelihood detection (MLD). It is shown that the proposed algorithm achieves near-MLD accuracy, while reducing complexity by 79.75% and 92.53% for the respective 4 × 4 16QAM and 4 × 5 16PSK HR-USTLD systems investigated.