Block-Orthogonal Space-Time Code Structure and Its Impact on QRDM Decoding Complexity Reduction

TL;DR

This paper introduces the block-orthogonal property of space-time codes, enabling significant QRDM decoding complexity reduction without performance loss, and designs new codes that maximize this benefit.

Contribution

It defines the block-orthogonal property for space-time codes and leverages it to reduce decoding complexity, including the creation of optimized full-rate BOSTC.

Findings

Block-orthogonal property enables complexity reduction in QRDM decoding.

New BOSTC codes outperform existing codes in reduced-complexity decoding.

Complexity saturation property observed in the proposed codes.

Abstract

Full-rate space time codes (STC) with rate = number of transmit antennas have high multiplexing gain, but high decoding complexity even when decoded using reduced-complexity decoders such as sphere or QRDM decoders. In this paper, we introduce a new code property of STC called block-orthogonal property, which can be exploited by QR-decomposition-based decoders to achieve significant decoding complexity reduction without performance loss. We show that such complexity reduction principle can benefit the existing algebraic codes such as Perfect and DjABBA codes due to their inherent (but previously undiscovered) block-orthogonal property. In addition, we construct and optimize new full-rate BOSTC (Block-Orthogonal STC) that further maximize the QRDM complexity reduction potential. Simulation results of bit error rate (BER) performance against decoding complexity show that the new BOSTC outperforms all previously known codes as long as the QRDM decoder operates in reduced-complexity mode, and the code exhibits a desirable complexity saturation property.