Use this URL to cite or link to this record in EThOS:
Title: Energy-efficient architectures for multi-gigabit MIMO detection
Author: Bello, Ibrahim Ahmad
ISNI:       0000 0004 6497 3514
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
The use of multiple antennas in wireless transmission, otherwise known as multiple-input multiple-output (MIMO), is an important technique for achieving the high datarates required by future communication systems. Already, MIMO technology has been adopted by the 3rd Generation Partnership Project Long Term Evolution, WiMAX and by recent Wireless Local Area Network standards, such as the IEEE 802.11ac. It is envisaged that multi-antenna systems will play an even more prominent role in future as more diverse platforms become interconnected and user data rates requirements increase. Although MIMO offers numerous advantages to communication systems, it also presents a number of challenges, in particular to the receiver, where the complexity of the signal detection is exacerbated by the interferences from the multiple transmit antennas. In the worst-case scenario, the signal detection in MIMO systems is an NP-hard problem, which makes its application to real-time systems impractical. As a result, low-complexity detection algorithms, with near-optimal performance, have been extensively studied in the literature in the past decade. In this thesis, a number of detection techniques for MIMO systems will be investigated, with particular focus on achieving high throughput and low power consumption. We begin by presenting the VLSI implementation of the sphere decoder (SD), which achieves the optimal maximum likelihood bit error rate (BER) performance. Although the SD has the potential of achieving a high throughput - specifically in high signal-to-noise ratios (SNR) - it also suffers a severe throughput degradation at low SNR, which is undesirable in a real-time system. This problem motivates us to investigate the K-best algorithm, which delivers a constant throughput irrespective of the channel condition. Two architectures for the K-best detector are considered: single and multi-stage architectures. The latter case is particularly interesting as the multiple stages can be utilised to achieve deeply pipelined detectors, which is attractive for high-throughput applications. The proposed multi-stage K-best detector is implemented in a 65 nm CMOS process, and achieves a throughput of 3.29 Gbps and a power consumption of 580 mW for a 64-QAM 4×4 MIMO configuration, which compares favourably with recent implementations in the literature.
Supervisor: Zwolinski, Mark Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available