Degree Name

Doctor of Philosophy (PhD)


School of Electrical, Computer and Telecommunications Engineering - Faculty of Informatics


For impending wireless communications networks, such as 3rd Generation Mobile networks and Wireless Local Area Networks, there is a greater emphasis on providing higher data rates and improved quality in terms of reliability over multipath fading channels. The area of Multiple Input Multiple Output (MIMO) communications systems has received enormous attention recently, as they can provide a roughly linear increase in data rate by using multiple transmit and receive antennas. In a future, where communication systems may implement MIMO technology, mobile user terminals are desired to be small, lightweight and have extended battery life. This seems at odds with MIMO systems because, as described in this thesis, Multiple In Multiple Out receivers require either a large number of receive antennas and associated electronic hardware, and/or powerful signal processing which has a computational complexity that varies significantly with Signal to Noise Ratio (SNR). The key objective of this thesis is to increase the flexibility of mobile units by reducing the number of receive antennas needed to obtain near optimal performance and a reasonable level of complexity. After presenting a background on problems affecting wireless communications, such as fading and noise, this thesis describes current and future methods of overcoming fading, such as receive and Space-Time diversity. This is followed by introduction into Multiple In Multiple Out systems that actually utilize the multiple fading to increase the data rate of wireless communication systems. To increase the performance of previously proposed suboptimal decoders for symmetric systems, where there is an equal number of transmit and receive antennas, this thesis first proposes a scheme, called Reduced constellation for Sorted QR Decomposition (RSQRD) that significantly improves performance by producing a list of possible combinations. This list is then used by the Maximum Likelihood decoder to determine which combination of symbols has the best performance. This thesis then provides a demonstration of how a large complex constellation, generated by multiple transmit and a single receive antenna, can be divided into groups or Asterisms. This concept of dividing the large complex constellation is then used to develop a MIMO decoder called Asterism decoding and is extended for any number of transmit and receive antennas. Also proposed is how Sphere decoding, described in Chapter 5, can be used in conjunction with Asterism decoding to reduce the overall computational complexity of Sphere decoding for systems where the number of transmit antennas is greater than the number of receive antennas. Finally it is shown how an Asterism decoding based decoder can be used in a simplified Turbo coded MIMO systems with the number of transmit antennas is greater than the number of receive antennas.

02Whole.pdf (1006 kB)