Multiple-input multiple-output (MIMO) wireless communication systems have been studied a lot in the last ten years. They have many promising features like array gain, diversity gain, spatial multiplexing gain, interference reduction, and improved capacity as compared to a single-input single-output (SISO) systems. However, the increasing demand of high data-rate in current wireless communications systems motivated us to investigate new rate-efficient channel coding techniques. In this dissertation, we study differential modulation for MIMO systems. Differential modulation is useful since it avoids the need of channel estimation by the receiver and saves valuable bandwidth with a slight symbol error-rate (SER) performance loss. The effect of channel correlation over differential MIMO system has not been studied in detail so far. It has been shown in the literature that a linear memoryless precoder can be used to improve the performance of coherent MIMO system over correlated channels. In this work, we implement precoded differential modulation for non-orthogonal and orthogonal space-time blocks codes (STBCs) over arbitrarily correlated channels. We design precoders based on pair-wise error probability (PEP) and approximate SER for differential MIMO system.
The carrier offsets, which result because of the movement of the receiver or transmitter and/or scatterers, and mismatch between the transmit and receive oscillators, are a big challenge for the differential MIMO system. The carrier offsets make the flat fading channel behave as a time-varying channel. Hence, the channel does not remain constant over two consecutive STBC block transmission time-intervals, which is a basic assumption for differential systems and the differential systems break down. Double-differential coding is a key technique which could be used to avoid the need of both carrier offset and channel estimation. In this work, we propose a double-differential coding for full-rank and square orthogonal space-time block codes (OSTBC) with M-PSK constellation. A suboptimal decoder for the double-differentially encoded OSTBC is obtained. We also derive a simple PEP upper bound for the double-differential OSTBC. A precoder is also designed based on the PEP upper bound for the double-differential OSTBC to make it more robust against arbitrary MIMO channel correlations.
Cooperative communication has several promising features to become a main technology in future wireless communications systems. It has been shown in the literature that the cooperative communication can avoid the difficulties of implementing actual antenna array and convert the SISO system into a virtual MIMO system. In this way, cooperation between the users allows them to exploit the diversity gain and other advantages of MIMO system at a SISO wireless network. A cooperative communication system is difficult to implement in practice because it generally requires that all cooperating nodes must have the perfect knowledge of the channel gains of all the links in the network. This is infeasible in a large wireless network like cellular system. If the users are moving and there is mismatch between the transmit and receive oscillators, the resulting carrier offset may further degrade the performance of a cooperative system. In practice, it is very difficult to estimate the carrier offset perfectly over SISO links. A very small residual offset error in the data may degrade the system performance substantially. Hence, to exploit the diversity in a cooperative system in the presence of carrier offsets is a big challenge. In this dissertation, we propose double-differential modulation for cooperative communication systems to avoid the need of the knowledge of carrier offset and channel gain at the cooperating nodes (relays) and the destination. We derive few useful SER/bit error rate (BER) expressions for double-differential cooperative communication systems using decode-and-forward and amplify-and-forward protocols. Based on these SER/BER expressions, power allocations are also proposed to further improve the performance of these systems.
List of papers included in the dissertation
This dissertation is based on the following five papers, referred to in the text by letters (A-E).
A. M. R. Bhatnagar, A. Hjørungnes, and L. Song, "Precoded Differential Orthogonal Space-Time Modulation over Correlated Ricean MIMO Channels", IEEE Journal of Selected Topics in Signal Processing, in the issue "MIMO-Optimized Transmission Systems for Delivering Data and Rich Content", volume 2, issue 2, pages 124 - 134, April 2008.
B. M. R. Bhatnagar, A. Hjørungnes, and L. Song, "Differential Coding for with Non-Unitary Constellations in Arbitrarily Correlated Rayleigh Channels," conditionally accepted for publication in IEEE Transactions on Wireless Communications (subject to reviewers’ and editors’ final approval of the revised version submitted 19-03-2008).
C. M. R. Bhatnagar, A. Hjørungnes, and L. Song, "Double Differential Orthogonal Space-Time Block Codes for Arbitrarily Correlated Rayleigh Channels with Carrier Offsets," conditionally accepted for publication in IEEE Transactions on Wireless Communications (subject to reviewers’ and editors’ final approval of the revised version under preparation).
D. M. R. Bhatnagar, A. Hjørungnes, and L. Song, "Cooperative Communications over Flat Fading Channels with Carrier Offsets: A Double-Differential Modulation Approach", EURASIP Journal on Advances in Signal Processing, in special issue on "Wireless Cooperative Networks", volume 2008, article ID 531786, 11 pages,
E. M. R. Bhatnagar, A. Hjørungnes, R. Bose, and L. Song, "Double-Differential Decode-and-Forward Cooperative Communications over Nakagami-m Channels with Carrier Offsets" in Proc. for Sarnoff Symposium 2008, pages 1-5, Princeton, NJ, USA, April 2008.