Multi-user wireless systems with multiple antennas can drastically increase the capac- ity while maintaining the quality of service requirements. The best performance of these systems is obtained at the presence of instantaneous channel knowledge. Since uplink-downlink channel reciprocity does not hold in frequency division duplex and broadband time division duplex systems, efficient channel quantization becomes important. This thesis focuses on different quantization techniques in a linearly precoded multi-user wireless system. Our work provides three major contributions. First, we come up with an end-to-end transceiver design, incorporating precoder, receive combining and feedback policy, that works well at low feedback overhead. Second, we provide optimal bit allocation across the gain and shape of a complex vector to reduce the quantization error and investigate its effect in the multiuser wireless system. Third, we design an adaptive differential quantizer that reduces feedback overhead by utilizing temporal correlation of the channels in a time varying scenario.

Multi-user wireless systems with multiple antennas can drastically increase the capac- ity while maintaining the quality of service requirements. The best performance of these systems is obtained at the presence of instantaneous channel knowledge. Since uplink-downlink channel reciprocity does not hold in frequency division duplex and broadband time division duplex systems, efficient channel quantization becomes important. This thesis focuses on different quantization techniques in a linearly precoded multi-user wireless system. Our work provides three major contributions. First, we come up with an end-to-end transceiver design, incorporating precoder, receive combining and feedback policy, that works well at low feedback overhead. Second, we provide optimal bit allocation across the gain and shape of a complex vector to reduce the quantization error and investigate its effect in the multiuser wireless system. Third, we design an adaptive differential quantizer that reduces feedback overhead by utilizing temporal correlation of the channels in a time varying scenario.

No description available.

Αντικείμενο της διπλωματικής αυτής εργασίας είναι η σχεδίαση, εξομοίωση και υλοποίηση συστήματος οδήγησης ψηφιακής συστοιχίας ηχείων με 5-bit PCM και 1-bit Σ-Δ σήματα εισόδου, δια μέσω διαφορετικών τεχνικών ανάθεσης των bits της πληροφορίας σε κάθε ηλεκτροακουστικό μετατροπέα της συστοιχίας, και ο υπολογισμός της απόκρισης πλάτους της συστοιχίας και των THD και THD+N σε διάφορες θέσεις ακρόασης. Τέλος γίνεται σύγκριση των αποτελεσμάτων μεταξύ εξομοίωσης και υλοποίησης. / The object of this thesis is the design, simulation and implementation of a system driving a digital transducer array loudspeaker with 5-bit PCM and 1-bit Sigma-Delta input signals, by applying different bit allocation methods on the transducers of the array, and the calculation of the array's amplitude response, THD and THD+N at different positions. Finally, simulation and implementation results are contrasted.

This thesis concerns communication across channels with multiple inputs and multiple outputs. Specifically, we consider the closed-loop scenario in which knowledge of the state of the multiple-input multiple-output (MIMO) channel is available at the transmitter. We show how this knowledge can be exploited to optimize performance, as measured by the zero-outage capacity, which is the capacity corresponding to zero outage probability. On at-fading channels, a closed-loop transmitter allocates different powers and rates to the multiple channel inputs so as to maximize zero-outage capacity. Frequency-selective fading channels call for a combination of orthogonal-frequency-division multiplexing (OFDM) and MIMO known as MIMO-OFDM. This exacerbates the allocation problem because it multiplies the number of allocation dimensions by the number of OFDM tones. Fortunately, this thesis demonstrates that simple allocations are sufficient to approach the zero-outage capacity. These simple strategies exploit the tendency for random MIMO channels to behave deterministically as the number of inputs becomes large.

Please read the abstract in the front matter this document / Dissertation (M Eng (Electronic Engineering))--University of Pretoria, 2006. / Electrical, Electronic and Computer Engineering / unrestricted

This thesis deals with the issue of the creation of a programme that would simulate the distortion that appears during the process of lossy audio coding. As the environment for the creation, the MATLAB programming language has been chosen. An encoder, which changes the subjective signal quality according to customer preferences for the bitrate, has been created as a practical part of this thesis. Its function is based on a dynamic bit allocation technique and includes an optional window switching algorithm. The theoretical background for the creation of the programme consists of an explanation of the main principles of lossy coding with emphasis on MPEG1 layer 3 operating principles. The practical chapter describes how the created programme and its parts work, and it includes results of the run quality testing. The testing was conducted using the objective assessment method PEMO-Q, and consisted of comparing the objective quality of the programme’s outputs to the quality of samples on which a regular MP3 encoder with identical settings was used.