Abstract
Since under the UMTS framework, the development of S-UMTS follows that of T-UMTS with the ultimate aim of achieving close integration between S-UMTS and T-UMTS air-interface designs, closed-loop transmit power control (TPC) remains a physical-layer procedure option for enhancing the performance of S-UMTS. In S-UMTS, however, closed-loop TPC should be modified to take into account large TPC command delays due to propagation. In this thesis, we study strength-based closed-loop TPC schemes for S-UMTS in order to develop TPC schemes that mitigate the effects of TPC command delays due to propagation. The study focuses on link-level analysis of the SW-CDMA based air-interface for the S-UMTS. Firstly, a mathematical model describing the dynamic behaviour of the closed-loop TPC has been developed as a stochastic difference equation, where the effects of TPC command delays in the presence of random disturbance are modelled as a random walk process, with a probability density function (pdf) of the TPC error process as its solution. The pdf is obtained by solving a corresponding Fokker-Planck equation whose derivate moments are obtained via statistical linearisation. The results show that although it is widely accepted that the TPC error process is lognormally distributed, the nonlinearity due to the fixed TPC step-size may flatten the pdf when the TPC delays and step-size are large. Secondly, predictive TPC schemes that mitigate the effects of delays and adaptive techniques that improve upon the performance offered by standard algorithms have been proposed and evaluated through Monte-Carlo simulations. In particular, a simple LMS algorithm has been chosen as an appropriate tracking algorithm for the predictive TPC schemes in S-UMTS, and in order to improve upon the performance offered by the standard LMS algorithm, a filter-shaped LMS algorithm for predictive TPC has been proposed for S-UMTS. Furthermore, the simulation results have shown that step-size adaptation, when complemented with power prediction, significantly improves the performance of the closed-loop TPC schemes even in the presence of TPC command delays. Finally, numerical results have shown that the accuracy of TPC has significant influence on the capacity. For example, a modest improvement in the standard deviation of the TPC error from 1.5 dB to 1.0 dB may yield a capacity improvement from 21 to 32 users per spotbeam: a relative gain of approximately 50%. Therefore, closed-loop TPC remains an important radio resource management procedure for mitigating slow-fading due to shadowing in S-UMTS, power prediction mitigates the effects of TPC command delays, and more accurate TPC schemes are necessary to realise high capacity S-UMTS networks.