The objective of this thesis is to study adaptive modulation algorithms for Single Input Single Output (SISO) and Multiple Input Multiple Output (MIMO) channels. The adaptive modulation technique has the advantages of flexibility, providing high transmission quality and throughput; and it can lead to a smaller volume of radio and more applications.The major contribution of the author is discussed in the following paragraphs.To the best of the author?s knowledge, the strategy and working procedures of the adaptive modulation technique have not yet been described in the literature. We propose strategies and work procedures for adaptive modulation in this dissertation. In addition, for the first time, it is proposed that the development of adaptive modulation consists of three stages, in which the adaptivity is taken to a higher level.We assert in this thesis that adaptive modulation algorithms involve two themes, i.e., channel and modulation. We analyze two types of channels: SISO and MIMO channels. We investigate the SISO channel in time, frequency, and space domains, which is efficient for designing adaptive modulation algorithms. Further, initially we propose three approaches as solutions to problems in each of the domains, i.e., adaptive techniques (time domain), OFDM (frequency domain), and MIMO technique (space domain). These three approaches are investigated in this thesis. We study the concept, structure, and characteristics of MIMO channels. The gains of three MIMO schemes and parameters influencing these gains are generalized. What is more, we introduce a new factor, the Rician K gain, to describe the MIMO gain with respect to that of SISO channels. In addition, we work out the SNR gains of these three MIMO schemes compared with SISO channels for a large SNR range, which have not been found in literature. This is indispensable for the design of adaptive modulation algorithms for MIMO channels.With regard to modulation, we list the parameters of OFDM and CDMA modulations for adaptive OFDM and adaptive CDMA algorithms to be designed. In order to obtain direct relationships between modulation parameters and throughput, we derive two throughput formulas, an OFDM formula and a CDMA formula. We show that the traditional commonly held belief that the subcarrier bandwidth influences transmission quality is true only if the channel RDS is constant. To explain this, we define a new parameter, PFR, as the product of subcarrier bandwidth and RDS. How this parameter works and by what factors it is limited are studied in this thesis.Subsequently, we propose three successful adaptive modulation algorithms, i.e., an adaptive OFDM algorithm for SISO channels, an adaptive OFDM algorithm for MIMO channels, and an adaptive CDMA algorithm for MIMO channels. Perfect channel estimation is assumed in our work. For the adaptive OFDM algorithms, we propose three subband settings. Remarkable throughput gains are obtained from all these adaptive modulation algorithms in simulations. We put forward the new idea of employing more than one MIMO scheme in a system and adapting these schemes using our adaptive modulation algorithms. Simulation results show that extra gains can be obtained by this method. The performance gains of our algorithms are dependent on the channel situation. The influence of channel parameters on these relative gains is investigated in this work as well. We introduce an SNR boundary matrix to set processing gains and modulation levels in our adaptive CDMA algorithm for MIMO channels.

Adaptive modulation and MIMO are two promising techniques to be integrated in future wireless communications for optimizing radio performance. No earlier work has been found in literature to combine adaptive CDMA with optimized MIMO parameters. In this paper we focus our attention on algorithm development and validation for adaptive CDMA with MIMO. This algorithm is based on CDMA modulation adaptation combined with MIMO parameter adaption meaning that it may utilize different MIMO schemes. Simulation results are are given to evaluate our algorithms.