Research On Techniques Of Statistical QoS Guarantees For MIMO Systems

This is an era when mobile internet is rapidly developing, and the users’expectation for Quality of Service (QoS) of wireless communication systems is growing.The research on using limited spectrum resources to provide QoS guarantees for theincreasingly heavy multimedia services has drawn great interests. Multiple-inputmultiple-output (MIMO) systems that employed multiple antennas at both transmitterand receiver are capable of exploiting extra spatial freedom gain and can providesignificant improvements in terms of capacity and reliability without requiring extrapower and bandwidth. It is chosen as the main physical layer transport technology forthe fourth generation of wireless communication (4G). On the other hand, it is infeasibleto provide deterministic QoS guarantees for multimedia services in wireless systemsbecause of the random and time-varying property of wireless channel. Therefore, thisdissertation researchs on the statistical QoS guarantees for MIMO systems basing on thetheory of effective capacity. The author’s main contributions in this dissertation are asfollows:1. For MISO systems in the presence of channel correlation, closed-formexpressions of effective capacity are derived under the assumption of Nakagami-m andRician fading channels. Some Monte Carlo simulation has verified theoretical analysisand showed the impact of antenna configuration, degree of spatial correlation, QoSrequirements and fading severity to effective capacity.2. For MIMO systems adopting MRT/MRC and M-array AMC scheme, firstly aclosed-form expression of effective capacity is derived under the policy of fixed powerallocation. For variable power allocation, the optimal power allocation scheme and theassociated outage probability and effective capacity are derived. Some Monte Carlosimulation has verified the theoretical analysis and showed the performance gain ofoptimal power allocation in terms of providing statistical QoS guarantees.3. For MIMO systems adopting TAS/MRC and M-array AMC scheme, firstly theprobability density function of receive SNR is derived, based on which a closed-formexpression of effective capacity is derived for fixed power allocation. For variablepower allocation, the optimal power allocation scheme and the associated outageprobability and effective capacity are derived. Both Monte Carlo simulation andtheoretical analysis suggest that the optimal power allocation scheme can greatlyimprove the system performance in statistical QoS guarantees. 4. For dual hop DF relaying MIMO systems adopting MRT/MRC scheme, firstly aexpression of optimal receive SNR is derived, based on which a closed-form expressionof outage probability is obtained. The closed-form expressions of average bit error rateare derived under M-PAM、M-QAM、M-PSK and M-PSK modulation respectively.Finally, a closed-form expression of effective capacity is derived. The related MonteCarlo simulation has verified the theoretical analysis and indicated that applyingMRT/MRC scheme can greatly improve the performance in terms of outage probability,average bit error rate and statistical QoS guarantees.