Performance evaluation of inter-cell interference mitigation techniques for OFDMA cellular networks

© 2010 Dr. Weiwei WuFor emerging cellular wireless systems, the mitigation of inter-cell interference is the key to achieve a high capacity and good user experience. This thesis is devoted to the performance analysis of interference mitigation techniques for the downlink in an orthogonal frequency division multiple access (OFDMA) network, with a focus on the Long Term Evolution (LTE) standard. We investigate two types of coordination techniques for interference mitigation, namely reuse partitioning and resource prioritization in the frequency domain. First, we assume best-effort elastic traffic for broad-band data networks and introduce a new metric, called the flow capacity, to indicate the maximum traffic intensity that can be supported by a base station sector while satisfying a minimum level of provided service. We develop a queueing theoretic methodology to analyse the flow capacity for standard reuse and reuse partitioning schemes with different scheduling algorithms. Using this analysis framework, we show how an improved cell-edge throughput can translate into an improvement in the flow capacity. We develop model variants for infinite (Poisson arrivals) and finite user populations; the infinite user population model is more tractable and yields simple, insightful expressions for the flow capacity, while the finite user population model has greater practical relevance. Furthermore, we develop a methodology to account for the effect of interference from neighbouring base stations with an arbitrary level of loading. Next, we propose possible distributed realizations of interference coordination schemes in a reuse-1 environment, which are based on setting allocation priority in the frequency domain. The proposed schemes are more suited to narrow-band services and can be implemented in a fractional loading scenario