Radio resource management and scheduling schemes in a multi-dimensional radio resource environment

Abstract

In this dissertation, we investigate efficient scheduling algorithms with medium access control and radio resource management in multi-dimensional radio resources. For a relatively simple space-domain scheduling, we propose a two-dimensional channel estimation scheme based on a moving average-level crossing rate ($\sl{MA-LCR}$) plane for classifying each user`s channel status in the respect of $\it{geometry}$ and mobility and also propose an adaptive multiple-input multiple-output (MIMO) transmission scheme that dynamically select an appropriate MIMO mode according to varying channel states. Based on the two-dimensional channel estimation, we derive an optimal MIMO mode selection criterion in single user MIMO systems and extend to apply the geometry and mobility information to the design of an optimal dynamic scheduler in multi-user MIMO systems. Furthermore, we propose a linear signal-to-interference-and-noise ratio (SINR) prediction scheme to alleviate performance degradation due to delayed channel state information (CSI) and the predicted CSI provides a significant increase in the throughput gain for $\it{opportunistic}$ scheduling. For a relatively simple time/frequency-domain power allocation, we consider a constant-power allocation (CPA) scheme and investigate the effect of $\it{cutoff}$ values on the performance. The proposed scheme provides near-optimum performance close to the optimal water-filling capacity and it reduces the peak-to-average power ratio. Further, we propose an iterative solution for allocating/assigning power, users, and subcarriers in order to ensure proportional fairness among users for orthogonal frequency division multiple access (OFDMA) systems. Based on the proportional fair (PF) criterion, we propose a near-optimal proportional fair allocation (PFA) scheme exploiting multi-user diversity as well as frequency diversity, and also propose a CPA scheme employing $\it{modified signal-to-noise ratio}$ (SNR) ordering and simple $\it{...