On the approach to capacity expansion after an initial C-RAN deployment

Abstract

This dissertation studies the feasible ways to capacity expansion after an initial cloud radio access network (C-RAN) deployment. An additional installation of fiber fronthaul and wireless fronthauling is realistically difficult to implement from a cost perspective and also from a geographical point of view. Therefore, the capacity expansion of initial C-RAN architecture is the practical challenges to address. In this dissertation, we propose two feasible approaches to increase capacity after an initial C-RAN deployment. First, we present a variant of C-RAN architecture, which we term partly wireless C-RAN (PW-CRAN). It is characterized by extra radio nodes connected through the existing remote radio heads (RRHs). Because of the newly connected nodes and the consequent additional delay, the operation of PW-CRAN requires different approaches compared with that of the typical C-RAN. Specifically, the effect of delayed channel state information (CSI) on the suitability of network cooperation should be identified. To tackle this problem, we introduce two representative cooperation strategies and evaluate the performance with regard to the delay via simulations. Numerical results suggest that it is better to exclude the extra nodes from cooperation if they incur excess delay. Furthermore, whether the delay is deemed excessive depends on the interference environment. Hence, we provide quantified guidelines on the cooperation strategy of PW-CRAN. Moreover, we carry out the extended performance evaluation for the special case of proposed PW-CRAN in which there exist the same number of extra w-RRHs as the number of c-RRHs supporting the unchanged user demands. To be specific, simulation results clearly show that the advantageous MIMO gain obtained from the use of supplementary w-RRHs and the impact of network size on the achievable spectral efficiency (SE) performance. Additionally, the impact of fronthaul capacity limitation on the PW-CRAN performance is evaluated, thus demonstrating that the cooperation strategy should be adjusted to the conditions of fronthaul capacity in PW-CRAN. Next, we consider the possibility of capacity expansion from cooperation between different C-RAN operators. In general, cooperation of co-located mobile network operators can provide potential benefits for the capacity expansion without further densification of radio nodes. However, such benefits need to be scrutinized for coexisting C-RAN operators because the inter-operator cooperation may compromise the superb interference coordination capability that each C-RAN has. Furthermore, altering C-RAN infrastructure for the cooperation may incur a high investment cost. In this context, we evaluate quantitative gain of inter-operator coordination strategies to provide the C-RAN operators with a guideline on their cooperation decisions. The coordination strategies encompass dynamic user association and the spectrum sharing which aim at maximizing the total user throughput. We propose a heuristic algorithm that reduces the computational burden of the coordination significantly. Numerical results suggest that the inter-operator cooperation is beneficial particularly when the network size of each operator tends to be highly asymmetric. It is also verified that the users who belong to smaller network attain more coordination gains.