| The ever growing popularity of powerful mobile devices such as smart phones and tablets has opened a new dimension in communication for effi-ciency of capacity intensive applications, consequently the recently deployed fourth generation (4G) networks could not meet the future capacity demands.To this effect, the attention of the mobile communication industry has now shifted towards the development of the fifth generation (5G) networks. The cloud radio access network (C-RAN) has been recognized as a key component of the 5G network, which integrates the cloud computing technology into the traditional cellular system to realize a large scale cooperative signal processing and networking. Although C-RAN can enhance spectrum efficiency and energy efficiency, a lot of key challenges remain, including analytic performance study and resource allocation. The main contributions of this thesis are as follows:(1). This thesis utilizes the centralized cooperative processing capabilities of C-RAN and proposes a user centralized joint coordinated transmission scheme(UC-JCT) scheme. The scheme requires all cooperating remote radio heads(RRHs) to transmit copies of the same data simultaneously towards a target user forming a distributed MIMO processing. Two RRH cooperation cases are con-sidered namely, 1) full coordination where all the RRHs located within a given UE' s coordination region contribute to jointly transmit to the served user; and 2) partial coordination in which only a subset of RRHs in the user coordination region can transmit to the user to lessen both the processing complexity at the BBU pool and the capacity burden on the fronthaul at the expense of reduce performance. For both cases the corresponding expressions of coverage prob-ability is derived analytically in closed-integral form. The proposed scheme have shown better performance compared with multi-antenna interference can-cellation and the conventional transmission scheme without RRH coordination by turning the most dominant interference into usable signals.(2). The fronthaul links in C-RAN are capacity and time delay constrained,this can affect the achievable capacity. To address these challenges, the sec-ond contribution consider the use of device-to-device (D2D) communication to offload RRHs. Specifically, H-CRAN with non-uniformly deployed D2D communication is proposed, in which D2D links are only utilized outside a specified distance from any HPN. Based on stochastic geometric theorems, the coverage and average ergodic rate of a randomly located typical user is pro-vided. Furthermore, the data traffic delay experience at both the fronthaul and RRH is modeled based on queuing theory and the average data delivery latency as a quality of service perceived by the users when served by RRHs evalu-ated. The theoretical analysis shows that when the fronthaul links are capacity constrained, H-CRAN with non-uniformly deployed D2D communication ob-tains lower average data delivery latency by comparing with H-CRAN with uniformly deployed D2D communication and simple H-CRAN.(3). To relieve the increased processing needs at the BBU and the capacity bur-den on the fronthaul, RRHs clustering is commonly utilized, however, it may not be suitable for heavily loaded networks because depending on the cooper-ation technique adopted, only a limited number of users can be simultaneously served within a given cooperative cluster, this will therefore leads inefficient spectral resource utilization. To this effect, the thesis studies the effect of in-tegrating D2D communication in H-CRAN system with the aim to improve network spectral utilization. Two cluster based cooperative transmission and coordination schemes are researched, namely: Zero force beamforming (ZFBF)and Non-coherent joint coordinated transmission (NC-JCT), and a correspond-ing analytical result is developed. The numerical results give the performance gains of spectral efficiency per unit area using the two cooperation techniques. |
PDF Full Text Request