Document Type
Dissertation
Date of Award
Summer 2017
Degree Name
Doctor of Philosophy in Electrical Engineering - (Ph.D.)
Department
Electrical and Computer Engineering
First Advisor
Osvaldo Simeone
Second Advisor
Joerg Kliewer
Third Advisor
Abdallah Khreishah
Fourth Advisor
Bipin Rajendran
Fifth Advisor
Gesualdo Scutari
Abstract
The recent growth of the number of smart mobile devices and the emergence of complex multimedia mobile applications have brought new challenges to the design of wireless mobile networks. The envisioned Fifth-Generation (5G) systems are equipped with different technical solutions that can accommodate the increasing demands for high date rate, latency-limited, energy-efficient and reliable mobile communication networks.
Mobile Cloud Computing (MCC) is a key technology in 5G systems that enables the offloading of computationally heavy applications, such as for augmented or virtual reality, object recognition, or gaming from mobile devices to cloudlet or cloud servers, which are connected to wireless access points, either directly or through finite-capacity backhaul links. Given the battery-limited nature of mobile devices, mobile cloud computing is deemed to be an important enabler for the provision of such advanced applications. However, computational tasks offloading, and due to the variability of the communication network through which the cloud or cloudlet is accessed, may incur unpredictable energy expenditure or intolerable delay for the communications between mobile devices and the cloud or cloudlet servers. Therefore, the design of a mobile cloud computing system is investigated by jointly optimizing the allocation of radio, computational resources and backhaul resources in both uplink and downlink directions. Moreover, the users selected for cloud offloading need to have an energy consumption that is smaller than the amount required for local computing, which is achieved by means of user scheduling.
Motivated by the application-centric drift of 5G systems and the advances in smart devices manufacturing technologies, new brand of mobile applications are developed that are immersive, ubiquitous and highly-collaborative in nature. For example, Augmented Reality (AR) mobile applications have inherent collaborative properties in terms of data collection in the uplink, computing at the cloud, and data delivery in the downlink. Therefore, the optimization of the shared computing and communication resources in MCC not only benefit from the joint allocation of both resources, but also can be more efficiently enhanced by sharing the offloaded data and computations among multiple users. As a result, a resource allocation approach whereby transmitted, received and processed data are shared partially among the users leads to more efficient utilization of the communication and computational resources.
As a suggested architecture in 5G systems, MCC decouples the computing functionality from the platform location through the use of software virtualization to allow flexible provisioning of the provided services. Another virtualization-based technology in 5G systems is Network Function Virtualization (NFV) which prescribes the instantiation of network functions on general-purpose network devices, such as servers and switches. While yielding a more flexible and cost-effective network architecture, NFV is potentially limited by the fact that commercial off-the-shelf hardware is less reliable than the dedicated network elements used in conventional cellular deployments. The typical solution for this problem is to duplicate network functions across geographically distributed hardware in order to ensure diversity. For that reason, the development of fault-tolerant virtualization strategies for MCC and NFV is necessary to ensure reliability of the provided services.
Recommended Citation
Al-Shuwaili, Ali, "Mobile cloud computing and network function virtualization for 5g systems" (2017). Dissertations. 33.
https://digitalcommons.njit.edu/dissertations/33
