Sharif Digital Repository

Due to the explosive growth of mobile data traffic and growth of the number of access points (APs), Wi-Fi data offloading is a promising solution to cope with the data explosion. In delayed WiFi offloading, mobile users (MUs) can intentionally bear some delay in order to increase the chance of offloading and remarkably improve offloading ratio. In this thesis, we propose a new decision method for delayed offloading in a network comprised of a base station (BS), some non-overlapping WLANs, and some MUs. We take into account the residual service time and the congestion level in WLAN as the key parameters for decision making in offloading process. Because these parameters are time-varying and... 

With the growing of smart device softwares, the speed of applications and energy consumption has become a concern for developers of smart device. The unobtrusive running of applications, when the smartphone processor is occupied, alongside the energy issue, when the user needs an urgent need to use a low power storage device, is of concerns for the use of these devices.In order to overcome this issue, cloud computing platforms, transfer smartphone tasks and duties to the cloud platforms and retrieve the results of the calculations to the phone.Obviously, communicating with the cloud requires a communication channel. The communication channel for the internet can either be the internet in... 

Based on fog computing paradigm, new applications have become feasible through the use of hardware capabilities of smart phones. Many of these applications require a vast amount of computing and real-time execution should be guaranteed. Based on fog computing, in order to solve these problems in is necessary to offload heavy computing to servers with adequate hardware capabilities. On the other side, the offloading process causes time overhead and endangers the real-timeliness of the application. Also, because of the limited battery capacity of the handheld devices, energy consumption is very important and should be minimized.The usual proposed solution for this problem is to refactor the... 

With the emergence of 5G networks, response time is becoming increasingly more important. 5G networks facilitate usage of Mobile Edge Computing. MEC provides computing capabilities at the edges of cellular networks. Since the computational capability in mobile devices is limited, running high performance applications using external resources is a way to overcome this limitation. Workload offloading in MEC is an approach that provides additional computation capability for users to meet the desired response time. In this study, we presented a Mixed Integer Non-Linear Programing model for offloading different workloads on a heterogeneous set of MEC servers to minimize the SLA violations. We... 

Collaborative edge computing (CEC) is a recently popular paradigm enabling sharing of computation resources among different edge devices. In CEC, multiple stakeholders (mobile users, mobile edge computing servers, cloud servers,...) collaborate to provide new computation capacity in order to perform computation-intensive tasks efficiently in the edge. In the thesis, each task, the data that should be computed in the cloud or edge, was modeled as a graph of dependent sub-tasks. Resource allocation for task offloading is an important problem to address in CEC as we need to decide when and where each subtask is executed. In this work, we mathematically formulate the problem of resource... 

In order to use the high performance capabilities of a computing system, first it is required to provide a proper modelling for the job and the system's environment. Second, it is required to design scheduling and offloading algorithms based on the job and the system modeling and third for evaluating the performance of these algorithms. It is needed to either simulate them or prove their approximation factors. This project aims to carry out these three parts for the Edge Computing environment. The laid out model of the system in this thesis consists of many devices that are distributed around the network, which they can execute tasks parallel to each other, and between each two devices there... 

The development of deep neural networks is making tremendous progress in various fields, including processing Image, speech processing and other areas. Despite this tremendous achivement, neural networks have a lot of computational overhead and memory access that prevent them from being used in resource-constrained devices. We also know that many neural network applications are of great importance in mobile devices, and it is desirable for us to use their power in this regard. Many efforts have been done at different levels to solve the problem of executing deep neural networks on these devices. In this research, an approach based on offloading is used in which two different small (on the... 

IoT-based systems use cloud computing servers that are generally far away from them to perform tasks. Whenever there is a computational task that requires heavy and complex processing, the problem of offloading decision is to identify the best computing server to perform the processing related to this task. Numerous dynamic factors affect the solution to this problem.Optimization in energy consumption of the device that has generated a task alongside reducing its calculation delay can be applied in solving the offloading problem. The optimization problem that is formed to minimize the energy consumption in the offloading decision is an integer programming problem that is generally difficult...