An Opportunistic Mobile Data Ooading Framework 

Get Complete Project Material File(s) Now! »

Mobile Broadband Internet

With the advance of smart devices and multimedia applications, users demand evennmore bandwidth from their cellular operators. Technologies to provide mobile broadband Internet has been evolving in the last years. The main advance relates to thendata rate oered by the network, which has evolved from few kilobits to gigabits in thenpast years.
The rst generation network (1G) introduced the concept of centralized cellularnarchitecture and applies Frequency Division Multiple Access (FDMA) to separate usersnin the frequency domain. However, 1G provides only voice services. Wireless datancommunication through mobile phones was introduced by the Global System for MobilennCommunications (GSM) standard, the second generation (2G) of mobile networks. 2G networks use Time Division Multiple Access (TDMA) and provide data rates of 9.6 Kbps. General Packet Radio Service (GPRS) and Enhanced Data rates for GSM Evolution (EDGE), known as generation 2.5, enhanced the data rate of GSM networks up to 237 Kbps. The third generation network (3G), dened by International Mobile Telecommunication (ITU), oers a peak data rate of at least 200 Kbps and up to 84Mbps. Nowadays, 3G networks are already widely deployed in the market. However, 3G networks are being replaced by the fourth generation, also called 4G, which uses two new standards: Long Term Evolution (LTE) [Sesia et al., 2009] and Worldwide Interoperability for Microwave Access (WiMax) [Vaughan-Nichols, 2004]. Both LTE and WiMax are based on Orthogonal Frequency Division Multiple Access (OFDMA). The standards dene peak rates of 100 Mbps for high mobility communication and 1 Gbps in low mobility scenarios. Nowadays, LTE networks dominate the market of 4G cellular data services.
The key dierence between 3G and 4G networks is that OFDMA increases the exibility of resource allocation by increasing the quantity of time and frequency slots [Ghosh et al., 2010]. For simplicity, we will refer to 3G and 4G networks as 3G, since 3G is largely deployed by the WISPs and has broader support by commercial o-the-shelf devices.
Academia and industry are working towards the fth generation of mobile network standards (5G), which aims to improve scalability in terms of cost, energy and resource eciency [Osseiran et al., 2013]. 5G is a work in progress and it is planned to be rolled out to the general public in 2020 or further.
As part of the 5G network, a work in progress group in 3rd Generation Partnership Project (3GPP), formed by several IT companies, attempts to extend the coverage of LTE antennas through Device-to-Device communication (D2D). The project, called Study on LTE Device to Device Proximity Services (ProSe), has identied the use cases and requirements, and now the physical layer is being rening to allow Deviceto- Antenna and Device-to-Device communications on the LTE radio [3GPPP, 2013]. This project can be seen as a promising solution for short-range communication in the mobile opportunistic data ooading problem.

Opportunistic Networking

Opportunistic networks (OppNets) are networks that, unlike classic networks, are prone to frequent disconnections and high communication delays. In some scenarios, it may be the case that node disconnection is the most frequent state since nodes may only communicate when a link is established (in OppNet jargon, this is called a contact). The frequent disconnections preclude the use of classic message forwarding paradigms, since these paradigms are based on the establishment of an instantaneous end-to-end path from source to destination. As a consequence, OppNets employ the store-carry-forward paradigm, where messages are stored in intermediate nodes until a suitable forwarding opportunity occurs. Each node selects a set of messages to be forwarded using the recently established link, using some sort of priority scheme [Zhang, 2006]. The process of storing a message for later transmission is also known in the literature as custody.

READ  Impact of PV generation on the primary frequency response 

WiFi availability and Ooading

IEEE 802.11 was originally designed with data rate of just 1 Mbps. However, the amendments 802.11a/b/g/n, broadly adopted in the market, increase the data rate up to 600 Mbps. Furthermore, IEEE 802.11ac aims to achieve 1 Gbps using multiple antennas, and the target data rate of IEEE 802.11ad, as proposed by the Wireless Gigabit Alliance, is about 7 Gbps. Nowadays, the deployment of IEEE 802.11 access points scattered around several cities leads the major eort to ooad overloaded cellular infrastructures.
WiFi coverage in metropolitan areas was characterized in [Bychkovsky et al., 2006]. Bychkovskyet al. conducted an experiment to evaluate the feasibility of using WiFi access points around Boston metropolitan area, driving vehicles at regular speeds in the city, during July 2005 and July 2006. They showed an average duration of link layer connectivity of 24 seconds, while only 3.2% of access points provided end-toend communication, which means that applications using only open WiFi connection should be delay-tolerant. A comparison between WiFi and 3G networks appears in [Gass and Diot, 2010] and [Chen et al., 2012]. In [Gass and Diot, 2010], the authors show that since the contact time 3G networks is greater than on WiFi networks, when the client is moving, the amount of transferred data (download) is larger in a 3G network. However, since the upload data rate is of the order of kilobits per second in 3G networks, WiFi networks outperform 3G network on data upload.

Table of contents :

List of Figures
List of Tables
List of Acronyms
1 Introduction 
1.1 Motivation
1.2 Denitions
1.3 Problem
1.4 Use Cases
1.5 Contributions
1.6 Document Organization
2 Fundamentals 
2.1 Mobile Broadband Internet
2.2 Opportunistic Networking
2.2.1 Types of Opportunistic Networks
2.3 Game Theory
2.3.1 Basics
2.4 Utility Theory
2.4.1 Denition
2.4.2 Utility Function: Single and Multi Criteria
2.5 Conclusion
3 Related Work
3.1 Methodology
3.2 Femtocell Ooading
3.3 WiFi availability and Ooading
3.4 Opportunistic Mobile Data Ooading
3.4.1 Selecting Best Relay Candidates
3.4.2 3GPP Device-to-Device Proximity Services
3.5 Selshness, Cooperation and Incentive Mechanisms
3.5.1 Classes of Incentive Mechanisms
3.6 Conclusions
4 On the Feasibility of WiFi Ooading 
4.1 Connectivity Categorization
4.1.1 Public WiFi Hotspot
4.1.2 WISP and Private Hotspots
4.1.3 WiFi Connectivity
4.1.4 3G Connectivity
4.1.5 Lessons Learned and Discussion
4.2 Ooading User Generated Data
4.2.1 User Localization Database
4.2.2 Evaluation
4.2.3 Towards WiFi Ooading
4.3 Conclusion
5 OppLite: An Opportunistic Mobile Data Ooading Framework 
5.1 Motivation
5.2 OppLite Framework
5.2.1 OppLite Framework
5.2.2 Opportunistic criteria
5.2.3 User Prole
5.2.4 Utility Function for Single Criterion
5.2.5 Multi-Criteria Aggregation Function
5.2.6 Decision Algorithm
5.3 Conclusions
6 OppLite Evaluation 
6.1 Application Scenarios
6.2 Methodology
6.2.1 Simulations
6.2.2 Parameters
6.2.3 Trac Model
6.2.4 Content Request Pattern
6.2.5 Traces
6.3 Trace Analysis
6.4 Simulation Results
6.4.1 Amount of Relays and Opportunistic Nodes
6.4.2 Opportunistic Relaying (OpR) Evaluation
6.4.3 Cache-and-Forward (CaF) Evaluation
6.4.4 Relay-as-Cache (RaC) Evaluation
6.5 Conclusions
7 Towards Incentive Mechanisms for Opportunistic Mobile Data Ofoading 
7.1 Incentive Mechanism for Opportunistic Forwarding
7.1.1 MINEIRO – Reciprocity based Incentive Mechanism
7.1.2 MINEIRO as a Bayesian Game
7.1.3 MINEIRO Evaluation
7.2 Engaging Cooperation in Opportunistic Ooading
7.2.1 Data Reward to Engage Users’ Cooperation
7.2.2 User Centric Incentive Mechanisms
7.3 Conclusions
8 Conclusions and Future Work 
8.1 Conclusions
8.2 Future Work and Research Perspective
8.2.1 WiFi Infrastructured Ooading
8.2.2 Opportunistic Mobile Ooading
8.2.3 Incentive Mechanisms Evaluation


Related Posts