الفهرس 
CHAPTER 1: INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 95 |  |  |
| | | from | 95 | | |
Abstract
Social network involves the use of social media platforms on the internet to connect with friends, family, etc. The most popular social networking sites are Facebook, Instagram and Twitter. The use of social behaviors and relationships between users define the structures and interactions between them and their affiliated organizations on social networks. One of the important services is the recommender system and one of its applications is the social recommendation of users. The recent applications are mostly created for mobile devices. In terms of mobile devices, social connection reflects the frequency of encounter, so that the users of these devices are socially connected if they communicate with each other frequently. We focus on the ad hoc network to enable the social recommendation based on mobile devices. Users can be described as a single device that can transfer and receive data to and from other devices nearby. Mobile ad hoc network (MANET) is a collection of mobile nodes that are dynamically situated in such a way that the interconnections between nodes can alter on a continuous basis. MANETs enable their users to communicate in an ad-hoc infrastructure-less environment. A Mobile Social Network (MSN) is introduced which consists of nodes that are related socially, in addition to their physical connection. MSN is introduced to merge the MANET with the social networking. The layer of social connectivity, combined with the physical connectivity of being in the communication range of each other, can help improve the routing
ii
performance. The aims of MSNs protocols are reducing end-to-end latency, maximizing the number of packets delivered and timely delivery of messages between two nodes. MSN does not have an end-to-end connectivity between nodes due to restricted radio coverage, network versatility and other factors. A routing protocol discovers pathways between nodes to make communication easier within the network. MSN faces several challenges including the limited energy resource, intermittent connectivity, and the limited storage. This work overcomes these challenges and improves routing efficiency to socially recommend users based on MSN, social metrics are exploited to carefully choose the candidate relays in MSNs. In this context, the multicast routing protocol is addressed through predefined criteria.
We introduce two protocols; the first protocol, Time-based Encounter of Socially Similar nodes, TESS, this protocol introduced to exploit users’ social profile and their interactions to improve relay recommendation and find target destinations. TESS tests social similarity of encounter nodes against the destinations. The second protocol, the Sustainability by using Socially Similar nodes, 3S, which uses the social profiles of registered users and their network connections to locate destinations from their social profiles. The social similarity between the desired destination profile and the encounter node is calculated, taking into consideration the remaining energy of the nodes. The objective is to reduce the number of dead nodes caused by mobile nodes overloading. This work aims to preserve and strengthen the multicast routing by reducing energy consumption and reducing the number of network
iii
transmissions. The recommended relays and destinations are established based on our measurements, and the main data is transferred to these nodes after validating their battery power value Using the ONE simulator, the proposed protocols have better results than the previous ones, according to number of relays, hop count, number of messages dropped, overhead ratio, and approximated results in delivery ratios. To sum up by exploiting the social connections between the mobile nodes, we can better recommend the nodes within the network.