|dc.description.abstract||Mobile devices have changed from being terminal equipment for making voice calls, video calls, playing games and sending messages to including differing functionalities such as acting as a server and client for sharing information. Information sharing directly between modern mobile phones can be done using Bluetooth or WiFi functionalities. Such a network does not need a fixed infrastructure, thereby allowing people to create Opportunistic mobile phone networks when they need the service, thereby avoiding charges from established service providers. Packet routing in opportunistic wireless networks is a key function to ensure that intermediate nodes act accordingly to achieve end-to-end communication in continuously changing peer-to-peer network nodes positions. Movement and mobility of connecting nodes plays a crucial role in the performance of such a network.
Undoubtedly, designing or identifying routing and forwarding mechanisms is a big challenge in this area. In this study, movement of cluster mobile phones of moving students at a university campus was described and mobility models were identified. The main objective was to understand the patterns of mobile phone node cluster movement and derive a model that can be used to evaluate Opportunistic mobile phone network performance metrics like throughput, coverage and delay. Using 802.11 WiFi connection to connect participating mobile phones of moving students, we analysed and compared routing protocols; their features, functionality and benefits, and identified those parameters relevant to the group mobility of mobile phone users identified at NUST campus.
This initial study established an understanding of geographical paths of interest for a further study on optimal routing schemes for practical opportunistic networks that may occur at a High Education institution campus when students are moving around and between lecture rooms. Informed by the findings, a free non-fixed infrastructure network model utilizing mobile phones that suit the NUST campus setting, which is fairly affordable and effective for students and staff to communicate locally, is recommended. Some of the most used Proactive and Reactive routing protocols in opportunistic networks were discussed. Simulated results showed that conditions like deterministic, semi- deterministic and random mobility patterns often occur in a network type of the NUST campus setting with Reference Point Group Mobility model closely matching how the inter-lecture movements of students are conducted. Reactive protocols complemented with the mobility pattern identified for the NUST campus, especially when students move in groups. Considering that the number of students always increases each year due to new intakes, we found that the performance differs across different network sizes, if we consider the movement patterns of other schools/faculties at the NUST campus.||en_US