| Summary: | Over the years WLAN Access Points (AP) have seen massive deployment across most of the modern cities in both domestic and commercial sectors, for example, in houses, shopping malls, airports, etc. WLAN APs were initially meant to support network services in indoor environments. Hence, their cover- age region is small and is typically restricted to a certain indoor area. Recently, the idea of using these already available APs from outdoor vehicular envir- onments has come under scrutiny. Among the various challenges associated with such WLAN-based vehicular communications, this thesis focuses on two main issues, namely disruption and handover latency. Although WLAN APs exist in large numbers across the roads in most de- veloped cities, the placement of these APs is highly unplanned. Due to this unplanned deployment, WLAN APs cannot support continuous connectivity over a large mobility domain. Consequently, WLAN APs offer irregular net- work services on the move. This irregularity or intermittency in the network services is called disruption. While various previous works have focused on tolerating disruption, this thesis explores a completely new research direction - mathematically modelling disruption. It has been argued that the first step to- wards effective disruption tolerance is to mathematically represent disruption. This work sets out to develop a mathematical model that can measure disrup- tion and comparatively analyze different areas of the city. Secondly, a vehicle spends very little time within the footprint of an AP. Most of this already small connection time is consumed in handing over to the APs. Because of the small outdoor coverage region of the AP and the high speed of the mobile node, han- dovers inevitably occur frequently. In order to achieve smoother transitions between the APs, the handover latency must be reduced. This work measures the delays involved in the handover procedure and presents a channel scan- ning scheme to reduce the scanning phase delay.
|
|---|
