SECURITY BASED VEHICULAR NETWORKS USING CONGESTION CONTROL ALGORITHM

Abstract

Supportive inter vehicular applications frequently on replace of transmit single-hop messages with vehicles on a particular power channel, which present in depth sequence on the vehicle location velocity, heading, quickening, and other numbers of importance In this thesis, they apply the network utility maximization (NUM) methodology with a different approach: First, since standards allow the use of different transmit power levels, they assume that vehicles an use a set of transmit power levels and select a particular beaconing rate for each power in the set, transmitting at multiple power levels with different rates during the cycle. Network interfaces for vehicular networks actually allow for setting the transmit power for each individual frame. Second, the optimization variable used in the utility function is not simply the
Beaconing rate but the beaconing rate used with one power multiplied by the number of neighbors reached with that power. This new variable counts the total number of copies of a beacon that are delivered (in the absence of errors) in its neighborhood, and hence, it provides a measure of the degree of dissemination of the state of a given vehicle. Thus, they call this new variable the beacon dissemination rate (BDR). Each vehicle seeks to maximize the BDR, which can also be seen as a measure of the awareness its neighbors have of it. From this model, they derive a particular distributed algorithm, with guaranteed convergence to a fair allocation and which is remarkably flexible since vehicles can independently and dynamically adapt the algorithm parameters to the requirements of a wide range of applications. For instance, it can be seamlessly used to implement prioritized congestion control, in the sense that vehicles with special needs are allocated higher rates to disseminate their status more frequently. In fact, the use of multiple power levels in a cycle, which is a novel approach compared with previous proposals, supports the parallel execution of multiple applications, with different quality-of-service requirements, on top of it. Replication consequences validate our approach and confirm that it provide fair rate allocation in realistic various and dynamic scenario by means of packet losses.