On Minimizing TCP Traffic Congestion Control in Vehicular Internet of Things

Vehicular communication is an emerging field and will experience a substantial rise in wireless exchange of information. Heavy usage of internet traffic introduces a higher likelihood of network congestion. Several internet congestion control algorithms are used to regulate transmission rates that continuously monitor packet losses and timeouts to identify congestion in the network. In the vehicular environment, packet loss eventuality happens due to mobility and channel losses. In this paper, a new algorithm called the Congestion Aware Rate Tuning (CART) is proposed to identify network congestion based on the estimation of Bottleneck Limit Threshold (BLT) parameter. This algorithm won’t respond to packet loss and timeout conditions. The parameters average throughput, average goodput and mean delay are taken as metrics to analyse the performance of this algorithm. Simulation results of the comparison between the existing Transmission Control Protocol Westwood algorithm and CART algorithm in NS-3 shows a considerable increment in throughput, goodput, and mean delay in vehicular conditions.