Research Of Congestion Control Mechanism Based On Cyber-Physical System In VANETs

Vehicular Ad-hoc Networks are a special kind of wireless Ad-hoc network, which is made of V2V(Vehicle to Vehicle) and V2I(Vehicle to Infrastructure). Vehicles periodically send status information to the around vehicles via VANETs, and warning messages to the surrounding vehicles to avoid traffic accidents when the accident happened. Vehicles can also provide in-car entertainment to improve traffic travel interests.In the Vehicular Ad-hoc Networks, it is very important to exchange information between vehicles. Exchanging information makes it possible to perceive each other and provide reference information for safety applications and non-safety applications in a vehicle. Because of the vehicle’s periodically sending the message, when the surrounding vehicles’number is excessive, the channel will become clogged if you don’t adopt the strategy of congestion control, leading to interference increasing and signal to interference ratio’s dropping of the receiver, which seriously affects the efficiency of the communication between the vehicles.The main work of this paper is organized as follows:By analyzing the impact of carrier sense threshold and contention window on the network performance, we proposed a joint carrier sense threshold and contention window control algorithm to guarantee the network performance. In the network, carrier sense threshold and contention window are very important parameters. The carrier sense threshold determines a node whether to transmit the information. The contention window can determine the node back off time when node discover the channel in busy. Therefore, according to the different status of the network, adaptive different adjustment of carrier sensing threshold and congestion control window, we can improve the network throughput and reduce the packet delay, etc. We use a parameter reflects the state of the network which divides the network into three states, On the basis of the different states of the network, we set different values of carrier sense threshold value and contention window and take different measures to adjust carrier sense threshold and contention window. The algorithm can improve the network throughput and the packet delay, etc. Second, we propose a congestion control method based on Linear Quadratic Gaussian (Linear Quadratic Gaussian, LQG).As we know, the actual network throughput is often different from the expected value. We monitor the network throughput, and dynamically adjust the carrier sense threshold base on the difference between the expected network throughput and the actual network throughput. This algorithm can make the network throughput maintained near the expected value. We use the Linear Quadratic Optimal control method to obtain the required carrier sense threshold. This algorithm can be divided into two steps:(1) Use Kalman filter to filter the noise in the network to obtain an accurate state variable. (2) Use the Linear Quadratic Optimal control method to obtain the optimal control variable. The optimal control variables will ensure the network is not congested. Simulation result shows that the proposed LQG algorithm can effectively improve the network throughput and ease the network congestion situation.