Outage performance prediction of cooperative vehicle network based on sparrow search algorithm based on back-propagation neural network

With the support of the sixth-generation mobile networks (6G) technology, the Internet-of-Vehicle (IoV) can realize the perception and monitoring of vehicle road information. However, due to the change of network topology and various environment, the reliable performance of the communication link is facing challenges. For the sake of improving communication quality, a cooperative vehicular network (CVN) system is established, which adopts cooperative communication and multiple input multiple output (MIMO) technology. According to the signal-to-noise ratio (SNR) threshold of relay vehicles, using hybrid decode-amplify-forward (HDAF) protocol and combining with antenna selection, the analytical expression of outage probability (OP) with Meijer-G function is obtained. For predicting the OP accurately, the sparrow search algorithm based on back-propagation neural network (SSA-BPNN) is put forward. The simulation results show that the cascade order of the channels has a negative effect on the OP. Meanwhile, the prediction accuracy of SSA-BPNN is 64.8% higher than that of BPNN, and 98.96% greater than that of general regression neural network, and the convergence rate is faster than ICS-BPNN.