Path Discovery Algorithm for Autonomous Area of Internet of Things Based on Artificial Intelligence Algorithm

In the process of path planning in the autonomous area of the Internet of Things, the uneven distribution of data traffic is easy to occur, which leads to the traditional path discovery algorithm in the autonomous area of the Internet of Things based on genetic particle swarm optimization. Because the shortest path is used to forward packets, the link is congested and the resource utilization rate is affected. Aiming at the problem of finding the best transmission path in the current Internet of Things application in a mixed fixed and mobile network, the most stable path is selected to transmit data in the frequently changing wireless links. Multipath routing is used to distribute traffic to underutilized areas, which realizes the function of load balancing and improves the utilization rate of network resources. By analyzing the environment within the autonomous region of the Internet of Things, a multi-path planning model is established, and hierarchical selection algorithms are integrated to search for key path planning attributes in different states. Taking the bandwidth, delay, jitter, and packet loss rate characteristics of the network path as the decision criteria for path selection, a judgment matrix of attributes is constructed. The weight of each judgment attribute is calculated using the characteristic root method, and normalization calculations are performed. The gray correlation coefficients of each candidate path are weighted and sorted by network path. And then complete the optimization of network path planning.