Link Transmission Stability Detection Based on Deep Learning in Opportunistic Networks

In order to solve the low throughput and high delay problems of traditional link transmission stability detection methods, a detection method of link transmission stability in opportunistic networks based on deep learning is proposed. Establish the network link blocking model. Considering the impact of path delay, analyze the network link information to adjust the hierarchical structure, divide the link data into data blocks, and complete the construction of the link model. According to the link transmission data, the ground point coordinates of the network links in the area are obtained. Under the constraint of link carrying capacity, obtain the barcode sent by network transmission. Calculate the number of packets sent by the network source during congestion, extract network level features using deep learning algorithm, select the number of network layers, set hidden layer nodes, implement network training according to the learning rate, achieve the construction of classification prediction model, and complete the link transmission stability detection. The experimental results show that the proposed link transmission stability detection method can effectively improve the throughput of opportunistic network links and reduce the communication delay of opportunistic networks.