Abstract: Aiming at the shortages that current routing algorithms for wireless sensor backbone networks could not balance the contradiction between energy consumption and data transmission, which led to large data transmission delay and low throughput of routing of wireless sensor backbone networks, in order to improve the overall performance of wireless sensor networks, we designed a new routing algorithm for wireless sensor backbone networks. Firstly, the working principle of wireless sensor networks was analyzed, and the corresponding routing model was established. Secondly, machine learning al
gorithm was introduced to predict the energy of wireless sensor nodes in the routing of wireless sensor backbone networks in real time. The energyintensive wireless sensor nodes were selected for data transmission, and the wireless sensor with the least energy consumption was constructed. Finally, the algorithm was compared with other routing algorithms for wireless sensor backbone networks. The test results show that the energy consumption of the routing algorithm for wireless sensor backbone networks is small, the reliability of data transmission in wireless sensor networks is high, and the speed of data transmission in wireless sensor networks is accelerated. The overall performance of the routing algorithm in wireless sensor blackbone networks is obviously better than that of comparative algorithms.

Key words: energy consumption of sensor nodes, machine learning algorithm, optimal data routing, network system throughput, data transmission success rate