Research on Consensus Method and System of Intelligent Traffic Toll Data Based on Blockchain and Deep Learning Model Training

This paper proposes a traffic toll data blockchain consensus algorithm (DL-TCC) based on deep learning optimization, aiming to solve the problems of low data processing efficiency and insufficient security in traditional traffic toll systems. The DL-TCC algorithm realizes efficient and secure processing of traffic toll data by integrating the powerful feature extraction ability of deep learning and the decentralized and tamper-proof characteristics of blockchain. Experimental simulation results show that the DL-TCC algorithm is superior to traditional PoW and PoS algorithms in consensus efficiency, data accuracy and system throughput. Specifically, the average consensus time of DL-TCC is only 50 Oms, which is significantly lower than 2500m <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$s$</tex> of PoW and 1200ms of PoS; in terms of data accuracy, the overall error rate of DL-TCC is as low as 1.5%, which is much lower than 5.2% of PoW and 3.8% of PoS. Moreover, as the number of nodes increases, the DL-TCC's consensus time increases slightly, showing good scalability, while the DL-TCC's throughput is always maintained at a high level, displaying excellent performance. The study provides a new way to optimize the intelligent transportation fee system, which is of great significance both in theory and practice.