Multi-objective federated learning traffic prediction in vehicular network for intelligent transportation system

Abstract

The spatial-temporal data of future freight traffic speed in the metropolitan region must be properly understood to develop freight-related traffic management strategies. This work introduces a new approach to traffic prediction using multi-objective federated learning. Instead of relying on a centralized cloud server for data processing, collaborative training is implemented among several participants. The proposed method utilizes the advantages of reinforcement learning in dynamic decision-making scenarios and the expressive capabilities of graphical models to identify traffic intensity. Furthermore, a new methodology integrates federated learning concepts with multi-objective optimization to forecast traffic patterns accurately. The proposed approach exhibits a higher level of performance than existing methods for estimating traffic speed. It achieves a communication delay of 23.4%, packet delivery ratio (PDR) of 92.45%, packet loss rate of 12.34%, prediction accuracy of 97.45%, and resource utilization of 89.56%. The visualisation findings demonstrate that this new approach is able to successfully capture interconnections of metropolitan areas in different neighboring cities.

Keywords: Federated learning; Intelligent transportation systems (ITS); Machine learning; Multi-objective optimization; Traffic prediction; Vehicular networks.

Figure 1. Proposed system model for VANET.

Figure 2. Graphical reinforcement learning for vehicular environment.

Figure 3. The timing diagram of several vehicles relative to the time in hours, days, and years, including dates for the planned approach at the junction.

(A) Shows the count of traffic by junction (2015–2017); (B) shows the weekly traffic patterns by junction; (C) shows the hourly traffic volume by junction; (D) shows the monthly traffic fluctuations by junction; (E) shows the seasonal traffic trends by junction; (F) shows the yearly traffic growth by junction.

Figure 4. Vehicle count at each junction by hour, day, month, and year.

(A), (B), (C), and (D) show the traffic prediction vs actual for the Junctions 1, 2, 3, and 4, respectively.

Figure 5. (A) The performance Metrics for the Junctions 1, 2, 3, and 4. (B) The confusion matrix for the Car hacking dataset.

Figure 6. (A and B) The precision-recall curve and the ROC curve for the car hacking dataset.

Figure 7. (A) The confusion matrix for the car hacking dataset training. (B) The analysis of communication latency.

Figure 8. (A) The analysis of communication latency. (B) The analysis of the packet loss ratio.

Figure 9. (A) The analysis of packet loss ratio. (B) The analysis of the packet loss ratio.