BlocksecRT-DETR: Decentralized Privacy-Preserving and Token-Efficient Federated Transformer Learning for Secure Real-Time Object Detection in ITS

TL;DR

This paper introduces BlockSecRT-DETR, a decentralized, privacy-preserving federated learning framework for real-time object detection in ITS, combining token-efficient transformers and blockchain-secured update validation to address heterogeneity, latency, and security challenges.

Contribution

It presents a novel decentralized federated learning approach integrating token pruning and blockchain validation for secure, efficient, and robust real-time object detection in ITS.

Findings

TEM reduces encoder FLOPs by 47.8%

Inference latency improves by 17.2%

Maintains 89.20% mAP@0.5 accuracy

Abstract

Federated real-time object detection using transformers in Intelligent Transportation Systems (ITS) faces three major challenges: (1) missing-class non-IID data heterogeneity from geographically diverse traffic environments, (2) latency constraints on edge hardware for high-capacity transformer models, and (3) privacy and security risks from untrusted client updates and centralized aggregation. We propose BlockSecRT-DETR, a BLOCKchain-SECured Real-Time Object DEtection TRansformer framework for ITS that provides a decentralized, token-efficient, and privacy-preserving federated training solution using RT-DETR transformer, incorporating a blockchain-secured update validation mechanism for trustworthy aggregation. In this framework, challenges (1) and (2) are jointly addressed through a unified client-side design that integrates RT-DETR training with a Token Engineering Module (TEM). TEM prunes low-utility tokens, reducing encoder complexity and latency on edge hardware, while aggregated updates mitigate non-IID data heterogeneity across clients. To address challenge (3), BlockSecRT-DETR incorporates a decentralized blockchain-secured update validation mechanism that enables tamper-proof, privacy-preserving, and trust-free authenticated model aggregation without relying on a central server. We evaluated the proposed framework under a missing-class Non-IID partition of the KITTI dataset and conducted a blockchain case study to quantify security overhead. TEM improves inference latency by 17.2% and reduces encoder FLOPs by 47.8%, while maintaining global detection accuracy (89.20% mAP@0.5). The blockchain integration adds 400 ms per round, and the ledger size remains under 12 KB due to metadata-only on-chain storage.