The ubiquitous use of Internet of Things (IoT) system in dynamic and real-life contexts presents a high level of challenges because of the constant changes in the data distributions, otherwise known as concept drift. The conventional machine learning models that are implemented in IoT systems are usually trained in static mode and cannot respond to changing trends in data, which causes deterioration in performance as time goes by. The drawback restricts the reliability and independence of intelligent IoT applications in the long run. In order to overcome these challenges, this paper will present a omprehensive lifelong learning model that can be used to enable drift-robust and autonomous IoT intelligence. The framework proposed combines the real-time drift detection, incremental model adaptation with memory-based knowledge retention into a single framework. It allows the IoT systems to continuously learn stream data and retain the past knowledge, thus eliminating catastrophic forgetting. The architecture is edge deportable, which means that it is low-latency inference and less reliant on centralized retraining. This is proven by on-the-job evidence of the proposed approach; it has been shown to be able to sustain stable performance even when data distribution changes, it converts to concept drift faster and it has a higher level of robustness than the traditional batch and online approaches to learning. The framework offers a scalable and effective way of facilitating self-adaptive, resilient, and autonomous IoT systems in the new generation smart environments.
T.Muthumanickam, L. S, D. Jayalakshmi et al.· 2026 6th International Confe...· 0 citations
The intersection of sixth-generation (6G) communication networks, edge computing, and federated learning offers a novel occasion regarding empowering trustful and scalable collaboration throughout distributed Internet of Things (IoT) ecosystems. Conventional centralized machine learning technology is plagued by severe constraints in IoT scenes, such as privacy issues, network congestion, and network bottlenecks. The paper has presented a new 6G-integrated federated learning system which builds on the native intelligence of 6G networks to support secure, efficient, and scalable cooperative learning among heterogeneous edge-IoT devices. The suggested architecture combines terahertz frequencies to synchronize model communication in real-time at ultra-low latency, reconfigurable intelligent surfaces to improve the quality of communication, and network slicing to provide differentiated quality-of-service assurances. Another new hierarchical federated learning system integrates intra-edge aggregation and inter-edge cooperation that can reduce communication overhead by 85-percent and achieve the same accuracy in models. This framework integrates blockchain-based trust management involving zero-knowledge proofs of verifiable model update, to provide integrity and accountability without impacting on privacy. Experimental analysis of massive scale edge-IoT applications has revealed that the suggested scheme attains 97.2% model precision and lowers communication expenses by 87 percent and convergence rate by 3.4 times that of traditional federated learning techniques. The framework has high-uniform performance in adversarial environments where 99.6 percent of malicious model updates are identified with a small false positive. The results define the 6G-integrated federated learning as a framework of reliable and scalable edge-IoT cooperation.
T.Muthumanickam, D. Jayalakshmi, Sathiyamoorthy M et al.· 2026 6th International Confe...· 0 citations
