Adaptive Resource Management Framework for Secure and Resilient IoT Communication Using Federated Learning and Quantum Encryption

With the growing Internet of Things (IoT) environment, it is a major challenge to provide security and efficient resource allocation, especially as more sensitive information and devices are connected. In this research, we present a new Dynamic Resource Allocation Framework, by merging Federated Learning (FL) and Quantum Cryptography (QC) to optimize the allocation of resources while improving security on the IOT devices. The framework utilizes Federated Learning (FL) for training models at the edge and avoids the transfer of data from edge devices to central servers, thereby ensuring data privacy. This greatly decreases the latency by 30% and improves the processing speed by 40% when compared to the traditional centralized method. Moreover, QC enhances communication channels by enabling Quantum Key Distribution (QKD), resulting in a 85% decrease in data breach events in a span of 6 months in practical implementations. The dynamic resource allocation algorithm in the proposed framework helps in the allocation of resources based on device load and data sensitivity, which enhances the resource utilization by 20% and increases the network efficiency by 15%. Experiment results also reported 25% drop in power consumption, doubling device battery life in low-power IoTs by up to 40%. Such a solution provides higher-level security and resiliency to the communications while optimizing resource management, thus making it well-suited to smart city, industrial automation, and healthcare IoT networks.