AoI-Aware Flow Allocation in IoT Systems Using Matching Theory for Edge Computing

The Internet of Things (IoT) enables intelligent monitoring and control through large-scale data sensing and real-time processing, leading to increasing computational demands. In this context, information freshness, commonly quantified by the Age of Information (AoI), plays a crucial role in evaluating system performance. To efficiently support data processing while reducing latency and network congestion, IoT monitoring systems increasingly rely on edge computing architectures, where computational resources are deployed close to end devices. This paper addresses the problem of computational workload distribution among edge nodes and proposes a flow allocation technique based on matching theory. The proposed approach enhances system performance by serving more devices with fewer edge nodes, achieving an approximate $45~\%$ gain over alternative solutions in terms of peak AoI.