ProLoC: An Efficient Probability-Based Forwarding Mechanism Considering Load Factor for Congestion Control in Named Data Networking

Named Data Networking (NDN) has become a highly promising architecture for the future of the Internet of Things (IoT). However, congestion remains a significant challenge in NDN, as it hinders real-time communication between network devices and lowers the Quality of Service (QoS). Developing effective congestion control mechanisms for NDN remains a persistent challenge, particularly in fully utilizing its multipath routing and router capabilities. Previous work has overlooked the essential role of NDN routers in congestion control. To address this gap, we propose an efficient Probability-based Forwarding Mechanism considering Load Factor for Congestion Control (ProLoC) in NDN. Our innovative approach incorporates the load factor into the probability-based forwarding decision process and dynamically updates congestion information. The mechanism actively manages traffic, distributes load across multiple paths to prevent congestion, and optimizes resource allocation. Simulation results demonstrate that ProLoC outperforms three state-of-the-art congestion control protocols for NDN (PCON, QSCCP, and AQM). Specifically, ProLoC demonstrates zero packet loss over 10,000 time steps, in contrast to AQM's 14,080 lost packets and PCON's 24. Throughput comparisons show that ProLoC achieves 53.8 % higher throughput than QSCCP and a 156.2 % improvement over PCON, while consistently maintaining a stable round-trip time (RTT) of 11 ± 1 ms and a near perfect fairness index of 0.9995. This highlights ProLoC's ability to prevent and respond to sudden network congestion while maintaining high performance across key metrics.