A GNN-based proactive caching strategy in NDN networks

As people spend more time watching and sharing videos online, it is critical to provide users with a satisfactory quality of experience (QoE). Leveraging the in-network caching named-based routing features in Named Data Networks (NDNs), our paper aims improve user through caching. We propose graph neural network-gain maximization (GNN-GM) cache placement algorithm. First, we use GNN model predict users’ ratings unviewed videos. Second, consider total predicted rating video as gain cached video. Third, algorithm maximize actively Cache replacement implemented based on ranking videos, higher replacing lower compare GNN-GM two state-of-the-art strategies, namely NMF-based strategy GNN-CPP. also compared traditional LCE LRU, FIFO. evaluate five strategies using real-world datasets tree network topology, topology GEANT, various random topologies. The experimental results show that policy significantly improves hit ratio, latency server load. Notably, achieves 25% rate, 5% 7% load than GNN-CPP GEANT.