Intelligent Stretch Reduction in Information-CentricNetworking towards 5G-Tactile Internet realization

TL;DR

This paper proposes a reinforcement learning-based strategy to reduce path stretch in Information-Centric Networking, aiming to improve latency and data rate in 5G and IoT environments.

Contribution

It introduces a novel stretch reduction approach for ICN routers using Q-Learning to optimize routing paths and enhance network efficiency.

Findings

Q-Learning effectively reduces path stretch in ICN.

Optimized parameters improve routing efficiency.

Reduced stretch leads to lower latency.

Abstract

In recent years, 5G is widely used in parallel with IoT networks to enable massive data connectivity and exchange with ultra-reliable and low latency communication (URLLC) services. The internet requirements from user's perspective have shifted from simple human to human interactions to different communication paradigms and information-centric networking (ICN). ICN distributes the content among the users based on their trending requests. ICN is responsible not only for the routing and caching but also for naming the network's content. ICN considers several parameters such as cache-hit ratio, content diversity, content redundancy, and stretch to route the content. ICN enables name-based caching of the required content according to the user's request based on the router's interest table. The stretch shows the path covered while retrieving the content from producer to consumer. Reduction in path length also leads to a reduction in end-to-end latency and better data rate availability. ICN routers must have the minimum stretch to obtain a better system efficiency. Reinforcement learning (RL) is widely used in networks environment to increase agent efficiency to make decisions. In ICN, RL can aid to increase caching and stretch efficiency. This paper investigates a stretch reduction strategy for ICN routers by formulating the stretch reduction problem as a Markov decision process. The evaluation of the proposed stretch reduction strategy's accuracy is done by employing Q-Learning, an RL technique. The simulation results indicate that by using the optimal parameters for the proposed stretch reduction strategy.