Deep learning inspired routing in ICN using Monte Carlo Tree Search algorithm

Dutta, Nitul; Patel, Shobhit K.; Samusenkov, Vadim; Vigneswaran, D.

doi:10.1016/j.jpdc.2020.12.014

Cited by 15 publications

(13 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The count of neighbouring vehicles is also considered to select a node. The protocol performance is examined in a simulation testbed designed in ndnSim-2.0 and compared with exiting approaches as stated in [36] and…”

Section: Discussionmentioning

confidence: 99%

“…To evaluate the performance of the proposed forwarding scheme, it is compared with the MCT based forward- ing as stated in [36] (depicted as FIB in graphs) and flooding. The [36] has explained a interest forwarding approach for ICN based on Monte Carlo Tree (MCT) search algorithm. This algorithm does not rely on FIB for interest forwarding.…”

Section: B Performance Observationmentioning

confidence: 99%

See 1 more Smart Citation

Optimized interest forwarding in information centric vehicular networks using general secretary problem

Samantha

Dutta

2023

Preprint

View full text Add to dashboard Cite

Vehicular networks of current era is not limited only to road safety and traffic management. It is been widely accepted as a network for information sharing, social networking and entertainment. With the advent of smart vehicles, the demand of vehicular network centric applications are getting popular. In one way, these applications demand faster data access and on the other hand, the underlying challenges of vehicular networks such as intermittent connectivity, low bandwidth and unpredictable vehicle movements impose considerable hindrance on communications. That is why, the vehicular communication is moving towards information centric approach so that the benefits of Information Centric Network (ICN) can be utilized. However, inherent challenges of vehicular communication and the lack of adequate protocols to implement content centric approach makes the extension of ICN into vehicular communications difficult. In ICN, a node generates interest packets to find a content. The interest is forwarded to an appropriate cache store to get the data. Hence, efficient forwarding of interest packets can considerably improve the content retrieval delay and can reduce network load. The interest forwarding techniques designed for wired ICN are not suitable for vehicular network due to computational complexity and bandwidth crisis. Hence, Information Centric Vehicular Network (ICVN) demands a new design for interest forwarding. Influenced by the need of the current time in vehicular communication, this research is aimed at developing an efficient interest forwarding mechanism for ICVN. The proposed algorithm models the interest forwarding task as a general secretary problem and finds the best node (or set of nodes) in the network to forward the interest packet. The protocol performance is examined in a simulation testbed designed in ndnSim-2.0 and compared with existing forwarding approaches. Observation demonstrates a better performance of the said vehicle with respect to content retrieval latency and network overhead.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: B Performance Observationmentioning

confidence: 99%

Optimized interest forwarding in information centric vehicular networks using general secretary problem

Samantha

Dutta

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…It executes on top of shortest path routing and works with any existing caching mechanism. The request forwarding mechanism driven by monte carlo tree search algorithm is introduced in [18] to send data requests to appropriate caches. The models of deep learning and machine learning have also been exploited by researchers to forward request in ICN.…”

Section: Existing Forwarding and Routing Approaches In Icnmentioning

confidence: 99%

“…This off policy RL based learning approach has the motive of identifying the best possible action to execute for given state. The problem statement is modelled with respect to single player game [18] by considering data requestor as an initial state and content store holding the desired content as a final state. Proposed protocol builds the path from data requestor to corresponding content store using Q-Learning based algorithm.…”

Section: Motivation and Objectivementioning

confidence: 99%

Reinforcement learning inspired forwarding strategy for information centric networks using Q-learning algorithm

Delvadia

Dutta

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Content interest forwarding is a prominent research area in Information Centric Network (ICN). An efficient forwarding strategy can significantly enhance the user level performance parameters such as data retrieval latency. It also helps to minimize the origin server load, network congestion and overhead. Reinforcement learning is widely accepted for taking efficient routing decisions in network. This paper introduces a Q-learning driven forwarding strategy in ICN for interest packets. We have investigated the feasibility of exploiting reinforcement learning mechanism named Q-Learning for Named Data Network (NDN) paradigm of ICN. By revising Q-Learning mechanism to address the inherent challenges related to overhead and latency for content retrieval, this paper introduces design and implementation of Q-Learning based forwarding mechanism. It aims to gain learning through historical events and selects best mechanism to forward interest. The performance investigation of proposed protocol is carried out using simulator named ndnSIM-2.0. Outcomes are compared by integrating proposed protocol with LCD (Leave Copy Down), LCE (Leave Copy Everywhere, CL4M (Cache Less for More) and ProbCache (Probability driven caching). Protocol behaviour is also compared against recent routing and forwarding mechanisms. The considered performance parameters are data retrieval delay, server hit rate, network overhead, network throughput and network load. The experimental outcomes conclude that integration of proposed forwarding strategy to state-of-the-art protocols lead to performance enhancement up to 10–35%. The integrated protocol variants are also sensitive to any changes in network compared to existing approaches.

show abstract

“…The request can be satisfied by any node/content router (CR) that has cached desired content without depending only on the content provider. The primary challenges in ICN are to provide efficient caching and routing strategies that improve content discovery latency by maximizing throughput [3]. The research work discussed in this paper adopts the named data network (NDN, 2014) architecture [4] of ICN.…”

Section: Introductionmentioning

confidence: 99%

CCJRF-ICN: A Novel Mechanism for Coadjuvant Caching Joint Request Forwarding in Information Centric Networks

2021

Self Cite

View full text Add to dashboard Cite

Information centric networking (ICN) shifts the focus of existing internet architecture from host-oriented to content-oriented model by enabling in-network caching and content-based forwarding. These ICN features help to increase network performance by decreasing content discovery delay, content server load, and network congestion. To route a content interest inside network such that content can be fetched with minimal time is a challenging task in ICN. The performance of the ICN routing protocol can be significantly improved if the decisions related to content chunk placement and request forwarding are taken in a cooperative fashion. This paper describes a novel strategy for co-operative caching joint request forwarding in ICN, focusing on decreasing content retrieval latency. To do so, the caching strategy leverages the concept of connected dominating set (CDS) for creating a virtual backbone network to eliminate caching redundancy and reduce content discovery delay. It considers content chunk placement and request forwarding tasks as strongly co-related procedures. It exploits the caching information so that the request is forwarded to a content router (CR) with the maximum likelihood of carrying needed data using the betweenness centrality (BC) of CR. The proposed approach also uses the Markov chain-based model to estimate CS hit likelihood and use it as decisional parameter while forwarding the interest packet. This mechanism helps to fetch the content within the shortest possible time duration. The CCJRF-ICN adopts the Dijkstra's shortest path routing and works in collaboration with the CDS-driven caching joint forwarding mechanism. The simulation study of CCJRF-ICN is done inside ns-3 based ndnSIM-2.0 simulator with performance measures like content store (CS) hit ratio, content discovery delay, mean hop distance, network load, and network overhead. The simulation outcomes demonstrate that CCJRF-ICN outperforms the state-of-the-art strategies for realistic topologies (GEANT, US-26, Euro-28) and shows improvement up to 5-35% against stated performance measures.

show abstract

Deep learning inspired routing in ICN using Monte Carlo Tree Search algorithm

Cited by 15 publications

References 21 publications

Optimized interest forwarding in information centric vehicular networks using general secretary problem

Optimized interest forwarding in information centric vehicular networks using general secretary problem

Reinforcement learning inspired forwarding strategy for information centric networks using Q-learning algorithm

CCJRF-ICN: A Novel Mechanism for Coadjuvant Caching Joint Request Forwarding in Information Centric Networks

Contact Info

Product

Resources

About