Congestion control is an important feature that directly affects network performance. Network congestion may cause loss of data or long delays. Although this problem has been studied extensively in the Internet, the solutions for Internet congestion control do not apply readily to challenged network environments such as Delay Tolerant Networks (DTN) where end-to-end connectivity may not exist continuously and latency can be high. In DTN, end-to-end rate control is not feasible. This calls for congestion control mechanisms where the decisions can be made autonomously with local information only. We use an economic pricing model and propose a rule-based congestion control mechanism where each router can autonomously decide on whether to accept a bundle (data) based on local information such as available storage and the value and risk of accepting the bundle (derived from historical statistics). Preliminary experimental results show that this congestion control mechanism can protect routers from resource depletion without loss of data.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    Autonomous Congestion Control in Delay-Tolerant Networks


    Contributors:

    Conference:

    AIAA 9th International Conference on Space Operations (SpaceOps ; 2006 ; Rome, Italy


    Publication date :

    2006-06-19


    Type of media :

    Preprint


    Type of material :

    No indication


    Language :

    English




    Autonomous Congestion Control in Delay-Tolerant Networks

    Burleigh, Scott / Jennings, Esther / Schoolcraft, Joshua | NTRS | 2006




    Autonomous Delay Tolerant Network Management Using Reinforcement Learning

    Buzzi, Pau Garcia / Selva, Daniel / Net, Marc Sanchez | AIAA | 2021


    Exploring Reinforcement Learning for Autonomous Delay Tolerant Network Management

    Garcia Buzzi, Pau / Selva, Daniel / Sanchez Net, Marc | AIAA | 2020