Klasifikasi Data Delay dengan LFID Strategi Forwarding menggunakan Machine Learning untuk Memaksimalkan Kinerja Jaringan NDN (Named Data Network)

Abstract

Named Data Network (NDN) is the future internet network that data-centric and adaptive to consumer requirement. Routing and forwarding systems on the NDN networks are different from IP networks due to the use of cache at each node on the network. The implementation of the Loop Free Inport-Dependent (LFID) routing protocol on NDN networks aims to eliminate loops on the network by eliminating the preferred routes or inefficient next hops. Forwarding strategies that can be implemented are Best Route, Access, Random, and Multicast. Therefore, machine learning technology is needed with various classification methods that can be implemented in machine learning so the output gives the recommendations that can be used to maximize the performance of the NDN network. The final result of this study recommends that the forwarding strategies of Best Route and Access provide good delay values, which in the range of 150 ms to 300 ms. Random forwarding strategy with a payload size> = 3072 kbps still provides a good delay value to the network, which in the 150 to 300 ms range. All forwarding strategies of Best Route, Access, Random, and Multicast provide delay values with a very good category of delay values, which is below 150 ms if the type of interest (data) that requested to the network is a popular interest.