Predicting Network Latency Using Adaptive Network Based Fuzzy Inference System

It looks at introducing an Adaptive network-based Fuzzy Inference System (ANBFIS) to predict community latency through an artificial neural community (ANN). This technique is used to build a fuzzy common sense structure that allows one to pick out the dynamic nature of the community. The ANBFIS comprises a three-layer feed-forward ANN and four fuzzy standard sense-based 'fuzzification' functions. The ANN has been educated using historical facts of network traits inclusive of bandwidth, packet loss, and latency from unique network topologies. The trained ANBFIS was then used to predict network latency with different input parameters. Consequences show that the anticipated latency values using the ANBFIS are very near the absolute measured latency values inside the networks. The ANBFIS version offers a unique technique to forecast network latency, which may be used to facilitate green and dependable network monitoring and control. The paper describes a singular Adaptive community-primarily based Fuzzy Inference device (ANFIS) version for predicting community latency. The proposed model employs an artificial neural community structure that can analyze the input statistics and predict approximate network latency with excessive accuracy. This adaptability is achieved using fuzzy inference techniques and adjusting the network parameters based on the input data., the model is evaluated and compared with traditional regression fashions, with statistically significant outcomes. The ANFIS version was shown to have advanced overall performance in predicting network latency compared to linear and non-linear regression fashions. This novel approach can improve the scalability and reliability of community monitoring and management systems..