Model to Improve Quality of Service in Wireless Sensor Network

Abstract

Communication channel is a transmitting medium by which two nodes can communicate with each other in a network. A channel can send the information also called as data in the form of packets to its receivers. If a channel is communicating in an open environment, there can be some environmental conditions that can affect the performance of communication. There can be parameters such as SNR, BNR or channels throughput for every channel which can be affected by the external environment. To improve the quality of service, in this research model, the proposed algorithms can recommend the best and healthy parameters for every channel and the healthy channel itself to have better communication between nodes. The proposed algorithms can be applied to the channel to reduce the dimensionality of data or feature selection from the available set of features automatically. This approach will automatically take care of suitable and useful features for every channel. The technique is further applied to recommend a single communication channel among all available channels, based on the most suitable features that are selected for every channel. The proposed model is mainly focused on wireless low throughput network with ISM band. Using machine learning techniques the model will be trained by itself so that it can return the one of the best-recommended channel with the best parameters to have the best communication in a network. One of the machine learning techniques is reinforcement learning, which will help the machines to learn by itself and give accurate results based on various channel selection parameters.