
Improved Reliability of Voice over Internet Protocol(VoIP) using Machine Learning
Author(s) -
Vinod Kumar,
Om Prakash Roy
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1020/1/012025
Subject(s) - voice over ip , computer science , computer network , packet loss , jitter , network packet , wireless network , quality of service , the internet , node (physics) , wireless , telecommunications , engineering , structural engineering , world wide web
Start Voice over Internet Protocol (VoIP) is the communication means inspired with the revolutionizing wireless internet technology to transfer voice signals. However, VoIP over wireless means faces several challenges in terms of data loss resulting in poor voice quality or communication delay. To deal with the practical aspects of voice communication over Internet Protocol(IP) we proposed a highly secure wireless network. The proposed VoIP as a secure wireless network for VoIP combines the advantage of Artificial Bee Colony (ABC) based network optimization based on node property. This is followed by implementation of two classifiers; Support Vector Machine (SVM) that identifies the affected route and Convolutional Neural Network (CNN) that detects the malicious nodes present in the affected route to offer secure data transmission. The quality of voice communication is evaluated in terms of dropped packets, delay and jitter to offer interactive communication service. The simulation analysis over 50 nodes had proved the effectiveness in achieving reliable quality voice calling service with average throughput of 98.65% with comparatively lower jitter, packet loss and latency of 2.325 ms, 1.35% and 1.616 ms, respectively.