Abstract
Qutaiba Ibrahim Ali, Ina'am Fathi, Jassim M. Abdul-Jabbar
With the revolutionary development of Wireless Sensor Network (WSN) technologies, researchers started to take benefits of integration embedded low-cost, low-power WSN technology in a various IoTs applications. Real-time voice transmission over IoTs is one interesting application that began to be explored by many researchers for a wide range of emergency scenarios. Thus, this paper presents a performance study for transmission of voice over WSN (VoWSN) with and without presence of Internet. A framework using a Raspberry Pi3 (RPi3) and open source FFmpeg technology for processing, compressing and streaming voice to a r emote computer is proposed, implemented and evaluated. The performance of the proposed framework is evaluated by studying its behaviour utilizing three audio encoding algorithms: AC3, MP3 and OPUS with different sampling rates. Furthermore, a set of evaluation metrics procedures such as: One-way delay, jitter, Bandwidth (B.W), CPU usage and packet losses are proposed and implemented effectively.
PDFShare this article