Implementation and evaluation of voice codec for zigbee

Abstract

Most of the current uses for ZigBee and IEEE 802.15.4 focus on control applications. However, there are areas that will benefit from standardisation, low cost and possibly low power of ZigBee. The project focuses on the use of ZigBee for audio applications. The theoretical limits of ZigBee for this kind of application are evaluated. Then algorithms are developed related to transfer of audio data. While streaming an audio, the most important factor is the available bandwidth where ZigBee has very tight constraints on bandwidth. By compressing the audio, alot of bandwidth can be saved. Two algorithms have been implemented at different data rates for transferring audio, one is MELP (Mixed Excitation Linear Prediction) and the other is CELP (Code Excited Linear Prediction). ZigBee was developed with low data rate in mind. Low energy consumption is mostly noticed for applications where data rate is low and devices can often go in power down mode. Software used is MATLAB. Original and synthesis signals are played in simulation and compared. Speech compression algorithms can be used in areas where there is need to reduce communication cost and use available bandwidth and storage space effectively. Real life applications include rescue mission, emergency scenarios, intra building communication and conferences.