Maral Salehi - Department of Computer Engineering, Amirkabir University of Technology, Tehran, Iran
Mehdi Dehghan

Providing a service with good quality for transmission and playing real-time voice conversations (voice streaming) over wireless ad-hoc networks is always a big challenge. Buffering and adjusting the playout time of packets is one of the methods used to overcome this challenge which can be deployed in receiver side. In this paper, a new adaptive playout adjustment algorithm to stream the voice conversations over wireless ad-hoc networks is proposed. This algorithm always tries to be aware of network's conditions, adapts itself with these conditions and adjusts the playout time of voice packets as good as possible. So, not only most of the packets must be received before their playout time which is scheduled in receiver, but also the playout time must not be too long as much that has a bad effect on the interactivity between sender and receiver. The main features of the presented method are: First, adjusting the threshold adaptively with respect to variant conditions of networks, in order to determine the state of system. Second, calculating the mean network jitter and especially alpha parameter dynamically based on the current conditions of networks, in order to calculate the playout delay of current packet. Third, being optimistic about the future state of networks and not using the history of delay, in order to calculate the mean network delay. The results of simulation show that the proposed algorithm adapts itself with the network's dynamic conditions and adjusts the playout delay of voice packets better than the other algorithms. other algorithms

Wireless mobile ad-hoc Networks, Voice streaming, Adaptive playout algorithm, Quality of service, Multimedia