Proposta e análise de um algoritmo adaptativo de ajuste de taxa de transmissão para sistemas VoIP

Abstract

With the economic benefits over conventional telephone communications, and the potential to offer new classes of services, Voice over IP (VoIP) has been attracting increasing commercial and academic interest. Currently available best effort data networks, such as Internet, do not offer guarantees on the quality of service experienced by its users. One approach to address this difficulty is to employ an adaptive system that estimates the network state and, on that basis, makes adjustments aiming to maintain an efficient utilization of the available resources. This work, however, proposes an algorithm for the adaptive adjustment of the transmission rate of a VoIP source based on the estimated voice quality at a receiver. This adjustment is achieved by switching voice encoders during the call in accordance with the network conditions and the voice quality being experienced. In the proposed validation, simulations have been conducted using a realistic architectural model for VoIP environments. This model considers VoIP sources based on Brady´s model of human conversation. The effects of the proposed model and algorithm on aggregate traffic of a data network has been analyzed and compared with the results of related works. The simulation results show that the proposed algorithm is able to make efficient use of available bandwidth, and presents superior performance in comparison with similar research in the literature that exhibit good performance for simplified architectural models.