Revista: Wireless Networks. Vol. 15. Pp. 227-239. March 2007

Autores: Alfonso Fernández-Durán and José I. Alonso

A common communications convergence scenario which is being adopted in personal communications relates to the combination of wireless and cellular networks by the use of multimode terminals. Since most of the wireless
networks were initially dimensioned only for data communications, this paper shows how voice over wireless LAN
dimensioning could be addressed under the optimal network throughput and the perspective of voice quality, using a simple approach. The maximum number of simultaneous users resulting from throughput is limited by the collisions taking place in the shared medium with the statistical contention protocol. The voice quality is conditioned by the delay and the packet loss in the contention protocol. Both approaches are analyzed within the scope of the voice codecs commonly used in voice over wireless LANs, to conclude that voice dimensioning based on network throughput and voice quality show complementary results. Additionally the use of low rate codecs in voice over wireless LANs is advantageous
for the network performance point of view but may produce poor voice quality results. Mid range codecs like G729 could
represent a trade-off for quality throughput. For these reasons, voice quality and wireless network throughput have
to be taken into account in the network admission control, design and deployment to ensure a satisfactory user experience. The impact of handoff interval of wireless convergent networks on the conversation quality need also be assessed for a proper network design.