This project is about how failover through wired and wireless links affects VoIP. The information of this thesis is relevant since VoIP is widely used and redundancy is a requirement in all types of enterprise networks. The goal of this study is to see whether a wireless link is equally good to a wired link when it comes to failover. Will the VoIP call experience more packet loss over a wireless link or is it possible to use one as a backup link for the wired network?

To find out, the difference between the two link types are measured by how many packets that are lost during a call - both in a scenario without failover and a scenario with failover. This is done by setting up two different topologies, one with a wired link and one with a wireless link to the backup device, with two IP-phones, one in each end of the network. The phones initiate a call and while this call is up, Wireshark is used to capture the packets for analysis. Measurements are be done both with and without a failover occurrence in both the wired and the wireless topology.

The results of this study show that the scenario with a wireless link have a higher average packet loss than the wired one. However, the difference is small enough to overlook since the wireless link doesn’t affect the call more than the wired link does. It can also be seen in the results that the average packet loss during a failover is lower if the call is initiated by an IP-phone that have a router between itself and the failover devices compared to when the call is initiated by an IP-phone with a switch between itself and the failover devices.