As one who spends a lot of time testing wireless and mobile devices (I've been knee deep in a particularly entertaining aspect of this lately, reviewing products for the annual Network World Holiday Gift Guide - stay tuned), I've realized the importance of having a fairly broad arsenal of tools for generating and evaluating workloads on the system under test. It's important that these be simple, reliable, reproducible, and quantifiable, and such is usually not an issue when evaluating digital quantities like throughput, error rates, coverage, and other criteria that boil down to numbers alone. But what about audio and video? These are a lot tougher, given that, while the storage and transmission of these media is digital, their consumption is analog. How the content looks and sounds is not something that machines have traditionally been very good at evaluating. And, given the lossy nature of video compression, direct frame-buffer or audio-buffer comparisons might not work very well. Consider, for example, Amimon's [http://www.amimon.com/] technique, which results in a very high error rate but video that is visually indistinguishable from its source.
Wi-Fi testing leader VeriWave just announced a product for evaluating the performance of video over Wi-Fi (or any network) called WaveVideo. The tool allows a high degree of flexibility and traffic types and patterns, and produces a large number of results including an automated indication of video quality known as MDI (Media Delivery Index). MDI evaluates network performance, not visual quality, but the latter can also be checked - manually (humans are still god for something...) - as well. I've not had the chance to use this product yet, but I am in the middle of a couple of video testing projects and I hope to have more on this development for you shortly. In the meantime, video testing is the last big frontier for all kinds of networks, including Wi-Fi, so I'm expecting that the demand for products of this type will be very heavy, from developers and end-users alike.