David Alvarez, Entry #4: Week 6

The past week has been really fast-paced and busy, which is always a good thing because that means progress. On the WebRTC analysis experiments, Rodda and I have begun to do a lot more complex experiments. The setup now has three machines involved; one sender, one receiver, and one monitor in order to quantify the physical data rate. We've also started having a YouTube video be streamed on a fourth machine that is on the same channel as the other three machines, in order to simulate real-world traffic and interference. Wireshark runs on all three machines, and the .pcap traces from them are used to create data sets, and therefore plots in order to see if any noticeable trends are appearing. Aside from the PHY data rate, the other metrics are throughput, packet loss, time delta, and packet size. We also began including cumulative distribution function (CDF) graphs more.

Examples of the plots we are generating through the Python script.

Varun has also been giving us some reading to do so that we may better understand what exactly WebRTC is doing. They haven't been directly relevant to the API definition, but rather to Google Congestion Control (GCC), which has been proposed for WebRTC. Also, I took some screen capture videos of the trials we conducted, and he sent me a guide to a video analysis software used by another professor working on the project. It's called BVQM, and essentially how it works is that it takes the original clip sent out, and compares it to the processed clip seen by the receiver. From here, it can gather more data to compile, as well as provide a more visual aid to go along with the graphs we've made. There have been some compatibility issues simply because the software hasn't been so friendly with my Peddie laptop, but I've still managed to extract some useful information. I've also begun work on my poster, and after the traumatic experience of reading through how wrong I was on my poster in the spring, I've begun to make some progress.

Outside of all of this, the Wireless and Mobile Networking Lab has continued to be a really great place to work. The structure of work in the lab has left me a lot of freedom to work at whatever pace I'd like, leaving my schedule very flexible. This helps a lot now that I have to start my summer work for all of my classes, as well as the Common App. Everyone I'm working with here has been understanding of that too, which has been even more helpful. Yesterday, we had to say good-bye to an undergraduate student who was here for the summer, Andy, and went out to go get dinner at an Ethiopian food place near the university's main campus. I'd never had it before, and was pleasantly surprised by how much I liked it. 

Time has flown by and I now only have one week left here. I hope to be able to do a lot of work and try to get as close as I can to a good stopping point before my last day.