Tracking Down Skype Traffic

Abstract

Abstract-Skype is beyond any doubt the most popular VoIP application in the current Internet application spectrum. Its amazing success drawn the attention of telecom operators and the research community, both interested in knowing Skype's internal mechanisms, characterizing traffic and understanding users' behavior.In this paper, we dissect the following fundamental components: data traffic generated by voice and video communication, and signaling traffic generated by Skype. Our approach is twofold, as we make use of both active and passive measurement techniques to gather a deep understanding on the traffic Skype generates. From extensive testbed experiments, we devise a source model which takes into account: i) the service type, i.e., voice or video calls ii) the selected source Codec, iii) the adopted transport-layer protocol, and iv) network conditions. Furthermore, leveraging on the use of an accurate Skype classification engine that we recently proposed, we study and characterize Skype traffic based on extensive passive measurements collected from our campus LAN.

Passive aggressive measurement with MGRP

“…We use a simple, custom fragmentation/reassembly protocol similar to IP fragmentation to manage fragmented chunks. 2 In the figure we can see that probe packets in (ii) carry a single, fragmented TCP segment, whose chunks are labeled 2a and 2b. If MGRP cannot piggyback an entire rider before the buffer timeout, MGRP simply sends the chunk in its own MGRP packet, shown as (iii) in the figure.…”

Tuning Skype's Redundancy Control Algorithm for User Satisfaction

“…The same authors later proposed OneClick [6], a lightweight framework for measuring network applications' quality of experience from users' perspective, in the hope to verify the passive measurement results by user experiments. Bonfiglio et al [3] analyzed how Skype adapts its traffic to different network conditions. They found that when available bandwidth is decreased, the bit rate and payload size of Skype traffic are also decreased.…”

“…How Skype adapts to changing bandwidth has been discussed in [3]. The authors have found that when they reduced the bandwidth, the payload size decreased, which suggests the codec switches to a lower bit rate.…”

“…The desired MOS score and the bandwidth required to achieve the desired audio quality are both set to 100. On the graphs, the left-hand side of the y-axis marks the normalized bandwidth utilization, and the right-hand side marks the normalized MOS score 3 . We observe that the bandwidth utilization and audio quality of Skype fluctuate under different network settings.…”

Passive aggressive measurement with MGRP

“…Broadly speaking, current techniques either passively observe existing traffic, or actively inject probe packets to see how the network responds. Using passive measurement, an application may observe the rate at which its buffers fill or empty [10] and monitor the rates of packet and frame loss [2,3]. Passive observations are particularly useful for signaling that a stream's rate should be reduced, but do little to suggest when its rate can be increased [2,3].…”