The purpose of this paper is to investigate the performance of four different algorithms for adaptively adjusting the playout delay of audio packets in an interactive packet-audio terminal application in the face of varying network delays.
The main contribution of this paper is that adaptive algorithms which adjust rapidly to changing delays can achieve a lower rate of packet loss for both a given average playout delay and a given maximum buffer size.
(1) The authors present four algorithms which perform delay estimation and
dynamic playout delay adaptation.
(2) An adaptive algorithm which explicitly adjusts to the sharp, spike-like
increases in packet delay which were observed in the authors' traces can
achieve a lower rate of packet loss for both a given average packet delay and
a given maximum buffer size.
I would rate this paper as a 3 because this paper does present a modest contribution in terms of an adaptive algorithm to deal with audio applications. I feel that this paper will be increasingly more important as additional applications, such as IP telephony, are created.
In justifying their results, the authors wrote a simulator which takes the received packet trace file generated by running NeVoT and simulates the behavior of the playout algorithms. By addressing an observed problem, the delay spike, I felt that the authors effectively addressed a previously untouched area of this problem.
One limitation of this approach is that I'm not sure how applications without frequent "talkspurts" perform using algorithm 4.
One lesson to take away from this work is that dynamic algorithms can often improve upon static algorithms.