Digital processing of speech signals

One of the fundamental difficulties with trying to characterize the Internet''s behavior lies in its immense diversity. To date, efforts to characterize end-to-end Internet dynamics such as packet delay and loss have studied at most a handful of traffic traces and Internet paths, making it difficult to assess the generality of the findings. Yet solid assessments of Internet path properties are crucial for engineering transport protocols for high performance. In this work, we attempt to tackle this problem using a distributed framework we developed for conducting end-to-end measurements. In the first part, we report on an end-to-end routing study conducted using the framework, in which 37 sites participated, yielding about 40,000 routing measurements along 1,000 Internet paths. We characterize routing pathologies and failures, stability over time, and symmetry. In the second part of the study, we analyze 20,000 TCP bulk transfers measured using the same framework. We discuss: 1) errors introduced by the packet filters used to make the measurements; 2) dealing with unsynchronized and skewed clocks; 3) separating out TCP behavior from network behavior; 4) the surprisingly large variations between different TCP implementations; 5) pathological network behavior, such as out-of-order delivery, replication, and corruption; 6) a robust algorithm for estimating bottleneck bandwidth; and 7) the resulting characterizations of Internet packet loss and delay, including general assessments of queueing time scales and available bandwidth.

/pdf/measurements-and-analysis-of-end-to-end-internet-dynamics-2g9j633ch4.pdf

Measurements and analysis of end-to-end Internet dynamics

Understanding and modelling packet loss in the Internet is especially relevant for the design and analysis of delay-sensitive multimedia applications. We present analysis of 128 hours of end-to-end unicast and multicast packet loss measurement. From these we selected 76 hours of stationary traces for further analysis. We consider the dependence as seen in the autocorrelation function of the original loss data as well as the dependence between good run lengths and loss run lengths. The correlation timescale is found to be 1000 ms or less. We evaluate the accuracy of three models of increasing complexity: the Bernoulli model, the 2-state Markov chain model and the k-th order Markov chain model. Out of the 38 trace segments considered, the Bernoulli model was found to be accurate for 7 segments, and the 2-state model was found to be accurate for 10 segments. A Markov chain model of order 2 or greater was found to be necessary to accurately model the rest of the segments. For the case of adaptive applications which track loss, we address two issues of on-line loss estimation: the required memory size and whether to use exponential smoothing or a sliding window average to estimate average loss rate. We find that a large memory size is necessary and that the sliding window average provides a more accurate estimate for the same effective memory size.

Measurement and modelling of the temporal dependence in packet loss

We review error resilience techniques for real-time video transport over unreliable networks. Topics covered include an introduction to today's protocol and network environments and their characteristics, encoder error resilience tools, decoder error concealment techniques, as well as techniques that require cooperation between encoder, decoder, and the network. We provide a review of general principles of these techniques as well as specific implementations adopted by the H.263 and MPEG-4 video coding standards. The majority of the article is devoted to the techniques developed for block-based hybrid coders using motion-compensated prediction and transform coding. A separate section covers error resilience techniques for shape coding in MPEG-4.

Error resilient video coding techniques

We investigate how FEC (Forward Error Correction) can be combined with ARQ (Automatic Repeat Request) to achieve scalable reliable multicast transmission. We consider the two scenarios where FEC is introduced as a transparent layer underneath a reliable multicast layer that uses ARQ, and where FEC and ARQ are both integrated into a single layer that uses the retransmission of parity data to recover from the loss of original data packets.To evaluate the performance improvements due to FEC, we consider different types of loss behaviors (spatially or temporally correlated loss, homogeneous or heterogeneous loss) and loss rates for up to 106 receivers. Our results show that introducing FEC as a layer below ARQ can improve multicast transmission efficiency and scalability and that there are substantial additional improvements when the two are integrated.

Parity-based loss recovery for reliable multicast transmission

We consider the problem of distributing audio data over networks such as the Internet that do not provide support for real-time applications. Experiments with such networks indicate that audio quality is mediocre in large part because of excessive audio packet losses. In this paper, we show using measurements over the Internet as well as analytic modeling that the number of consecutively lost audio packets is small unless the network load is very high. This indicates that open loop error control mechanisms based on forward error correction would be adequate to reconstruct most lost audio packets.

Hugues Crépin

Papers

Analysis of Audio Packet Loss in the Internet