Providing statistical quality-of-service guarantees introduces the conflicting requirements for both deterministic traffic models to isolate and police users and statistical multiplexing to efficiently utilize and share network resources. We address this issue by introducing two schemes for providing statistical services to deterministically policed sources: (1) adversarial mode resource allocation in which we bound the stochastic envelopes of policed streams and provide a statistical service for adversarial or worst case sources and (2) non-adversarial mode allocation in which we approximate the stochastic envelopes of policed, but non-worst-case streams in order to exploit a further statistical multiplexing gain in the typical case. Our key technique is to study the problem within the domain of deterministic and stochastic traffic envelopes, which allows us to explicitly consider sources with rate variations over multiple time scales, obtain results for any deterministic traffic model, and apply accurate admission control tests for buffered priority schedulers. We evaluate the scheme’s performance with experiments using traces of compressed video and show that substantial statistical multiplexing gains are achieved.

We present a comprehensive model for variable-bit-rate MPEG video streams. This model captures the bit-rate variations at multiple time scales. Long-term variations are captured by incorporating scene changes, which are most noticeable in the fluctuations of I frames. The size of an I frame is modeled by the sum of two random components: a scenerelated component and an AR(2) component that accounts for the fluctuations within a scene. Two random processes of i.i.d. rvs are used to model the sizes of P and B frames, respectively. The complete model is then obtained by intermixing the three sub-models according to a given GOP pattern. It is shown that the composite model exhibits longrange dependence (LRD) in the sense that its autocorrelation function is non-summable. The LRD behavior is caused by the repetitive GOP pattern which induces periodic crosscorrelations between different types of frames. Using standard statistical methods, we successfully fit our model to several empirical vi...

We study caching strategies for proxies that cache VBR encoded continuous media objects for highly interactive streaming applications. First, we develop a model for streaming VBR encoded continuous media objects. This model forms the basis for a stream admission control crite- rion and our study of caching strategies. We find that unlike conventional web caches, proxy caches for continuous media objects need to replicate or stripe objects to achieve high hit rates. We develop novel caching strategies that either implicitly or explicitly track the request pattern and cache (and replicate) objects accordingly. Our numerical results indicate that our caching strategies achieve significantly higher hit rates than caching without object repli- cation.

Broadband networks will increasing carry prerecorded traffic, such as high-fidelity audio, short multimedia clips, and full-length movies. We study how to manage the transmission and transport of traffic from prerecorded VBR sources so that network resources are efficiently utilized and end users receive satisfactory service. Specifically, we study Piecewise Constant-Rate Transmission and Transport (PCRTT), whereby the server transmits and the network transports each connection's packets at different constant rates over a small number of intervals. We show how dynamic programming (DP) can be applied to find optimal PCRTT rates and intervals for a wide variety of optimization criteria, including criteria which explicitly account for delays to user interaction (pause, fast forward, and rewind). We also introduce two admission policies for PCRTT: peak-rate admission and packing admission. For peak-rate admission, for each video the network allocates an end-to-end CBR connection; the rate ...

Streaming of continuous media over wireless links is a notoriously difficult problem. This is due to the stringent Quality of Service requirements of continuous media and the unreliability of wireless links. We develop a streaming protocol for the real--time delivery of gprerecorded continuous media from a central base station to multiple mobile clients within a wireless cell. Our protocol prefetches parts of the ongoing continuous media streams into prefetch buffers in the clients. Our protocol prefetches according to a Join--the--Shortest--Queue policy. By exploiting rate adaptation techniques of wireless data packet protocols, the Join--the--Shortest--Queue policy dynamically allocates more transmission capacity to clients with small prefetched reserves. Our protocol uses channel probing to handle the location--dependent, time--varying, and bursty errors of wireless links. We evaluate our prefetching protocol through extensive simulations with VBR MPEG encoded video traces. Our simulations indicate that for bursty VBR video with an average rate of 64 kbit/sec and typical wireless communication conditions our prefetching protocol achieves client starvation probabilities on the order of and a bandwidth efficiency of 90 % with client buffers of 128 kBytes.

krunzQece.arizona. edu

We develop a practical, multiple time scale model for MPEG video traffic whose accuracy and relatively low computational complexity make it well suited for real-time traffic generation experiments on broadband networks. The major feature of our approach is the decomposition of the frame size sequence into simple slow and fast time scale components. This accurately captures aspects of queueing behavior that are difficult to model otherwise. The model also exploits the existence of deterministic patterns that are due to the MPEG coding scheme. We also present a novel modeling approach based on spatial renewal processes (SRP). This model gives exact matches to any desired marginal distribution and any convex non-increasing autocorrelation function. In particular, it can match subexponentially decaying autocorrelations (i.e., can capture long range dependence), something no other model of comparable complexity can do. A SRP is suited for on-line model construction, since it involves no sea...

The recently developed H.264/AVC video codec with Scalable Video Coding (SVC) extension, compresses non-scalable (single-layer) and scalable video significantly more efficiently than MPEG–4 Part 2. Since the traffic characteristics of encoded video have a significant impact on its network transport, we examine the bit rate-distortion and bit rate variability-distortion performance of single-layer video traffic of the H.264/AVC codec and SVC extension using long CIF resolution videos. We also compare the traffic characteristics of the hierarchical B frames (SVC) versus classical B frames. In addition, we examine the impact of frame size smoothing on the video traffic to mitigate the effect of bit rate variabilities. We find that compared to MPEG–4 Part 2, the H.264/AVC codec and SVC extension achieve lower average bit rates at the expense of significantly increased traffic variabilities that remain at a high level even with smoothing. Through simulations we investigate the implications of this increase in rate variability on (i) frame losses when transmitting a single video, and (ii) on the number of supported video streams in a bufferless statistical multiplexing scenario with restricted link capacity and information loss. We find increased frame losses, and rate-distortion/rate-variability/encoding complexity tradeoffs. We conclude that solely assessing bit rate-distortion improvements of video encoder technologies is not sufficient to predict the performance in specific networked application scenarios. Index Terms Frame loss ratio, H.264/AVC, hierarchical B frames, rate-distortion (RD), rate variability-distortion (VD), single-layer video, statistical multiplexing, SVC, video traffic. I.

The Internet of the future and next generation wireless systems are expected to carry to a large extent video of heterogeneous quality and video that is scalable encoded (into multiple layers). However, due to a lack of long traces of heterogeneous and scalable encoded video, most video networking studies are currently conducted with traces of single layer (non–scalable) encoded video. In this technical report we present a publicly available library of traces of heterogeneous and scalable encoded video. The traces have been generated from over 15 videos of one hour each, which have been encoded into two layers using the temporal scalability and spatial scalability modes of MPEG–4. We provide both the frame sizes as well as the frame qualities (PSNR values) in the traces. We study the statistical characteristics of the traces, including their long–range–dependence and multi–fractal properties.

Statistical evidence suggests that the autocorrelation function ae(k) (k = 0; 1; : : :) of a compressedvideo sequence is better captured by ae(k) = e \Gammafi p k than by ae(k) = k \Gammafi = e \Gammafi log k (long-range dependence) or ae(k) = e \Gammafik (Markovian). A video model with such a correlation structure is introduced based on the so-called M jGj1 input processes. In essence, the M jGj1 process is a stationary version of the busy-server process of a discrete-time M jGj1 queue. By varying G, many forms of time dependence can be displayed, which makes the class of M jGj1 input models a good candidate for modeling many types of correlated traffic in computer networks. For video traffic, we derive the appropriate G that gives the desired correlation function ae(k) = e \Gammafi p k . Though not Markovian, this model is shown to exhibit short-range dependence. Poisson variates of the M jGj1 model are appropriately transformed to capture the marginal distribution of a ...