Herman D. Hughes

Statistical characteristics and multiplexing of MPEG streams

This paper presents a study of the statistical characteristics and multiplexing of variable-bit-rate (VBR) MPEG-coded video streams. Our results are based on 23 minutes of video obtained from the entertainment movie, The Wizard of Oz. The experimental setup which was used to capture, digitize, and compress the video stream is described. Although the study is conducted at the frame level (as opposed to the slice level), it is observed that the inter-frame correlation structure for the frame-size sequence involves complicated forms of pseudo-periodicity that are mainly affected by the compression pattern of the sequence. A simple model for an MPEG traffic source is developed in which frames are generated according to the compression pattern of the original captured video stream. The number of cells per frame is fitted by a lognormal distribution. Simulations are used to study the performance of an ATM multiplexer for MPEG streams.

Factors affecting the performance of ad hoc networks

Mobile ad hoc networks (MANETs) are an emerging class of network architectures that are characterized by their highly dynamic topology, limited resources (i.e., bandwidth and power), and lack of fixed infrastructure. The primary motivation for such networks is increased flexibility and mobility. Random node mobility along with various other factors such as network size and traffic intensity may be very dynamic, resulting in unpredictable variations in the overall network performance. This study centers on investigating and quantifying the effects of various factors and their two-way interactions on the overall performance of ad hoc networks. This study contributes to the modeling and development of adaptive ad hoc protocols (routing, medium access control, scheduling and buffer management). Using 2/sup k/r factorial experimental design, we isolate and quantify the effects of five factors: node speed, pause-time, network size, number of traffic sources, and type of routing (source versus distributed), that affect the performance of ad hoc networks. Specifically, this paper evaluates the impact of these factors on the following performance metrics: throughput, average routing overhead, and power consumption. Our study was conducted using a library-based simulator for sequential and parallel simulation of wireless networks.

A survey on quality-of-service support for mobile ad hoc networks

The general field of mobile ad hoc networking is still in its infancy. Particularly, the challenge of providing Quality-of-Service (QoS) support for ad hoc networks is an open problem and remains relatively uncharted territory. Providing a complete QoS solution for the ad hoc networking environment requires the interaction and cooperation of several components. These components include: (1) a QoS routing protocol, (2) a resource reservation scheme and (3) a QoS capable medium access control (MAC) layer. In this paper, we present a survey of the current research that has addressed each of these components in the context of ad hoc networks. This work is intended to provide a broad and comprehensive view of the various components and protocols required to provide QoS support in computer networks, focusing primarily on ad hoc networks. First, we introduce the unique characteristics of mobile ad hoc networks, which distinguishing this new network architecture from traditional infrastructured wired and wireless networks (i.e. cellular-based networks). We also discuss the impact of these characteristics on QoS provisioning. Next, we describe the first QoS model proposed for mobile ad hoc networks and its relationship to QoS models proposed for the Internet. We then present a review of the proposed algorithms for each QoS component (e.g. QoS routing, resource reservation and the MAC layer). Copyright

A traffic for MPEG-coded VBR streams

Compression of digital video is the only viable means to transport real-time full-motion video over BISDN/ATM networks. Traffic streams generated by video compressors exhibit complicated patterns which vary from one compression scheme to another. In this paper we investigate the traffic characteristics of video streams which are compressed based on the MPEG standard. Our study is based on 23 minutes of video obtained from an entertainment movie. A particular significance of our data is that it contains all types of coded frames, namely: Intra-coded (I), Prediction (P), and Bidirectional (B) MPEG frames. We describe the statistical behavior of the VBR stream using histograms and autocorrelation functions. A procedure is developed to determine the instants of a scene change based on the changes in the size of successive I frames. It is found that the length of a scene can be modeled by a geometric distribution.A model for an MPEG traffic source is developed in which frames are generated according to the compression pattern of the captured video stream. For each frame type, the number of cells per frame is fitted by a lognormal distribution whose parameters are determined by the frame type. The appropriateness and limitations of the model are examined by studying the multiplexing performance of MPEG streams. Simulations of an ATM multiplexer are conducted, in which traffic sources are derived from the measured VBR trace as well as the proposed model. The queueing performance in both cases is found to be relatively close.

Adaptive QoS routing based on prediction of local performance in ad hoc networks

We propose an adaptive QoS routing scheme based on the prediction of the local performance in ad hoc networks. It is implemented by a link performance prediction strategy. Integrated QoS performance in each local area is estimated based on translating the effects of the lower layer parameters into the link state information. Corresponding to the prediction approach, several mechanisms are built to complete the location information management process (i.e., information monitoring, collecting and updating functions). The node movement is characterized by the probabilities of the link state and the prediction of local QoS performance. The QoS routing proposed is adaptive to its nodes mobility, and also scalable due to the distributed structure. Theoretical computation and simulation results are presented and discussed.

A performance study of loss priorities for MPEG video traffic

We study the performance of a loss priority mechanism for MPEG-coded video traffic. This mechanism, known as partial buffer sharing (PBS), is implemented at an ATM multiplexer. The input to the multiplexer consists of several MPEG streams which are derived from the frame-size trace of a full-motion movie. Each stream contains three types of frames: I, P, and B. The significance of the information carried in MPEG frames differs according to the frame type; with I frames being more important than P and B frames. Hence, PBS with two priority classes is used to reduce the cell loss rate for I frames at the expense of higher loss rates for P and B frames. It is observed that the cell loss rates for the two priority classes are greatly influenced by the level of synchronization among the video streams.

Design and analysis of a buffer management scheme for multimedia traffic with loss and delay priorities

Future broadband networks (e.g., BISDN/ATM) will provide transport for a wide range of applications that vary in their quality-of-service (QoS) requirements. Such requirements are measured by the cell loss rate and the end-to-end delay (or delay variation). This paper presents an analysis of a buffer management scheme known as nested threshold cell discarding with multiple buffers. NTCD/MB implements loss and delay priority queueing strategies in order to satisfy a wide range of quality-of-service requirements. We consider a heterogeneous mix of traffic which consists of real-time and nonreal-time traffic types. A proportion of the real-time traffic is loss-sensitive. Cell arrivals are modeled using the fluid-flow approach. The system we analyze consists of two buffers interacting via a common server. The steady-state probability distribution for the content of the buffers is reported along with other performance statistics. Numerical results are presented to demonstrate the effectiveness of NTCD/MB in simultaneously providing loss and delay performance guarantees.

Performance analysis of a shortest-delay protocol

Abstract A generalized shortest-delay access method (SDAM) protocol for local networks is defined and evaluated. This protocol differs from a previously reported SDAM in that it accommodates a branching-bus topology instead of a single-bus network. It is shown that for small bus-delays, SDAM performs very close to that of M/D1- with perfect scheduling. In this paper, the performance evaluation of SDAM is more pragmatic in that the effects of various protocol overheads (e.g., decoding, turnaround time, initializing packets, etc.) are taken into account. An analysis of the tradeoffs between exhaustive and nonexhaustive transmission disciplines is also presented.

Video traffic modeling over wireless networks

This paper studies the statistical model of MPEG video over wireless networks. A real-time MPEG video traffic workload was generated and transmitted over a wireless network. The model parameters of I, B and P frames were measured before and after wireless transmission. The results we got were very interesting. Intuitively, larger frames are more sensitive to network errors. The parameters associated with the statistical model should change after transmission. However, the results of the experiments showed that the statistical models of all I, B and P frames remain almost unchanged after transmission. We repeated the experiment under different wireless link layer error rates and got the same results. To simulate errors, we designed a controllable wireless link layer error model by modifying the WaveLAN device driver which is a Linux kernel module.

On the multiplexing effects of the ATM traffic

Traffic measurements taken from real data are highly desirable for validating traffic models, designing congestion control and flow control algorithms, implementing communication protocols, and developing switch architectures. This paper investigates the traffic characteristics of several applications in an ATM network with the goal of studying bursty traffic behavior and realizing the multiplexing effects of variable bit rate (VBR) traffic. Specifically, three types of traffic (ftp, voice, and video) are obtained from an ATM testbed. These applications are reported to be important applications that tend to dominate the wide-area networks. The multiplexing effects of FTP traffic are studied to identify empirical bandwidth requirements. The characteristics of the multiplexing effects can be applied to develop an efficient call admission algorithm based on the measured data. These experimental results designed to investigate traffic characteristics provide insights to bursty traffic behavior and performance requirements of practical real-time loads.