Consider the link bandwidth allocation for transport of correlated traffic on a single link with finite buffer capacity under a maximum allowable delay constraint dmax . In our measurement architecture, traffic is decomposed into three frequency regions: low-frequency (LF) traffic in 0 ! j!j ! L , high-frequency (HF) traffic in j!j ! H and mid-frequency (MF) traffic in ! L ! j!j ! ! H . The zero-frequency component (dc term) of the traffic provides the average input rate which corresponds to the minimum link bandwidth requirement. As one will see, traffic in each frequency region plays a fundamentally different role in link bandwidth allocation. Subject to the delay constraint dmax , we develop an engineering guideline (! L ; ! H ) = ( 0:01ß dmax ; 2ß dmax ) for the frequency division in measurement. Hence, the transport of LF traffic exceeds the limit of dmax -constrained buffer capacity; its link bandwidth is essentially captured by the peak rate of LF traffic. In contrast,...

TCP Westwood (TCPW) is a sender-side only modification of TCP Reno congestion control, which exploits end-to-end bandwidth estimation to properly set the values of slow-start threshold and congestion window after a congestion episode. This paper aims at showing via both mathematical modeling and extensive simulations that TCPW significantly improves fair sharing of high-speed networks capacity and that TCPW is friendly to TCP Reno. Moreover, we propose EASY RED, which is a simple Active Queue management (AQM) scheme that improves fair sharing of network capacity especially over high-speed networks. Simulation results show that TCP Westwood provides a remarkable Jain's fairness index increment up to 200% with respect to TCP Reno and confirm that TCPW is friendly to TCP Reno. Finally, simulations show that Easy RED improves fairness of Reno connections more than RED, whereas the improvement in the case of Westwood connections is much smaller since Westwood already exhibits a fairer behavior by itself. 1.

SMAQ is a measurement-based tool for integration of traffic modeling and queuing analysis. It can be used in a variety of network design areas. For instance, it can be used as a traffic generator to generate various traces for network testing. It also provides numerical solutions of the queue length and loss rate performance for transport of multimedia traffic. Several application modules are built into the tool for the evaluation of statistical multiplexing, buffer dimensioning, and link bandwidth allocation. Other examples include the evaluation of traffic shaping, local congestion control, and the modeling of wireless channel dynamics. As one will find, the SMAQ tool indeed provides a solution technique for network engineers to solve many of the current design issues. A trial version of the Web-based SMAQ tool can be found on our Web site at

The multipath fading channel modeling traditionally focuses on physical-level dynamics such as signal strength and bit error rate. In this paper we characterize multipath fading channel dynamics at the packet-level and analyze the corresponding data queueing performance in various environments. The integration of wireless channel modeling and data queueing analysis provides us a unique way to capture important channel statistics with respect to various wireless network factors such as channel bandwidth, mobile speed and channel coding. The second order channel statistics, i.e., channel power spectrum, is found to play an important role in the modeling of multipath fading channels. The data queueing performance is largely dependent on the interaction between the channel power spectrum and the data arrival power spectrum; whichever has more low frequency power will have more impact on queueing performance. Note that the data arrival power spectrum provides a measure of burstine...

Today TCP/IP congestion control implements the additive increase/multiplicative decrease (AIMD) paradigm to probe network capacity and obtain a "rough" but robust measurement of the best effort available bandwidth. Westwood TCP proposes an additive increase/adaptive decrease paradigm that adaptively sets the transmission rate at the end of the probing phase to match the bandwidth used at the time of congestion, which is the definition of best-effort available bandwidth in a connectionless packet network. This paper addresses the challenging issue of estimating the best effort bandwidth available for a TCP/IP connection by properly counting and filtering the flow of acknowledgments packets using discrete-time filters. We show that in order to implement a low-pass filter in packet networks it is necessary to implement an anti ACK compression algorithm, which plays the role of a classic anti-aliasing filter. Moreover, a comparison of time-invariant and time-varying discrete filters to be used after the anti-aliasing algorithm is developed.

Consider a class of queueing systems which can be modeled by a finite Quasi-Birth-Death (QBD) process. In this paper we develop a powerful computational technique for spectral analyses (i.e. second-order statistics) of output, queue and loss. Emphasis is placed on output power spectrum and input-output coherence function in response to various input power spectral properties and system parameters. The coherence function is defined to measure linear relationship between input and output processes. A key technical contribution of this paper is the exploration of linearity of low frequency traffic flow. Through the evaluation of the coherence function, one can identify a so-called nonlinear break frequency, ! b , under which the low frequency traffic stay intact via a queueing system. Such a low frequency I/O linearity plays an important role in characterizing the output process, which may form a partial input to other "downstream" queues of the network. After all, it is the "upstream" ou...

In this paper, we propose a refinement of TCP Westwood allowing the management of the Efficiency/Friendliness-to-NewReno tradeoff. We show that the refined TCP Westwood is able to achieve higher efficiency yet at the same time maintain friendliness. TCP Westwood (TCPW) implements a novel window congestion control algorithm based on available bandwidth estimation. The performance of TCPW has been promising, exceeding that of TCP NewReno in high speed and/or wired/wireless networks. However, under certain circumstances, TCP NewReno may experience some performance degradation because TCPW possesses more information and thus can take better advantage of available bandwidth. In this paper we propose combining the original TCPW sampling strategy that produces available Bandwidth Estimates (BE), with a new strategy that produces Rate Estimates (RE). Our studies show that RE works best when packet losses are mostly due to congestion. If, on the other hand, the packet losses are mostly due to link errors, BE gives better performance. To achieve the “best of all worlds”, we introduce a method we call Combined Rate and Bandwidth estimation (CRB). A connection first infers the predominant cause of packet losses, and then uses the most appropriate estimation method. We also introduce the Efficiency/Friendliness Tradeoff Graph that provides better tradeoff visualization. In our experiments, we found that CRB provides a better compromise between efficiency and friendliness, and the means to manage such a tradeoff. 1.

In packet-switched network traffic management and control, efficiently allocating bandwidth to provide quantitative packet-level QoS to aggregate traffic has been difficult due to unpredictable, unknown statistical characteristics of the aggregate traffic. With inaccurate traffic information, using static bandwidth allocation results in the network being underutilized, or the QoS requirement not being satisfied. An alternative is to use Adaptive Bandwidth Control (ABC), whereby the allocated bandwidth is regularly adjusted over the packet-level time scale to attain a given QoS requirement. This paper provides a literature review of ABC algorithms that guarantee aggregate traffic packet-level QoS metrics, such as the average queue length, packet loss, and packet delay. We describe different ABC algorithms, identify their advantages and shortcomings, and provide some comparative performance evaluation results. Open issues in ABC for future research directions are also discussed.

In this paper, we present the time-dependent solutions of various stochastic processes associated with a finite Quasi-Birth-Death queueing system. These include the transient queueing solutions, the transient departure and loss intensity processes and certain transient cummulative measures associated with the queueing system. The focus of our study is the effect of the arrival process correlation on the queueing system before it reaches steady-state. With the aid of numerous examples, we investigate the strong relationship between the time scales of variation of the arrival process and those of the transient queueing, loss and departure processes. These time-dependent solutions require the computation of the exponential of the stochastic generator matrix G which may be of very large order. This precludes the use of known techniques to solve the matrix exponential such as the eigenvalue decomposition of G. We present a numerical technique based on the computation of the Laplace Transfor...

This paper presents an ABR control scheme (ABR 2 - DP) which maximises the utilisation of information available about the network traffic. Firsfly, it takes into account the long-range dependence of the self-similar background network traffic, and utilises it for more accurate multi-step predictions of the explicit rate (ER). Secondly, prediction errors for connections with greater network delay are compensated for in the ER calculations for connections with smaller delay. Hence, the available bandwidth is distributed both in temporal and spatial dimensions. The scheme allows a reduction in buffer requirements and queueing delay. Performance results are presented and compared with a previously proposed scheme which uses digital filtering.