Franco Russo

A wavelet-based approach to the estimation of the Hurst parameter for self-similar data

In this paper we analyse a wavelet based method for the estimation of the Hurst parameter of synthetically-generated self-similar traces, widely used in a great variety of applications, ranging from computer graphics to parsimonious traffic modelling in broadband networks. The aim of this work is to point out the efficiency of multiresolution schemes in the analysis of fractal processes, characterized by similar statistical features over different time scales. To this end we generated a huge amount of data using the random midpoint displacement (RMD) algorithm, a well-known fast technique for the generation of fractional Gaussian noise (fGn) traces. We then evaluated the Hurst parameter of such sequences in the wavelet domain and compared the results with those obtained with more traditional methods, based on the estimation of the fractal dimensional (Higuchi method) and the moments of the aggregated series.

Statistical multiplexing of self-similar VBR videoconferencing traffic

ATM based telecommunication networks are designed to obtain high resource utilization by means of statistical multiplexing of traffic sources corresponding to heterogeneous multimedia services. As highlighted by recent measurements and statistical studies, these applications are quite different from traditional narrowband services. Particularly, self similarity and long range dependence, that characterise the bursty nature of the traffic pattern, have been pointed out as a very predominant characteristic in the design of new networks. In order to achieve an efficient resource utilisation, peak rate allocation must be substituted by a type of dynamic rate allocation. This paper presents a dynamic bandwidth allocation method that employs linear prediction of the input rate based on self-similar modelling of data. The performances of the proposed strategy, in terms of multiplexing gain for reasonable values of cell loss rate and queuing delay, have been evaluated considering experimental data collected over a DQDB MAN during a multiparty videoconferencing session.

On the use of pipesize estimators to improve TCP transient behavior

This paper presents a simulative analysis of a modification to the TCP congestion control mechanism called ESSE (early slow start exit), designed to improve the TCP startup phase by setting the slow start threshold according to a pipesize estimation based on the observation of few ACK arrival times. We evaluate the performance of ESSE by using various methods to estimate the pipesize as the ratio between the round trip time and the spacing between ACK. This algorithm is easy to implement and preserves the compatibility with the standard protocol since it requires changes to the sender side only. Simulative experiments show that ESSE allows us to speed-up TCP connections and drastically reduce the packet drop rate under several working conditions and load levels. Better performance of TCP can be observed for any of the considered estimators, which indicates that the algorithm is robust against estimation errors. Further, the characteristics of fairness and friendliness (towards Newreno) of the algorithm are investigated. According to our simulations, ESSE-modified protocols guarantee fair utilization of bandwidth among homogeneous and heterogeneous (Newreno) connections sharing a common link.

Comparison of LRD and SRD traffic models for the performance evaluation of finite buffer systems

The paper presents a comparison of several traffic models in order to investigate their adequacy to approximate the behaviour of actual traffic in a single server queue under realistic working conditions. Hence, considering queueing systems with finite buffers, models parsimoniously fitting the long range dependence (LRD) property are compared to intrinsically short range dependent models and to flexible ones, capturing both LRD and short range dependence (SRD). The paper highlights an optimistic queueing behaviour evaluated by means of models capturing only LRD property, when realistic buffer sizes are considered and a strong SRD correlation component is present. The simulation results point out that the circulant modulated Poisson processes (CMPP) and fractional autoregressive integrated moving average (p,d,0) (f-ARIMA(p,d,0)) are suitable models to estimate the performance of the queueing system under heavy load conditions. On the contrary, when the system works under low-medium load conditions, the relevance of the marginal distribution to queueing performance and the inaccuracy of the CMPP model on fitting this statistic in the case of actual traffic data lead to a reduction of the accuracy of this model on system performance evaluation.

Performance Analysis of a Single Server Queue Loaded by Long Range Dependent Input Traffic

Recent studies on broadband networks show that real traffics exhibit Long Range Dependence (LRD). Because of its dramatic consequences on queueing performances this property has to be taken into account in network design and management. To these aims we analysed the behaviour of a single server queueing system loaded by an LRD arrival process modelled by fractional Gaussian noise. Our goal was to obtain an analytical estimation of the complementary probability, that represents an upper bound for Cell Loss Ratio. The approach is based on the Large Deviations Theory and we successfully validated the results by means of simulation studies.

DiffServ Aggregation Strategies of Real Time Services in a WFQ+ Schedulers Network

The paper presents an analysis in a DiffServ network scenario of the achievable Quality of Service performance when different aggregation strategies between video and voice real time services are considered. Each network node of the analyzed DiffServ scenario is represented by a WF2Q+ scheduler. The parameters setting of the WF2Q+ scheduler is also discussed. In particular, the necessary network resources, estimated by the WF2Q+ parameters setting obtained considering the aggregation of traffic sources belonging to the same service class, are compared with those estimated on a per flow basis. The higher gain achievable using the first approach with respect to the second one, is also qualitatively highlighted. The simulation results evidence the possible problems that can be raised when voice traffic is merged with video service traffic. As a consequence, the paper results suggest to consider in different service class queues the two kinds of traffic.

A Novel High-Speed Micro-Flows Classification Algorithm Based on Perfect Hashing and Direct Addressing

The high level of performance achieved by todays traditional PCs makes x86 personal computers effective and inexpensive platforms for the development of high performance network devices. In particular, such devices can replace, especially in the access portion of the network, traditional devices for operations like management, monitoring and accounting. This new possibility pushes the research towards the development of new algorithms and functionalities on top of the x86 architecture. In this framework, the purpose of this work is to design a new high-speed longest prefix classification algorithm for packet accounting to be integrated on top of Linux based PCs for traffic characterization and accounting on high speed (> 1Gbps) links, regardless of the detail level of the rule set.

A prediction based UPC mechanism for VBR video traffic

Preventive actions have to be used to avoid congestion in ATM networks. These actions are represented by call admission control (CAC) and usage parameter control functions (UPC). Two UPC algorithms are compared and an enhancement of the leaky bucket mechanism is presented. The comparison is based on the results of several trace driven simulations. The simulations, realised on a CAMAD tool, are driven by MPEG1 video traces as input traffic. The effectiveness of the proposed algorithm has been evaluated in terms of response time and transparency. These performance indexes have been obtained considering the number of non-conforming cells and the mean value of the conforming cells path.

Self similar progressive image coding scheme for multimedia packet switching communications

Connectionless services provided by best effort networks, such as SMDS and CBDS, are emerging in support of multicast real time communication over a global environment. A major role in the development of these services will be played by voice and video coding. The paper presents a novel coding approach for still images that combines a fractal approach with wavelet transform and multiresolution decomposition. A progressive coding is achieved by finding self-similarities between different scale bands of the wavelet pyramid. This coding technique has been successfully adopted over a regional wide IP environment where some users were connected by dial-up lines, several by N-ISDN and others through a LAN to LAN interconnection service provided by a regional DQDB MAN.

Efficient estimation of buffer occupancy in ATM systems loaded by self-similar traffic

B-ISDN networks based on asynchronous transfer mode (ATM) are characterized by very low cell loss probability. This results in an enormous computational burden when determining the systems performance via Monte Carlo simulation. To improve the computational efficiency different methods (for example Restart and GEVT) have been proposed in literature to evaluate the occurrence of rare events. We show how importance sampling (IS) allows one to obtain useful results in this field. While IS techniques have been largely exploited in the evaluation of bit error rate in digital communication systems and of false alarm rate in radar devices very little has been done to develop efficient IS-based schemes to simulate a queueing system fed with self-similar traffic. With this work we show how IS may be applied to simulate a queue fed both with fractional Gaussian noise (FGN) and with synthetic traces with characteristics close to the ones of real traffic, resulting from videoconferencing sessions.