Antoine Scherrer

Non-Gaussian and Long Memory Statistical Characterizations for Internet Traffic with Anomalies

The goals of the present contribution are twofold. First, we propose the use of a non-Gaussian long-range dependent process to model Internet traffic aggregated time series. We give the definitions and intuition behind the use of this model. We detail numerical procedures that can be used to synthesize artificial traffic exactly following the model prescription. We also propose original and practically effective procedures to estimate the corresponding parameters from empirical data. We show that this empirical model relevantly describes a large variety of Internet traffic, including both regular traffic obtained from public reference repositories and traffic containing legitimate (flash crowd) or illegitimate (DDoS attack) anomalies. We observe that the proposed model accurately fits the data for a wide range of aggregation levels. The model provides us with a meaningful multiresolution (i.e., aggregation level dependent) statistics to characterize the traffic: the evolution of the estimated parameters with respect to the aggregation level. It opens the track to the second goal of the paper: anomaly detection. We propose the use of a quadratic distance computed on these statistics to detect the occurrences of DDoS attack and study the statistical performance of these detection procedures. Traffic with anomalies was produced and collected by us so as to create a controlled and reproducible database, allowing for a relevant assessment of the statistical performance of the proposed (modeling and detection) procedures

Description and simulation of dynamic mobility networks

During the last decade, the study of large scale complex networks has attracted a substantial amount of attention and works from several domains: sociology, biology, computer science, epidemiology. Most of such complex networks are inherently dynamic, with new vertices and links appearing while some old ones disappear. Until recently, the dynamics of these networks was less studied and there is a strong need for dynamic network models in order to sustain protocol performance evaluations and fundamental analyzes in all the research domains listed above. We propose in this paper a novel framework for the study of dynamic mobility networks. We address the characterization of dynamics by proposing an in-depth description and analysis of two real-world data sets. We show in particular that links creation and deletion processes are independent of other graph properties and that such networks exhibit a large number of possible configurations, from sparse to dense. From those observations, we propose simple yet very accurate models that allow generate random mobility graphs with similar temporal behavior as the one observed in experimental data.

Revisiting an old friend: on the observability of the relation between long range dependence and heavy tail

Taqqu’s Theorem plays a fundamental role in Internet traffic modeling, for two reasons: First, its theoretical formulation matches closely and in a meaningful manner some of the key network mechanisms controlling traffic characteristics; Second, it offers a plausible explanation for the origin of the long range dependence property in relation with the heavy tail nature of the traffic components. Numerous attempts have since been made to observe its predictions empirically, either from real Internet traffic data or from numerical simulations based on popular traffic models, yet rarely has this resulted in satisfactory quantitative agreements. This raised in the literature a number of comments and questions, ranging from the adequacy of the theorem to real world data to the relevance of the statistical tools involved in practical analyses. The present contribution aims at studying under which conditions this fundamental theorem can be actually seen at work on real or simulated data. To do so, numerical simulations based on standard traffic models are analyzed in a wavelet framework. The key time scales involved are derived, enabling a discussion of the origin and nature of the difficulties encountered in attempts to empirically observe Taqqu’s Theorem.

Automatic phase detection for stochastic on-chip traffic generation

(NoC) prototyping is used for adapting NoC parameters to the application running on the chip. This prototyping is currently done using traffic generators which emulate the SoC components (IPs) behavior: processors, hardware accelerators, etc. Traffic generated by processor-like IPs is highly non-regular, it must be decomposed into program phases. We propose an original feature for NoC prototyping, inspired by techniques used in processor architecture performance evaluation: the automatic detection of traffic phases. Integrated in our NoC prototyping environment, this feature permits to have a completely automatic toolchain for the generation of stochastic traffic generators. We show that our traffic generators emulate precisely the behavior of processors and that our environment is a versatile tool for networks-on-chip prototyping. Simulations are performed in a SystemC-based simulation environment with a mesh network-on-chip (DSPIN) and a processor running MP3 decoding applications.

Long-range dependence and on-chip processor traffic

Long-range dependence is a property of stochastic processes that has an important impact on network performance, especially on the buffer usage in routers. We analyze the presence of long-range dependence in on-chip processor traffic and we study the impact of long-range dependence on networks-on-chip. long-range dependence in communication traces of processorips at the cycle-accurate level. We also study the impact of long-range dependence on a real network-on-chip using the SocLib simulation environment and traffic generators of our own. Our experiments show that long-range dependence is not an ubiquitous property of on-chip processor traffic and that its impact on the network-on-chip is highly correlated with the low level communication protocol used.

Software defined radio architecture survey for cognitive testbeds

In this paper we present a survey of existing prototypes dedicated to software defined radio. We propose a classification related to the architectural organization of the prototypes and provide some conclusions about the most promising architectures. This study should be useful for cognitive radio testbed designers who have to choose between many possible computing platforms. We also introduce a new cognitive radio testbed currently under construction and explain how this study have influenced the test-bed designers choices.

Testing fractal connectivity in multivariate long memory processes

Within the framework of long memory multivariate processes, fractal connectivity is a particular model, in which the low frequencies (coarse scales) of the interspectrum of each pair of process components are determined by the autospectra of the components. The underlying intuition is that long memories in each components are likely to arise from a same and single mechanism. The present contribution aims at defining and characterizing a statistical procedure for testing actual fractal connectivity amongst data. The test is based on Fishers Z transform and Pearson correlation coefficient, and anchored in a wavelet framework. Its performance are analyzed theoretically and validated on synthetic data. Its usefulness is illustrated on the analysis of Internet traffic Packet and Byte count time series.

Cycle accurate simulation model generation for SoC prototyping

We present new results concerning the integration of high level designed ips into a complete System on Chip. We first introduce a new computation model that can be used for cycle accurate simulation of register transfer level synthesized hardware. Then we provide simulation of a SoC integrating a data-flow ip synthesized with MMAlpha and the SocLib cycle accurate simulation environment. This integration also validates an efficient generic interface mechanism for data-flow ips.

Analysis of Dynamic Sensor Networks: Power Law Then What?

Recent studies on wireless sensor networks (WSN) have shown that the duration of contacts and inter-contacts are power law distributed. While this is a strong property of these networks, we will show that this is not sufficient to describe properly the dynamics of sensor networks. We will present some coupled arguments from data mining, random processes and graph theory to describe more accurately the dynamics with the use of a random model to show the limits of an approach limited to power law contact durations.

A Generic Multi-Phase On-Chip Traffic Generation Environment

In the process of mapping compute-intensive algorithms onto arrays of processing elements (PEs) an efficient usage of channels between PEs and registers within PEs is crucial for achieving a significant algorithm acceleration. In this paper this problem is solved for algorithms represented as systems of uniform recurrence equations. We address an optimization problem in order to realize the algorithmic data dependencies within the processor array (PA) with minimum cost for channels and registers. There, we use a new mapping approach which allows a direct mapping of the algorithm onto the PA by a partitioning method. In contrast to existing approaches, the authors consider the issue of avoiding redundant usage of channels and registers, which can appear if one instance of a variable has to be transferred from a source PE to several sink PEs. Further, a solution of the optimization problem determines the schedule for the transfer of the variable instances in the channels and their storage in registers as well as the inner schedule for the operations in the PEs. We illustrate our method on the edge detection algorithm