Giovanni Schembra

An accurate and treatable Markov model of MPEG-video traffic

An accurate and treatable Markov-based model for MPEG video traffic is proposed. Before reaching a definition of the model, a large number of MPEG video sequences are analyzed and their statistical characteristics are highlighted. Besides the well-known autocorrelation function, testifying the pseudo-periodicity of the sequence in the short term, and the gamma-shaped probability density function the correlation between different frames belonging to the same group of pictures (GOP) is also accurately studied. Then a structured model is proposed. Thanks to its two-level structure it is able to capture both the inter-GOP and the intra-GOP correlation. In order to obtain the first level of the model, the well-known inverse eigenvalue problem is solved in the discrete-time domain. Finally, the model is used to evaluate the loss probability in an ATM multiplexer loaded by an MPEG video source and an aggregate of external traffic. The accuracy of the model is evaluated by comparing the analytically obtained loss probability with the loss probability calculated by simulating the system using real traffic sequences as driven traces.

Energy saving and network performance: a trade-off approach

Power consumption of the Information and Communication Technology sector (ICT) has recently become a key challenge. In particular, actions to improve energy-efficiency of Internet Service Providers (ISPs) are becoming imperative. To this purpose, in this paper we focus on reducing the power consumption of access nodes in an ISP network, by controlling the amount of service capacity each network device has to offer to meet the actual traffic demand. More specifically, we propose a Green router (G-router) implementing a congestion control technique named Active Window Management (AWM) coupled with a new capacity scaling algorithm named Energy Aware service Rate Tuner Handling (EARTH). The AWM characteristics allow to detect whether a waste of energy is playing out, whereas EARTH is aimed at invoking power management primitives at the hardware level to precisely control the current capacity of access nodes and consequently their power consumption. We test the benefits of the AWM-EARTH mechanism on a realistic scenario. Results show that the capacity scaling technique can save up to 70% of power consumption, while guaranteeing Quality of Service and traffic demand constraints.

Wireless mesh networks to support video surveillance: architecture, protocol, and implementation issues

Current video-surveillance systems typically consist of many video sources distributed over a wide area, transmitting live video streams to a central location for processing and monitoring. The target of this paper is to present an experience of implementation of a large-scale video-surveillance system based on a wireless mesh network infrastructure, discussing architecture, protocol, and implementation issues. More specifically, the paper proposes an architecture for a video-surveillance system, and mainly centers its focus on the routing protocol to be used in the wireless mesh network, evaluating its impact on performance at the receiver side. A wireless mesh network was chosen to support a video-surveillance application in order to reduce the overall system costs and increase scalability and performance. The paper analyzes the performance of the network in order to choose design parameters that will achieve the best trade-off between video encoding quality and the network traffic generated.

Performance evaluation of an adaptive-rate MPEG encoder matching IntServ traffic constraints

Promoting the evolution of the Internet from a simple data network to a true multiservice network today constitutes a challenging task. To this end, the Internet Engineering Task Force (IETF) has set up the Integrated Services (Intserv) and Differentiated Services (DiffServ) Working Groups, with the goal of defining a next-generation Internet, in which traditional best-effort datagram delivery and additional enhanced quality of service delivery classes coexist. The IntServ framework, in particular, is designed to be used in the access network, and requires a traffic source with the capability of matching the traffic characteristics declared to the network, TSpec. In this paper we propose to use the rate control facility usually implemented in the current MPEG encoders to provide a constant bit rate, to shape the output traffic according to the declared TSpec while maintaining an acceptable perceived image quality. In order to assess this scenario, we introduce an SBBP/D/N/K queueing system, where the SBBP emission process is varied according to the quantizer scale parameter chosen by the addressed rate control mechanism. The analytical framework allows us: 1) to evaluate system performance in terms of both the marking probability of nonconforming output traffic and the quantization distortion introduced by the encoder and 2) to choose the TSpec parameters to be declared such that given performance parameters are respected.

Modeling superposition of ON-OFF correlated traffic sources in multimedia applications

In this paper a multimedia source is modeled as an arrival process defined as the superposition of heterogeneous correlated arrival processes, each of which models one monomedia source. To this purpose each monomedia source is modeled as an ON-OFF process where the transition rates are functions of the states of the other monomedia sources. Auto- and cross-correlation functions are derived for each monomedia source. The analytical model is used to develop results for single media performance in a multimedia multiplexer structure.

An analytical paradigm to calculate multiplexer performance in an ATM multimedia environment

In this paper a finite-buffer discrete-time approach is presented in order to analyze the performance of a multiplexer loaded by N multimedia sources. For this purpose the emission process of each multimedia source is defined as the superposition of heterogeneous correlated emission processes, each of which models one monomedia source as a Switched Batch Bernoulli Process (SBBP). In order to model the intermedia relationships, the Markov chain underlying each SBBP consists of a 2-state process, where the transition probabilities are functions of the states of the other monomedia sources. Correlation functions are derived for each monomedia source as well as for the multimedia source as a whole. The proposed analytical model is used to calculate the loss probability and the delay jitter pdf for each media in an ATM multimedia multiplexer structure. The effectiveness of the proposed paradigm is shown by means of various examples in a case study.

Active Window Management: An Efficient Gateway Mechanism for TCP Traffic Control

Many techniques have been proposed in the last few years to address performance degradations in end-to-end congestion control. Although these techniques require parameter tuning to operate in different congestion scenarios, they miss the challenging target of both minimizing network delay and keeping goodput close to the network capacity. In this paper we propose a new mechanism, called Active Window Management (AWM), which addresses these targets by stabilizing the queue length in the network gateways. AWM acts on the Advertised Window parameter in the TCP segment carrying the acknowledge, but it does not affect the TCP protocol. The proposed technique is implemented in the network access gateways, that is, in the gateways through which both the incoming and outgoing packets related to a given TCP connection are forced to go, whatever the routing strategy used in the network. We show that when the access gateways implementing AWM are the bottleneck in the networks, TCP performance is very close to that of a pseudo constant bit rate protocol providing no loss, while network utilization is close to one.

Performance analysis of an ATM multiplexer loaded with VBR traffic generated by multimode speech coders

Multimode coders are able to exploit the different characteristics of the speech waveform and to take into account the different peculiarities of background noise, thus allowing improvements in both signal reconstruction and network-offered load. In this context the variable rate code excited linear prediction (VR-CELP) coding, that is, a multimode variable bit rate (VBR) coding based on the CELP technique, has been introduced in the literature and is currently being considered for use in various applications, especially in the third-generation UMTS cellular systems. The target of the paper is to introduce an efficient and accurate framework allowing a network designer to analyze the impact of multimode VBR speech coding on the quality of service (QoS) provided by a wireless/wired ATM network. In order to capture the coder output characteristics, we propose to model a VR-CELP voice source by using a switched batch Bernoulli process (SBBP). More specifically, three models are introduced and compared in terms of accuracy and simplicity in determining network performance. As a result of the comparison, a four-state model has been chosen as the best tradeoff. The model is then used to analytically derive the loss probability and the jitter probability density function of an ATM multiplexer loaded by a number of VR-CELP sources. Finally, the proposed paradigm has been assessed in a case study where we demonstrate that, for a given output ATM link capacity and for a number of telecommunication services involving voice transmission, VR-CELP coding performs better than traditional on-off coding.

Measuring and modeling energy consumption to design a green NetFPGA giga-router

One of the most weighty matter in the Internet today is the waste of energy due to the fact that consumption of network nodes is not tuned with the input traffic. For this reason, the implementation of rate adaptation facilities in the routers constitutes a challenging problem to make the network energy efficient. Rate adaptation in the green routers is usually achieved by scaling the processing power according to the data rate the router has to manage; at this purpose the clock frequency driving the router processes can be modified according to the input data rate. In this context this paper, starting from a measurement study of the Reference Router implemented on the NetFPGA platform, defines a model of the consumed power as a function of both the managed input traffic and the available clock frequencies. As demonstrated in the paper, the model can be applied by router designers to choose the main router platform parameters, i.e. the number of clock frequencies and the clock frequency switching time, while respecting a given tradeoff between the percentage of energy saved and the maximum tolerated loss probability due to frequency switch.

Traffic specifications for the transmission of stored MPEG video on the Internet

Guarantees of quality-of-service (QoS) in the real-time transmission of stored video on the Internet is a challenging task for the success of many video on demand (VoD) applications. Two QoS classes have been specified by the IETF Integrated Services (intserv) Working Group: Guaranteed Services and Controlled-Load Services. For both of them, it is necessary to provide traffic sources with the capability of calculating the traffic characteristics to be declared to the network, Tspec, on the basis of a limited set of parameters statistically characterizing the traffic and the required level of QoS. The target of this paper is to develop an algorithm for the evaluation of the Tspec parameters which characterize the video stream when a given QoS is required. To this end an analytical framework modeling an MPEG stored-video server and the access network node is introduced. The video emission process is modeled with a switched batch Bernoulli process (SBBP), and performance in the video-server smoother is analytically evaluated. Then the token bucket at the network access point, loaded by the output traffic of the video-server smoother, is modeled to calculate the probability of marking nonconforming data packets.