Distributed Computer and Communication Networks

Frontmatter

Next-Generation Internet Projects

Abstract

The paper gives an introduction to some projects and initiatives that are connected with the new-generation Internet. US-Ignite partnerships, the European commission Future Internet Program, the European Fire projects, the Global Environment for Network Innovation (GENI), and WiMAX, are presented. New technologies and protocols, such as Software-defined networking (SDN), OpenFlow, a route configuration mechanism and the constituent protocols to add redundant packet forwarding capabilities that will provide high reliability communication for critical applications, are described. The goal of the paper is to understand intricacies, and nuances of some of these techniques and show some of the possibilities of next-generation high-speed networking and their possible applications in the field of the Internet and the applications to education, libraries and museums.

John Geske, Peter Stanchev

Design and Scheduling in 5G Stationary and Mobile Communication Systems Based on Wireless Millimeter-Wave Mesh Networks

Abstract

The paper presents the concept of local and metropolitan wireless mesh networks operating in millimeter-wave band. These self-organizing networks may be used as the high-throughput backbones for emerging 5G communication systems, as well as standalone networks providing extensive QoS, mobility, reliability and ultra-high throughputs for the connected users. The paper observes the current state and the prospects of the development of the millimeter-wave mesh networks. The paper outlines the network logical structure including MAC level design, routing and resource allocation. The problem of delay-optimal scheduling within Spatial TDMA is formulated, and a simple fast algorithm for scheduling is proposed. The experimental results provided prove the algorithm allows to utilize the network resources at very high rates while building scheduling sufficient for delay-critical applications.

Vladimir Vishnevsky, Andrey Larionov, Sergey Frolov

A Survey of Congestion Control Mechanisms in Linux TCP

Abstract

The Transmission Control Protocol (TCP) is used by the vast majority of Internet applications. Since its introduction in the 1970s, a lot of variants have been proposed to cope with the different network conditions we can have (e.g., wired networks, wireless networks, satellite links) and nowadays Linux OS includes 13 different TCP variants. The aim of this paper is to provide a complete survey of the different congestion control mechanisms used by the variants of the TCP implemented in the Linux Kernel 2.6.x.

Christian Callegari, Stefano Giordano, Michele Pagano, Teresa Pepe

Analysis of the Resource Distribution Schemes in LTE-Advanced Relay-Enhanced Networks

Abstract

Nowadays with the introduction of smart phones along with the development of audio and video applications increase people’s demands for the ubiquitous high-data rate coverage. One of the cost-effective solutions adopted in the Long-Term Evolution-Advanced (LTE-Advanced) standard that extends coverage and enhances throughput is the placement of heterogeneous nodes: low-power Relay Nodes (RNs) are deployed to assist transmissions between an evolved Node B (eNB) and multiple User Equipments (UEs). In this framework resource allocation becomes an important and crucial problem that directly influences the potential capacity and coverage improvements. In this paper we overview the concepts of basic relaying and resource allocation in a relay-enhanced LTE-Advanced network. We investigate the performance of an analytical model for a relay-enhanced LTE-Advanced downlink channel, which is presented in detail. Various resource allocation schemes are analyzed, among which, according to the experimental analysis, the proportional one with constraints achieves the best performance.

Tatiana Efimushkina, Konstantin Samuylov

Modelling Transient States in Queueing Models of Computer Networks: A Few Practical Issues

Abstract

The article summarises author’s experience in two problems related to the use of queueing models in performance evaluation of computer networks: modelling transient states of queues and computations for queueing network models having large number of nodes. Both issues are not well represented in classical queueing theory, yet important to applications, because the observed traffic is time dependant and network topologies that should be considered become larger and larger. The article discusses two approaches: diffusion approximation and fluid-flow approximation that can cope with much larger models that are attainable with the use of Markov chains.

Tadeusz Czachórski, Monika Nycz, Tomasz Nycz

New Generation Computer Networks Survivability Analysis and Optimization

Abstract

The problem of survivability analysis of new generation computer networks is considered. Survivability index for computer networks is suggested and method for its estimation is described. The problem of computer networks structure optimization by survivability indices is also considered and method for its solution described. The experimental investigations of the suggested methods for computer networks survivability analysis and optimization are presented.

Yurij Zaychenko, Helen Zaychenko

Optimisation of Extended Generalised Fat Tree Topologies

Abstract

Extended generalised fat tree (XGFT) are interconnection networks with bidirectional multistage properties, (BMIN) which can be extended or scaled to accommodate different system sizes and requirements. However, these extended topologies do not address power consumption and traffic constraints. In this paper, we extract a sub-set of the generalised fat tree topologies that are power consumption and performance aware. We called this sub-set optimised OXGFT. The cost which is proportional to the relative power is the objective function that is minimised based on the traffic constraints to maintain a lower delay and a higher throughput. The simulation results show that the extracted OXGFT topologies perform well under various load conditions.

Adamantini Peratikou, Mo Adda

The Problems of Digital TV and Radio Broadcasting Systems Implementation

Abstract

Implementation of digital broadcasting systems requires the solution of a set of problems, including national and international standardization, advances in video and channel coding, measurement methods and equipment for broadcasting systems. Different systems could be effective for highly populated and sparsely inhabited regions. New digital terrestrial multimedia broadcasting system may solve such problems. Some key characteristics of the RAVIS system are presented and the prospects of the system are discussed.

Viktor P. Dvorkovich, Alexander V. Dvorkovich

Cross-Layer Adaptation-Based Video Downlink Transmission over LTE: Survey

Abstract

The recent technological growth in the field of video communications has caused a rise of video applications along with an emergence of new large screen mobile devices. Note that unlike the second and the third generation wireless technologies, the Long-Term Evolution (LTE) is able to support high data rate video applications based on its increased capacity. A Cross-Layer Adaptation (CLA) approach has become very popular due to its ability to achieve an acceptable video quality and satisfy delay requirements over the error-prone LTE networks. The CLA design implies the interaction between the network layers that results in adaptation of the source compression and channel coding toward an optimized video delivery. This paper presents a survey on the recent CLA advances which claim to provide considerable improvements for video communication over LTE. State-of-the-art CLA techniques are studied based on the type of adaptation, distortion estimation and optimization problem formulation and solution.

Tatiana Efimushkina, Moncef Gabbouj

A Discrete Waiting Time Model for Optical Signals

Abstract

We consider a discrete time queueing system where the service of a customer may start at the moment of arrival or at moments differing from it by the multiples of a given cycle time. One finds the distribution of waiting time and its mean value. These results give possibility for the numerical optimization of cycle length. The original model was raised in connection with the landing process of airplanes, but it appears to be an exact model to describe the functioning of a node at the transmission of optical signals.

Laszlo Lakatos, Dmitry Efrosinin

Analytical Model of the TCP Reno Congestion Control Procedure through a Discrete-Time Markov Chain

Abstract

An analytical model of the Reno congestion control procedure for Transmission Control Protocol is presented, and its theoretical predictions are compared with real TCP traces and some well-known results [1]. The model is based on Discrete-Time Markov Chain, and it covers slow start, congestion avoidance, fast recovery, fast retransmit, cumulative and selective acknowledgments, timeouts with exponential back-off and appropriate byte counting features of TCP. The model provides a way to estimate Reno TCP performance as a function of round trip time and loss rate for bulk transfer TCP flow. The model allows to compare SACK performance with traditional cumulative acknowledgments. The model can be used to find conditions beneficial for advertised window adjustment to improve performance.

Vladimir Kokshenev, Sergey Suschenko

Asymptotic Method of Traffic Simulations

Abstract

A method of simulation and predicting the behavior of daily traffic in computer or mobile communication networks based on differential equations of nonlinear oscillations with one degree of freedom and with a small parameter was proposed. Small perturbations are modelled using the sum of delta functions. The asympototic method was used for solving the differential equations. Some numerical experiments are presented.

Ivanna Dronjuk, Maria Nazarkevych, Olga Fedevych

New Generation of Safety Systems for Automobile Traffic Control Using RFID Technology and Broadband Wireless Communication

Abstract

Broadband wireless data transmission network for providing of automobile transport system safety is considered. The network operates under IEEE802.11n-2012 protocol that guarantees high-speed transmission of multimedia information between automatic stationary and mobile systems of traffic control. The model of stochastic network with dependent service time for the problem solution is used. Based on product form representation of the steady state probabilities of such system the formulas and algorithms for calculating the main system characteristics such as: mean marginal queue length and mean sojourn times in the nodes, as well as mean packages delivery time in the whole network are proposed. Results of some numerical experiments are presented.

Vladimir Vishnevsky, Dmitry Kozyrev, Vladimir Rykov

Parameters Estimator of the Probabilistic Model of Moving Batches Traffic Flow

Abstract

A probabilistic model for time characteristic of a traffic flow moving on a motorway is proposed and investigated in this paper. The time intervals between consecutive cars are supposed to be dependent and have different distribution. Cars with the slow and fast movement are distinguished in the traffic flow. The mathematical model of such traffic flow is represented as a control cybernetic system of a certain class. The methods to derive estimate for the parameters of the control cybernetic system are proposed in order to select the adequate traffic flow model in the form of batches flow. These methods are approved processing the statistical data of the Bartlett traffic flow. Effectiveness of suggested methods for the parameters estimation and algorithms for splitting a real traffic flow into the batches is demonstrated.

Michael Fedotkin, Maria Rachinskaya

Simulation of Wireless Broadband Network Traffic in the Electromagnetic Interference

Abstract

The article describes the authors’ simulation system and the results of heterogeneous network traffic modeling in the physical channel of the wireless network for different values of the signal/noise ratio in a range of different broadband noise and intensity.

Vladimir S. Zhdanov, Victor M. Churkov

On the Effective Envelopes for Fluid Queues with Gaussian Input

Abstract

Thanks to their flexibility and compact characterization, Gaussian processes have emerged as popular models to describe the traffic dynamics in a wide class of the modern telecommunication networks. A relatively new characterization of traffic flows is based on the effective envelopes, which represent a probabilistic generalization of the arrival curve of Network Calculus. In this paper, we analyse the effective envelopes for a general Gaussian process and use these results to derive non-asymptotic performance bounds for a fluid queuing system. To highlight the effectiveness of the proposed approach, numerical results are shown taking into account heterogeneous traffic flows as well as different correlation structures.

Oleg Lukashenko, Evsey Morozov, Michele Pagano

Algorithmic Analysis of Dual Tandem Queue with Batch Markovian Arrival Process and Repeated Attempts at the First Station

Abstract

A dual tandem queueing system with a Batch Markovian Arrival Process (\(BMAP\)) is considered. The first station is represented by a retrial multi-server queueing system with infinite orbit. General form of dependence of the total retrial intensity on the current number of customers in orbit is suggested. The second station is a multi-server queue with a finite buffer and impatient customers. The stationary distribution of the system states and the main performance measures of the tandem queue under consideration are calculated.

Valentina Klimenok, Alexander Dudin

Performance Characteristics for DD Priority Discipline with Semi-Markov Switching

Abstract

Priority Discretionary Discipline (DD) with an arbitrary number of priority classes and with switching losses at switching process of service from one class of priority to another one is considered. Such characteristics as distribution of busy periods, queue lengths, probability of states, as well as various supplementary distributions are presented. Symmetric priority discipline named Relative DD discipline also was analyzed. Even through, the main characteristics are obtained in the terms of Laplace, Laplace-Stieltjes transforms and recurrent functional equations, the obtained relations are convenient for algorithmization.

Gheorghe Mishkoy, Lilia Mitev

Multi-rate Loss Model for Optical Network Unit in Passive Optical Networks

Abstract

Nowadays, extensive research on passive optical network (PON) technology is being conducted. PON technology facilitates the use of inexpensive passive components and helps reduce the total length of optical fiber required. PON is a future optical technology that enables high-speed data transfer of multiservice traffic using optical fibers. This paper is concerned with multi-rate loss model for optical network units in PON like a queue system with a limited capacity buffer. The results obtained in the paper are applied to the calculation of the quality of service parameters of the network.

Gelii Basharin, Nadezhda Rusina

Investigation of the Queueing Network $$GI-(GI|\infty )^K$$ G I - ( G I | ∞ ) K by Means of the First Jump Equation and Asymptotic Analysis

Abstract

Analysis of the open non-Markovian queueing network with renewal arrival process, Markovian routing, infinite servers count and general service time distribution is presented in the paper. Equation for characteristic function of the joint distribution of the number of customers in the nodes of the network is derived. Approximations for this characteristic function are presented under asymptotic condition of an infinite growth of the arrival rate. Both stationary and non-stationary cases are considered. It is shown that the multi-dimensional distributions under study can be approximated by the multi-dimensional Gaussian distributions. Expressions for parameters of these distributions are obtained.

Anatoly Nazarov, Alexander Moiseev

Performance Analysis and Monotone Control of a Tandem Queueing System

Abstract

A controllable tandem queueing system consists of two nodes in tandem of the type \(M/M/n_i\) and a controller. Customers arrive to the controller, who allocates them between the nodes. After service completion at node 2 the controller can allocate the customer waiting at node 1 to node 2. With probability \(p\) after a service completion at node 1 a failure occurs. In this case the customer from node 1 joins node 2. With complement probability \(1-p\) the service completion at node 1 is successful. For the given cost structure we formulate an optimal allocation problem to minimize the long-run average cost per unit of time. Using dynamic-programming approach we show the existence of thresholds which divides the state-space into two contiguous regions where the optimal decision is to allocate the customers to node 1 or to node 2. Some monotonicity properties of the dynamic-programming value function are established.

Dmitry Efrosinin, Mais Farhadov, Saule Kudubaeva

Application of RFID-Technology to the Problem of Traffic Control through Roadway Intersections

Abstract

This paper is devoted to an actual problem: the optimization of traffic light to control on roadway intersection. In order to reduce the total delay of vehicles, minimize the queues at junctions and optimize traffic flow through signalized intersection a model of stochastic polling systems is proposed. On order to determine the most effective service disciplines a comparative analysis of two different service disciplines was conducted. Exhaustive and gated service disciplines were analyzed, compared and simulated for the purpose to choose the most effective one. The viability and efficiency of the proposed algorithm using computer simulation are also demonstrated. Moreover the possibility of using RFID (Radio Frequency Identification) technology in order to reduce the total delay of vehicles was analyzed.

Olga Semenova, Stanislav Lykov

Backmatter