Quality-of-Service Analysis of Queuing Systems with Long-Range-Dependent Network Traffic and Variable Service Capacity

Abstract

Many high-quality measurement studies have demonstrated that wireless network traffic exhibits the noticeable Long-Range-Dependent (LRD) property. Moreover, the fading nature of wireless channels can lead to variable service capacity. Due to the inherent difficulty and high complexity of modelling the fractal-like LRD traffic, existing analytical models of queuing systems with LRD arrival processes have been primarily limited to the simplified scenarios where the service capacity is assumed to be constant. Given the time-varying nature of wireless channels in the real-world working environments, it is very important and necessary to investigate system performance in the presence of variable service capacity. To this end, this paper presents a comprehensive analytical model for queuing systems subject to LRD traffic and variable service capacity. We extend the application of a Large Deviation Principle and derive the closed-form expressions of Quality-of-Service (QoS) metrics. The accuracy of the model validated through extensive simulation experiments makes it a cost-effective evaluation tool for performance analysis of communication networks. To illustrate its applications, the model is adopted to investigate the effects of LRD traffic and variable service capacity on the system performance and resource configuration.

Analysis of Packet Loss in Next Generation Networks Using Large Derivation

“…The Hurst exponent estimate is conducted with the help of three widely spread approaches, the correlation function slope analysis in log-log scale, the variance-time plot analysis and R/S[]. The wide sense stationary stochastic process X(t) has mean variance and autocorrelation function .The autocorrelation function can be represented as (1) where 0 <β <1 and slowly vary to infinite If , the process is called asymptotically second-order self-similar and for exactly second order self similar processes , this implying that the sum of auto correlation diverge. The covariance function decays hyperbolically Main features of self similarity are 1.…”

“…There for FARIMA is better choose for VBR traffic model [8].FARIMA model can represent both short range dependency and long range along with heavy tailed distribution traffic with parameters (p, q) and d − 1/α [7] [8]. The heavy-tailed behavior of marginal distribution and the sub-exponential decay of autocorrelation function exhibited by video traffic have a significant impact on queuing performance [1], [8], hence real time traffic with high persistence had to be modeled III. TRAFFIC MODELING AND PREDICTION USING TIME SERIES MODELS F-ARIMA is an extension of classic ARMA time series model and this is flexible to model both short term and long term correlation structure of a series[] [8].Both lower frequency components as well as higher frequency components can be fitted using one or two parameters of FARIMA and long memory that can explicitly account for persistence to incorporate the long term correlation in the data.…”

“…When congestion occurs in the link, queues in routers build up to create long delays and packet losses that will decrease the quality of service (QoS) provided to the users [6] [1]. In the case of 'conventional' traffic packet losses due to congestion are expected to have less impact on the QoS than in the case of bursty traffic where a burst of packets arriving at a full router is likely to be partially or entirely lost, this situations persist may harm the performance.…”

Analysis of Finite Buffer Queue: Maximum Entropy Probability Distribution With Shifted Fractional Geometric and Arithmetic Means

“…However, the traffic in broadband network contradicts the Markovian assumption and the traffic exhibits a long-term correlation which decays as a power law [2]. The quality-of-service (QoS) parameters in the presence of the power law characteristics are significantly different from the system governed by the SRD [2], [3]. An appropriate mathematical framework for the analysis of such systems is generally based on either M/G/1 type system with service time exhibiting the power law or the stochastic differential equation (SDE) driven by the fractional Brownian motion (fBM) [3].…”

“…The quality-of-service (QoS) parameters in the presence of the power law characteristics are significantly different from the system governed by the SRD [2], [3]. An appropriate mathematical framework for the analysis of such systems is generally based on either M/G/1 type system with service time exhibiting the power law or the stochastic differential equation (SDE) driven by the fractional Brownian motion (fBM) [3]. Modeling input traffic by fBM, Cheng, Zhuang and Wang [4] have calculated the loss probability in finite-size partitioned buffer.…”