Home About us Contact
|
Home About us Contact | ||
|
|
||
Effective Bandwidth (effective + bandwidth)
Selected AbstractsPerformance issues of bandwidth management in ATM networksINTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS, Issue 2 2003Christos Bouras Abstract In our days, efficient management of the available network resources becomes a critical issue, both from a functional point of view (so that users can be provided with the bandwidth they need), and an economical point of view (so that carriers can satisfactorily and efficiently serve as many customers as possible and at the same time increase their revenue). In this paper we consider a bandwidth control scheme (i.e. managed bandwidth service) for an ATM network infrastructure which is applied to the Greek research and technology network (GRNET). We present some methods that we have tested (in a simulation setting) in order to increase the efficiency of the system and the utilization of the available bandwidth. More specifically, we consider a bandwidth-resizing algorithm for virtual paths, in order to keep the allocated bandwidth very close to the bandwidth actually used. This leads to an increased number of accepted requests and better network utilization. We, also, use the simulation results in order to get an estimation of the effective bandwidth for VBR paths that can be used in call admission. Finally, we consider a semi-offline scheme where requests are gathered and considered for acceptance in regular intervals. Simulation results show an increase in the utilization of resources. As a further improvement, we allow connections to be allocated a little before or after the time initially requested. This leads to further improvement in network utilization. All the improvement schemes were tested with the ATM-TN simulator and the results look promising. Copyright © 2003 John Wiley & Sons, Ltd. [source] Novel broadband asymptotic waveform evaluation method and its applications in parameter extractionINTERNATIONAL JOURNAL OF RF AND MICROWAVE COMPUTER-AIDED ENGINEERING, Issue 3 2007Guo-Bing Han Abstract A new technique to estimate the effective bandwidth of the asymptotic waveform evaluation (AWE) method is proposed, which guarantees the reliable application of AWE in a certain region. Furthermore, based on the effective bandwidth estimation, we develop a novel broadband AWE method, which is an extension and improvement of the conventional AWE method. The new method selects the multiple expansion points automatically by a simple compare algorithm and then implements the AWE process on every expansion point, thereby expanding the variable bandwidth of the AWE greatly. The variable here can be frequency, structure dimension, material parameter, etc. In the end, two examples of parameter extraction of microwave structures are given, which demonstrate that the new method has high efficiency and good accuracy in a broad band. © 2007 Wiley Periodicals, Inc. Int J RF and Microwave CAE, 2007. [source] Precision 1H,1H distance measurement via 13C NMR signals: utilization of 1H,1H double-quantum dipolar interactions recoupled under magic angle spinning conditionsMAGNETIC RESONANCE IN CHEMISTRY, Issue 2 2004Yoh Matsuki Abstract We applied the POST-C7 DQ-dipolar recoupling pulse sequence to the measurement of 1H,1H distances with high precision. The spectral resolution is enhanced by detecting the 1H magnetization via 13C signals. A least-squares fitting of the build-up curve of the transferred magnetization to the exact numerical simulations yielded a 1H,,1H, distance of 248 ± 4 pm for fully 13C-labeled L -valine. This distance agrees with the neutron diffraction study. The negative transferred magnetization clearly indicates that the direct DQ 1H,1H dipolar couplings have the largest effect. The signal for the magnetization transfer builds up rapidly by the direct 1H,1H dipolar coupling, and decreases to zero at longer mixing time when the relayed magnetization transfer becomes significant. This large intensity change of the signal leads to the high precision in the distance measurement. We inspected factors that limit the effective bandwidth of the POST-C7 recoupling for the 1H and 13C homonuclear spin systems. The spin interactions at times shorter than the cycle time of the C7 sequence were also evaluated to measure the distances. The carbon-detected 2D 1H DQ mixing experiment was demonstrated for the measurement of multiple 1H,1H distances. Copyright © 2004 John Wiley & Sons, Ltd. [source] Conductance distribution at criticality: one-dimensional Anderson model with random long-range hoppingANNALEN DER PHYSIK, Issue 12 2009A. Méndez Abstract We study numerically the conductance distribution function w(T) for the one-dimensional Anderson model with random long-range hopping described by the Power-law Banded Random Matrix model at criticality. We concentrate on the case of two single-channel leads attached to the system. We observe a smooth transition from localized to delocalized behavior in the conductance distribution by increasing b, the effective bandwidth of the model. Also, for b < 1 we show that w(ln T/Ttyp) is scale invariant, where Ttyp = exp , ln T , is the typical value of T. Moreover, we find that for T < Ttyp, w(ln T/Ttyp) shows a universal behavior proportional to (T/Ttyp) -1/2. [source] Modelling multi-dimensional QoS: some fundamental constraintsINTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS, Issue 3 2004Nelson X. Liu Abstract In this paper, we model multi-dimensional QoS in a unified framework, and study some fundamental constraints from the network and the traffic on realizing multiple QoS goals. Multi-dimensional QoS requirements are quantitatively represented using a QoS region. Based on the theory of effective bandwidths, the framework connects the throughput, the delay, and the loss rate in a uniform formula. Important traffic and network factors, namely, the burst size and the link speed, are involved. With this framework, it is found that the burst size sets hard limit on the QoS region that can be achieved, and that the matching between the link speed and the node processing power can greatly improve the limit. It is also made clear that while pure load imbalance among links does not affect the QoS region, the heterogeneities of burst size or link speed may severely degrade the QoS performance. Applying the theory to real-time services in differentiated services architecture, we show it provides a useful tool for QoS prediction and network dimensioning. Copyright © 2004 John Wiley & Sons, Ltd. [source] | ||||||||||