Short-term Quality Of Service Research Articles

LA-LMRBF: Online and Long-Term Web Service QoS Forecasting

We propose a <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L</u> ong-term Quality of Service (QoS) forecasting approach using <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">A</u> dvertisement and <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L</u> evenberg- <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">M</u> arquardt improved <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">R</u> adial <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</u> asis <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</u> unction (LA-LMRBF)—a novel online QoS forecasting approach. LA-LMRBF aims to accurately predict QoS attributes of Web services in the form of multivariate time series via three stages. First, the phase space reconstruction theory is employed to restore multi-dimensional and nonlinear relations among the multivariate QoS attributes. Second, short-term QoS advertisement data is incorporated to enable long-term QoS forecasting. Finally, an optimized Radial Basis Function (RBF) neural network is constructed to forecast long-term multivariate QoS values, where the Affinity Propagation clustering algorithm is used to determine the number of hidden nodes and the Levenberg-Marquardt (LM) algorithm is utilized to dynamically update some parameters of the RBF neural network. A series of experiments are performed on a mixture of public and self-collected data sets. The results show that LA-LMRBF is superior to the other approaches and more suitable for long-term QoS forecasting.

Joint Power-Rate-Slot Resource Allocation in Energy Harvesting-Powered Wireless Body Area Networks

Wireless body area network (WBAN) has become a promising network for continuous health monitoring of various diseases. The limited energy of sensors in WBAN cannot support the long-term work with the high requirements of Quality of Service (QoS) for health applications. Energy harvesting (EH)-powered WBAN, which can provide uninterrupted work, has attracted more attention from both academic and industry. However, the time-varying and heterogeneous EH states of different sensors become an important factor when designing the resource allocation schemes in EH-powered WBAN. In this paper, we propose a novel two-phase resource allocation scheme, which optimizes the allocation of the transmission power, the source rate and the time slots to improve the QoS performances of EH-powered WBAN. In the first phase, we investigate that the statistical knowledge of energy harvesting to measure the long-term QoS performances, and then a joint power-rate control scheme is proposed to optimize the transmission power and the source rate for ensuring the long-term QoS performance. Moreover, we design a QoS aware slot allocation scheme to dynamically adjust the time slot allocation to cope with the time-varying and heterogeneous EH states for obtaining better short-term QoS performances in the second phase. Finally, numerical simulation results demonstrate that the proposed joint power-rate-slot resource allocation scheme of EH-powered WBANs can efficiently exploit the time-varying EH to improve both the long-term and short-term QoS performances.

A Software-Defined Networking Framework to Provide Dynamic QoS Management in IEEE 802.11 Networks

In this paper, the concept of SDN (Software Defined Networking) is extended to be applied to wireless networks. Traditionally, in a wired SDN environment, the OpenFlow protocol is the communication protocol used to configure the flow table of forwarding elements (i.e., switches and Access Points). However, although in IEEE 802.11 networks there is no concept of forwarding, the SDN paradigm could also be applied to set up the wireless network dynamically, in order to improve the performance. In this case, not only the network elements, that is the Access Points, but also the mobile elements should configure their link and physical layers parameters following the guidelines of a centralized SDN controller. In particular, we propose a mechanism called DEDCA (Dynamic Enhanced Distributed Channel Access) to manage the channel access in wireless networks, and a framework that enables its implementation in 802.11-based wireless networks using SDN technology. The key aspect of this alternative solution is the control over the contention window size of the wireless terminals. Thus, an adequate response to dynamic and short-term Quality of Service (QoS) requirements can be offered to services running on these networks. DEDCA mechanism relies upon the use of a scalar parameter called gain. The mathematical model which has allowed us to obtain this parameter is presented and evaluated in this paper. Finally, the usefulness of the proposed solutions have been evaluated by means of their implementation in an example case.

Scheduling self-similar traffic in packet-switching systems with high utilisation

The authors address the short-term quality of service (QoS) issue caused by the unexpected nature of traffic and network operators&apos; expectations of a high network utilisation. The authors develop a short-term QoS guaranteeing scheduler, called the short-term QoS deficit round robin (SQ-DRR), which adapts the allocated bandwidth to guarantee delay constraints. The key feature of the scheduler is that it guarantees the delay constraint for the short-term aggregate burst traffic violating pre-contracted descriptors without losing long-term fairness. The performance of the scheduler with and without an admission control scheme is evaluated for non-stationary long-range dependent traffic. The simulation results indicate that the SQ-DRR performs well in dynamic burst traffic conditions.