Chanintorn Jittawiriyanukoon

A design fuzzy control policing mechanisms on quality of service support in wireless networks

The performance communication networks including wireless network have been conceived to carry traffic sources and support a continuum of transport rates ranging from low bit-rate to high bit-rate traffic. As a number of telecommunications traffic (bursty traffic) fluctuates as such a network then it results congestion. The traditional policing mechanisms are finite-sized buffers with queue management techniques and fixed leak rate. Most queue management schemes employ fixed thresholds or a limited number of arrival frames, to determine when to allow or discard the entry of frames. So it is not properly controlled, some sessions transported by the wireless network may not meet their quality-of-service (QoS) requirements. However, traditional policing mechanisms have proved to be inefficient in coping with the conflicting requirements of ideal policing mechanisms, that is, low dropping frames and high conforming frames. An alternative solution based on artificial intelligence techniques, specifically, in the field of fuzzy systems is introduced. In this paper, we propose fuzzy control policing mechanisms that aims at detecting violations in parameter negotiation. We evaluate and compare the performance of the fuzzy control policing mechanisms, namely: FuzzyLeaky Bucket(FLB) Fuzzy Jumping Window(FJW) and Fuzzy Triggered Jumping Window(FTJW) with conventional policing mechanisms, namely: Leaky Bucket(LB), Jumping Window(JW) and Triggered Jumping Window(TJW). Simulation results show that on wireless frames, the fuzzy control policing mechanism helps improve high throughputs and dropped frames compared to other policing mechanisms. In any case, fuzzy control policing mechanism in wireless network meets its quality-of-service (QoS) requirements if source traffic is at high peak rate.

QoS sensitive fuzzy backoff schemes in policing mechanisms

Fuzzy controllers provide an efficient method to handle inexact information as a basic of reasoning. In fuzzy control, the controller can be represented by if-then-else rules. The controller is processing exact input data and is producing exact output data in a deterministic way. However, backoff time computation schemes, namely: fuzzy control backoff time (FB), pseudorandom backoff (PB) time, exponential backoff (EB) time and random backoff (RB) time that are applicable in waiting time re-arrangement in queue. None of these backoff schemes are good for all traffic rates. One scheme may carry well at one certain rate while no longer applicable if rate is changeable. In this paper, we propose a fuzzy backoff schemes in policing mechanism (FP) is introduced. FP involves the policing mechanisms which function prior to the waiting room. The performance of FP is evaluated vis-a-vis ones from pseudorandom backoff scheme (PB), random backoff scheme (RB), exponential backoff scheme (EB) and fuzzy control backoff scheme (FB). Simulation results show that the FP helps improve the amount of conforming frames as well as helps reduce non-conforming frames compared to the other four backoff schemes. In conclusion, FP is appropriate for the QoS required media while RB is only good for bandwidth sensitive case

Performance evaluation of reliable data scheduling for Erlang multimedia in cloud computing

Cloud computing emerges huge number of servers connected via a real-time computer network such as broadband network. Cloud computing refers to a distributed network as it has a capability to run an application on many servers simultaneously. In this paper we present a queuing model that takes reliable data scheduler into account and computes performance characteristics like mean waiting time, mean traversal time, utilization, throughput, and mean queue length. Reliable data will be aware and dynamically scheduled in order to make the decision after the arrival of requested job. In addition to these scheduling algorithms we investigate the resource allocation solutions with a distributed Erlang multimedia data. A queuing model is employed in order to study, analyze, and evaluate the performance of each scheduling algorithm. An approximation method based on Mean Value Analysis (MVA) in queuing theory will be introduced. Results regarding performance metrics from both simulation and approximation then are compared. The difference between simulations and approximation method is insignificant. Results as well confirm that schedulers can be effectively applied in a Cloud Computing environment in terms of fairness and load balancing centric.

A rapid anomaly detection technique for big data curation

Anomaly detection (outlier) using simulation helps us analyze the anomaly instances from big data source. As the hasty explosion of todays data stream, outlier detection technique will be an analytical tool to be employed for evaluating massive unstructured datasets. In order to speed-up the processing time to handle enormous datasets, this research will conduct experiments of advanced distant-based outlier detection algorithms to investigate the most effective algorithms using MOA. The algorithms used in this study are Continuous Outlie Detection (COD), Micro-Cluster based COD or MCOD, and STream OutlierR Miner (STORM). The results demonstrate MCOD algorithm can outperform other two algorithms in terms of processing time and accurate anomalies.

QoS sensitive backpressure scheme over wireless network

Congestion in wireless network has been occurring when the traffic sources are uncoordinated and communication bandwidth is finite. Links congestion & links introduce have been losses and delays. The wireless network might not meet its quality of service (QoS) requirements because it has enough congestion, delays grow, queues overflow, and service degrades. However, an alternative way to solve this problem, we has proposed a backpressure algorithm under which no frames are dropped inside a wireless network frame, when congestion has builds up. If congested nodes have sent backpressure feedback to neighboring nodes for informing them to unavailable buffering capacity and stopping them from forwarding another frames until sufficient buffer becomes available. While, the potential advantages in backpressure networks has not allow frames dropping. In this paper, we have proposed backpressure algorithm that aims to detecting violations in parameter negotiation. We has evaluated and compared performance of backpressure with policing mechanisms, namely is Leaky Bucket (LB); Jumping Window (JW); and Triggered Jumping Window (TJW): Simulation results has show that wireless frames and backpressure helps to improve throughputs and non dropped frames compared with other policing mechanisms. In any case, backpressure meets its quality-of-service (QoS) requirements when source traffic was low peak rate.

Performance evaluation of adaptive rate control (ARC) for multimedia traffic

The paper presents an Adaptive Rate Control (ARC) mechanism to improve the performance of high-speed network to handle multimedia traffic by guaranteeing the cell loss ratio (CLR) for all cell streams. First, the cases in which a Tahoe, Reno, New Reno, SACK and Plain schemes are applicable in peak-cell-rate (PCR) are discussed. The ARC improves the performance by regulating the increment (up) and the decrement (down) of window size (flow control). Incoming traffic rate, number of cell drop, preset size of the window and estimated delay time are taken into account for this regulation. Simulations are used to investigate how Tahoe, Reno, New Reno, SACK and “Plain” can conduct, as congestion existed. Then we compare these results from four schemes to the “Plain” scheme (no flow control applicable) and to the proposed ARC. By altering windows size for the mentioned six schemes, we can obtain the supportive results.