Taekhyun Kim

A wavelet-based approach to detect shared congestion

Per-flow congestion control helps endpoints fairly and efficiently share network resources. Better utilization of network resources can be achieved, however, if congestion management algorithms can determine when two different flows share a congested link. Such knowledge can be used to implement cooperative congestion control or improve the overlay topology of a P2P system. Previous techniques to detect shared congestion either assume a common source or destination node, drop-tail queueing, or a single point of congestion. We propose in this paper a novel technique, applicable to any pair of paths on the Internet, without such limitations. Our technique employs a signal processing method, wavelet denoising, to separate queueing delay caused by network congestion from various other delay variations. Our wavelet-based technique is evaluated through both simulations and Internet experiments. We show that, when detecting shared congestion of paths with a common endpoint, our technique provides faster convergence and higher accuracy while using fewer packets than previous techniques, and that it also accurately determines when there is no shared congestion. Furthermore, we show that our technique is robust and accurate for paths without a common endpoint or synchronized clocks; more specifically, it can tolerate a synchronization offset of up to one second between two packet flows.

LED Lamp Flicker Caused by Interharmonics

The current IEC flickermeter standard only considers light flicker of incandescent lamps. Hence, the standard flicker-meter has limitations regarding assessing flicker of lamp types other than the incandescent lamp, particularly in the presence of interharmonics. In this paper, the flicker responses of various lamps (e.g., incandescent, compact fluorescent, and LED (light-emitting-diode) lamps) are investigated using a laboratory interharmonic flicker station and a photo detector. The experimental results demonstrate that LED lamps exhibit flicker response similar to that of compact fluorescent lamps, which are sensitive to high-frequency interharmonics located close to the odd order harmonics. Our analysis indicates that a diode bridge rectifier, which converts AC to DC for both compact fluorescent and LED lamps, is a source of flicker due to high-frequency interharmonics.

Detection of Flicker Caused by Interharmonics

International standard bodies have specified an instrument capable of detecting and assessing the severity of lighting flicker, which is called a flickermeter. However, a deficiency with regard to flicker caused by high-frequency interharmonics has been reported in the recent literature. In this paper, another deficiency of the International Electrotechnical Commission (IEC) flickermeter with regard to low-frequency interharmonics, whose frequencies are below the difference between the fundamental frequency and the cutoff frequency of a bandpass filter of the IEC standard flickermeter, will be discussed and illustrated based on analysis and numerical experiments. A new approach based on down-up sampling in the discrete-time sample domain is also proposed to address the flicker-detection problem associated with interharmonics. We demonstrate the efficacy of the proposed method by a comparison with the current flickermeter standard in the presence of interharmonics.

Scalable Clustering of Internet Paths by Shared Congestion

Internet paths sharing the same bottleneck can be identified using several shared congestion detection techniques. However, all of these techniques have been designed to detect shared congestion between a pair of paths. To cluster N paths by shared congestion, a straightforward approach of using pairwise tests would require O(N) time complexity. In this paper, we present a scalable approach to cluster Internet paths based on DCW (Delay Correlation with Wavelet denoising) which does not require a common end point between paths. We present a function to map each path’s measurement data into a point in a multidimensional space such that points are close to each other if and only if the corresponding paths share congestion. Because points in the space are indexed using a tree-like structure, the computational complexity of clustering N paths can be reduced to O(N log N). The indexing overhead can be further improved by reducing dimensionality of the space through wavelet transform. Computation cost is kept low by reusing for dimensionality reduction the same wavelet coefficients obtained in DCW. Our approach is evaluated by simulations and found to be effective for a large N . The tradeoff between dimensionality and clustering accuracy is shown empirically.

ASD System Condition Monitoring Using Cross Bicoherence

In this paper, we propose a novel condition monitoring method for adjustable speed drives (ASD) and ASD-driven motors. The key idea is to diagnose the status of the load side of an ASD from observations made at the source side. The use of a higher-order spectral metric, the cross bicoherence, is proposed for the detection and analysis of anomalies associated with nonlinear devices in ASD systems. In this paper, we apply our method to detect phase imbalance at the load side of an ASD. Our experimental results demonstrate that the proposed method provides a unique signature of phase imbalance for detection and classification of asymmetric impedance associated with the phase imbalance. Furthermore, the proposed method shows a more sensitive detection performance compared to the conventional imbalance measurement method, which enables early detection of potential faults.

Detection of Flicker Caused by High-frequency Interharmonics

International standard bodies specified have an instrument capable of detecting and assessing the severity of lighting flicker, which is called a flickermeter. However, the recommended flickermeter did not consider flicker caused by high-frequency interharmonics. In recent studies, the deficiency of the current flickermeter standard is reported based on real field case examples. In this paper, we review the deficiency of the current flickermeter standard regarding interharmonics, and propose a new approach based on down-up sampling in the discrete time sample domain. We demonstrate the efficacy of the proposed method by qualitative and quantitative comparison with the current flickermeter standard in the presence of interharmonics.

Real and reactive power analysis for interharmonics

Due to the non-harmonic nature of interharmonics, power flow analysis associated with interharmonic frequencies has not been intensively studied. Hence, we propose the use of the cross power spectrum technique for interharmonic power analysis. The cross power spectrum allows one to determine the magnitude and direction of real and reactive power flow at every frequency of interest. Therefore, the method appears to be an appropriate tool for interharmonic power analysis. The efficacy of the proposed method is demonstrated through an investigation of voltage and current waveforms from a simulated current source type inverter drive system based on ATP-EMTP. It is shown that power flow associated with interharmonic events can be better understood in a quantitative manner with the use of the cross power spectrum method.

Application of wavelet denoising to the detection of shared congestion in overlay multimedia networks

The overlay network approach is an emerging technique to satisfy the strict requirements for various real-time multimedia services. However, overlay networks suffer from a shared congestion problem since each unicast flow may interfere with each other in the common underlying links. Most previous techniques to detect shared congestion have limitations when applied as a general solution, since they assume perfect synchronization between probing packets. However, our recent work shows that a technique based on wavelet denoising can overcome the limitations by mitigating the interfering effects such as synchronization offset and the random fluctuations of queueing delay; the proposed technique provides a more robust and accurate detection in the presence of a large amount of synchronization offset In this paper, wavelet denoising is tailored to the characteristics of queueing delay on packet networks. The wavelet denoising based technique is verified through extensive simulations. The efficacy of the proposed approach is demonstrated by the detection accuracy and convergence speed.