Grid-connected distributed generation power systems (DGPS) based on inverters require the employed controller to include an islanding detection algorithm in order to determine the grid status and operate properly. In ...
详细信息
Grid-connected distributed generation power systems (DGPS) based on inverters require the employed controller to include an islanding detection algorithm in order to determine the grid status and operate properly. In certain cases, such as low-power low-cost current-controlled inverters used in residential photovoltaic (PV) systems, the inverter must be stopped once the islanding condition is detected according to the standard and grid-code limits. Passive, active and monitoring-based algorithms for islanding detection have been analyzed in literature considering one DGPS connected to the local electrical power system (EPS). Only in case of active methods, such as active frequency drift (AFD), the effect of two DGPS in the same local EPS on the islanding capabilities has been analyzed. This paper analyzes the performance of diverse passive and active islanding detection methods (IDM) in multiple inverters connected at the same local EPS interact.
This paper proposes spatial outliers detection method of studying multiple non-spatial attributes based on special objects. SOFMF algorithm is presented and its implementation has been discussed in detail in this arti...
详细信息
This paper proposes spatial outliers detection method of studying multiple non-spatial attributes based on special objects. SOFMF algorithm is presented and its implementation has been discussed in detail in this article. Simultaneously analyze and summarize this algorithm: overcome the insufficiency of many clustering algorithms, be able to find clusters in different shapes, be non-sensitive to the input data sequence, process noise data and multi-dimensional data well, and have multi-resolution. A novel idea for spatial data clustering is proposed by the author, emphatically numerous experiments prove this idea can be applied to spatial clustering quite well.
Current outlier detection schemes typically output a numeric score representing the degree to which a given observation is an outlier. We argue that converting the scores into well-calibrated probability estimates is ...
详细信息
Current outlier detection schemes typically output a numeric score representing the degree to which a given observation is an outlier. We argue that converting the scores into well-calibrated probability estimates is more favorable for several reasons. First, the probability estimates allow us to select the appropriate threshold for declaring outliers using a Bayesian risk model. Second, the probability estimates obtained from individual models can be aggregated to build an ensemble outlier detection framework. In this paper, we present two methods for transforming outlier scores into probabilities. The first approach assumes that the posterior probabilities follow a logistic sigmoid function and learns the parameters of the function from the distribution of outlier scores. The second approach models the score distributions as a mixture of exponential and Gaussian probability functions and calculates the posterior probabilites via the Bayes' rule. We evaluated the efficacy of both methods in the context of threshold selection and ensemble outlier detection. We also show that the calibration accuracy improves with the aid of some labeled examples.
This paper presents and shares excerpts from our implementation of near real-time anomaly detection algorithms on the IBM InfoSphere Streams platform. The purpose of this article is to: 1) Describe how to design and i...
详细信息
This paper presents and shares excerpts from our implementation of near real-time anomaly detection algorithms on the IBM InfoSphere Streams platform. The purpose of this article is to: 1) Describe how to design and implement known anomaly detection algorithms on IBM InfoSphere Streams. 2) Present some performance optimization capabilities of IBM InfoSphere Streams platform and propose a method to use them in anomaly detection applications. 3) Present some IBM InfoSphere Streams best practices and describe how their adoption in the context of anomaly detection application. The document describes the architecture and design of anomaly detection algorithms developed on IBM InfoSphere Streams. Although the solution was designed to be used for cyber security, the implemented algorithms are agnostic regarding the data type that they monitor and therefore can detect anomalies in data from various industries such as healthcare, finance and retail. The document describes the implementation of two anomaly detection algorithms: KOAD and PCA. The KOAD algorithm performs online anomaly detection with incremental learning and the PCA algorithm in performs offline anomaly detection. The solution was designed to provide near real-time insight into low latency on large data volume observation.
This paper compares the performance of parametric and non-parametric sequential change detection algorithms for detecting in-band wormholes in wireless ad hoc networks. The algorithms considered are the non-parametric...
详细信息
This paper compares the performance of parametric and non-parametric sequential change detection algorithms for detecting in-band wormholes in wireless ad hoc networks. The algorithms considered are the non-parametric cumulative sum (NP-CUSUM) and the repeated sequential probability ratio test (R-SPRT). Theoretical performance of the two is compared using metrics that take into account the algorithms' repeated nature, and the advantage of the parametric method is illustrated. On the other hand, connections between the parametric and non-parametric methods are made in the proposed worst case adversary model, where the non-parametric method is shown to be more robust to attack strategy changes. Experimental evaluation of wormhole detection schemes based on the two algorithms is presented. This work has implications for both the theoretical understanding and practical design of wormhole detection schemes based on parametric and nonparametric change detection algorithms.
Epilepsy is a dynamic disease in which the brain transitions between different states. In this paper, we focus on the problem of identifying the time points, referred to as change points, where the transitions between...
详细信息
ISBN:
(纸本)9781479923915
Epilepsy is a dynamic disease in which the brain transitions between different states. In this paper, we focus on the problem of identifying the time points, referred to as change points, where the transitions between these different states happen. A Bayesian change point detection algorithm that does not require the knowledge of the total number of states or the parameters of the probability distribution modeling the activity of epileptic brain in each of these states is developed in this paper. This algorithm works in online mode making it amenable for real-time monitoring. To reduce the quadratic complexity of this algorithm, an approximate algorithm with linear complexity in the number of data points is also developed. Finally, we use these algorithms on ECoG recordings of an epileptic patient to locate the change points and determine segments corresponding to different brain states.
Detecting the road area and ego-lane ahead of a vehicle is central to modern driver assistance systems. While lane-detection on well-marked roads is already available in modern vehicles, finding the boundaries of unma...
详细信息
ISBN:
(纸本)9781479929153
Detecting the road area and ego-lane ahead of a vehicle is central to modern driver assistance systems. While lane-detection on well-marked roads is already available in modern vehicles, finding the boundaries of unmarked or weakly marked roads and lanes as they appear in inner-city and rural environments remains an unsolved problem due to the high variability in scene layout and illumination conditions, amongst others. While recent years have witnessed great interest in this subject, to date no commonly agreed upon benchmark exists, rendering a fair comparison amongst methods difficult. In this paper, we introduce a novel open-access dataset and benchmark for road area and ego-lane detection. Our dataset comprises 600 annotated training and test images of high variability from the KITTI autonomous driving project, capturing a broad spectrum of urban road scenes. For evaluation, we propose to use the 2D Bird's Eye View (BEV) space as vehicle control usually happens in this 2D world, requiring detection results to be represented in this very same space. Furthermore, we propose a novel, behavior-based metric which judges the utility of the extracted ego-lane area for driver assistance applications by fitting a driving corridor to the road detection results in the BEV. We believe this to be important for a meaningful evaluation as pixel-level performance is of limited value for vehicle control. State-of-the-art road detection algorithms are used to demonstrate results using classical pixel-level metrics in perspective and BEV space as well as the novel behavior-based performance measure. All data and annotations are made publicly available on the KITTI online evaluation website in order to serve as a common benchmark for road terrain detection algorithms.
Automatic voicing-decision algorithms depend on thresholds which are dependent on speaker, channel, S/N ratio, etc. Low-frequency energy (LFE) is one of the best voicing statistics when properly thresholded; it is eve...
详细信息
Automatic voicing-decision algorithms depend on thresholds which are dependent on speaker, channel, S/N ratio, etc. Low-frequency energy (LFE) is one of the best voicing statistics when properly thresholded; it is even better if two thresholds are set, one for onset of voicing and one for offset. Two schemes are proposed for adaptive, estimation of thresholds. The first is finding stretches that are "surely" voiced or unvoiced, finding boundaries by heuristic algorithms, and setting thresholds consistent with these boundaries, in the second, one finds segments that are "surely" voiced or unvoiced according to voicing statistics other than LFE, using these to form estimates of the distribution of LFE in voiced and unvoiced cases. Both schemes successfully determine speaker-dependent thresholds in about 15 seconds, during which "standard" thresholds can be used. Overall voicing error rate using LFE with adaptive thresholds is about 1%.
Spectrum sensing is a critical function for enabling dynamic spectrum access (DSA) in wireless networks that utilize cognitive radio (CR). In DSA networks, unlicensed secondary users can gain access to a licensed spec...
详细信息
Spectrum sensing is a critical function for enabling dynamic spectrum access (DSA) in wireless networks that utilize cognitive radio (CR). In DSA networks, unlicensed secondary users can gain access to a licensed spectrum band as long as they do not interfere with primary users. Spectrum sensing is subject to errors in the form of false alarms and missed detections. False alarms cause spectrum under-use by secondary users, and missed detections cause interference to primary users. Although existing research has demonstrated the utility of a Markov chain for modeling the spectrum access pattern of primary users over time, little effort has been directed toward spectrum sensing based upon such models. In this paper, we develop soft-input sequence detection algorithms of Markov sources in noise for spectrum sensing in DSA networks. We assign different Bayesian cost factors for missed detections and false alarms, and we show that a suitably modified Forward-Backward sequence detection algorithm is optimal in minimizing the detection risk. Along the way, we observe new fundamental limitations that we call "Risk Floor" and "Limiting ROC" for energy detection and coherent detection due to the PU's spectrum access pattern.
暂无评论