• International Journal of Control, Automation, and Systems
• /
• v.5 no.2
• /
• pp.155-160
• /
• 2007

A fault detection and isolation (FDI) problem is considered for inertial sensors, such as gyroscopes and accelerometers and a new FDI method for double faults is proposed using reduced-order parity vector. The reduced-order parity vector (RPV) algorithm enables us to isolate double faults with 7 sensors. Averaged parity vector is used to reduce false alarm and wrong isolation, and to improve correct isolation. The RPV algorithm is analyzed by Monte-Carlo simulation and the performance is given through fault detection probability, correct isolation probability, and wrong isolation probability.

• Journal of the Korean Society of Safety
• /
• v.19 no.1
• /
• pp.49-55
• /
• 2004

This study presents an analysis of comparison of P-type fire detection system with fuzzy logic-applied fire detection system. The fuzzy logic-applied fire detection system has input variables obtained by fire experiment of small scale with K-type temperature sensor and optical smoke sensor. And the antecedent part of fuzzy rules consists of temperature and smoke density, and the consequent part consists of fire probability. Also triangular fuzzy membership function is used for input variables and fuzzy rules. To calculate the final fire probability a centroid method is introduced. A fire experiment is conducted with controlling wood crib layer, cigarette to simulate actual fire and false alarm situation. The results show that peak fire probability is 25[%] for non-fire and is more than 80[%] for fire situation, respectively. The fuzzy logic-applied fire detection system suggested here is able to distinguish fire situation and non-fire situation very precisely.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.4
• /
• pp.380-386
• /
• 2002

Conventional post-processing as like confidence measure (CM) proposed by Rahim calculates phones' CM using the likelihood between phoneme model and anti-model, and then word's CM is obtained by averaging phone-level CMs[1]. In conventional method, CMs of some specific keywords are tory low and they are usually rejected. The reason is that statistics of phone-level CMs are not consistent. In other words, phone-level CMs have different probability density functions (pdf) for each phone, especially sri-phone. To overcome this problem, in this paper, we propose normalized confidence measure. Our approach is to transform CM pdf of each tri-phone to the same pdf under the assumption that CM pdfs are Gaussian. For evaluating our method we use common keyword spotting system. In that system context-dependent HMM models are used for modeling keyword utterance and contort-independent HMM models are applied to non-keyword utterance. The experiment results show that the proposed NCM reduced FAR (false alarm rate) from 0.44 to 0.33 FA/KW/HR (false alarm/keyword/hour) when MDR is about 8%. It achieves 25% improvement of FAR.

• Journal of Information Processing Systems
• /
• v.4 no.3
• /
• pp.113-120
• /
• 2008

Most enterprises have controlled claim data related to marketing, production, trade and delivery. They can extract the engineering information needed to the reliability of unit from the claim data, and also detect critical and latent reliability problems. Existing method which could detect abnormal quality unit lists in early stage from claim database has three problems: the exclusion of fallacy probability in claim, the false occurrence of claim fallacy alarm caused by not reflecting inventory information and too many excessive considerations of claim change factors. In this paper, we propose a process and methods extracting abnormal quality unit lists to solve three problems of existing method. Proposed one includes data extraction process for reliability measurement, the calculation method of claim fallacy alarm probability, the method for reflecting inventory time in calculating claim reliability and the method for identification of abnormal quality unit lists. This paper also shows that proposed mechanism could be effectively used after analyzing improved effects taken from automotive company's claim data adaptation for two years.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.11A
• /
• pp.1063-1071
• /
• 2008

This paper proposes a prior maximum likelihood (ML) detection verifier which has an ability to verify if the zero forcing (ZF) detection results are identical to the ML detection results. Since more than 90% of ZF detection results are identical to ML detection results, the proposed verifier makes it possible to omit the computationally complex ML detection in 90% cases of MIMO signal detections. The proposed verifier is designed by using the diversity gain obtained from converting MIMO signal into single input multiple output (SIMO) signals. In the proposed method, single input multiple output (SIMO) signals for each transmit antenna are separated from MIMO signals after the MIMO signals are detected by ZF method. Computer simulations show that the true alarm probability of the proposed verifier is more than 80% and the false alarm probability is less than $10^{-4}$.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.49 no.3
• /
• pp.37-46
• /
• 2012

In RADAR and SONAR detection systems, noise environment can be classified into homogeneous and heterogeneous environment. Especially heterogeneous environments are modelled as target masking and clutter edge. Since the variability-index (VI) CFAR, a composed CFAR algorithm, dynamically selects one of the mean-level algorithms based on the VI and the MR (mean ratio) test, it is robust to various environments. However, the VI CFAR still suffers from lowered detection probabilities in heterogeneous environments. To overcome these problems, we propose an improved VI CFAR processor where TM (trimmed mean) CFAR and a sub-windowing technique are introduced to minimize the degradation of the detection probabilities appeared in heterogeneous environments. Computer simulation results show that the proposed method has the better performance in terms of detection probability and false alarm probability compared to the VI CFAR and single CFAR algorithms.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.6
• /
• pp.153-162
• /
• 2009

This paper suggests the novel light-weight signal processing algorithm for wireless sensor network applications which needs low computing complexity and power consumption. Exponential average method (EA) is utilized by real time, to process the magnetometer signal which is analyzed to understand the own physical characteristic in time domain. EA provides the robustness about noise, magnetic drift by temperature and interference, furthermore, causes low memory consumption and computing complexity for embedded processor. Hence, optimal parameter of proposal algorithm is extracted by statistical analysis. Using general and precision magnetometer, detection probability over 90% is obtained which restricted by 5% false alarm rate in simulation and using own developed magnetometer H/W, detection probability over 60~70% is obtained under 1~5% false alarm rate in simulation and experiment.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.29 no.12
• /
• pp.68-76
• /
• 2015

We examine the standalone lightning warning system (LWS) and its warning performances for three years. This system acquires and analyzes the data of cloud-to-ground strike (CG), intra-cloud discharge (IC) and electrostatic field (EF) to produce prior warnings with respect to the impending arrival of CG in the area of concern (AOC). The warnings in this system are carried out based on the fixed two areas method. To evaluate warning performances, we analyzed the statistics of warnings with probability of detection (POD) and false alarm ratio (FAR). Based on the previous study, we revised the trigger and clear conditions of lightning warning for improving the performances of the system. As a result of this revision, POD increased from 0.18 to 0.44 and FAR decreased from 0.96 to 0.78 during the summer of 2014. However, the LWS was not possible to trigger effective alerts (EA) because there was no effective lead time (LT) for the fixed two areas method. Problems related to the low detection efficiency of IC and the use of EF data for warnings still decreased POD and increased FAR. Hence, we proposed the development method of a new LWS (NLWS) that would be composed of integrated weather data, the flexible two areas and the user software in order to trigger EA and improve warning performances.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1653-1661
• /
• 2022

In this study, the classification of cold water and normal water based on Geo-Kompsat 2A images was performed. Daily mean surface temperature products provided by the National Meteorological Satellite Center (NMSC) were used, and convolution neural network (CNN) deep learning technique was applied as a classification algorithm. From 2019 to 2022, the cold water occurrence data provided by the National Institute of Fisheries Science (NIFS) were used as the cold water class. As a result of learning, the probability of detection was 82.5% and the false alarm ratio was 54.4%. Through misclassification analysis, it was confirmed that cloud area should be considered and accurate learning data should be considered in the future.

• International Journal of Safety
• /
• v.11 no.1
• /
• pp.6-10
• /
• 2012

This paper presents a study on the analysis for reducing the number of false alarms in fire detection system. In order to intelligent algorithm fuzzy logic is adopted in developing fire detection system to reduce false alarm. The intelligent fire detection algorithm compared and analyzed the fire and non-fire signatures measured in circuits simulating flame fire and smoldering fire. The algorithm has input variables obtained by fire experiment with K-type thermocouple and optical smoke sensor. Also triangular membership function is used for inference rules. And the antecedent part of inference rules consists of temperature and smoke density, and the consequent part consists of fire probability. A fire-experiment is conducted with paper, plastic, and n-heptane to simulate actual fire situation. The results show that the intelligent fire detection algorithm suggested in this study can more effectively discriminate signatures between fire and similar fire.