• Journal of the Korea Institute of Military Science and Technology
• /
• v.9 no.1 s.24
• /
• pp.24-32
• /
• 2006

This paper is to analyze the effectiveness of multistatic sonar network based on detection performance. The multistatic sonar network is a distributed detection system that places a source and multi-receivers apart. So it needs a detection technique that relates to decision rule and optimization of sonar system to improve the detection performance. For this we propose a data fusion procedure using Bayesian decision and optimal sensor arrangement by optimizing a bistatic sonar. Also, to analyze the detection performance effectively, we propose the environmental model that simulates a propagation loss and target strength suitable for multistatic sonar networks in real surroundings. The effectiveness analysis on the multistatic sonar network confirms itself as a promising tool for effective allocation of detection resources in multistatic sonar system.

• The Journal of the Korea Contents Association
• /
• v.3 no.4
• /
• pp.38-47
• /
• 2003

Distributed problem solving is the cooperative solution of problem by a decentralized and loosely couped collection of knowledge-sources (KS's), located in a number of distinct processor nodes. The contract net protocol has been developed to specify problem-solving communication and control for nodes in a distributed problem solver. Task distribution is affected by a negotiation process, a discussion carried on between nodes with tasks to be executed and nodes that may be able to execute tasks In this paper, we present the coordination method among distributed intrusion detection system and firewall by the contract net protocol. The method enhances the intrusion detection performance and provides the communication methods. To mode IDS and firewall, security models hue been hierarchically constructed based on the DEVS (Discrete Event system Specification) formalism. Each ID agent cooperates through the contract net protocol for detecting intrusions. The IDS which detects the intrusion informs to firewall, so the harmful network traffic is blocked. If an agent detects infusions, the agent transfers attacker's information to a firewall. Using this mechanism attacker's packets detected by In can be prevented from damaging the network.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.9
• /
• pp.362-369
• /
• 2012

Recently, a target detection based on a distributed sensor network has been much studied in active sonar. Zhou et al. proposed a target localization method using line fitting based on a distributed sensor network which consists of low complexity sensors that only report binary detection results. This method has three advantages relative to ML estimator. First, there is no need to estimate propagation model parameters. Second, the computation is simple. Third, it only use sensors with "detection", which implies less data to be collected by data processing center. However, this method has larger target localization error than the ML estimator. In this paper, a target localization method which modifies Zhou's method is proposed for reducing the localization error. The modified method shows the performance improvement that the target localization error is reduced by 40.7% to Zhou's method in the point of RMSE.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.4
• /
• pp.139-147
• /
• 2009

In this paper, optimum design of energy-aware distributed detection is considered for a parallel sensor network system consisting of a fusion center and two passive sonar nodes. AND rule and OR rule are employed as the fusion rules of the sensor network. For the fusion rules, it is shown that a threshold rule of each sensor node has uniformly most powerful properties. Optimum threshold for each sensor is investigated that maximizes the probability of detection under a constraint on energy consumption due to false alarms. It is also investigated through numerical experiments how signal strength, an energy constraint, and the distance between two sensor nodes affect the system detection performances.

• Proceedings of the KIEE Conference
• /
• 2001.05a
• /
• pp.42-45
• /
• 2001

The islanding protection for distributed resources (DR) becomes an important and emerging issue in power system protection since the distributed generator installations are rapidly increasing and most of the installed systems are interconnected with distribution network. In order to avoid the negative impacts on distributed network resulting from islanding operations of DR, it is necessary to detect the loss of mains (LOM) effectively. This paper presents a new LOM detection method using the rate of change in total harmonic distortion (THD) of current. The proposed method effectively detects LOM of the DR unit operating in parallel with the distribution network. We also verified the efficiency of the proposed method using the radial distribution network of IEEE 34 bus model.

• The Transactions of the Korean Institute of Electrical Engineers C
• /
• v.50 no.7
• /
• pp.346-350
• /
• 2001

We present a surface-micromachined polysilicon capacitive accelerometer using unevenly distributed comb electrodes. The unique features of the accelerometer include a perforated proof-mass and the inner and outer comb electrodes with uneven electrode gaps. The perforated proof-mass reduces stiction between the structure and the substrate and the unevenly distributed electrodes shorten the electrode length required for a given sensitivity. The polysilicon accelerometer has been fabricated by the conventional 6-mask surface-micromachining process and showes a sensitivity of 1.03mV/g with a hybrid detection circuitry.

• Journal of Institute of Control, Robotics and Systems
• /
• v.14 no.10
• /
• pp.983-990
• /
• 2008

This paper presents the fault detection and diagnosis(FDD) algorithm to enhance reliability of a longitudinal controller for an autonomous All-Terrain Vehicle(ATV). The FDD is designed to monitor and identify faults which may occur in distributed hardware used for longitudinal control, e.g., DSPs, CAN, sensors, and actuators. The proposed FDD is an integrated approach of decentralized and centralized FDD. While the former is processed in a DSP and suitable to detect faults in a single hardware, it is sensitive to noise and disturbance. On the other hand, the latter is performed via communication and it detects and diagnoses faults through analyzing concurrent performances of multiple hardware modules, but it is limited to isolate faults specifically in terms of components in the single hardware. To compensate for disadvantages of each FDD approach, two layered structure including both decentralized and centralized FDD is proposed and it allows us to make more robust fault detection and more specific fault isolation. The effectiveness of the proposed method will be validated experimentally.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.18 no.4
• /
• pp.118-123
• /
• 2004

This paper studies islanding detection of DG(distributed generation) considering fault location. Through the past studies, we found that islanding detection has been studied that DG disconnected when power islanding was detected by power state change and output change of DG. But, fault location was not considered. For example, fault in adjacent distribution line, fault on interconnection line fault, load shave by overload and normal operation were not considered. In this paper, We distinguish these considerations through power state change. Also, we proved islanding detection algorithm through PSCAD/EMTDC simulation.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.20 no.3
• /
• pp.247-254
• /
• 2015

Unintentional islanding results in safety hazards, power quality degradation, and many other issues. Thus, islanding detection of grid-connected distributed generation system is a key function for standards compliance. Many anti-islanding schemes are currently being studied; however, existing anti-islanding schemes used in inverters have power quality degradation and non-detection zone issues. Therefore, this paper analyzes existing anti-islanding schemes by using frequency drift in accordance with both islanding detection performance and power quality. This paper also proposes a new anti-islanding scheme by using frequency drift. Both simulation and experimental results show that the proposed scheme has negligible power quality degradation and no non-detection zones compared with other existing schemes.

• Journal of Internet Computing and Services
• /
• v.24 no.6
• /
• pp.1-11
• /
• 2023

Federated learning (FL) is a ground breaking machine learning paradigm that allow smultiple participants to collaboratively train models in a cloud environment, all while maintaining the privacy of their raw data. This approach is in valuable in applications involving sensitive or geographically distributed data. However, one of the challenges in FL is dealing with heterogeneous and non-independent and identically distributed (non-IID) data across participants, which can result in suboptimal model performance compared to traditionalmachine learning methods. To tackle this, we introduce FedGCD, a novel FL algorithm that employs Graph Neural Network (GNN)-based community detection to enhance model convergence in federated settings. In our experiments, FedGCD consistently outperformed existing FL algorithms in various scenarios: for instance, in a non-IID environment, it achieved an accuracy of 0.9113, a precision of 0.8798,and an F1-Score of 0.8972. In a semi-IID setting, it demonstrated the highest accuracy at 0.9315 and an impressive F1-Score of 0.9312. We also introduce a new metric, nonIIDness, to quantitatively measure the degree of data heterogeneity. Our results indicate that FedGCD not only addresses the challenges of data heterogeneity and non-IIDness but also sets new benchmarks for FL algorithms. The community detection approach adopted in FedGCD has broader implications, suggesting that it could be adapted for other distributed machine learning scenarios, thereby improving model performance and convergence across a range of applications.