• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.2
• /
• pp.35-44
• /
• 1998

This paper describes multiple fault diagnosis method in complex system with hierarchical structure. Complex system is divided into subsystem, item and component. For diagnosing this hierarchical complex system, it is necessary to implement special neural network. We introduced Modular Artificial Neural Network(MANN) for this purpose. MANN consists of four level neural network, first level for symptom classification, second level for item fault diagnosis, third level for component symptom classification, forth level for component fault diagnosis. Each network is multi layer perceptron with 7 inputs, 30 hidden node and 7 outputs trained by backpropagation. UNIX IPC(Inter Process Communication) is used for implementing MANN with multitasking and message transfer between processes in SUN workstation. We tested MANN in reactor system.

• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2005.11a
• /
• pp.147-151
• /
• 2005

CAN(Controller Area Network) is a simple and efficient network system for real time control and measurement. As it is not only good at error detection but also strong in electromagnetic interference, CAN has been widely used all over the industries. Basically, CAN needs a master node in charge of sensor data collection, node scheduling for data transmission to a monitoring system and error detection. According to the number of mater nodes, the CAN system is classified into two type of master system. One is a single master system that has only one master node and the other is a multi-master system where any sensor node can become a master node depending on the system's conditions. While it has the advantage of its fault tolerance, the multi-master system will suffer form the overall performance degradation when a defect is found in the master node. It is because all sensor nodes pertaining to a defective master node lose their position. Moreover, it is difficult and expensive to implement. For a single master system, the whole system will be broken down when a problem happens to a single master. In this paper, a dynamic master system is presented that there are several sub-master nodes of which basic functions are those of other sensor nodes at ordinary times but dynamically changed to replace the failing master node. An effective scheduling algorithm is also proposed to choose an appropriate node among sub-master nodes, where each sub-master node has its precedence value. The performance of the dynamic master system is experimented and analyzed.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.10 no.1
• /
• pp.69-74
• /
• 2010

Now the CAN(controller Area Network) is using electronic modules as a serial communication which is very robust to noise. Especially the CAN is using for automotive part that very popular in which automotive electronic control module, engine controller unit, sensor modules, etc. but the CAN has the order of priority to linking node and also has fault confinement so using in these features that is applied to in factory automation product line. The CAN communication is basically very robust to electric noise so varied applying to others part. In this paper, we suggest to CAN interface for embedded system that is possible for error detection using two CAN nodes on Hi-speed, full-CAN.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.2E
• /
• pp.43-49
• /
• 1995

In this paper a space search technique is used to detect and locate the faults of the circuit interconnections. The circuit interconnections are represented by the tree structure and the tree space is searched to detect and locate the open faults of the circuit interconnections. The breadth search is used to detect the open faults and reduce the space size. The depth search is used to locate the open faults.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.34C no.8
• /
• pp.39-48
• /
• 1997

The fast redundancy identification is required to perform an efficient test pattern genration. Due to the reconvergent fanouts which make the dependency among objectives and the fault propagation blocking, there may exist redundnat faults in the cirucit. This paper presents the isomorphism identification and the pseudo dominator algorithms which are useful to identify redundant faults in combinational circuits. The isomorphism identification algorithm determines whether mandatory objectives required for fault detection cannot be simultaneously satisfied from primary input assignments or not using binary decision diagrma. The pseudo dominator algorithm determines whether faults propagation is possible or not by considering all paths at a given fanout node. Several experiments using ISCAS 85 benchmark circuits demonstrate the efficiency and practicability of the algorithms.

• Journal of Communications and Networks
• /
• v.16 no.6
• /
• pp.667-676
• /
• 2014

Connectivity is a crucial quality of service measure in wireless sensor networks. However, the network is always at risk of being split into several disconnected components owing to the sensor failures caused by various factors. To handle the connectivity problem, this paper introduces an in-advance mechanism to prevent network partitioning in the initial deployment phase. The approach is implemented in a distributed manner, and every node only needs to know local information of its 1-hop neighbors, which makes the approach scalable to large networks. The goal of the proposed mechanism is twofold. First, critical nodes are locally detected by the critical node detection (CND) algorithm based on the concept of maximal simplicial complex, and backups are arranged to tolerate their failures. Second, under a greedy rule, topological holes within the maximal simplicial complex as another potential risk to the network connectivity are patched step by step. Finally, we demonstrate the effectiveness of the proposed algorithm through simulation experiments.

• The KIPS Transactions:PartA
• /
• v.16A no.1
• /
• pp.17-26
• /
• 2009

Large-scale wireless sensor network are expected to play an increasingly important role for the data collection in harmful area. However, the physical fragility of sensor node makes reliable routing in harmful area a challenging problem. Since several sensor nodes in harmful area could be damaged all at once, the network should have the availability to recover routing from node failures in large area. Many routing protocols take accounts of failure recovery of single node but it is very hard these protocols to recover routing from large scale failures. In this paper, we propose a routing protocol, which we refer to as LSFA, to recover network fast from failures in large area. LSFA detects the failure by counting the packet loss from parent node and in case of failure detection LSFAdecreases the routing interval to notify the failure to the neighbor nodes. Our experimental results indicate clearly that LSFA could recover large area failures fast with less packets than previous protocols.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.05a
• /
• pp.385-388
• /
• 2006

비선형 주성분 분석은 기존에 널리 알려져 있는 주성분 분석기법과 유사한 다변수 데이터 분석을 위한 새로운 접근 방법이다. 비선형 주성분 분석은 AANN(Auto Associative Neural Network)으로 PCA와 마찬가지로 변수들 간에 존재하는 상관관계를 제거함으로써 고차의 다변수 데이터를 정보의 손실을 최소화하면서 최소 차원의 데이터로 변환하는 기법이다. AANN 기반 센서노드 고장검출 기법을 실제 센서 네트워크에 적용하여 봄으로써 센서 드리프트 등과 같은 센서 고장의 검출 및 유효한 센서 보정 성능을 확인하였다.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.42 no.12
• /
• pp.91-102
• /
• 2005

Recently, concurrent error detection for the processor becomes important. But it imposes too much overhead to adopt concurrent error detection capability on the system. In this paper, a new approach to resolve the problems of concurrent error detection is proposed. A loop detection scheme is introduced to reduce the repetitive loop iteration and memory access. To reduce the memory overheat an offset to calculate the target address of branching node is proposed. Performance evaluation shows that the new architecture has lower memory overhead and frequency of memory access than previous works. In addition, the new architecture provides the same error coverage and requires nearly constant memory size regardless of the size of the application program. Consequently, the proposed architecture can be used as an cost effective method to detect control flow errors in the commercial on the shelf products.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.39 no.2
• /
• pp.129-137
• /
• 2016

To identify the cause of the error and maintain the health of system, an administrator usually analyzes event log data since it contains useful information to infer the cause of the error. However, because today's systems are huge and complex, it is almost impossible for administrators to manually analyze event log files to identify the cause of an error. In particular, as OpenStack, which is being widely used as cloud management system, operates with various service modules being linked to multiple servers, it is hard to access each node and analyze event log messages for each service module in the case of an error. For this, in this paper, we propose a novel message-based log analysis method that enables the administrator to find the cause of an error quickly. Specifically, the proposed method 1) consolidates event log data generated from system level and application service level, 2) clusters the consolidated data based on messages, and 3) analyzes interrelations among message groups in order to promptly identify the cause of a system error. This study has great significance in the following three aspects. First, the root cause of the error can be identified by collecting event logs of both system level and application service level and analyzing interrelations among the logs. Second, administrators do not need to classify messages for training since unsupervised learning of event log messages is applied. Third, using Dynamic Time Warping, an algorithm for measuring similarity of dynamic patterns over time increases accuracy of analysis on patterns generated from distributed system in which time synchronization is not exactly consistent.