• 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.

• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.2
• /
• pp.189-199
• /
• 1999

This paper presents an optimization algorithm for a stable Self Dynamic Neural Network(SDNN) using genetic algorithm. Optimized SDNN is applied to a problem of controlling nonlinear dynamical systems. SDNN is dynamic mapping and is better suited for dynamical systems than static forward neural network. The real-time implementation is very important, and thus the neuro controller also needs to be designed such that it converges with a relatively small number of training cycles. SDW has considerably fewer weights than DNN. Since there is no interlink among the hidden layer. The object of proposed algorithm is that the number of self dynamic neuron node and the gradient of activation functions are simultaneously optimized by genetic algorithms. To guarantee convergence, an analytic method based on the Lyapunov function is used to find a stable learning for the SDNN. The ability and effectiveness of identifying and controlling a nonlinear dynamic system using the proposed optimized SDNN considering stability is demonstrated by case studies.

• Journal of Biomedical Engineering Research
• /
• v.11 no.1
• /
• pp.157-162
• /
• 1990

In this study, a peripheral auditory model is used as a frequency feature extractor and a recurrent network which has recurrent links on input nodes is constructed in order to show the reliability of the recurrent network as a recognizer by executing recognition tests for 4 Korean place names and syllables. In the case of using the general learning rule, it is found that the weights are diverged for a long sequence because of the characteristics of the node function in the hidden and output layers. So, a refined weight compensation method is proposed and, using this method, it is possible to improve the system operation and to use long data. The recognition results are considerably good, even if time worping and endpoint detection are omitted and learning patterns and test patterns are made of average length of data. The recurrent network used in this study reflects well time information of temporal speech signal.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1994.10a
• /
• pp.790-795
• /
• 1994

This paper describes multiple fault diagnosis method in complex system with hierarchical structure. Complex system is divided into subsystem, item, component. For diagnosing this hierarchical complex system, it is necessary to implement special neural network. We introducd to Hierarchical Artificial Neural Network(HANN) for this purpose. HANN 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 trainined by backpropagation. UNIX IPC(Inter Process Communication) is used for implementing HANN with multitasking and message transfer between processes in SUN workstation. We tested HANN in reactor system.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1650-1653
• /
• 2002

In this paper, we propose the initial optimized structure of the Radial Basis function Networks that is simple in the part of the structure and fast converges more than neural networks with the analysis method using Time- Frequency Localization. We construct the hidden node with the Radial Basis functions their localization are similar with approximation target function in the plane of the Time and Frequency. We finally make a good decision of the initial structure for function approximation using genetic algorithm

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.6B
• /
• pp.875-887
• /
• 2010

The workgroup of IEEE 802 proposed the IEEE 802.16 standard, also known as WiMAX, to provide broadband wireless access (BWA). The standard specifies two operational modes, one is popular PMP mode, and the other is optional mesh mode. In the mesh mode, the network entry process-NetEntry is the pivotal procedure for mesh network topology formulation and thus, influences the accessibility of whole mesh network. Unfortunately, the NetEntry process suffers from the hidden neighbor problem, in which new neighborship emerges after a new node comes in and results in possible collisions. In this paper, we propose a new SN-protected NetEntry process to address the problem. Simulation results show that the new proposed NetEntry process is more stable compared with the standard-based NetEntry process.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.4
• /
• pp.419-427
• /
• 1999

We propose an algorithm for obtaining the optimal node number of hidden units in dynamic neural networks. The dynamic nerual networks comprise of dynamic neural units and neural processor consisting of two dynamic neural units; one functioning as an excitatory neuron and the other as an inhibitory neuron. Starting out with basic network structure to solve the problem of control, we find optimal neural structure by multiplication and combining dynamic neural unit. Numerical examples are presented for nonlinear systems. Those case studies showed that the proposed is useful is practical sense.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.17 no.7
• /
• pp.656-665
• /
• 1992

In this paper, the performance of the various multiple access techniques for the mobile computer network has been studiedi in the consideration of the charactersitics of the mobile cimmunication channel. In the case of the hidden node occurring. It could be seen that the performance of the code division multiple access (CDMA) technique with simultaneous access function is better than that of the other packet access methods such as carrier sendsed multiple access (CDMA), busy tone multiple access (BTMA) and idle signal multiple access (ISMA) in the view of the throughput and mean delay time. Also, it has been shown that the performance of the CDMA method is superior to that of other packet access techniques such as multiple access (CSMA), etc. when the fading effect or impulsive noise exists in the mobile channel, Especially, in the case of the distributed mobile network it has been shown that the receivertransmitter based CDMA method using the characteristics of CDMA effectively has better throughput and less mean delay time than the commontransmitter based CDMA technique.

• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.1
• /
• pp.29-35
• /
• 2024

TCP is considered as one of the major candidate transport protocols even for constrained IoT networks..In our previous work, we investigated the congestion control mechanism of the uIP TCP. Since the uIP TCP sets the window size to one segment by default, managing the retransmission timer is the primary approach to congestion control. However, the original uIP TCP sets the retransmission timer based on the fixed RTO, it performs poorly when a radio duty cycling mechanism is enabled and the hidden terminal problem is severe. In our previous work, we proposed a TCP retransmission timer adjustment scheme for uIP TCP which adopts the notion of weak RTT estimation of CoCoA, exponential backoffs with variable limits, and dithering. Although our previous work showed that the proposed retransmission timer adjustment scheme can improve performance, we observe that the scheme often causes a node to set the retransmission timer for an excessively too long time period. In this work, we show that slightly modifying the dithering mechanism of the previous scheme is effective for improving TCP fairness.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.3
• /
• pp.459-466
• /
• 2006

Numerical values of thermodynamic properties such as temperature, pressure, dryness, volume, enthalpy and entropy are required in numerical analysis on evaluating the thermal performance. But the steam table itself cannot be used without modelling. From this point of view the neural network with function approximation characteristics can be an alternative. the multi-layer neural networks were made for saturated vapor region and superheated vapor region separately. For saturated vapor region the neural network consists of one input layer with 1 node, two hidden layers with 10 and 20 nodes each and one output layer with 7 nodes. For superheated vapor region it consists of one input layer with 2 nodes, two hidden layers with 15 and 25 nodes each and one output layer with 3 nodes. The proposed model gives very successful results with ${\pm}0.005%$ of percentage error for temperature, enthalpy and entropy and ${\pm}0.025%$ for pressure and specific volume. From these successful results, it is confirmed that the neural networks could be powerful method in function approximation of the steam table.