• Proceedings of the IEEK Conference
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(3)
• /
• pp.173-176
• /
• 2001

This paper proposes training algorithms of neuro-fuzzy systems. First, we introduce a structure training algorithm, which produces the necessary number of hidden nodes from training data. From this algorithm, initial fuzzy rules are also obtained. Second, the parameter training algorithm using evolution strategy is introduced. In order to show their usefulness, we apply our neuro-fuzzy system to a nonlinear system identification problem. It was found from experiments that proposed training algorithms works well.

• Journal of Institute of Control, Robotics and Systems
• /
• 제5권3호
• /
• pp.281-287
• /
• 1999

In this paper, a design method based on evolutionary programming for feedforward neural networks which have a single hidden layer is presented. By using an evolutionary programming, the network parameters such as the network structure, weight, slope of sigmoid functions and bias of nodes can be acquired simultaneously. To check the effectiveness of the suggested method, two numerical examples are examined. The performance of the identified network is demonstrated.

• Journal of the Korean Graphic Arts Communication Society
• /
• 제14권2호
• /
• pp.1-20
• /
• 1996

we describes a method for realizing the color conversion from the tristimulus (X, Y, Z) values to the print ink signals (C, M, Y) by using neural networks. The realized nonlinear color conversion system consists of two hidden layers those have seventeen nodes. We determined the C, M, Y values of the input control signals to compensate the printer nonlinearity of real systems. Experimental results showed that the described method is useful and valid to realized the nonlinear color conversion.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 한국퍼지및지능시스템학회 2001년도 춘계학술대회 학술발표 논문집
• /
• pp.57-60
• /
• 2001

In this paper, we propose growing algorithm of wavelet neural network. It is growing algorithm that adds hidden nodes using wavelet frame which approximately supports orthogonality in wavelet neural network based on wavelet theory. The result of this processing can be reduced global error and progresses performance efficiency of wavelet neural network. We apply the proposed algorithm to approximation problem and evaluate effectiveness of proposed algorithm.

• ETRI Journal
• /
• 제16권1호
• /
• pp.35-43
• /
• 1994

In this paper, the robustness of the artificial neural networks to noise is demonstrated with a multilayer perceptron, and the reason of robustness is due to the statistical orthogonality among hidden nodes and its hierarchical information extraction capability. Also, the misclassification probability of a well-trained multilayer perceptron is derived without any linear approximations when the inputs are contaminated with random noises. The misclassification probability for a noisy pattern is shown to be a function of the input pattern, noise variances, the weight matrices, and the nonlinear transformations. The result is verified with a handwritten digit recognition problem, which shows better result than that using linear approximations.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 한국해양정보통신학회 2004년도 SMICS 2004 International Symposium on Maritime and Communication Sciences
• /
• pp.1-5
• /
• 2004

In this paper, we propose a novel approach for evolving the architecture of a multi-layer neural network. Our method uses combined ART1 algorithm and Max-Min neural network to self-generate nodes in the hidden layer. We have applied the. proposed method to the problem of recognizing ID number in student identity cards. Experimental results with a real database show that the proposed method has better performance than a conventional neural network.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.1042-1045
• /
• 2003

In this paper, we propose the speaker identification system that uses vowel that has speaker's characteristic. System is divided to speech feature extraction part and speaker identification part. Speech feature extraction part extracts speaker's feature. Voiced speech has the characteristic that divides speakers. For vowel extraction, formants are used in voiced speech through frequency analysis. Vowel-a that different formants is extracted in text. Pitch, formant, intensity, log area ratio, LP coefficients, cepstral coefficients are used by method to draw characteristic. The cpestral coefficients that show the best performance in speaker identification among several methods are used. Speaker identification part distinguishes speaker using Neural Network. 12 order cepstral coefficients are used learning input data. Neural Network's structure is MLP and learning algorithm is BP (Backpropagation). Hidden nodes and output nodes are incremented. The nodes in the incremental learning neural network are interconnected via weighted links and each node in a layer is generally connected to each node in the succeeding layer leaving the output node to provide output for the network. Though the vowel extract and incremental learning, the proposed system uses low learning data and reduces learning time and improves identification rate.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제5권1호
• /
• pp.76-82
• /
• 2005

Speech signal has various features of speakers. This feature is extracted from speech signal processing. The speaker is identified by the speaker identification system. In this paper, we propose the speaker identification system that uses the incremental learning based on neural network. Recorded speech signal through the microphone is blocked to the frame of 1024 speech samples. Energy is divided speech signal to voiced signal and unvoiced signal. The extracted 12 orders LPC cpestrum coefficients are used with input data for neural network. The speakers are identified with the speaker identification system using the neural network. The neural network has the structure of MLP which consists of 12 input nodes, 8 hidden nodes, and 4 output nodes. The number of output node means the identified speakers. The first output node is excited to the first speaker. Incremental learning begins when the new speaker is identified. Incremental learning is the learning algorithm that already learned weights are remembered and only the new weights that are created as adding new speaker are trained. It is learning algorithm that overcomes the fault of neural network. The neural network repeats the learning when the new speaker is entered to it. The architecture of neural network is extended with the number of speakers. Therefore, this system can learn without the restricted number of speakers.

• Journal of Communications and Networks
• /
• 제13권5호
• /
• pp.421-427
• /
• 2011

Problems encountered in IEEE 802.11 medium access control (MAC) design are interferences from neighboring or hidden nodes and collision from simultaneous transmissions within the same contention floors. This paper presents the selective decoding schemes in MAC protocol for multiple input multiple output ad-hoc networks. It is able to mitigate interferences by using a developed minimum mean-squared error technique. This interference mitigation combined with the maximum likelihood decoding schemes for the Alamouti coding enables the receiver to decode and differentiate the desired data streams from co-channel data streams. As a result, it allows a pair of simultaneous transmissions to the same or different nodes which yields the network utilization increase. Moreover, the presented three decoding schemes and time line operations are optimally selected corresponding to the transmission demand of neighboring nodes to avoid collision. The selection is determined by the number of request to send (RTS) packets and the type of clear to send packets. Both theoretical channel capacity and simulation results show that the proposed selective decoding scheme MAC protocol outperforms the mitigation interference using multiple antennas and the parallel RTS processing protocols for the cases of (1) single data stream and (2) two independent data streams which are simultaneously transmitted by two independent transmitters.

• Journal of the Institute of Convergence Signal Processing
• /
• 제4권4호
• /
• pp.30-39
• /
• 2003

In this paper, we carried out the performance evaluation of HM-Net(Hidden Markov Network) speech recognition system for Korean speech databases. We adopted to construct acoustic models using the HM-Nets modified by HMMs(Hidden Markov Models), which are widely used as the statistical modeling methods. HM-Nets are carried out the state splitting for contextual and temporal domain by PDT-SSS(Phonetic Decision Tree-based Successive State Splitting) algorithm, which is modified the original SSS algorithm. Especially it adopted the phonetic decision tree to effectively express the context information not appear in training speech data on contextual domain state splitting. In case of temporal domain state splitting, to effectively represent information of each phoneme maintenance in the state splitting is carried out, and then the optimal model network of triphone types are constructed by in the parameter. Speech recognition was performed using the one-pass Viterbi beam search algorithm with phone-pair/word-pair grammar for phoneme/word recognition, respectively and using the multi-pass search algorithm with n-gram language models for sentence recognition. The tree-structured lexicon was used in order to decrease the number of nodes by sharing the same prefixes among words. In this paper, the performance evaluation of HM-Net speech recognition system is carried out for various recognition conditions. Through the experiments, we verified that it has very superior recognition performance compared with the previous introduced recognition system.