• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.10
• /
• pp.140-148
• /
• 1994

In this paepr, the modified neocognitron, which has capability of recognizing multi-patterns in input image in one pass, is proposed. It is the hierarchical neural network composed of S and V layer which are able to extract features and of C layer with little effect from deformation, changes in size, shifts in position. S and V cells extract the features of all patterns in input image by applying DCC(don't care condition) to those cells. S and C cells also have position informations of extracted patterns. Position information is used in extracting good features without extracted features beting interfered one another. The proposed method is shorter in recognition time than the selective attention method with backward connection, because of recognizing multi-patterns in one passe. The modified neocognitron can recognizze attached multi-patterns because of using DCC and position informations.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.33A no.6
• /
• pp.99-107
• /
• 1996

When GPS (global positioning system) is used as synchronous signal in CDMA digital cellular base station system and high speed digital synchronous communication network, antenna cable length is increased, comparing with other GPS application such as positioning or car navigation. In this paper, it is proposed that a type of new GPS antenna system including RF stage for reduction of cable loss in case of long cable.The antenna system with TMPA(truncated-corners microstrip patch antenna) is designed and fabricated because GPS signal has RHCP (right-hand circular polarization), consequently antenna size can be made small size. LNA (low noise amplifier) is designed by using HEMT(high electron mobility transistor)which has lower noise figurae and better AGC characteristics at low voltage than GaAs FET, and we equiped mixer, in GPS antenna unit, which converts from 1575.42MHz to 75.42MHz. As result of comparing between typical system and proposed system when cable length is 60m, 63dB, 55dB and 25dB gain are obtained for RG-316/U, RG-58C/U and RG-213/U, and better characteristics are achieved than typical system as far as cable length is longer.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.6
• /
• pp.88-98
• /
• 1998

Subspace classifier is a popular pattern recognition method based on Karhunen-Loeve transform. This classifier describes a high dimensional pattern by using a reduced dimensional subspace. Because of the loss of information induced by dimensionality reduction, however, a subspace classifier sometimes shows unsatisfactory recognition performance to the patterns having quite similar principal components each other. In this paper, we propose the use of multiple novelty neural network classifiers constructed on novelty vectors to adopt minor components usually ignored and present a method of improving recognition performance through combining those with the subspace classifier. We develop the proposed classifier on handwritten numeral database and analyze its properties. Our proposed classifier shows better recognition performance compared with other classifiers, though it requires more weight links.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.32A no.8
• /
• pp.31-38
• /
• 1995

In this paper, we consider the leaky bucket parameter dimensioning problem in the presence of the cell delay variation(CDV) which arises at the customer premises network dud to the multiplexing with other traffic streams. We consider an ATM multiplexer in which a single CBR stream and several heterogeneous VBR traffic streams are multiplexed. Choosing an MMPP model for the bursty traffic streams, we derive an (MMPP+DD)/D/1/K queueing model for the evaluation of the CDV experienced by the CBR stream. We first evaluate the equilibrium queue length distribution embedded at tagged-cell arrival-time instants, based on whcih we calcuate the inter-cell time distribution and the distribution kof the number of tagged-cell departures in an arbitrary interval. Then we apply the analysis to the dimensionging problem of the leaky bucket parameters, examining how the employed traffic model affects the determination of the bucket size. Through numerical examples, we confirm that the Poisson traffic model can underestimate the bucket size, thus causing a considerable blocking probability for compliant use cells while the MMPP model can optimally design the bucket size which keeps the blocking probability under the target value.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.34S no.4
• /
• pp.1-11
• /
• 1997

In this paper, modeling of the communication channel occupancy characteristics at the base station in the wide area radio trunking syste has been performed. Using the results, mathematical analysis for the switching capacity at the network switch has also been done. Specificatlly, we propose a G/M/m queueing model for the single service area modeling, and analyze changes in call blockingprobabilities accoridng to changes in the burstness characteristics of group paging calls. As a result, we have observe dthat the channel occupancy variations become larger as the burstness characteristics become more apparent, to make the call blocking probability higher. Next, based on the single service area analysis, we have anayzed mathematically the average switching capacity required to serve mobile subscribers for a wide area radio trunking sytem, where multiple switching port assignments are required when the people in the same group are distributed over several base stations. Accordingly, we have observed that the average and the variance of switching channel occupancy are closely related to the mobility characteristics of subscribers, and that we need a netowrk switch having bigger capacity as subscribers show wider distribution. Especially, with the call dropping probability within 0.2%, a switch having about 5 to 6 times bigger can be required, compared with the one when the mobility of subscribers is mostly restricte dto a single service area.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.1
• /
• pp.170-177
• /
• 1996

In this paper, we propose a neural network for extracting dashed lines from map images based on the feature extracting path be found in a cat and monkey's visual cortex. The proposed system converts a color image to a binary image and extracts the lines per orientation. Next, the system removes the lines longer than the dashed line and extracts the dashed lines by finding the consecutive short lines. The proposed system for extracting the dashed lines was tested using 1/25,000 scaled maps published by the National Geographic Institute of Korea. This paper shows that the proposed system can extract the dashed-lines from maps.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.1
• /
• pp.187-194
• /
• 1996

In this paper, we propose regression neural networks based on regression trees. We map regression trees into three layered feedforward networks. We put multi feature split functions in the first layer so that the networks have a better chance to get optimal partitions of input space. We suggest two supervised learning algorithms for the network training and test both in single feature split and multifeature split functions. In experiments, the proposed regression neural networks is proved to have the better learning performance than those of the single feature split regression trees and the single feature split regression networks. Furthermore, we shows that the proposed learning schemes have an effect to prune an over-grown tree without degrading the learning performance.

• Proceedings of the Korea Contents Association Conference
• /
• 2004.05a
• /
• pp.522-528
• /
• 2004

Mobile internet service based on CDMA network evolution and multifunction mobile terminal is popular in Korea. Mobile internet service is classified as the information service using browser and mobile multimedia service using download platform. Killer applications based on multifunction mobile terminals have been developed for new business in mobile internet services. One of killer applications is location based seurvice (LBS) using mobile location information. In this paper, we develop mobile map browser using WIPI platform that is the main function in LBS. Mobile map browser interworking with GIS server provides the map information. The developed mobile map browser can be working on PDA and telematics terminals when WIPI is porting on these devices.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.28B no.11
• /
• pp.852-858
• /
• 1991

An adaptive hierarchical classifier(AHCL) for Korean character recognition using a neural net is designed. This classifier has two neural nets: USACL (Unsupervised Adaptive Classifier) and SACL (Supervised Adaptive Classifier). USACL has the input layer and the output layer. The input layer and the output layer are fully connected. The nodes in the output layer are generated by the unsupervised and nearest neighbor learning rule during learning. SACL has the input layer, the hidden layer and the output layer. The input layer and the hidden layer arefully connected, and the hidden layer and the output layer are partially connected. The nodes in the SACL are generated by the supervised and nearest neighbor learning rule during learning. USACL has pre-attentive effect, which perform partial search instead of full search during SACL classification to enhance processing speed. The input of USACL and SACL is a directional edge feature with a directional receptive field. In order to test the performance of the AHCL, various multi-font printed Hangul characters are used in learning and testing, and its processing its speed and and classification rate are compared with the conventional LVQ(Learning Vector Quantizer) which has the nearest neighbor learning rule.

• Journal of the Korean Institute of Telematics and Electronics T
• /
• v.35T no.1
• /
• pp.105-110
• /
• 1998

When communication network is designed, total Quantity of lines is decided by expected capacity to be caculated based on busy hour traffic. But this method is wasted many a portion of communication networks because of caculation by busy hour traffic. In this paper, it is proposed the solution that applied CDMA to the wired lines and it is shown changing the volume of networks via BER. In example, when a channel of 13kbps voice was transmitted to El at 2% BER, we can get 120 channels and the higher BER makes the bigger channels. So, it is shown that communication networks is designed by BER instead of busy hour traffic.