• The Journal of the Acoustical Society of Korea
• /
• v.13 no.1
• /
• pp.5-13
• /
• 1994

This paper presents a study on Neural Networks for Phoneme Recognition and performs the Phoneme Recognition using TDNN (Time Delay Neural Network). Also, this paper proposes training algorithm for speech recognition using neural nets that is a proper to large scale TDNN. Because Phoneme Recognition is indispensable for continuous speech recognition, this paper uses TDNN to get accurate recognition result of phonemes. And this paper proposes new training algorithm that can converge TDNN to an optimal state regardless of the number of phonemes to be recognized. The recognition experiment was performed with new training algorithm for TDNN that combines backpropagation and Cauchy algorithm using stochastic approach. The results of the recognition experiment for three phoneme classes for two speakers show the recognition rates of $98.1\%$. And this paper yielded that the proposed algorithm is an efficient method for higher performance recognition and more reduced convergence time than TDNN.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37B no.12
• /
• pp.1208-1218
• /
• 2012

Disseminating broadcast information stably and reliably in IEEE 802.11-like CSMA wireless networks requires that a source should seek collision-free transmission to multiple receivers and keep track of the reception state of the multiple receivers. We propose a simple yet efficient feedback scheme for stable reliable broadcast in wireless networks, called 2-polling feedback, where the state of two receivers are checked by a source before its broadcast transmission attempt We present a performance analysis of the class of reliable broadcast feedback schemes in terms of two performance metrics (packet transmission delay and packet stable time). The analysis results show that the proposed 2-polling feedback scheme outperforms the current existing classes of feedback schemes in the literature, i.e., all-polling feedback and 1-polling feedback. The 2-polling feedback scheme has lower asymptotic complexity than the all-polling feedback, and has the same asymptotic complexity as the 1-polling feedback but exhibits almost 50 % reduction in packet stable time.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.12B
• /
• pp.1236-1243
• /
• 2010

The perimeter defense system was created and its characteristics were evaluated. It was designed to utilize the fiber sensing device, namely OTDR(Optical Time Domain Reflectometer) which has been used for the maintenance of the optical communication network. An OTDR was constituted by a pulse laser with the nature of 1310nm, +15dBm for the observation of 400 meter optical fence. The high-speed 32-bit processor(S3C2440) has applied to MPU(Main Processor Unit) which helps to improve the performance of OTDR algorithms. Consequently, the maximum error was 0.84 meter on the performance test of the 10km monitoring and the pass criteria of ${\pm}1m$ satisfied in all the sections. The alarm delay time was under 3 sec after detecting the disorder. For the case of secondary trespassing after primary trespassing, the optical switch was installed in OTDR to monitor the secondary trespassing and to measure the multi-point detection. Therefore, this paper shows that the detections of secondary trespassing and multi-point is possible by means of optical switch.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.10 no.4
• /
• pp.71-77
• /
• 2011

In this paper, we propose a conversational quality measurement (CQM) system for providing the objective QoS of high quality mobile VoIP voice telecommunication. For measuring the conversational quality, the VoIP telecommunication system is implemented in two smart phones connected with VoIP. The VoIP telecommunication system consists of echo cancellation, noise reduction, speech encoding/decoding, packet generation with RTP (Real-Time Protocol), jitter buffer control and POS (Play-out Schedule) with LC (loss Concealment). The CQM system is connected to a microphone and a speaker of each smart phone. The voice signal of each speaker is recorded and used to measure CE (Conversational Efficiency), CS (Conversational Symmetry), PESQ (Perceptual Evaluation of Speech Quality) and CE-CS-PESQ correlation. We prove the CQM system by measuring CE, CS and PESQ under various SNR, delay and loss due to IP network environment.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.5
• /
• pp.44-56
• /
• 2009

Multimedia streaming service is susceptible to loss and delay of data as it requires high bandwidth and real time processing. Therefore QoS cannot be guaranteed due to data loss caused by heavy network traffic and error of wireless channel. To solve these problems, studies about algorithms which improve the quality of multimedia by serving differently according to the priority of packets in multimedia stream. Two algorithms are proposed in this paper. The first algorithm proposed is WMS-1(Wireless Multimedia Scheduling-1) algorithm which acts like IWFQ when any wireless loss is occurred but assigns channels first in case of urgent situation like when the running time of multimedia runs out. The second algorithm proposed is WMS-2(Wireless Multimedia Scheduling-2) algerithm that assigns priority to multimedia flow and schedules flow that has higher priority according to type of frame first. The comparison with other existing scheduling algorithms shows that multimedia service quality of the proposed algorithm is improved and the larger the queue size of base station is, the better total quality of service and fairness were gained.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.2B
• /
• pp.83-89
• /
• 2007

In this paper, we have proposed the power detection (PD)-based TDM-PON architecture, which is implemented with a hardware-based decentralized DBA (Dynamic Bandwidth Allocation) algorithm among ONUs, without OLT's intervention. Of particular importance in the proposed approach is that the hardware-based power detection scheme is used to manage the distributed MAC algorithm without any complex, time-consuming and costly software-processing elements for DBA in ONUs and OLT. We evaluate the performance of average packet end-to-end delay in a statistical analysis and numerical analysis. In addition, through simulations with various traffic models, we verified the superior performance of the proposed approach by comparing with the results of other E-PONs.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.9A
• /
• pp.844-851
• /
• 2010

Recently, AODV routing protocol which one of the table driven method for the purpose data transmission between nodes has been broadly used in mobile wireless sensor networks. An existing AODV has a little overhead of routing packets because of keeping the routing table for activity route and re-routes to recovery the routes in route discontinuation. However that has faults in that excesses useless of the network bandwidth to recovery the route and takes a lone time to recovery the route. This paper proposes an efficient route recovery method for AODV based on wireless sensor networks in connection breaks. The proposed method. The propose method controls the number of RREQ message considering the energy's node and distance between nodes to restrict the flooding range of RREQ message while expanding the range of local repair. In test results, the proposed method are compared to existing method, the number of drops decrease 15.43% and the delay time for re-route decrease 0.20sec.

• KIPS Transactions on Computer and Communication Systems
• /
• v.1 no.3
• /
• pp.219-228
• /
• 2012

As the wireless Internet become more widely accessible, variety of Internet services can be used without limitation of location. However, existing mobility management methods such as MIP and PMIP of IETF requires the load of heavy protocol stack on the mobile nodes or the addition of components such as LMA and MAG. In this paper, we proposed the location management technique in the PMIPv6 and Fast Handover technique. according to the moving pattern of the node, the location management technique proposed in order to adjust the paging area dynamically. The Fast Handover technique applied MIH technology and it reduced the handover signal processing time between heterogeneous network. The location management cost in the environment which the node moves in order to evaluate this and handover delay time was calculate. The proposal technique was efficiently more evaluated than PMIPv6 with the smallest 29% and maximum 83%.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.1
• /
• pp.1-11
• /
• 2001

This paper presents a learning-based approach for the construction of license Plate recognition system. The system consist of three modules. They are respectively, car detection module, license plate recognition module and recognition module. Car detection module detects a car in the given image sequence obtained from the camera with simple color-based approach. Segmentation module extracts the license plate in detect car image using neural network as filters for analyzing the color and texture properties of license plate. Recognition module then reads characters in detected license plate with support vector machine (SVM)-based characters recognizer. The system has been tested from parking lot and tollgate, etc. and have show the following performances on average: Car detect rate 100%, segmentation rate 97.5%, and character recognition rate about 97.2%. Overall system performances is 94.7% and processing time is one sec. Then our propose system does well using real world.

• The KIPS Transactions:PartC
• /
• v.18C no.6
• /
• pp.423-432
• /
• 2011

Recently, OpenFlow has been paid attention to as a fundamental technology which provides a function of virtualization and programmability in network. In Korea, deployment of OpenFlow networks in campuses and the interconnection between them through tunneling in layer 3 has been performed. However, the performance of the interconnected networks is decreased due to delay in IP layer. In this paper, we design and deploy nation-wide, not local, OpenFlow networks in a pure layer 2 environment over KREONET. After that, we do end-to-end Round-trip Time measurements and TCP/UDP performance tests in OpenFlow and normal networks, and do comparison and analysis on the test results. The results show that the nation-wide OpenFlow networks provide equal performance to normal networks except for the initial packet loss for UDP streaming. In regards to the performance decrease due to early UDP packet loss, we can mitigate it by implementing exceptional procedures in a controller which deal with the same continuous "Packet_in" events.