• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.05a
• /
• pp.431-432
• /
• 2018

MRC(기계독해)는 Passage, Question, Answel 로 이루어진 Dataset 으로 학습된 모델을 사용하여 요청한 Question 의 Answer 를 같이 주어진 Passage 내에서 찾아내는 것을 목적으로 한다. 최근 MRC 시스템의 성능 측정 지표로 활용되는 SQuAD Dataset 을 활용하여 RNN 의 한 분류인 match-LSTM과 R-NET 알고리즘의 성능을 비교 분석하고자 한다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.8 no.4
• /
• pp.297-306
• /
• 2007

Model reference control is control method such that overall response of a plant plus controller approaches that of a given reference model. The reference model provides desired trajectory the plant should follow. There are many kinds of control methods in MRC. However, this paper focuses on Model Reference Sliding Mode Control. The plant of these controls is an uncertain and linear system varying in time, of second order, and with SISO. In this paper, a design scheme of reference model is proposed for MRSMC. The scheme determines reference model based on the information on bounded control inputs, reference inputs and system parameters. It is used to choose a Fixed Reference Model in the process of controller design, to update Variable Reference Model when stepwise reference inputs change and to update Instant Reference Model at every sampling time. The simulation results show that the proposed method yields better control performance than the conventional MRC subject to the stepwise reference input when applied to the position control system for motor system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39A no.12
• /
• pp.777-779
• /
• 2014

In this paper, we consider IOTA-OFDM/OQAM system that improves the transmission power and rate efficiency of conventional OFDM system and propose DF relaying cooperative communication scheme based on the IOTA-OFDM/OQAM system. In the proposed scheme, the destination receives orthogonal signals from source and relay simultaneously and combines using MRC. We demonstrated that the proposed schemes get better BER performance than conventional system because of the diversity gain.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.4
• /
• pp.145-152
• /
• 2020

MRC (Machine reading comprehension) is the AI NLP task that predict the answer for user's query by understanding of the relevant document and which can be used in automated consult services such as chatbots. Recently, the BERT (Pre-training of Deep Bidirectional Transformers for Language Understanding) model, which shows high performance in various fields of natural language processing, have two phases. First phase is Pre-training the big data of each domain. And second phase is fine-tuning the model for solving each NLP tasks as a prediction. In this paper, we have made the Patent MRC dataset and shown that how to build the patent consultation training data for MRC task. And we propose the method to improve the performance of the MRC task using the Pre-trained Patent-BERT model by the patent consultation corpus and the language processing algorithm suitable for the machine learning of the patent counseling data. As a result of experiment, we show that the performance of the method proposed in this paper is improved to answer the patent counseling query.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.3
• /
• pp.355-362
• /
• 2003

In this paper, we analyzed the DS/CDMA System with Hybrid SC/MRC diversity, as a method to achieve a good performance by reducing the complexity of system, when PLL gain value was revised a phase gulf among branches to make perfect coherent receive signal. The channel is a multipath Nakagami-m fading channel, we consider user(K), processor gain(N), number of paths(Lp) in DS/CDMA system with hybrid SC/MRC diversity. We found that PLL gain values exist to make perfect coherent receive signal. We verified that DS/CDMA system with Hybrid SC/MRC diversity can improve performance when the proper PLL gain values is used in the system. As a result, we verified that the upper limit of PLL gain value is 5dB to make perfect coherent receive signal in the DS/CDMA system with hybrid SC/MRC diversity.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.5 no.4
• /
• pp.645-652
• /
• 2001

In this paper is adopted the One-Ray Rician channel model for AVHS's platoon to short-range IVC (Inter-Vehicle Communication System) and analyzed packet probability characteristics in 60㎓ millimeter wave with very powerful to MP(multipath-wave). Both Convolution coding and MRC diversity is adopted that Multimedia service is satisfied following user's desire increase in the next and analyzed packet probability characteristics of DS-CDMA/QPSK systems.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.2B
• /
• pp.176-183
• /
• 2011

In this paper, we propose the implementation and performance analysis of power line communication based on SISO/MIMO-OFDM which focuses on high speed data transmission in smart grid and future power line grid. We employ Zimmermann frequency model and Middleton Class A model as the multipath power line fading channel and impulse noise channel, respectively. In this paper, in order to improve the three-phase or single-phase PLC performance, we introduce a new MRC (called a&f-MRC) which effectively sums up multiple antenna diversity gain and multipath fading diversity gain. Via simulation, we prove the performance advantage over existing SISO/MIMO systems. In addition, we offer the tradeoff on system design through comparing with MRC, EGC and SC.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.335-339
• /
• 2021

기계독해(MRC)는 인공지능 알고리즘이 문서를 이해하고 질문에 대한 정답을 찾는 기술이다. MRC는 사전 학습 모델을 사용하여 높은 성능을 내고 있고, 일반 텍스트문서 뿐만 아니라 문서 내의 테이블(표)에서도 정답을 찾고자 하는 연구에 활발히 적용되고 있다. 본 연구에서는 기존의 사전학습 모델을 테이블 데이터에 활용하여 질의응답을 할 수 있는 방법을 제안한다. 더불어 테이블 데이터를 효율적으로 학습하기 위한 데이터 구성 방법을 소개한다. 사전학습 모델은 BERT[1]를 사용하여 테이블 정보를 인코딩하고 Masked Entity Recovery(MER) 방식을 사용한다. 테이블 질의응답 모델 학습을 위해 한국어 위키 문서에서 표와 연관 텍스트를 추출하여 사전학습을 진행하였고, 미세 조정은 샘플링한 테이블에 대한 질문-답변 데이터 약 7만건을 구성하여 진행하였다. 결과로 KorQuAD2.0 데이터셋의 테이블 관련 질문 데이터에서 EM 69.07, F1 78.34로 기존 연구보다 우수한 성능을 보였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.5
• /
• pp.970-976
• /
• 2004

In this paper, we analyze channel estimation algorithms in a RAKE receiver with MRC. There are 3 popular channel estimation algorithms, which are WMSA(Weighted Multi-Slot Averaging) algorithm, EGE(Equal Gain Estimation) algorithm, SSE(Symbol-to-Symbol Estimation) algorithm. We analyze asynchronous IMT-2000(3GPP) which employ 3 different channel estimation algorithms by HP-ADS(Advanced Design System) simulation tool. We used lakes fading channel model for the analysis. from simulation results, we could observe that the performance of WMSA algorithm is better than others in low Doppler effect(3Km/h). However, in the case of high Doppler effect(120km1h), the EGE algorithm is more efficient. In this case the simple estimator with EGE algorithm seems to be more useful.

• The KIPS Transactions:PartC
• /
• v.11C no.6 s.95
• /
• pp.835-842
• /
• 2004

In this paper, we have analyzed the MC-DS/CDMA system with input signal synchronized completely through adjustment of the gain in the PLL loop, by using the hybrid SC/MRC-(2/3) technique, which is said to one of the optimal diversity techniques under the multi-path fading environment, assuming that phase error is defined to the phase difference between the received signal from the multi-path and the reference signal in the PLL of the receiver. Also, assuming that the regarded radio channel model for the mobile communication is subject to the Nakagami-m fading channel, we have developed the expressions and performed the simulation under the consideration of various factor, in the MC/DS-CDMA system with the hybrid SC.MRC-(2/3) diversity method, such as the Nakagami fading index(m), $the\;number\;of\;paths\;(L_p),$ the number of hybrid SC.MRC-(2/3) $diversity\;branches\;(L,\;L_c),$ the number of users (K), the number of subcarriers (U), and the gain in the PLL loop. As a result of the simulation, it has been confirmed that the performance improvement of the system can be achieved by adjusting properly the PLL loop in order for the MC/DS-CDMA system with the hybrid SC/MRC-(2/3) diversity method to receive a fully synchronized signal. And the value of the gain in the PLL loop should exceed 7dB in order for the system to receive the signal with prefect synchronization, even though there might be a slight difference according to the values of the fading index and the spread processing gain of the subcarrier.