스마트 빌딩을 위한 가시광 통신 시스템의 성능 향상 기법

서성일,Seo, Sung-Il 한국인터넷방송통신학회 2020 한국인터넷방송통신학회 논문지 Vol.20 No.5

본 논문은 스마트 빌딩을 위한 가시광 통신 시스템 (VLC, Visible Light Communication)에서 통신의 안정성 및 데이터의 신뢰성을 향상시키는 다중 채널 간섭 제거 알고리즘에 대해 연구하였다. 가시광 통신 시스템은 LED (Light Emitting Diode)를 통해 방사되는 가시광을 이용한 차세대 통신 시스템으로 스마트 빌딩을 위한 유비쿼터스 네트워크 서비스 환경에서 에너지 절감 효과를 가져 올 수 있다. 또한 기존 인프라를 활용하여 고출력 전송이 가능하며 유지 보수비용을 절감할 수 있다. 실내에서 다수의 사용자가 통신을 할 경우, 채널 간섭으로 인해 성능의 열화가 발생하게 된다. 간섭 제거를 위해 먼저 MMSE (Minimum Mean Square Error) 기법을 통해 간섭을 제거한 후, 순차 연속 간섭 제거(SIC, Successive Interference Cancellation) 알고리즘을 이용하여 추가적으로 주변의 영향력이 큰 간섭원을 제거하여 성능을 향상시켰다. 채널 모델로서 실내 가시광 통신 모델을 적용 하였고, 비트 오류율 (BER, Bit Error Rate)을 평가하여 성능을 검증하였다. 모의실험을 통해 제안된 시스템의 성능을 검증하였고 기존의 시스템보다 향상된 결과가 나타냈다. 제안한 시스템 모델은 가시광 통신뿐만 아니라 일반적인 통신 시스템에서도 신호의 품질을 향상시킬 수 있도록 다양하게 적용이 가능하다. In this paper, we propose the multi-channel interference cancellation algorithm for visible light communication (VLC) in smart building. The VLC system is communication technology using visible rays that come out in Light Emitting Diode (LED) device. It has energy curtailment effect and possible to use in ubiquitous network service applications. When a large number of users communicate indoors, the performance can be reduced due to channel interference. To remove interference, at the first, the minimum mean square error (MMSE) scheme as interference cancellation methods used, and then the successive interference cancellation (SIC) is applied to obtain additional diversity gain and improve interference cancellation performance. Indoor VLC channel model is employed. The performance is evaluated in terms of bit error rate (BER). From the simulation results, it is confirmed that the proposed scheme has better BER performance compared to the previous systems. As a result, the proposed interference cancellation improves the signal quality of VLC systems by effectively removing the channel noise. The results of the paper can be applied to VLC for smart building and general communication systems.

딥러닝 기반 가시광 통신 시스템의 성능 향상 기법

서성일,Seo, Sung-Il 한국인터넷방송통신학회 2021 한국인터넷방송통신학회 논문지 Vol.21 No.4

본 논문은 스마트 빌딩을 위한 가시광 통신 시스템에서 데이터 신뢰성을 향상시키는 딥러닝 기반의 간섭 제거 알고리즘에 대해 연구하였다. 본 논문에서 제안한 기법은 딥러닝 기술을 적용하여 채널에서 발생하는 잡음을 예측하여 제거하는 기술로서 수신단에서 딥러닝에 의해 학습된 잡음들을 활용하여 효과적으로 잡음을 제거함으로써 신호의 품질을 향상시킬 수 있다. 딥러닝 기술의 잡음 예측 정확도를 향상시키기 위해 기존의 잡음 형태를 데이터베이스화하여 활용하였다. 모의실험을 통해 간섭 제거 기법이 적용된 시스템 모델의 성능을 검증하였으며, 제안하는 시스템이 잡음을 효과적으로 제거하여 신호의 품질 성능을 향상시킬 수 있음을 확인하였다. 제안한 시스템 모델은 가시광 통신뿐만 아니라 일반적인 통신 시스템에서도 신호의 품질을 향상시킬 수 있도록 다양하게 적용이 가능하다. In this paper, we propose the deep learning based interference cancellation scheme algorithm for visible light communication (VLC) systems in smart building. The proposed scheme estimates the channel noise information by applying a deep learning model. Then, the estimated channel noise is updated in database. In the modulator, the channel noise which reduces the VLC performance is effectively removed through interference cancellation technique. The performance is evaluated in terms of bit error rate (BER). From the simulation results, it is confirmed that the proposed scheme has better BER performance. Consequently, the proposed interference cancellation with deep learning improves the signal quality of VLC systems by effectively removing the channel noise. The results of the paper can be applied to VLC for smart building and general communication systems.

사물 인터넷 기기들의 이종 네트워크 간 간섭 완화 기법

서성일(Sung-Il Seo) 한국통신학회 2017 韓國通信學會論文誌 Vol.42 No.6

MIMO 시스템에서 낮은 복잡도를 지니는 QRD-M

서성일(Sung-Il Seo) 한국통신학회 2017 韓國通信學會論文誌 Vol.42 No.9

한국 전립샘암 데이터베이스를 활용한 근치적 전립샘절제술 후 생화학적 재발 예측

유성혜 ( Sung Hye Yu ),최문주 ( Mun Joo Choi ),이선정 ( Sun Jung Lee ),김청수 ( Choung Soo Kim ),서성일 ( Seong Il Seo ),변석수 ( Seok Soo Byun ),정창욱 ( Chang Wook Jeong ),박용현 ( Yong Hyun Park ),홍준혁 ( Jun Hyuk Hong ),최인 한국보건정보통계학회(구 한국보건통계학회) 2019 한국보건정보통계학회지 Vol.44 No.1

Objectives: As the surgical techniques are developed such as adopting robotic surgery etc. It has been reduced postoperative side effects such as urinary incontinence and sexual dysfunction etc., but also remain constant risk of recurrence after the treatment. It is very important to figure out the predictive factors, whichincreasing the probability of long-term survival, by Analyzing the progression of the disease according to biochemical recurrence (BCR). This study aims to figure out the influential predictive factors on BCR by using Korean Prostate Cancer Registry (KPCR) database, which constructedwith providing data from Korean manifold, because of therecurrence-free survival rates are reported to vary according to geographical location and race. Methods: KPCR database consists 7,394 patients’ data which collected from six medical institutions, and final analysis was conducted with 5,119 patients’ data which collected from December 2003 to December 2014. Statistical analysis was conducted with the multivariable Cox proportional hazard regression models to figure out the correlation of clinical factors which affect BCR. The biochemical recurrence free survival rate (BCR-FS) was calculated using with Kaplan-Meier method. Results: The 5 year BCR occurred in 23.6% (1,209 patients). As a result of the recurrence rate, prostate specific antigen (PSA) value ≥20.1 ng/mL was 49.3%, the clinical T3 stage was 40.4%, Gleason score ≥8 was 54.4%, Gleason score 7 was 21.8%. The mean duration of BCR was 20.7 months and the strongest predictor of BCR was Gleason scoreand PSA value ≥20.1 ng/mL. Conclusions: It has been verified that, Gleason scores and PSA value are the most influential factors for biochemical recurrence after radical prostatectomy. These results will be helpful fordetermining the treatment direction in postoperative follow-up and able to contribute to improving survival outcome in the future.

다기관 한국 전립샘 암 등록(KPCR) 데이터베이스를 이용한 근치적 전립샘 절제술의 수술 결과 비교

유성혜 ( Sung Hye Yu ),최문주 ( Mun Joo Choi ),김청수 ( Choung-soo Kim ),서성일 ( Seong Il Seo ),정창욱 ( Chang Wook Jeong ),변석수 ( Seok-soo Byun ),이지열 ( Ji Youl Lee ),홍준혁 ( Jun Hyuk Hong ),최인영 ( In Young Choi ) 한국보건정보통계학회(구 한국보건통계학회) 2018 보건정보통계학회지 Vol.43 No.3

목적: 전립샘암은 한국 남성에서 발병률이 급격히 증가하는 추세이다. 국소 전립샘암의 표준 치료로 근치적 전립샘 절제술이 시행되고 있으며 수술 기법은 개복, 복강경 그리고 로봇 수술로 나뉜다. 여러가지 수술 기법은 삶의 질, 경제적 측면, 수술 및 예후의 결과에서 차이가 있다. 이에 저자는 다기관 한국 전립샘 암 등록(KPCR) 데이터베이스를 이용하여 국소 전립샘 암 수술을 시행한 환자들의 결과와 예후에 대한 결과를 비교하고자 한다. 방법: KPCR 데이터베이스에는 5개 의료기관에서 수집한 6,032명의 데이터가 수집되어 있고 데이터의 무결성 보장을 위해 데이터 표준화 및 품질 관리를 시행하였다. KPCR 데이터베이스로부터 분석에 필요한 데이터를 추출하였다. 추출된 데이터는 2006년 1월부터 2011 년 12월까지 수집된 3,262명으로 최종 분석을 시행하였다. 이 데이터를 이용하여 수술 기법에 따른 국소 전립샘 암 환자의 인구학적 특성, 수술 결과 및 예후에 대해 분석하였다. 통계 분석방법은 one-way ANOVA, chi-square test, post-hoc 분석을 실시하였다. 무재발 생존율 분석을 위해 Kaplan-Meier와 multivariable Cox proportional hazard regression model을 통해 분석을 실시하였다. 결과: 분석 결과, 초기 전립샘 특이 항원치의 평균 값은 10.30 ng/mL이었다. 총 환자의 56.7%는 임상 병기 T2를, 43.3%는 병기 T1을 가지고 있었고 환자의 64.4%는 글리슨 스코어가 7 이상이었다. 절제면 양성률의 결과는 개복수술이 33.6%, 복강경 수술이 27.0%, 로봇수술이 28.8%로 나타났다(p <0.001). 로봇수술은 수술 후 합병증 및 예상 출혈량 결과에서 우수한 결과를 나타냈다. 5년 생화학 무 재발 생존율은 전체 환자에서 77.0%의 결과를 보였고 세 수술 기법은 임상 결과에 있어서 차이가 없었다. 또한 Cox 회귀분석 결과 세 수술기법은 생존율에 영향을 미치지 않았다. 결론: 본 연구결과는 국소 전립샘 암 환자의 예후와 특성을 설명한다. KPCR 데이터베이스를 통한 본 결과는 환자와 의료진의 올바른 의사 결정에 도움을 줄 것이라는 것을 보여준다. Objectives: Prostate cancer, commonly is rapidly increasing the incidence of prostate cancer among Korean men. Primary treatment of localized prostate cancer involves radical prostatectomy, and surgery techniques are divided into the abdominal, laparoscopic, and robotic surgery. Different surgical techniques differ in quality of life, economic aspect, and outcome of surgery and prognosis. Therefore we compared the outcomes and prognosis of patients who underwent surgery for localized prostate cancer using the multicenter Korean Prostate Cancer Registry (KPCR) database. Methods: The KPCR database collected data of 6,032 patients from five medical institutions, and data standardization and quality management were performed to ensure data integrity. We extracted data necessary for analysis from the KPCR database. We analyzed data for 3,262 patients during the period from January 2006 to December 2011. The demographic characteristics, surgical outcomes, and prognosis of patients with localized prostate cancer were analyzed. All statistical analyses were performed using one-way ANOVA, chi-square test, and post-hoc analysis. We used the Kaplan-Meier method and multivariable Cox proportional hazard regression model for biochemical recurrence-free survival rate (BCR-FS) analysis. Results: The analysis results showed that the average initial prostate specific antigen value was 10.30 ng/mL. A total of 56.7% of the patients had clinical stage T2, 43.3% had stage T1, and 64.4% of the patients had a Gleason score of more than 7. The rate of positive surgical margin was radical retropubic prostatectomy, laparoscopic radical prostatectomy, robotic-assisted laparoscopic radical prostatectomy (RALP) as 33.6%, 27% and 28.8%, respectively (p<0.001). The RALP gave excellent results for postoperative complication, and estimated blood loss (EBL). The 5-year overall BCR-FS was 77.0%, and the three surgical techniques did not differ in clinical outcomes. Cox regression analysis showed that the surgical technique did not affect the biochemical recurrence rate. Conclusions: Our study describes the characteristics of patients with localized prostate cancer and their prognosis. These results indicate that patients and medical staff can rely on information from the KPCR database for decision-making.

Predictive Factors for Acute Urinary Retention After Transperineal Template-Guided Mapping Biopsy

Si Hyun Sung(성시현),Chung Un Lee(이충언),Jae Hoon Chung(정재훈),Wan Song(송완),Minyong Kang(강민용),Hyun Hwan Sung(성현환),Byong Chang Jeong(정병창),Seong Il Seo(서성일),Seong Soo Jeon(전성수),Hyun Moo Lee(이현무),Hwang Gyun Jeo 대한비뇨기종양학회 2021 대한비뇨기종양학회지 Vol.19 No.3

Purpose: We investigated the predictive factors for acute urinary retention (AUR) after transperineal template-guided mapping biopsy (TTMB). Materials and Methods: We retrospectively reviewed the records of 459 patients who had undergone TTMB between May 2017 and July 2020. Overall complications after TTMB were analyzed and categorized according to the Clavien-Dindo classification. Factors that were likely to affect AUR were analyzed using a logistic regression model. Results: Overall complications after TTMB were observed in 95 of the 459 patients (20.7%), of which AUR was the most commonly reported (17.4%, n=80), followed by hematuria (3.1%, n=14). Hematuria in one patient was categorized as Clavien-Dindo grade IIIa. All remaining complications were Clavien-Dindo grade I. In the multivariate regression model, age ≥65 (odds ratio, 2.44; 95% confidence interval [CI], 1.42–4.17; p=0.001), prostate volume ≥30 mL (odds ratio, 3.72; 95% CI, 1.19–11.62; p<0.02), and number of biopsy cores ≥30 (odds ratio, 2.89; 95% CI, 1.29–6.43; p=0.01) were identified as the predictors for AUR after TTMB. Conclusions: AUR is the most common complication after TTMB. Age ≥65 years, prostate volume ≥30 mL, and number of biopsy cores ≥30 were significant predictors of AUR following TTMB.

인간 피부섬유아세포에서 자외선 B 조사에 대한 Epigallocatechin Gallate 의 영향

서병욱(Byung Wook Seo),서성일(Sung Il Suh),백원기(Won Ki Baek),서민호(Min Ho Suh),김상표(Sang Pyo Kim),정태영(Tae young Jung),조재위(Jae We Cho),김병천(Byung Chun Kim),이규석(Kyu Suk Lee),류영욱(Young Wook Ryoo) 대한피부과학회 2001 대한피부과학회지 Vol.39 No.5

N/A Background : The main polyphenol components in green tee are (-)-epicatechin, (-)-epigallocatechin, (-)-epicatechin gallate and (-)-epigallocatechin gallate (EGCG). It is well known that flavonoids such as catechins can be protective against inflammatory and cancer and cardiovascular diseases. These protective effects are largely due to their inhibition of some enzymes and antioxidative activities by scavenging free radicals. Ultraviolet(UV) exposure of the skin, particulary UVB (290-320nm), causes adverse biological effects, including alterations in cutaneous immune cells, photoaging and photocarcinogenesis. Several studies have shown that EGCG afforded protection against UVB-induced inflammatory responses and photocarcinogenesis in murine models. Objective & Methods : In this study, we investigated the effects of EGCG on UVB irradiated human skin fibroblasts using viability test, thiobarbituric acid assay, propidium iodide(PI) stain, and western blot analyses and RT-PCR. Results : Cell survival curves after UVB irradiation showed dose dependent decrement pattern by trypan blue exclusion assay. Only 42% of dermal fibroblasts survived at 150 mJ/cm2 UVB irradiation. The damage was associated with cell membrane lipid peroxidation, as shown by accumulation malondialdehyde(MDA). By pre-cultivation with EGCG (50nmol), a significant preventive effect was noted on the increase in the absolute number of surviving cells(up to 81.5% of cells survived) and the levels of MDA markedly decreased, Morphological changes associated with apoptotic cell death were easily distinguished by PI stain. Bases on our finding, we investgated the regulation of p53, p21, bax, bc1-2, cyclin Dl, E, Cdk2, and PARP proteins by western blot analyses. The expression p53 protein was elevated by following UVB exposure which was inhibited by EGCG treatment. Using RT-PCR, the transcription of p53, fas and jun gene showed similar results which obtained by western blot analyses. Conclusion : EGCG, which have newly accepted as a potential UV protection properties, is effective membrane peroxidation inhibitor and prevent apoptotic changes when present in relevant concentration at the site of action beginning and during UVB irradiation. And the protective mechanism of EGCG against UVB-induced cell damage maybe, at least in part, related with p53, fas and jun pathway.

아날로그 주파수공용통신 시스템과 디지털 주파수공용통신 시스템간의 양립성 분석에 관한 연구

김은철(Eun Cheol Kim),서성일(Sung Il Seo),류충상(Chung Sang Ryu),오성백(Seong Taek Oh),김진영(Jin Young Kim) 한국통신학회 2007 한국통신학회 학술대회논문집 Vol.2007 No.11

단일기관에서 시행한 1,002예의 로봇보조복강경하 부분신장절제술의 주술기 및 종양학적 결과

김태진(Taejin Kim),정재훈(Jae Hoon Chung),성현환(Hyun Hwan Sung),전황균(Hwang Gyun Jeon),정병창(Byong Chang Jeong),전성수(Seong Soo Jeon),이현무(Hyun Moo Lee),강민용(Minyong Kang),서성일(Seong Il Seo) 대한비뇨기종양학회 2021 대한비뇨기종양학회지 Vol.19 No.3

Purpose: To analyze and compare the results of robotic partial nephrectomy (RPN) at a single center with the previous large-scale studies in terms of perioperative and oncological outcomes. Materials and Methods: We retrospectively evaluated 1,013 cases of RPN in our center database from December 2008 to August 2018. Total 11 cases were excluded in final analysis. We evaluated perioperative outcomes as the Trifecta achievement, which is defined as no positive surgical margin (PSM), no perioperative complications greater than Clavien-Dindo classification I and a warm ischemia time of <25 minutes. In addition, we analyzed pathological and oncological outcomes; recurrence, metastasis, all-cause deaths, cancer-specific deaths, and 5-year survival rates. Results: In 1,002 cases, the Trifecta achievement was 61.1% (n=612). The postoperative complication was 18.4% (n=184) but most were grade 2 or less (14.9%, n=145). Ninety-three cases (9.28%) had benign and 907 cases (90.5%) had malignant pathologies. A local recurrence were 14 cases (1.54%) and distant metastasis were 20 cases (2.2%) during follow-up periods. All-cause death rate was 1.2% (n=11) and cancer-specific death rate was 0.2% (n=2). The median follow-up period was 39 months. A 5-year recurrence-free survival rate, cancer-specific survival rate, and overall survival rate were 95.2%, 99.7%, and 98.4%. Conclusions: In summary, our data shows comparable perioperative outcomes to other large-scale studies of RPN in terms of the Trifecta achievement with similar baseline characteristics. In terms of oncological outcomes, there was lower rate of PSM and similar recurrence free survival rate.