Earticle

차세대 무선 네트워크 환경에서는 다양한 이기종 무선 엑세스 기술들이 존재하여 서비스 되고 다양한 서비스를 하나의 단말로 지원하는 멸티 인터페이스 시스템의 출현이 예상되고 있다. 이러한 멀티 인터페이스 특성상 많은 연구들이 주로 전력제어 측면에서만 이루어지고 있다. 그러나 통합 무선네트워크 안에서 다양한 서비스가 원활하게 운용되기 위해서는 QoS를 고려한 이기종망간 네트워크접속에 대해서 연구가 필요하다. 본 논문에서는 전력제어 측 면 외에 QoS를 고려한 사전인증 및 네트워크 접속 메커니즘을 제안한다.

In next generation wireless network environment, various different wireless access technologies exist and multi-interface systems which can provide various services in one terminal are required. So far there have been many studies only on the power control part of multi-interface systems. Therefore, the study of network connection considering Qos between different networks is necessary in order for various services to be operated in integrate wireless networks. This paper proposes a pre-authentication and network connection mechanism rather than power control part.

Nowadays the distributed Micro Inverter system for PV (Photo Voltaic) generation is regarded as the one of the effective way for improving the power efficiency. The Micro Inverter system is a distributed system which is composed of PCSs (Power Conversion Systems) on each PV cells. In this paper, we've developed ZigBee network module and monitoring system of the Micro Inverter system for gathering PV generation and PCS conditions. The major packet loss of the wireless network due to the packet collision was reduced by 0.1% during 3,000 times data communication based on 100 ZigBee modules using our new packet protocol. Also we performed the Micro Inverter control and information analysis using our Solar Gateway and web based monitoring server.

In this paper, the research on the AC copper loss of Interior Permanent Magnet Synchronous Motor(IPMSM) for Electric Vehicle(EV) is carried out according to the number of conductors. For this research, the analysis method of AC copper loss is introduced, and the loss characteristics are analyzed by classifying the components of AC copper loss according to the number of conductors. The analysis is carried out from the low speed to high speed region(5,000rpm, 7,500rpm, 10,000rpm). The analysis model is EV traction IPMSM of 8 poles 72 slots. To verify reliability of the analysis, 2D-Finite Elements Analysis(2D-FEA) software tool is employed.

본 논문은 실종사건이 발생하였을 때 신속하게 실종자를 찾기 위해 실종자의 휴대전화가 꺼진 위치에서 수식 이 착륙드론(V-Tol)을 활용하여 실종 위치를 순찰하면서 실종자 얼굴을 Deep Learning 기술 중 GAN(Generative Adversarial Nets)기술을 활용하여 5G통신망을 이용 지상 시스템에서 학습 후 드론에게 XML코드를 전송해서 고해 상도의 카메라 기능을 통해 실종자의 위치를 더욱 빠르게 파악하기 위한 목적으로 연구하였다.

This study combines the drone and deep learning technology to improve the efficiency of missing people and to search for missing people, especially young girls, in Golden Time. In order to precisely search a small area, the rotor drone is more efficient, and the Fixed wings drone is more effective when searching a wider and wider area. The proposed technique is implemented and implemented. In the future, this technology is expected to be implemented and demonstrated faster when inter networking with existing 5G communication networks.

In this paper, using the water level of the reservoir and the water pressure on pipe networks, a decision making technology for estimating the virtual water pressure on arbitrary points in pipe netwrorks and managing the water distribution networks was proposed. We derived the regression equations between the water level in reservoir and the water pressure on pipe networks. The results of the actual application show that the event was detected and the correlation between the measured water pressure at the evaluation point and the estimated value of virtual water pressure related to the change in the water level of the distributing reservoir satisfied the evaluation criterion “5% or less” in all cases. In addition, the result of evaluating the opening and closing of the valve of water pipe show that all events were detected, and the relevant solutions according to that events were suggested by the proposed system