환자영상 전송을 위한 Proxy기반의 Transcoder

Abstract

Objective: The progress in computer and communication technologies is making the Internet increasingly heterogeneous in terms of network, hardware and software capacities. Moreover, this has made it possible for emergency telemedicine services to provide high quality medical services. However, resource availability on the Internet varies unexpectedly. Thus, providing an efficient access to emergency telemedicine services requires that medical multimedia streams be adapted according to the environment constraints. One approach to this issue is based on the use of intermediate nodes within the network to perform such adaptations (media transformations and data transcoding). For this purpose, we have designed and implemented a proxy server for Quality of Service adaptations of medical multimedia streams. Methods: We have organized a new emergency telemedicine system by designing a proxy server to execute transcoding. The proxy server is located between a patient system and a doctor system over heterogeneous networks. Before a patient system can deliver medical video streams to a doctor system, the proxy server measures uplink bandwidth which is one of the Quality of Service factors, from the proxy server to the doctor system. At this moment, frame rates are determined according to the measured bandwidth, and the proxy server transmits medical video streams modified for new frame rates to the doctor system. We describe the implementation of this proxy server on top of the Microsoft DirectShow? environment and report on a performance evaluation which demonstrates the effectiveness of the approach. Results: The quality of requested medical video streams can be predicted when they are adapted to the receiver. With this prediction, adapted medical video streams which meet the frame rates constraints of the receiver can be delivered without additional measurements of bandwidth. Conclusion: This study represents a proxy server of a hybrid multimedia telemedicine system over heterogeneous networks. We expect that the designed proxy server can provide not only dynamic Quality of Service monitoring functions along bandwidth measurement, but also medical video adaptations to the receiver in heterogeneous network environments.