| Summary: | 碩士 === 佛光人文社會學院 === 資訊學研究所 === 92 === In this paper, we try to develop a real-time stock price broadcasting system based on multi-channel communication. In a broadcasting data delivery system, stock information provider (SIP) transmits stock price to their clients over the Internet. In any client/server information system, user real-time data is one of the most important factors to evaluate the system’s quality of service. In order to provide more reliable and added value services (i.e., more information must be transmitted), the SIP should be to have a high-bandwidth (i.e., high cost) communication channel. The SIP processes high quality of stock information system with high operating cost.
Instead of processing a high-bandwidth (cost) channel, SIP could use multiple low-bandwidth channels to reduce the operating cost. In such situation, we will partition the transaction data of stock market into data segments with different significance and priority. Here, a segment with lower priority or less significance means it can be ignored occasionally or delayed transmission. The financial news is an example of low priority segment.
In the past, many algorithms have been proposed for multi-channel data broadcasting. However, most of them are focus on the applications of mobile computing. In this paper, we are facing two challenges. One is to find an adaptive algorithm for data segments dispatching over multi-channel. The other is to design an effective scheduling algorithm for data segments transmission of stock price information.
In this paper, therefore, we try to build an adaptive dispatching algorithm and an efficient scheduling algorithm for the delivery of stock information through multi-channel, with low communication cost. The results of this paper can be used in any types of information dissemination where transmission efficiency is a major concern.
|
|---|
