Design of Channel Allocation Strategies for the GPRS System

博士 === 國立臺灣科技大學 === 資訊工程系 === 98 === As time goes and technologies evolve, multimedia transmission has gradually become one of the main topics in the contemporary and future research areas. Therefore, transmission technology has the trend to integrate voice, video, and data simultaneously in one sys...

Full description

Bibliographic Details
Main Authors: Yi-Chou Tsai, 蔡逸舟
Other Authors: Huei-Wen Ferng
Format: Others
Language:en_US
Published: 2010
Online Access:http://ndltd.ncl.edu.tw/handle/34560085363169758775
Description
Summary:博士 === 國立臺灣科技大學 === 資訊工程系 === 98 === As time goes and technologies evolve, multimedia transmission has gradually become one of the main topics in the contemporary and future research areas. Therefore, transmission technology has the trend to integrate voice, video, and data simultaneously in one system, but not offers voice service only. At present, GSM (Global System for Mobile Communication) wireless network systems that provide voice services and a small amount data services are most widely used in many nations/regions. The systems have not satisfied users' requirements. Consequently, GPRS (General Packet Radio Service) network system is evolved from GSM network infrastructure as a result of both GSM and GPRS shared channel resource. As for data transmission method, GSM employs fixed channel allocation while GPRS employs dynamic channel allocation. Hence, data transmission rate of GPRS is faster than GSM, and channel allocation is efficient for GPRS. With the traffic of data service increase, quality of service (QoS) of voice service is surely affected. Therefore, how to obtain balance point between voice and data services and how to efficiently manage channel resource are important issues. In this dissertation, we take a comprehensive approach for channel allocation schemes of GPRS compared with the past studying. The schemes employ the following techniques including buffering priority, service priority, threshold control, de-allocation, preemption strategies etc. We use Markov chain approach to build an analytic model. The performance metrics include blocking/forced-termination/dropping probabilities, and delay times. Analytical results are validated by simulation. In addition, we observe the impact of performance when changing parameters, e.g., buffer size, the values of threshold, data service time and data size, mobility rate etc. At last, better schemes suitable for the GPRS system are suggested after a thorough comparison on system performance.