USER-PERCEIVED QoS PERFORMANCE STUDY OF WEB-BASED CONTENT MOBILE INTERNET APPLICATION

• Main Author: PENSISKA (NIM 23205001), FINNY
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/7923
• Institution: Institut Teknologi Bandung

Since the rapid growth of cellular phone and WWW industry, cellular service providers offer many content web-based service which adapted to fit the limited bandwidth, display, memory and processing power of small cellular devices. A reliable content internet mobile service and fitable the user-perceived QoS also become a must for service provider to exist in strict cellular industry competition nowadays. <br /> <br /> <br /> <br /> Industrial competition and costumer needed for high quality services become a big challenge for operator in order to provide QoS-based high quality service. Moreover cellular network have air-interface characteristic which high sensitivity of noise and interference. Hence user-perceived QoS estimation for mobile internet application become an important thing in cellular packet network planning. <br /> <br /> <br /> <br /> In this thesis, cellular packet network modeled as an ON/OFF traffic, then content service users are grouped into several group due to their channel condition. End performance felt by interactive data application users in packet cellular network can be measured with ECR (Equivalent Circuit Rate) method. This method also useful as addmission control by determining maximum tolerable user which can still be accomodated by netwotk to fit the standard QoS. <br /> <br /> <br /> <br /> <br /> The model which had obtained can be used for calculating network parameters' value, specifically mean page delay. Model tested by comparing the value which obtained from model and simulation. These comparing process done by visualization plots and statistical tests. The statistical tests showing that model’s mean page delay value equal with one of simulation, hence model can describing the mean page delay pattern very well. These model can be used as a first step in cellular network planning process. Furthermore model can be developed so that the cellular network physical aspects can be measured too. <br /> <br />