Related Experiment Video
Updated: Jun 8, 2026
06:49
Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
Published on: December 11, 2015
Large-scale performance evaluation of e-homecare architectures using the WS-NS simulator
S Van Hoecke1, B Volckaert, B Dhoedt
1Department of Information Technology, Ghent University, Gaston Crommenlaan 8, Bus 201, 9050 Gent, Belgium. sofie.vanhoecke@intec.ugent.be
Methods of Information in Medicine
|October 7, 2010
Summary
Evaluating e-homecare network performance is challenging. A new simulator, WS-NS, enables accurate modeling, leading to optimized architectures with over 45% faster response times for better patient care.
Area of Science:
- Computer Science
- Telemedicine
- Network Simulation
Background:
- E-homecare offers convenient, cost-effective patient care but requires high reliability and low response times.
- Characterizing network load and usage behavior is crucial for time-critical e-homecare services.
- Building realistic large-scale testbeds for e-homecare evaluation is difficult.
Purpose of the Study:
- Design and evaluate the Network Simulator for Web Services (WS-NS).
- WS-NS accurately models service-oriented architectures for e-homecare performance evaluation.
- Assess network-aware vs. network-unaware service selection algorithms.
Main Methods:
- Applied WS-NS to the Coplintho e-homecare use case (diabetes, multiple sclerosis patients).
- Developed and tested network-unaware and network-aware service selection algorithms.
- Optimized the e-homecare broker design from centralized to hierarchical.
Main Results:
- Service selection based solely on properties can be suboptimal.
- Incorporating network link information improves service selection strategies.
- Optimized broker design significantly reduced average response times.
Conclusions:
- WS-NS effectively analyzes load and response times in large-scale e-homecare architectures.
- Coplintho project's e-homecare architecture optimization reduced network overhead.
- Optimized architecture achieved over 45% reduction in average response times.
