HPC-Cloud-based monitoring of crowds
- /
- POC
- /
- HPC-Cloud-based monitoring of crowds
DFRC is a center for geospatial data fusion and analysis, operating LBASense that measures crowd behavior. The system requires “fingerprinting” of cellular network radio signals to calculate mobile phone location. DFRC plans to expedite the costly mapping process by utilizing Cloud-based-HPC simulations.
Start date: 01/07/2013
Duration in months: 18
Problem Description
LBASense uses Amazon EC2 cloud infrastructure for cellular network radio signals to calculate mobile phone location. But the most expensive part is the process of taking measurements at specific points. The aim of the project is to find Cloud-based-HPC simulations to speed up this expensive mapping process.
Goals
Reduce time to solution
Challenges
The challenge of this experiment was to improve the process of “fingerprinting” a city by offering a near real-time simulation to determine the best locations at which to take measurements. To respond to this challenge, the mapping algorithm, would need to be ported to a Cloud-based HPC system
Innovation results
Existing simulation tools have been ported to enable them to run on an HPC system with many processors. The outcome of the experiment has been a high-performance combined simulation tool that reduces the deployment time of LBASense.
Business impact
Simulation cost reduction by 250k€. More efficient service makes DFRC more competitive expecting an increment in installation from 20 to 200 in 4 years. This would mean a cost saving of 8M€. HLRS benefits from the experiment through an increased knowledge about commercially relevant scenarios..