Disaster Monitoring using Grid Based Data Fusion Algorithms

Journal Title: INCAS BULLETIN - Year 2010, Vol 2, Issue 4

Abstract

This is a study of the application of Grid technology and high performance parallel computing to a candidate algorithm for jointly accomplishing data fusion from different sensors. This includes applications for both image analysis and/or data processing for simultaneously tracking multiple targets in real-time. The emphasis is on comparing the architectures of the serial and parallel algorithms, and characterizing the performance benefits achieved by the parallel algorithm with both on-ground and in-space hardware implementations. The improved performance levels achieved by the use of Grid technology (middleware) for Parallel Data Fusion are presented for the main metrics of interest in near real-time applications, namely latency, total computation load, and total sustainable throughput. The objective of this analysis is, therefore, to demonstrate an implementation of multi-sensor data fusion and/or multi-target tracking functions within an integrated multi-node portable HPC architecture based on emerging Grid technology. The key metrics to be determined in support of ongoing system analyses includes: required computational throughput in MFLOPS; latency between receipt of input data and resulting outputs; and scalability, processor utilization and memory requirements. Furthermore, the standard MPI functions are considered to be used for inter-node communications in order to promote code portability across multiple HPC computer platforms, both in space and on-ground.

Authors and Affiliations

Catalin NAE

Keywords

Related Articles

INCAS TRISONIC WIND TUNNEL

The 1.2 m x 1.2 m Trisonic Blowdown Wind Tunnel is the largest of the experimental facilities at the National Institute for Aerospace Research - I.N.C.A.S. "Elie Carafoli", Bucharest, Romania. The tunnel has been designe...

A Semi-Empirical Airborne Particle Erosion Model for Polyesteric Matrix Fiberglass Composites

The paper deals with the mathematical modeling of the airborne solid particle erosion rate of composite materials, in particular non-oriented fiberglass reinforced polyesteric matrices. Using the mathematical tool of non...

Static and Dynamic Design Methods for Hybrid Bolted Joints in a Satellite Structure

A detailed analytical design for hybrid (metal-composite) bolted joints in static and dynamic (random vibrations) analysis is presented in this paper. Due to the fact that a space structure must be lighter and stiffened,...

Dichotomic Structure of DAEs Solutions for the Aircraft Control

The paper has its roots in earlier studies focused on DAEs solutions, for the aircraft flight control and intends to be a synthesis of them. The main goal is to structure the solution for the control laws so as to derive...

Numerical Investigations of Dynamic Stall Control

In this paper we investigated numerically the dynamic stall phenomenon and the possibilities to control it, with application to vertical axis wind turbines (for urban users). The Phenomenon appear at low tip speed ratio...

Download PDF file
  • EP ID EP124291
  • DOI 10.13111/2066-8201.2010.2.4.19
  • Views 109
  • Downloads 0

How To Cite

Catalin NAE (2010). Disaster Monitoring using Grid Based Data Fusion Algorithms. INCAS BULLETIN, 2(4), 143-152. https://europub.co.uk/articles/-A-124291