Online data reduction with a DSP-FPGA multiprocessor system

Author

Balzer, M. ; Stripf, H.

Author_Institution

Inst. for Data Process. & Electron., Forschungszentrum Karlsruhe, Germany

Volume

2

fYear

2002

fDate

2002

Firstpage

819

Abstract

For the detection of cracks and corrosion in pipelines an ultrasonic measuring system (pig) was developed. During the pipeline inspection there is no communication to the outside world, the accumulated data have to be stored on a mass storage device. The amount of data for a 250 kilometre run is in the range of about 700 TBytes at a rate of 2.8 GByte/s. This amount of data has to be filtered, reduced and compressed by a factor of 7000 to store it on a storage device. To overcome these difficulties a massive parallel multiprocessing system with ASICs and DSPs was developed. Each of the system modules has a computing power of more then 1200 MOPS. The complete computing power is installed in a volume the size of a shoe box.

Keywords

corrosion testing; crack detection; data reduction; digital signal processing chips; field programmable gate arrays; multiprocessing systems; parallel processing; ultrasonic materials testing; 2.8 GByte/s; 250 km; ASIC; DSP-FPGA multiprocessor system; corrosion detection; crack detection; data compression; mass storage device; massive parallel multiprocessing system; online data reduction; pig; pipeline inspection; system modules; ultrasonic measuring system; Corrosion; Data processing; Digital signal processing; Inspection; Multiprocessing systems; Petroleum; Pipelines; Pulse measurements; Testing; Ultrasonic variables measurement;

fLanguage

English

Publisher

ieee

Conference_Titel

Digital Signal Processing, 2002. DSP 2002. 2002 14th International Conference on

Print_ISBN

0-7803-7503-3

Type

conf

DOI

10.1109/ICDSP.2002.1028216

Filename

1028216