For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Mapping QR Decomposition on Parallel Computers: A Study Case for Radar Applications
Antonio d'ACIERNO Michele CECCARELLI Alfonso FARINA Alfredo PETROSINO Luca TIMMONERI
IEICE TRANSACTIONS on Communications
Publication Date: 1994/10/25
Print ISSN: 0916-8516
Type of Manuscript: PAPER
Category: Electronic and Radio Applications
Radar, sidelobe canceler, QR decomposition, parallel processing, systolic array, MIMD computers, SIMD computers,
Full Text: PDF(688KB)>>
The sidelobe canceler in radar systems is a highly computational demanding problem. It can be efficiently tackled by resorting to the QR decomposition mapped onto a systolic array processor. The paper reports several mapping strategies by using massive parallel computers available on the market. MIMD as well as SIMD machines have been used, specifically MEIKO Computing Surface, nCUBE2, Connection Machine CM-200, and MasPar MP-1. The achieved data throughput values have been measured for a number of operational situations of practical interest.