NO-EMI

 

The NOEMI project ("Nulling obstructive Electromagnetic Interferers") investigated the merits of modern array signal processing algorithms in the context of electromagnetic interference rejection in radio interferometry, where the main source of RFI is communication signals. The aim of the project was to provide an in-depth study of such algorithms, ultimately leading to a field-test demonstrator for an application suite provided by ASTRON.

Start:	1 February 1998
End:	1 February 2003
Partners:	TU Delft and ASTRON
Sponsor:	STW

Introduction

The problem addressed by NOEMI is the rejection of electromagnetic interference from man-made sources that prevent radio astronomers from observing at certain frequency bands of interest. The main output of the existing telescope hardware is a sequence of correlation matrices, which are are stored for off-line spectral analysis and imaging. RF interference usually enters the antennas through the sidelobes of the main beam. It can be stronger or weaker than the system noise. An important property is that it has a certain spatial signature (or directivity), so that it does not average out in the correlation process. In this project, we are mostly interested in interference removal for the Westerbork radio telescope (WSRT).

The NOEMI project

Our research proposal addresses the question: How can a array signal processing techniques be used to mitigate interference in radio astronomy. From all possibilities of interference cancellation, we have focused on the following two schemes:

• Multichannel interference detection and excision. The interference is detected at short time scales (ms), and contaminated samples are removed from the averaging process. This works well if the interference is concentrated in frequency and time, as e.g. in the GSM system.
• Spatial filtering. This more ambitious scheme is also suitable for continuously present interference such as TV stations. After detection, we estimate the spatial signature of the interferer and project out that dimension or otherwise subtract the signal coming from that direction.

Objectives

• The selection of a viable interference rejection application scenario, including the specification of the complete system, the benchmarks and performance requirements.
• The characterization of the signals of interest and the sources of interference, and the recording and analysis of interfering signals as received by the telescope array.
• The selection of appropriate signal detection and rejection algorithms, based on modern high-resolution subspace techniques.
• Design and implementation of a small-scale demonstrator for RFI mitigation based on standard DSP solutions, as an add-on to WSRT.
• Conducting field tests, leading to the verification of the used methodologies and strategies, as well as the recommendation of preferred beamforming hierarchies with respect to real-time RFI processing.

Results

In summary, the results of the project have been:

• Scientific: introduction of array signal processing techniques to radio astronomy, new algorithms for interference rejection and array calibration; documented in 6 journal papers, 1 patent application, and 15 conference publications,
• Technological: we developed the NOEMI 8-channel data recorder, the NOEMI short-term correlator, a small-scale demonstration of temporal blanking and spatial filtering, a demonstration of spatial filtering using a reference antenna (the NOEMI THEA tile),
• Utilization:
  • application of our algorithms for array calibration and factor analysis to radar systems (utilized by Thales);
  • implementation of our spatial filtering algorithms in an operational system will be further explored by the management of Westerbork Radio Telescope;
  • our scientific work is the main input for the RFI mitigation workpackage of the design of LOFAR, a future radio telescope consisting of 13,000 antennas;
  • our algorithms have made an impact in the international scientific literature, and our PhD candidate has won the 2002 KIVI Telecommunication Prize for the best thesis research in telecommunications in the Netherlands.

Links

• Final report
• List of publications
• People involved in the project

Contact address

Mail:	prof.dr.ir. Alle-Jan van der Veen
	Delft University of Technology
	Fac. EWI/Electrical Engineering
	Mekelweg 4
	2628 CD Delft
	The Netherlands
Phone:	(+31 15) 2786240
Fax:	(+31 15) 2786190
E-mail:	allejan@cas.et.tudelft.nl