Radio Frequency Monitoring for Radio Astronomy

Radio Frequency Monitoring for Radio Astronomy Purpose, Methods and Formats Albert-Jan Boonstra IUCAF RFI-Mitigation Workshop Bonn, March 28-30, 2001

Contents Monitoring goals in radio astronomy Operational radio telescopes Radio telescope design Spectrum characterization Spectrum characterization parameters Characteristic frequency scales Characteristic time scales Monitoring approaches Automatic parameter extraction Databases and formats Conclusions and plans 2

Monitoring goals in radio astronomy Operational radio telescopes Monitoring goals for a radio observatory: Scheduling purposes online / short term scheduling observation planning Detection and reporting out-of-band emissions internal - external RFI system adaptions (filtering, scheduling) due to changes in spectrum use Diagnostics 3

Monitoring goals in radio astronomy Radio telescope design Monitoring goals for radio telescope design: Overall spectral occupancy attainable bandwidth, frequency range, telescope location Specifically: Spectrum occupancy of strongest transmitters Linearity (intermods, noise increase, gain compression,...) Spectrum occupancy of moderately strong and weak transmitters Sensitivity (cw, broadband, spatial correlation,...) 4

Monitoring goals in radio astronomy Radio telescope design Telescope design areas affected by the spectral environment: analog (HTS) filtering LNA design (e.g. feed forward concept) receiver design (mixing scheme, RF/IF/video digitization) beamforming (phased arrays, RFI nulling) Pre correlation mitigation (spatial filtering, blanking, sidelobe canc.) Post correlation mitigation (spatial filtering, blanking, sidelobe canc.) We need an overall RFI mitigation system design 5

Monitoring goals in radio astronomy Radio telescope design Linearity: Sensitivity: 0 60 Power at LNA input (dbm) 20 40 60 80 100 120 140 160 180 largest RFI power at WSRT (TV) max. allowable RFI power for LNA input using the " 70dB" criterium intermodulation product level when RFI input power is 25 dbm system noise (40 K, 20 MHz bw) " 70 db" criterium for WSRT (calculated from RA769) Spectral power flux density db(wm 2 Hz 1 ) 80 100 120 140 160 180 200 220 240 260 RFI observed at the WSRT SKA RFI 2ms det. limit SKA, 300 km att. SKA, 6.4 km att. SKA, 36 m att. SKA continuum 200 " 90 db" criterium for SKA 280 10 8 10 9 10 10 10 11 10 12 10 13 frequency (Hz) 10 7 10 8 10 9 10 10 10 11 10 12 frequency (Hz) 6

Monitoring goals in radio astronomy Radio telescope design System design approach w.r.t. RFI mitigation astronomical requirements RFI monitoring data RFI mitigation techniques RFI propagation models telescope design? ff analysis: linearity, sensitivity available bandwidth???? system change proposals? 7

Spectrum characterization Scope: RFI monitoring gives an impression of how the situation currently is. We also need: Information from allocation and assignment tables Information from system descriptions Information on future trends (allocation and assignment tables) Propagation models 8

Spectrum characterization Characterization parameters High power RFI duty cycle RFI duty cycle in the time domain RFI spectral occupancy Time-frequency block size duty cycle at the second to one hour level Power flux density measurements Modulation characteristics and system information Polarization identification Location of the RFI source 9

Spectrum characterization Characteristic frequency scales Receiver (LNA) input frequency range, typically a few MHz to over 100 MHz IF frequency range, typical a few MHz to order 100 MHz Video band frequency range, typical a few MHz to about 20 MHz Spectral line resolution, order khz RFI mitigation based on RFI modulation characteristics: sub-khz resolution 10

Spectrum characterization Characteristic timescales Timescales Rationale 11 years Solar cycle 1 year Spectrum occupancy changes, long term trends 1 month Seasonal changes, propagation 1day Propagation 1 hour Changes in spectrum use over the day, weather influences 1 minute Relevant for duty cycle of observation integration process 1 second Relevant for duty cycle of observation integration process 1 milli s Relevant for duty cycle of observation integration process 1 micro s Less relevant for LOFAR monitoring 11

Monitoring approaches Parameter extraction Two approaches: Fill database with spectra with an observing duty cycle as high as possible Fill database with processed data with an observing duty cycle as high as possible Extract characterization parameters automatically from this database using (flexible) filters such as: one-hour average RFI duty cycle daily median RFI power 12

Monitoring approaches Parameter extraction time (hours) 50 40 30 20 10 Observed spectrogram time (hours) 600 610 620 630 640 650 660 670 680 690 700 Detected RFI 50 40 30 20 10 600 610 620 630 640 650 660 670 680 690 700 Fraction of the time that RFI is present 1 RFI fraction 0.5 0 600 610 620 630 640 650 660 670 680 690 700 frequency (MHz) 13

Monitoring approaches Parameter extraction RFI power (dbm) RFI power (dbm) 10 2 Observed maximum RFI power 10 3 10 4 10 5 600 610 620 630 640 650 660 670 680 690 700 Observed minimum RFI power 10 2 10 3 10 4 10 5 600 610 620 630 640 650 660 670 680 690 700 Observed power rms 10 3 power rms 10 4 10 5 10 6 600 610 620 630 640 650 660 670 680 690 700 frequency (MHz) 14

Monitoring approaches Database and formats - RMDF data format (1) Format developed in the robust receiver project (TMR-LSF RTD) by ASTRON, MPIfR, NRAL/JBO, and IRA-CNR. Purpose of the project: development of an RFI robust receiver system Purpose of the format: making easy exchance and inspection of RFI monitoring data possible (RFI monitoring is part of the project allowing semi-automatic RFI detection possible for use at radio observatories 15

Monitoring approaches Database and formats - RMDF data format (2) Format requirements: the format should allow automatic RFI detection and reporting (e.g. CRAF database reporting) the format should be self-contained: all relevant monitoring (a.o. calibration) data should be present the data should be calibrated header and data files (separate files) should be readable by standard text processing and mathematical packeges (e.g. Matlab) datasets should be rectangular, time increases with row number, frequency increqases with column number format should allow merging of different datasets with the same observational parameters the datasets should be complete (no holes) 16

Monitoring approaches Database and formats - RMDF data format (3) RMDF format - file header data format: RMDF version 1.0 station code: Wb station name: WSRT, ASTRON, The Netherlands scanning (y or n): y receiver: AR5000 antenna type: R&S HE202 antenna height (m): 25 antenna geographic coordinates: +525508.76-0063615.01 antenna location: WSRT construction hall tower antenna polarization: Horizontal, N-S dipole direction azimuth (dd:mm:ss) no elevation (dd:mm:ss) no level type: standard merge type: no number of frequency bins 100 lowest frequency (Hz): 600e6 highest frequency (Hz): 700e6 transition frequency (Hz): no number of spectra: 53 start date (UT): 1999-03-30 start time (UT): 06:16:07.173 duration of each scan (s): 3600 channel integration time (s): 18.000 channel bandwidth (Hz): 1e6 rfi signal unit: W ^ m (-2) rfi measurement accuracy (% or db): 3dB rfi id: Strongest RFI is from TV Smilde tower, channels 44 fri id: (655 MHz) and 47 (680 MHz). rfi id: TV channel 44 is switched off during the early night. rem: Normal propagation conditions. scan time stamps (s): no 17

Monitoring approaches Database and formats - RMDF data format (4) The RMDF data file is a rectangular (ASCII) data file, readable by curret spreadsheeds and mathematical packages (e.g. Matlab). RMDF format - data file 5.65e-2 2.00e-1 1.65e-1 6.75e-1 1.47e+0... etc (32 columns) 5.49e-2 2.52e-1 2.80e-1 6.15e-1 9.87e-1 4.98e-2 2.50e-1 3.60e-1 3.49e-1 1.14e+0 5.67e-2 1.13e-1 1.63e-1 1.05e+0 1.85e+0 5.98e-2 1.78e-1 3.39e-1 4.70e-1 9.38e-1 5.59e-2 1.44e-1 1.63e-1 9.25e-1 1.81e+0... etc. (128 rows) 18

Monitoring approaches Database and formats CRAF EMI and spectrum occupancy database (*) (1) provide authorities with quantitative information on degradation of Purpose: radio astronomical observations due to RFI Supply radio astronomical community with RFI information for scheduling purposes CRAF EMI database, based on radio telescope measurements Data bases: RFI database Spectrum Occupancy database, based on data from dedicated RFI monitoring observations usually using separate monitoring stations (*) Information supplied by ESF frequency manager Dr. T.A.Th.Spoelsta 19

Monitoring approaches Database and formats CRAF EMI and spectrum occupancy database (2) RFI report data format ASCII file with a record-length of 80 characters Same format for both databases Supplying EMI report to the CRAF databases: Via email (can be automated) Database parameter extraction Online via a web based questionnaire Database access In principle for governmental frequency managers and the radio telescope community Requests for database access password requests to the CRAF / ESF frequency manager, spoelstra@astron.nl 20

Monitoring approaches Database and formats CRAF EMI and spectrum occupancy database (3) CRAF database format: 1. DATE 2. STATION 3. START 4. END 5. ANTENNA 6. RFIFREQ 7. BANDWIDTH 8. REP INTERVAL 9. INTENSITY 10. INT UNIT 11. RFI AZ 12. RFI EL 13. TYPE (broadband or spectral) line 14. ANT AZ 15. ANT EL 16. DEG 17. EOR total: 80 bytes CRAF RFI database format available via spoelstra@astron.nl 21

Monitoring approaches Database and formats CRAF EMI and spectrum occupancy database (4) Query options to the CRAF EMI database: Degradation of observation as a function of time in the day / days in the week /frequency Developments of interference-intensity or degradation of observations as a function of requested time interval Interference-occurrence as a function of time in the day / days in week / frequency Developments of interference-occurrence as a function of requested time interval Query options to the SPECTRUM OCCUPANCY database: Averages and maxima of signal-intensity as a function of time in the day /days in week / frequency Developments of signal-intensity as a function of requested time Signal-occurrence as a function of time in the day /days in the the week / frequency Developments of signal-occurrence as a function of requested time 22

Conclusions and plans Conclusions: The RMDF format is used by the robust receiver groups and outside (West Australia monitoring, accepted by the LOFAR site evaluation group) Request for the RMDF format and related information to boonstra@astron.nl CRAF RFI reports from radio observatories are welcome/necessary (request for the format and information to the CRAF secretary (spoelstra@astron.nl) Requests for access to CRAF RFI databases to the CRAF secretary (spoelstra@astron.nl) Semi automatic parameter extraction software is developed by AS- TRON (Perl-based); is available but not supported 23

Conclusions and plans Plans: Plans are to develop Matlab software (LOFAR project) for RFI parameter extraction based on the RMDF format In the near future the RMDF format header will be extended allowing the representation of statistical data 24