Recent Results with the UAV-based Array Verification and Calibration System Giuseppe Virone POLITECNICO DI TORINO DIATI
Framework Research contract between INAF and CNR-IEIIT Title: Power Pattern Measurements on the Modern Low-Frequency Arrays for Radioastronomy (AAVSx, LOFAR, SAD) and Feasibility of Phase Pattern Measurements INAF PI: Jader Monari and Pietro Bolli CNR PI: Giuseppe Virone Partner : Politecnico di Torino- DIATI, Resp. Prof. Andrea M. Lingua Official initial date: September 2015 TECNO INAF 2014 Title: Advanced calibration techniques for next generation low-frequency radio astronomical arrays PI: Pietro Bolli (OAA) co-pi: Giuseppe Pupillo (IRA-MED) and Tonino Pisanu (OAC) Official initial date: April 2015 Research grant (18 months) to Salvo Pluchino (IRA-Noto) External collaborators: Stefan Wijnholds, Andrea Lingua and Giuseppe Virone
Flying Far-field Test Source A micro hexacopter is used as far-field RF source flying over the AUT RF transmitter UAV equipped with a continuous-wave RF transmitter and a dipole antenna The UAV autonomously performs quasi-rectilinear, constant height paths (GPS navigation) e.g. E-plane or H- plane cuts Automatic take-off and landing Differental GNSS system to track the position (accuracy few cm) Dipole antenna On-board IMU to measured attitude (pitch, roll, yaw)
Compass Verification DIATI Two scatterers on the UAV Two total station pointing the UAV (accuracy 1 cm) The other angles could be verified as well mean = - 0,547, std = ±0,867 4
AUT gain (dbi) SKALA 2.0 Pattern from 50 to 650 MHz Courtesy of Spin Flight at 50 MHz 10 E-plane @ 650 MHz 5 SKALA dual logperiodic antenna working from 50 to 650 MHz. Height is about 1.8 m -20 Co-polar meas. -25 Co-polar sim. X-polar meas. X-polar sim. -30-60 -40-20 0 20 40 60 Zenith angle (deg) 0-5 -10-15 E-plane at 650 MHz
LOFAR campaign April 2016 in collaboration with Stefan Wijnholds and Menno Norden Aerial View of a LOFAR station (The Netherlands) UAV to perform an end-toend system verification LBA inner Three Arrays in Three Days with Multi-frequency TX (almost 300 Patterns/Flight) HBA half LBA outer Embedded Element Patterns Array Calibration and Pattern Antenna positions (SW s talk) Courtesy of LBAinner: 48 dual-pol elements (random quasi-dense) 10-80 MHz LBAouter: 48 dual-pol elements (random sparse) 10-80 MHz HBA: 48 tiles having 16 dual-pol bow-ties (dense) 120-240 MHz
Normalized gain (dbi) LOFAR LBA antenna in Turin LOFAR Low-Band Antenna in Turin Courtesy of: LBA simulation by Michel Arts using WIPL-D 5 Normalized pattern @ 60 MHz - E-plane 0-5 -10-15 LBA antenna consists of a pair of crossed inverted-v dipoles operating from 10 to 80 MHz. Antenna height is 1.7 m, width is about 3 m -20-25 -30-35 -40 co-polar meas. co-polar sim. x-polar meas. -45-60 -40-20 0 20 40 60 (deg)
Gain Measurement
Gain Measurement
AUT AUT Pattern Pattern (dbi) (dbi) AUT AUT Pattern Pattern (dbi) (dbi) Embedded Element Pattern LBA inner Extracted AUT pattern @4.450950e+01 MHz, volo2_autocorr_44p5mhz 10 10 5 5 0 0 Extracted AUT pattern @4.450950e+01 MHz, volo2_autocorr_44p5mhz Extracted TW1, x scan, AUT average pattern @4.450950e+01 bearing: 88, RMS MHz, dev. volo2_autocorr_44p5mhz from ideal plane: 0.4 TW1, x CH scan, A: average H-plane bearing: cross-polar; 88, CH RMS B: dev. E-plane from co-polar ideal plane: 0.4 CH A: H-plane cross-polar; CH B: E-plane co-polar Element 0: E-plane 44.5 MHz -5-5 -10-10 Meas. A Sim. Meas. A A -15 Meas. Sim. A B -15 Sim. Meas. B B -20 Extracted AUT pattern @4.450950e+01 MHz, volo2_autocorr_44p5mhz Sim. B -20-60 -50-40 Extracted TW1, -30 x scan, AUT -20pattern average -10 @4.450950e+01 @5.722650e+01 bearing: 88, 0 RMS MHz, 10 dev. volo2_autocorr_44p5mhz volo2_autocorr_57mhz from ideal 20 plane: 300.4 40 50 60-60 -50-40 TW1, -30 x scan, CH -20 A: average H-plane -10 bearing: Zenith cross-polar; Angle 88, 0 CH RMS (Deg) B: dev. 10 E-plane from co-polar ideal 20 plane: 30 0.4 40 50 60 10 CH A: H-plane cross-polar; Zenith Angle CH (Deg) B: E-plane co-polar 10 5 5 0 0-5 -5 57 MHz -10-10 Meas. A Meas. A Sim. Meas. A A -15 Sim. A Meas. Sim. AB -15 Meas. B Sim. Meas. B B -20 Sim. B Sim. B -20-60 -50-40 -30-20 -10 0 10 20 30 40 50 60-60 -50-40 -30-20 -10Zenith Angle 0 (Deg) 10 20 30 40 50 60 Zenith Angle (Deg)
AUT gain (dbi) AUT gain (dbi) AAVS0.5 Embedded Element Patterns at 10 5 0-5 50 and 350 MHz -10-15 Phi=0-50 MHz Mullard Radio Astronomy Observatory (UK) Courtesy of -20 pol A meas pol A sim -25 pol B meas pol B sim -30-50 -40-30 -20-10 0 10 20 30 40 50 Zenith Angle (deg) 10 5 0-5 -10-15 Phi=0-350 MHz -20 co-pol meas co-pol sim -25 x-pol meas x-pol sim -30-50 -40-30 -20-10 0 10 20 30 40 50 Zenith Angle (deg)
Phase (deg) Phase of Cross Correlation Coefficient at 44.5 MHz UAV at zenith above central element of the LBA inner Baseline 0-40 (length 13.7 m) 49 48 47 No Average Averaged Measurement -Correction+Avg(UnCorr)+Avg(Correction) K(R 0 -R 40 ) 46 45 44 43 42 0 5 10 15 20 25 30 35 40 Time (sec)
Deg -4.3-4.4 Residual Phase Value Baseline 0-40 Averaged Trace (256 Sample) Average +/- estimated std Average +/- measured std Measured RMS -4.5-4.6 RMS estimated from GNSS precision -4.7-4.8-4.9 0 5 10 15 20 25 30 35 40 Time (sec) For shorter baselines where the elements see the UAV at the same observation angle, this residual correspond to the calibration coefficient
Second Day: Wind Speed 35 km/h Cause not yet fully understood The day before the wind was even faster. The UAV was not able to follow the preset path but was still able to land automatically The values of attitude angles (pitch about 20 Deg) are less accurate All the attitude traces are more noisy
Ideas for the Upcoming Cambridge pre-aavs1 Campaign Near-field scan and Relative Phase Measurements Phase Reference Correlator Measured Phase φ out = φ array φ ref K(R array -R ref )
Ongoing: Phase-Locked UAV Transmitter RF oscillator locked with the GNSS PPS signal Independent of Reference Antennas Copy of the transmitted signal Correlator Trimble: 10MHz + PPS Valon: 10 MHz ref. + 400 MHz fout 16
Ongoing: UAV-based Sensitivity Measurements by G. Pupillo and S. Pluchino The proposed method allows to measure the Sensitivity (A eff / T sys ) of the system, including all its parts, in the operating environment. UAV The method estimates the Power Flux Density at AUT G Tx θ, φ PFD = P Tx G Tx θ, φ 4πR 2 [Wm 2 ] AUT R P meas PFD is defined as the transmitted UAV energy, impinging upon the AUT per unit area Soil
P meas [W] UAV-based Sensitivity measurements by G. Pupillo and S. Pluchino P high PFD sys = PFD highp low PFD low P high P high P low P low The sensitivity can be estimated by mean of PFD sys as follow: PFD sys O PFD low PFD high Where: P meas = measured received power [W] PFD low, PFD high = Surface Power Density Wm 2 corresponding to two different power level trasmissions by UAV Assumptions: Rx chain linearity Negligible T sys and gain variations PFD [Wm 2 ] A eff T sys = kb m 2 PFD sys K Tasks to be done: Error budget investigation Simulations Test on the field
Conclusion Good results obtained on single element, embedded-element patterns from 50 to 650 MHz UAV flying test source as an end-to-end verification and calibration system Advantages of UAV-based antenna measurements: Measurements in the real installation conditions Beam pattern measurements on arrays and instrument calibration are possible (see G. Pupillo, et al. Medicina Array Demonstrator: calibration and radiation pattern characterization using a UAV-mounted radio-frequency source, Experimental Astronomy, pp. 1-17, Apr. 2015 Low cost, portability, no infrastructures
S21 (deg) S21 (db) Phase measurements - CABLES 35 LMR400 cable 1 Phase-shift variation with respect to meas. 1 at 10:15 AM 0.12 LMR400 cable 1 Attenuation variation with respect to meas. 1 at 10:15 AM 30 25 20 15 10 5 0 Meas. 12 at 16:15 Meas. 11 at 15:45 Meas. 10 at 15:15 Meas. 9 at 14:15 Meas. 8 at 13:45 Meas. 7 at 13:15 Meas. 6 at 12:45 Meas. 5 at 12:15 Meas. 4 at 11:45 Meas. 3 at 11:15 Meas. 2 at 10:45 0.1 0.08 0.06 0.04 0.02 0-0.02-0.04 Meas. 12 at 16:15 Meas. 11 at 15:45 Meas. 10 at 15:15 Meas. 9 at 14:15 Meas. 8 at 13:45 Meas. 7 at 13:15 Meas. 6 at 12:45 Meas. 5 at 12:15 Meas. 4 at 11:45 Meas. 3 at 11:15 Meas. 2 at 10:45-5 10 2 10 3 Frequency (MHz) -0.06 10 2 10 3 Frequency (MHz) 400 MHz 400 MHz 21
Relative Phase Measurements Reference characterization In a vertical Flight φ array is constant θ 1 max -θ 1 min θ 1 Reference will be characterized in the range from θ 1 min to θ 1 max Correlator φ out = φ array φ ref K(R array -R ref )