A versatile compact array calibrator for UHE

A versatile compact array calibrator for UHE neutrino acoustic detection S. Adrián-Martínez, M. Ardid, M. Bou-Cabo, I. Felis, G. Larosa et al. Citation: AIP Conf. Proc. 1535, 190 (2013); doi: 10.1063/1.4807546 View online: http://dx.doi.org/10.1063/1.4807546 View Table of Contents: http://proceedings.aip.org/dbt/dbt.jsp?KEY=APCPCS&Volume=1535&Issue=1 Published by the American Institute of Physics.

Additional information on AIP Conf. Proc. Journal Homepage: http://proceedings.aip.org/ Journal Information: http://proceedings.aip.org/about/about_the_proceedings Top downloads: http://proceedings.aip.org/dbt/most_downloaded.jsp?KEY=APCPCS Information for Authors: http://proceedings.aip.org/authors/information_for_authors

Downloaded 29 May 2013 to 158.42.144.82. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

A versatile compact array calibrator for UHE neutrino acoustic detection S. Adrián-Martínez*, M. Ardid, M. Bou-Cabo*, I. Felis, G. Larosa, C. Llorens, J.A. Martínez-Mora and M. Saldaña. Instituto de Investigación para la Gestión Integrada de las zonas Costeras (IGIC), Universitat Politècnica de València. C/Paranimf,nº1 46730 Gandia, Spain. *Multidark fellow. Abstract. In situ acoustic calibration devices are necessary to assure the optimal performance of sensors and detectors for the acoustic detection of ultra-high energy neutrinos in underwater telescopes. Moreover, they provide evidences for the feasibility evaluation of the technique and for the determination of the entire detector efficiency. Following previous studies related to parametric acoustic sources, a first prototype of a compact acoustic array able to mimic the acoustic neutrino signal (a transient bipolar signal with ‘pancake’ directivity) is presented. The compact array developed has practical features such as easy handling and operation, and versatile functionality. In the latter sense, the transmitter is able to work in different frequency ranges for different application modes, and thus to carry out several tasks related to acoustics in underwater neutrino telescopes: emission of neutrino-like signals, calibration of sensor sensitivities and responses, emission of signals for positioning, etc. The design, construction and characterization of the prototype are described. A simulation study is also discussed, where experimental signals were propagated over distances in the kilometre range. Keywords: Underwater neutrino telescopes; acoustic neutrino detection; parametric acoustic sources; underwater acoustic array sensors. PACS: 07.64.+z, 43.25.Lj , 43.38.+n , 95.55.Vj .

well for positioning purposes it could be optimized an acoustic system that performs both tasks: positioning and acoustic detection. Even though all acoustic sensors can be characterized in the lab, it would be desirable to deploy a compact calibrator that in situ, as the sensors may change its properties over time, calibrates the detection system emitting neutrino like signals, in order to test and validate the technique, to determine the efficiency of the system and to train and tune it in order to improve its performance [3]. In this work a compact acoustic calibrator using a parametric acoustic source technique is considered. Parametric generation of acoustic signals is a wellknown nonlinear effect that was first studied by Westervelt in the 1960s [4]. This technique has been implemented in many applications in underwater acoustics, especially to obtain very directive acoustic sources. The acoustic parametric effect occurs when two intense monochromatic beams, with two close frequencies, travel together through a medium. Under these conditions, in the region of nonlinear interaction, secondary harmonics of these frequencies are produced: the sum and difference of the frequencies and the double frequencies of both beams. In this case the difference of frequencies is used to generate a secondary beam signal. The main advantage of this technique is that the secondary parametric beam has a similar directivity pattern as the primary beam, enabling low frequency beams (difference frequency)

1. INTRODUCTION Ultra-high energy (UHE) neutrino acoustic detection is a promising technique to extent the energy range of neutrino astronomy. Towards this sense, ANTARES (Astronomy with a Neutrino Telescope and Abbys Enviromental RESearch) [1], the first undersea neutrino telescope, is studying the feasibility of the technique. ANTARES comprises the AMADEUS (ANTARES Modules for the Acoustic DEtection Under the Sea) system [2], an acoustic test setup which is operational and taking data. The latter consists of six “acoustic clusters”, each comprising six acoustic sensors that are arranged at distances of roughly 1m from each other. Two vertical mechanical structures (so-called lines) of the ANTARES detector host three acoustic clusters each with spacing between the clusters ranging from 14.5 to 340m. Each cluster contains custom-designed electronics boards to amplify and digitize the acoustic signals acquired by the sensors. An on-shore computer cluster is used to process and filter the data stream and store selected events. AMADEUS can be considered a prototype array to evaluate the feasibility of the neutrino acoustic detection. In case of positive results, a large-scale detector that combines optical and acoustical detection techniques into a hybrid underwater neutrino telescopes may be developed. Considering that the optical neutrino technique needs acoustic sensors as

5th International Workshop on Acoustic and Radio EeV Neutrino Detection Activities AIP Conf. Proc. 1535, 190-194 (2013); doi: 10.1063/1.4807546 © 2013 AIP Publishing LLC 978-0-7354-1159-3/$30.00

190 Downloaded 29 May 2013 to 158.42.144.82. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

the secondary beam signal shape generated in the medium and, on the other hand, reproducing the desired ‘pancake’ pattern of emission [7]. The results obtained, in the lab and in a larger pool for longer distances [8], are in agreement with the expectations from parametric theory and previous measurements, but now being able to generate the signals for almost cylindrical propagation. Following these first results, the parametric acoustic sources technique can be considered a good tool to develop a transmitter able to mimic the acoustic signature of a UHE neutrino interaction. Moreover, although the primary application of this transmitter is generating the acoustic signature of a neutrino, the possibility of working in two frequency ranges (5-20 kHz and 400 kHz) allows the system to carry out several acoustic-related tasks in an underwater neutrino telescope: the aforementioned acoustic detection calibration, calibrating of receivers and monitoring emitters of an acoustic positioning system and acting as transceivers for such a system. Certainly, this could reduce the overall costs of the necessary calibration systems and facilitate the deployment and operation in the deep sea. In that sense, besides the studies of acoustic parametric generation, some tests have been made to select signals to carry out all the acoustic tasks including broadband signals that offer good results on signals processing techniques, such as cross correlation techniques. This kind of signals can be used in both modes of operation: generated directly at low frequency, or generated parametrically at 400 kHz. Figure 1 shows some tests with a parametric sweep (400 kHz signal modulated by a 10-50 kHz sweep signal). Using this signal we are able to obtain a low frequency sweep signal from parametric generation, which is highly directive. On the left side of the figure, we show the emitted signal and the expected parametric signal (signal to be used for the cross-correlation). On the right, the received signal and the signal obtained after correlation are shown. The first peak in the correlated signal corresponds to the direct path between emitter and receiver (this will be the reference peak in positioning tasks), the remaining peaks are due to reflections on the pool walls and on the surface. The time for the different peaks agrees with the geometric dimensions of the setup. The measurements were made in a pool of 6.3 m length, 3.6 m width and 1.5 m depth. A FFR-SX83 transducer was used as emitter, and the receiving hydrophone used to measure the acoustic waveforms was a FFRSX30, both of them are from the same company. The distance between emitter and receiver was 4.3 m. The following DAQ system was employed for emission and reception. To drive the emission, a 14-bits arbitrary waveform generator (National Instruments, PXI-5412) was utilized with a sampling frequency of

with high directivity (primary beam). Since the emission is made at high frequencies, it is possible to obtain narrow directional patterns using a transducer with small overall dimensions. However, as the neutrino acoustic signal is a transient signal with broadband frequency content and cylindricalsymmetric directivity, there are some difficulties when applying this method to a compact acoustic calibrator. To deal with transient signals, theoretical and experimental studies [5] indicate that it is possible to generate a signal with a ‘special’ modulation at a larger frequency in such a way that the pulse interacts with itself while it is travelling through the medium, thereby generating the desired signal. If the signal has to travel long distances primary high frequency signal will be absorbed and only low frequency signal will be available. In this case, the secondary parametric signal generated in the medium is related to the second time derivative of the envelope of the primary signal, following the equation 1:

p ( x, t )

B · P2S w 2 ª § x ·º § f t  ¸» ¨1  ¸ 4 2 « ¨ © 2 A ¹ 16SUc Dx wt ¬ © c ¹ ¼

2

(1)

where P is the pressure amplitude of the primary signal, S is the surface area of the transducer, f(t-x/c) is the envelope of the primary signal, x is the propagation distance, t is the time, B/A is the nonlinear parameter of the medium (~5.3 for seawater),  is the density, c the sound speed and  is the absorption coefficient. Previous research studies were done to evaluate the possibility of using the parametric acoustic sources technique to reproduce the acoustic signature of a UHE neutrino interaction. Our first studies were done in the lab using planar transducers. This work concluded with positive results regarding the feasibility of the bipolar pulse generation with the parametric technique [6].

2. CYLINDRICAL TRANSDUCER TESTS The next step was to deal with the cylindrical symmetry. For this purpose a Free Flooded Ring (FFR) transducer model FFR-SX83 manufactured by Sensor Technology Ltd. (Collingwood, Canada) was selected as emitter. This transducer, with cylindrical symmetry, usually works in the 5-20 kHz frequency range, but in this application a second resonance peak at about 400 kHz is used, which is the frequency used for the primary beam in the parametric studies. Using this transducer a new setup was configured in order to, on the one hand, study in more detail the influence of the signal envelope used in emission with respect to

191 Downloaded 29 May 2013 to 158.42.144.82. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

10 MHz. A linear RF amplifier (1040L, 400W, +55 dB, Electronics & Innovation Ltd.) was handled to amplify the emitted signal. For the reception, an 8-bit

digitizer (National Instruments, PXI-5102) applied with a sampling frequency of 20MHz. Received signal

1

1

0.5

0.5

Amplitude [V]

Amplitude [V]

Emitted signal

0 0.5 1

0 0.5 1

0

0.2

0.4 0.6 Time [ms]

0.8

1.5

1

2

3

Parametric expected signal

5

6

5

6

400

Correlation amplitude

Amplitude [V]

4 Time [ms]

Correlated signal

1 0.5 0 0.5 1

was

0

0.2

0.4 0.6 Time [ms]

0.8

200 0 200 400 600

1

2

3

4 Time [ms]

FIGURE 1. Emitted and received signal for the parametric sweep test. The correlated signal is the result of correlation between parametric expected signal and received signal. Atenuation

Directivity 1

1

0.8

Normalized Amplitude

Normalized Amplitude

0.9 0.8 0.7 0.6

0.6

0.4

0.5

Sweep 0.2

Parametric Sweep

0.4 0.3 200

Bipolar Pulse 250

300 350 400 ER distance [cm]

450

0 15

500

10

5

0 Angle [º]

5

10

FIGURE 2. Attenuation and directivity comparisons of the three different kinds of signals that the calibrator can generate. The error bars refer to statistical errors. parametric signals have a clear directive pattern whereas the direct sweep signal is non-directive since Figure 2 shows the results of a comparative study of the transducer is not directive for low frequencies. The the different signals used that was made in the pool: results must be taken with caution due to the geometric Sweep signal from 5 kHz to 25 kHz (low frequency limitations of the tank, especially for the case of the signal), parametric sweep signal, i.e. 400 kHz signal sweep. For this case we have a non-very directive modulated by a 10-50 kHz sweep signal (the low transducer and long signals, and the reflections on the frequency directive signal is generated parametrically bottom and on the water surface might affect the in the medium), and a bipolar pulse generated behaviour of the correlation. The different behaviour parametrically (so a transient and directive signal). On of parametric and non-parametric signals is evident, the left, the measured attenuation is shown for each however. The possibility of having different kinds of signal. On the right, a comparison between the signals will allow to perform different kind of directivity patterns of some signals is shown. Both

192 Downloaded 29 May 2013 to 158.42.144.82. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

calibratioons and/or to do calibrationn in different steps. s This aspeect is described in the sea caampaign sectioon.

me, it is expeccted to receivve a bipolar nonnlinear regim pulse of, at leaast, 75 mPa ppeak-to-peak, which is a MADEUS thrreshold (20 vallue clearly aabove the AM mP Pa) [2], and coorresponds to tthe pressure reeference for a a10 a 20 eV neutrrino interactionn at 1 km distaance.

3. PROTOTY P YPE ARRA AY DESIGN N The pparametric acooustic sources technique preesents the advaantage of a ccompact desiggn with respeect to other classsical solutionns [9]. To obtaain directive bbeams with a linear phaseed array, the use of hhigher ments with sm maller frequenciies implies thhat fewer elem spatial seeparation are needed. n Such a device wouuld be easier to install and ddeploy in an undersea neuutrino telescopee.

33.1. Mechan nical Design n The array prototype is composed of three traansducers moddel FFR-SX83 (see Figure 44). Each one hass a diameter of 11.5 cm and is 5 cm m high. The traansducers are fixed on a rigid r axis usiing flexible polyurethane (E EL110H, Robnnor Resins Lttd). Due to thee nature of tthe cured pollymer, this ooffers water ressistance and eelectrical insullation for highh frequency andd high voltagge applicationss. The resultinng compact dessign has a lenngth of 17.5 ccm for the active region. Divverse studies,, including diirectivity studdies, carried outt with the firsst prototype off the array aree detailed in [8]].

FIGURE 3. Simulated signal obtaained after 1 km propagatioon.

In orrder to estim mate the optiimum numbeer of elements needed, to achieve both basic b requirem ments, that are,, reproducingg the pancakke directivity and obtainingg the necessarry acoustic levvel in the recceiver sensors, signals recorrded experimeentally have been m. An propagateed theoreticallly to a distaance of 1 km algorithm m which workks in the frequency domaain is used. Thiis algorithm ttakes into accoount the frequuency dependennce of the hyddrophone’s sennsitivity. Moreeover propagatees each specttral componennt consideringg the radial divvergence of thhe pressure fieeld, as 1/r, annd the absorptioon of the meedium [10], calculated forr the typical seea conditions at the ANTA ARES site. Figgure 3 shows a signal propagated to 1 km ((the original ssignal was measured in the lab with a disttance of 0.7 m m). In this casse, no addittional filter is applied, the propagatiion medium aacts as a natuural filter, andd at 1 km only low frequenccies arrive. Too be exact, theere is mall high-freqquency compoonent which iis not still a sm observedd at distances oof 1.2 km (or hhigher). Noticee that the highh-frequency signal was three orderrs of magnitudde higher thann the seconddary beam at a1m distance. As coonclusion of thhis study, usinng a single eleement it is expeected to obtainn a bipolar pulsse of 25 mPa peakp to-peak amplitude a at 1 km distance ffrom the emittter, if the emittter and receiiver are alignned (0º). Forr this reason, cconsidering the array configguration with three elements fed in phase with enough ppower to reacch the

FIG GURE 4. Com mpact array prototype (bottom rright) and the meechanical structuure necessary too hold and operrate it from a boaat (left).

3.2. Elecctronics Eleectronics has bbeen developeed with the aim m of having an autonomous and optimizedd compact sysstem able to woork in differrent frequenccy ranges foor different appplications. T The electronic device coontrols the traansmitter, gennerates and aamplifies the signals in ordder to have enough acouustic power to achieve em mission in a nonlinear n regiime. This is an a essential reqquirement for parametric generation and it is neccessary for caalibration and//or positioninng purposes, in the last cases is not necessaary to reach thhe nonlinear reggime but may be it is necesssary the ampllification of thee signals in ordder to be deteccted from distaant acoustic recceivers. Pulse Width Modullation techniquue (PWM) is used for thesse purposes. This T techniquue has been mented for tthe acoustic transceiver alrready implem eleectronics of thhe positioningg systems forr the future KM M3NeT neutrrino telescopee [11-13]. Soome of the

193 Downloaded 29 May 2013 to 158.42.144.82. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

shape in time and ‘pancake’ directivity. Moreover, due to the transmitter system versatility, it could be implemented to carry out several tasks related to acoustic emission for underwater neutrino telescopes: acoustic detection calibration, positioning emissions and calibration of hydrophone sensitivities. This may simplify the set of acoustic transmitters needed in a neutrino telescope, and thus may reduce the complexity of the system and the costs with respect to the use of multiple devices. A selection of signals and an analysis strategy based on cross correlation techniques for several acoustic tasks have been presented. The future work will consist in testing the complete compact transmitter prototype (array and electronics) and characterizing it in the lab, in shallow water and in situ. For the last part, tests of the transmitter using the AMADEUS system are foreseen, either using the array from a vessel in a sea campaign or maybe integrated in the ANTARES infrastructure. A measurement strategy has been developed in order to face the sea campaign in three stages, gradually increasing the difficulty of the tests and adding control signals to facilitate post processing work.

advantages this technique offers are the efficiency, it needs minimum power consumption in stand-by mode, the simplicity of design, and the low and the compactness of the resulting system, the latter is fundamental for the final compact device [14].

4. SEA CAMPAIGN: MEASUREMENT PLAN In order to test the transmitter in situ, on the ANTARES site during a sea campaign, a mechanical structure has been built (Figure 4). The structure allows fixing the device to a boat’s hull for handling it and facilitates the control on the rotation angle. This is a very important point in order to be able to orientate the emitted signal pointing to the direction of the receivers, which is not easy due to the high directivity of the bipolar pulse (opening angle of ~10º). The detection of this weak directive signal emitted from a boat and detected by the AMADEUS system several kilometres away is very challenging. To help on this, some devices are being implemented monitoring the emitted signal and its associated location information: current GPS position and time of emission, together with the orientation information from a tiltmeter and compass. Moreover, a measurement strategy has been developed taking into account the array versatility, that is the two operation modes. A three step calibration is foreseen, increasing progressively the difficulty. Firstly, a long broadband low frequency non directive signal (sweep for example) can be emitted. Since it is directly emitted, so it could be of high amplitude. It is low frequency signal, so attenuation is small. It is not directive, so orientation is not difficult. Finally it has broadband, so processing techniques may be applied. For all of this it should be quite easy to detect it. In a second step, a long parametric signal can be emitted. Incorporating the directivity challenge, so high frequency and lower amplitude, but it could be still taken advantage of processing techniques. Finally, the parametric bipolar signal, that is transient and directive, can be emitted. The emission of the last one might be tagged, that is preceded and followed by signals of the previous modes. In this way it will be easier to look for the bipolar signal during the post processing, looking at the correlation peaks of the received signal and the known expected tag signals.

ACKNOWLEDGMENTS This work has been supported by the Ministerio de Ciencia e Innovación (Spanish Government), project references FPA2009-13983-C02-02, ACI2009-1067, Consolider-Ingenio Multidark (CSD2009-00064). It has also being funded by Generalitat Valenciana, Prometeo/2009/26 and Gerónimo Forteza FPA/2012/096.

REFERENCES 1. M. Ageron et al., Nucl. Instrum. Methods. A, 651, 11-38, (2011). 2. J.A. Aguilar et al., Nucl. Instrum. Methods. A, 626-627, 128-143, (2011). 3. M. Ardid, Nucl. Instrum. Methods. A, 604, S203-S207, (2009). 4. P. J. Westervelt, J. Acoust. Soc. Am., 35, 535-537, (1963). 5. M.B. Moffett and P. Mello, J. Acoust. Soc. Am., 66, 1182-1189, (1979). 6. M. Ardid et al., Nucl. Instrum. Methods. A, 604, S208-S211, (2009). 7. M. Ardid et al., Nucl. Instrum. Methods. A, 662, S206-S209, (2012). 8. M.Ardid et al., Sensors, 12, 4113-4132, (2012). 9. W.Ooppakaew et al., Mobile Adhoc and Sensor Systems (MASS), 2011 IEEE 8th International Conference on, 910-915, ( 2011). 10. R.E. Francois, G.R. Garrison, J. Acoust. Soc. Am., Part I. 72(3), 896-907, (1982). Part II. 72(6), 1879-1890, (1982). 11. M. Ardid et al., Nucl. Instrum. Methods. A, 617, 459-461, (2010). 12. M. Ardid et al., Nucl. Instrum, Methods. A, 626-627, S214-S216, (2011). 13. C.D. Llorens, et al., J. Instrum., 7, C01001, 1-8, (2012). 14. S. Adrián-Martínez et al. Nucl. Instrum. Methods. A,. In press. http://dx.doi.org/10.1016/j.bbr.2011.03.031

5. CONCLUSIONS AND FUTURE WORK The solution proposed based on parametric acoustic sources can be considered a good candidate to generate acoustic neutrino-like signals, with both specific characteristics of the predicted signal: bipolar

194 Downloaded 29 May 2013 to 158.42.144.82. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions