Introduction to the SETI program, criticalities and future ...sait.oats.inaf.it/MSAIt890318/PDF/2018MmSAI..89..316M.pdfIntelligence) program whose aim is to look for artiﬁcial radio

Mem. S.A.It. Vol. 89, 316c© SAIt 2018 Memorie della

Introduction to the SETI program, criticalitiesand future strategies

S. Montebugnoli

SETI Advisor to the INAF Scientific Directorate, Rome – Institute of Radioastronomy, ViaGobetti 101, Bologna, e-mail: [email protected]

Abstract. In this paper, we will analyze the evolution of the SETI (Search for Extra-TerrestrialIntelligence) program whose aim is to look for artificial radio signals from advanced civiliza-tions in the space with the radiotelescopes (there is also SETI program in the optical band buthere we consider only the search in the radio band). We will investigate the intrinsic criticali-ties and difficulties of the program considering anyway that it has a low probability of success:we don’t know where to point the large antennas, the tuning frequency, what to search for, atwhich moment to observe, what sensitivity is required to receive an ET signal. It is extremelydifficult to exactly set the just mentioned parameters and correctly answer all the questions. Inthe future, SETI observational activities will have to be based on the synergy between differ-ent groups with competences in mathematics, Physic and engineers introducing new researchparadigms and related algorithms. The SETI program represents a challenge to the limit of theimpossible for both our current technologies and knowledge! We search for an unknown ana-log or digital signal buried in an ocean of man-made radio interferences, self-generated noiseand natural noise from the radio sources that populate the universe.

Key words. SETI – Radio Communication – Exoplanets – Data processing

1. Introduction.

The formal birth of the SETI program was justafter the launch of the Russian Sputnik, thefirst artificial satellite in the world. In 1959,Giuseppe Cocconi and Philip Morrison sug-gested using the microwaves in the region ofthe Hydrogen line emission at 21 cm, to lookfor radio communications from other inhab-ited worlds or to communicate between differ-ent planetary systems (Cocconi 1959), as de-scribed in an article on the prestigious US mag-azine ”Nature”.

Radio searches for extraterrestrial signalsofficially began in 1960, when Frank Drake(who wrote the Drake Equation) was based at

the Greenbank (NRAO, West Virginia). Drakeused the 85-foot Tatel radio telescope in hisProject Ozma (named after the queen figurein the Wizard of Oz), which examined sig-nals coming from two nearby stars, Tau Cetiand Epsilon Eridani. Looking at the hydro-gen frequency (1,420 MHz), Drake searchedfor both repeated sequences of uniformly pat-terned pulses, and sequences of prime num-bers; the detection of either of these would in-dicate the presence of an extraterrestrial civ-ilization, but with no success (Drake 1961).On the other hand, he exploited the technol-ogy of that time using noisy receiver and avery poor data processing (analog domain).Since late 1970 NASA started to be inter-

Montebugnoli: Introduction to the SETI program, criticalities and future strategies 317

ested in the program, which became a re-ality in late the 80ies. The first SETI ob-servations started in 1992 in the 500th an-niversary of the discovery of the new worldby ”Christopher Columbus”. Unfortunately, inlate 1993 the American Congress definitivelycanceled the SETI program, probably for lackof funds or due to other unknown reasons(Garber 1999). SETI activities were taken overby the new SETI Institute, based in MountainView (California) today fully operative withfunds coming from donations from individu-als, industries of the Silicon Valley or other.The University of Berkeley also participates inthe SETI program with a group of researchers.Serendip II- V. (Backus 1992; Serendip 1988;Werthimer 1995; Cas 2016) were a revolution-ary spectrometer able to operate in parallel atthe ongoing observation without interfere withthe observer. At the Medicina dish antenna aSerendip IV system acquired data from 1998till 2008 in the frame of the SETI-Italia pro-gram as reported in Tab. 1.

It is very likely that in our galaxy exists alarge number of extrasolar planets, similar tothe earth, hosting advanced civilizations. Thesecould able to communicate through electro-magnetic signals in certain segments of thespectrum. The possible bands of frequenciesthat ET could exploit for the communications,especially towards its spacecraft traveling inthe galaxy, are those to which his atmosphereturns out to be transparent. From our obser-vation point, the frequency windows open to-ward/from the universe are the radio and theoptical one. The current SETI research in Italyis mainly conducted in the radio window anduntil today has been based on the hypothesisthat if an extraterrestrial civilization wants tonotify us of its presence, it probably irradiatesa monochromatic signal (radio carrier) towardthe outer space:

1. it is not strongly affected by interstel-lar/interplanetary dispersion and scattering.

2. searching for monochromatic signalsis computationally efficient with a FastFourier transform (complexity scales asNs log(Ns)) where Ns is the number ofsignal samples.

3. simple to interpret: a narrowband beaconcarries only one bit of information, but it isenough to tell us that an alien civilizationsomewhere exists.

4. is very efficient from an energetic point ofview because all the available energy is con-centrated in the radio carrier.

5. it is easily distinguishable from natural ra-dio emission.

In a near future, optical SETI (OSETI) pro-grams could start in some national optical ob-servatories still operating but not completelyoverscheduled. In this case, we would lookfor very short laser pulses of the order of ananosecond. These two types of signals (radiocarrier and laser pulses) have the advantage ofbeing easily distinguishable from those of nat-ural origin. Due to the Earth atmosphere trans-parency at the radio and optical bands, theycan easily reach the ground based observato-ries. If the transmission is not intentional andthe radio waves arriving from space representthe residue of electromagnetic activities of ET(communications, spacecraft control etc..), theproblem becomes really difficult. Even if theantenna is pointed in the right direction, tunedon the exact frequency, with the appropriatesensitivity and observing at the right time butthe received signal is totally unknown, it is verydifficult to use the right detector. For this rea-son it is unlikely to notice a possible hiddeninformation in an ocean of noise. In fact, thedemodulator on board of the receiver must becompatible with the modulation of the trans-mitted signal, otherwise the information is lost.It is likely that ET has long been disseminatingunknown radio signals in space, which almostcertainly risk being never detected by our mod-ern radio systems.

With the current radio telescopes, wesearch for sinusoids (monochromatic signals)shifting linearly in frequency (Doppler effect)due to variation of the radial component corre-lated to the relative Earth-Exoplanet motion. Inthis approach, the search for a monochromaticradio signal is done with a high-resolution FFT(Fast Fourier Transform) performed by verypowerful low cost computers that are availabletoday on the market. However, in the future,

318 Montebugnoli: Introduction to the SETI program, criticalities and future strategies

Table 1. SETI-Italia activities at the Medicina (Bologna) 32 m dish antenna.

Institution Station Program name Back-End Periods FundsCNR/INAF Medicina SETI ITALIA Serendip IV 1998-2008 CNR/INAFINAF/ASI Medicina ITASEL Spectra-1 2005-2013 INAF/ASI

it will be necessary to change the searching al-gorithms to considerably increase the probabil-ity of success of the program. Maybe we arelooking for an alien signal that does not actu-ally exist (see the above mentioned CW signalsearched till today), or there are indeed alienradio signals wandering in space that we areunable to reveal with our current technology.We could be unable to detect because we donot know what we have really to search for. Itis a truly complex scenario!

2. Main difficulties of the program.

The SETI Program assumes being possibleto establish a radio communication link, asschematically shown in Fig. 1, between an ex-oplanet and the planet Earth.

To increase the power (Pout) at the receiveroutput connector, it is clear that the effects ofthe impressive attenuation due to the FSPL(Free Space Path Loss) have to be strongly re-duced. FSPL is defined as the loss incurredby a radio signal as it travels in a straightline through a vacuum space with no absorp-tion or reflection of energy from nearby ob-jects (It is why distant headlights are more faintthan close ones). Signal detection becomes ex-tremely complicated if it is time-varying withsome type of unknown modulation. The SETIsearch space is characterized by many param-eters and it is a very hard task to set them withthe right value in order to increase the programsuccess possibilities. A list is reported in Tab.2 with some hypothesis on how to facilitate thedemand reported in the first column:

In the case the alien transmissions are notintentional and then not specifically aimed atour planet, it means that these are residuesfrom their own radio communication activities.

At this point, the problem to extract it from thebackground noise becomes extremely compli-cated because we don’t know anything aboutthe morphology of such a signal (almost surelydigitally modulated).

Every single row of the above reported ta-ble, is here commented:

2.1. Antenna Pointing.

It is easy to point the antenna to already cat-aloged exoplanets, or single sources as well,because their positions on the sky are known(targeted mode). In the case the source positionis unknown, we have to systematically scanall the sky in order to search for it (sky sur-vey mode). In this case a very high number ofpossible antenna displacements is required. Inpractice, such a number depends on both theoperating frequency and the antenna size thatdetermine the size of its FoV (Field of View).Anyway, this remains one of the most complexparameters to be defined. Beyond our assump-tions, it is unlikely that ET will intentionallypoint his antenna towards the earth by sendinga radio signal and, at the same time, it is un-likely that just by chance we orientate one ofour antennas to his direction.

Considering that a space radio link is com-posed by two antennas of the SRT class (64m), as an exercise, it can be computed by howmany telescope position is composed the celes-tial dome. The following formula (1) gives thenumber of beams:

Nbeam =4

[tan(28.6D−1

λ )]2 (1)

Considering a working frequency of 1420MHz, the number of antenna beams needed tocover all the visible sky is 1.5106. This means


Fig. 1. Basic radio link block diagram for the SETI Program.

Table 2. Simplification hypothesis for some of the search parameters.

Search Parameters: HypothesisIn which direction the antenna The antenna can be pointed to a certain set of exoplanetshas to be pointed? similar to the earth (targeted observations).At which frequency the radio The more suitable frequency band is from 1 GHz to 10 GHztelescope has to be tuned? where the galactic noise contribute is very low: the

preferred band is that one of the emission of neutralhydrogen (1420 MHz), or more generally the band of the so-

called water hole spanning from the neutral Hydrogen (1420MHz) to the OH at 1720 MHz).

What signal do we look for? The radio signal searched today, as a proof of the ETAnalog? Digital? Modulated? In presence, is a CW radio signal probably intentionally sent bythis case how is modulated? ET.What is sensitivity with which do Observing with the largest available antennas probably thewe have to observe? ”sensitivity” requirement is met.Are we in the right time window Regarding the time window of the alien ”electromagneticof the ”electromagnetic visibility”, nothing can be hypothesized. . . .visibility” of that aliencivilization?

that if ET already points its antenna toward us,randomly moving our dish we have one chancein 1.5106 to point it in the right direction!! IfET doesn’t know anything about us, the proba-bility that it points our planet with its 64 m an-tenna considered in this example, is the sameof 1.5106. Then the probability that ET pointshis antenna in our direction while we point

our one in his direction is represented by 1part in 2.251012. Just for a practical example toevaluate such probabilities, we assume F=1420MHz and some different configurations of ETtransmitter and Earth base receiver composedby FAST (500 m), ARECIBO (300 m), SRT(64 m) and MEDICINA (32 m) antennas as re-ported in Tab. 3.


Table 3. Chance of success with different configurations.

ET TX Gain Num. of RX Earth RX antenna Num. of UNINTENTIONAL INTENTIONALantenna (dB) Beam antenna Gain (dB) Beam ET doesn’t take ET is pointing(m) (TX) (m) (RX) (care of the Earth.) Earth. One

One chance in: chance in:

64 57 1.5 · 106 32 51 3.7 · 105 5.6 · 1011 3.7 · 105

64 57 1.5 · 106 64 57 1.5 · 106 2.2 · 1012 1.5 · 106

4300 71 3.2 · 107 64 57 1.5 · 106 4.8 · 1013 1.5 · 106

500 75 9.0 · 107 300 71 3.2 · 107 29 · 1014 3.2 · 107

500 75 9.0 · 107 500 75 9.0 · 107 481 ·1014 9.0 · 107

2.2. Frequency at which theradiotelescope has to be tuned.

In the book ”Communication with ExtraTerrestrial Intelligence (CETI)” edited in the70s, Barnard M. Oliver, basing on the article byCocconi and Morrison appeared in Nature in1959, proposed the use of the lower part of mi-crowaves (1-10 GHz) as the most suitable seg-ment of the electromagnetic spectrum for thesearch for extraterrestrial radio signals (Fig.2)for the following reasons:

– Galactic noise (synchrotron radiation) at aminimum.

– Thermal noise (isotropic backgroundnoise) at a minimum.

– Quantum noise (spontaneous emission orshot noise) at low level.

– Star noise at low level.

In particular, within this ”Silent Valley”, thesearch for signals is limited to a segment thatlies between the emission of the neutral hy-drogen H (1420 MHz) and that one of theOH (1720 MHz) referred as the waterhole. Inthe case ET wanted to be noticed (a ques-tionable supposition that meets, however, thefavorable opinion of many SETI researchers)could transmit a signal within a band includedin such a waterhole.

The name given to this particular band offrequencies would like to remember the water-hole in the desert, around which the animalsgather to quench their thirst. If in the universeexist several civilizations, they would probablymeet from an electromagnetic point of view,each other around the frequencies related tothe waterhole to get noticed through their ra-dio emission!

In the 0.9-10 GHz frequency segment thereare 9.1 109 1Hz wide channels. This means thatthe chance to detect a CW signal in this bandis one in 9.1 109.

2.3. What signal are we searching for.

Which type of radio signals could we expectfrom an extraterrestrial technologically ad-vanced civilization? This is a core question forthe SETI program and it is extremely difficultto answer because we do not know anythingabout their technology, their way of being andhow (or if) they communicate through radiowaves or with other unknown mode. In the casewe wanted to reveal our presence to the aliens,we would send out to space a monochromaticsignal, then we suppose that even ET couldbehave in this way. Such a type of signal isconsidered to be very efficient because all theavailable energy is concentrated on the radiocarrier but it doesn’t carry information since nomodulation is present. Anyway, if detected, itshould be considered absolutely a valuable de-tection because makes us aware that someonemust have generated and sent it into the space!Anyway, it is realistic to think that it is unlikely(not impossible) that some extraterrestrial civ-ilization wants to indicate its presence to us. Itcould be instead more likely to receive alien ra-dio signals coming from their normal planetaryradio activities (if any). In this case, what kindof signal could we receive? Analog? Digital?Modulated in which way? This is another ex-tremely complicated question which is difficultto answer. The type of modulation must be ab-solutely known because the receiver needs tobe equipped with a demodulator compatiblewith the modulation mode used at the trans-


Fig. 2. Microwave window.

mitter stage, otherwise, the information couldbe completely lost.

In Fig. 3 is shown a very simple schematicblock diagram of a radio communication linkwhere it is clear that if the transmitter, for in-stance, is PSK modulated (Phase Shift Keyingis a digital modulation process), the receivermust be equipped with a detector suitable forthe PSK and this occurs for any type of modu-lation. This is a great challenge because if wea priori don’t know the modulation mode wecannot apply for the correct demodulation al-gorithm. And it is even more complicated if wedeal with an absolutely unknown alien modu-lation mode. For many decades, on our planetwe transmitted analog signals (radio, TV, etc.)then, with the advancement of technology, wehave moved to the digital techniques (satellites,military, TV, DAB, etc.). A similar evolutioncould be thought for the alien civilization radiotransmissions as well. The digital field, in fact,offers many advantages as to concentrate most

of the energy in the information. If the mod-ulation mode is not known, the digital signalsreceived increasingly resemble the noise mak-ing them even more undetectable! The post-processing phase is one of the most criticalstage of the SETI ”chain”.

Worthy of note is the Spread Spectrummode, which is considered to be the most suit-able modulation mode for interstellar radiocommunications, being insensitive to radio in-terference in the environment in which the re-ceiver operates.

In Fig. 4 are reported the different types ofboth analog and digital basic modulations usedby modern terrestrial technology. Fig. 5 shows,in a more detail, only the various types of dig-ital modulation used today. Since any kind ofanalysis and signals detection must be carriedout with suitable algorithms and procedures,

we report in Tab. 4 the main processing al-gorithms (ITU-R 2011). To date, the main goalof the modulation techniques is to compress as


Table 4. Main analysis tools for different operational mode.

Parameter to be Analysis tool Modulation type Radio EnvironmentmeasuredPresence of a radio Cross-Correlation of Any modulation type Anycommunication signal Q-I signal or of but especially for

instantaneous AI with known TDMA, CDMAreference signal and DSSS signal

Spectral power density Any modulation type Medium and high SNR

Auto-correlation and OFDM, SC-FDMA, Anycyclic auto-correlation SC-FDE

Spectrum correlation Unknown DSSS and Anyanalysis weak signals

PRF or burst length Amplitude time OOK, Radar, IFF, other Medium and high SNRsignal analysis bursted signals

Carrier frequency and Spectral power density Any modulation type Medium and high SNRsubcarrier frequency

Instantaneous frequency FSK Medium and high SNRFi istogram

Instantaneous frequency FSK Medium and high SNRFi average

Spectrum of I-Q signal PSK, QAM, CPM Positive SNRraised to power N(=)M(MPSK), 4(QAM) or14 for CPM)

Spectrum correlation Any linear modulation Anyanalysis and especially ASK,

BPSK, QPSK

The spectrum of signal Pi/2DBPSK, Positive SNRmodule raised to power Pi/4DQPSK, SQPSK Any2 or 4 with severefiltering

Emission bandwidth Spectral power density Any modulation type Medium and high SNRand channelization compared with mask or

limit line functionFrequency distance Spectral power density. FSK, OFDM, COFDM Medium and high SNRbetween subcarriers Harmonic search and/or(Shift for FSK) harmonic markers

Spectral power Medium and high SNRdensiti. Harmonicsearch and/orharmonic markers


Fig. 3. Example of a radio link (for example with a PSK modulation) architecture. The demodulationmethod at the receiver side has to be strictly related to the Tx modulation.

Fig. 4. A simple list of both analog and digital modulation.

much data as possible into the least amount ofspectrum. We could speculate that ET, look-ing for the maximum efficiency, can transmit aCW signal in the analog domain and somethingsimilar to the OFDM (Orthogonal FrequencyDivision Multiplexing) in the digital domain as

shown in Tab. (5) where the spectral efficiencyfor a set of known modulation methods is re-ported (Frenzel 2012).

The digitally modulated signals occupy acertain bandwidth, as reported in Tab. 6 for areduced set of different types of common digi-


Fig. 5. A more detailed list of the digital modulation mode used to date.

Table 5. Spectral efficiency for the more common modulation methods.

Type of modulation Spectral Efficiency (bit/s/Hz)FSK <1 (related with the modulation index)GMSK < 1.35BPSK 1QPSK 28PSK 316QAM 464QAM 6OFDM >10 (depends on the type of modulation and

the number of carrier

tal signals. It is clear that it is difficult to detectand extract information from a digital commu-nication buried into the noise implementing asole spectrum analysis. This type of digital sig-nal could be transmitted by the aliens towardtheir spacecraft wandering in the space

Different types of modulation correspondto different bandwidths allowing a certain datatransmission velocity. This is expressed by theShannon Hartley equation that relates the max-imum capacity C of a given communication

channels with a certain noise and bandwidth.Such a capacity is given by:

C = Bw log2

(1 +

SN

)bit/sec (2)

where

– Bw is the bandwidth– S/N is the signal to noise ratio

Considering, for example, a ZigBee type signal(Bw = 5 MHz) and S/N ratio of 10, the capac-


Table 6. Band occupation for a set of modern digital signals.

Type of signals Channel bandwidthGSM 200 KHzCDMA (IS-95) 1.25 MHzCDMA2000 1.25 MHz (Channel bonding@1xEx-DO Rev.B;C)3GPP WCDMA 5 MHz3GPP TD-CDMA 5 MHz3GPP LTE 1.4, 3.5, 10, 15, 20 MHzWIMAX IEEE 802.16xxx 3.5, 5, 7, 8.75, 10, 20 MHzTETRA 25, 50, 100, 150 KHzWLAN & WIFI 20, 22 MHzDECT 1.728 MHzZigBee 5 MHzATSC 6 MHzDVB-H 5, 6, 7, 8 MHzT-DBM 1.536 MHzOFDM 1.25 to 20 MHz @128 to 2048 sub-carriers

ity of the channel is C = 17 Mbit/sec. This pa-rameter is important in the case the signal hasto be demodulated. The study of algorithms forthe computation of the entropy (or the agnos-tic entropy) of a very noisy information, couldplay an important role in the design of new pro-cedures for searching for information buried ina noisy band. We recall that Shannon helps tomodel virtually all the systems that store, pro-cess or transmit information in digital form,from compact disks to computers, from faxesto probes for deep space investigation! At themoment is not foreseen to find out the meaningof a code of an alien message. When this willbe required in the future, surely the difficultieswill increase in an unpredictable way!

2.4. What sensitivity do we need.

Having exactly set the antenna pointing di-rection, the frequency, the type of signal, thedemodulation algorithm etc it is necessary tohave the right sensitivity to receive the signal!Since we don’t know the minimum requestedvalue, we have to perform the observations atthe maximum available sensitivity. This is ob-tained through large radio telescopes and in thefuture by means of the large array. The sensi-tivity is the result of many factors that charac-

terize the radio link. Here following we dealwith main aspects of a hearth-exoplanet link inorder to compute the Pout at the receiving an-tenna terminal. To do this, we need to evaluatethe link budget taking into account all the gainsand losses (including the FSPL -Free SpacePath Losses-) being present in such a link. Ittakes the form of the equation below:

Rx power (dBm) = Tx power (dBm)

+∑

Gains (dB) −∑

Losses (dB) (3)

It is mainly a matter of accounting for all thedifferent losses and gains between the transmit-ter and the receiver, in other words:

– transmitter power– antenna gains (receiver and transmitter)– antenna feeder, cables, connector (receiver

and transmitter) and FSPL losses

Considering a given transmitted power levelPTX, the above reported parameters and thePout are here estimated in a practical example:an Earth-Trappist-1 (40 Ly) and Earth-Kepler-452 (1400 Ly) link assuming f=1.42 GHz, a300 m Tx and a 64 m Rx dishes antennas (65%efficiency).

Recalling the base block diagram of Fig. 1,we compute the antenna gains.


2.4.1. TX and RX antenna gains

The formula is reported in (4).

G(dB) = 10 log(η4πAcoll

λ2

)(4)

where

- η is the antenna efficiency- Acoll is the collecting area- λ is the wavelength

The Tx and Rx antenna gains are reportedin Tab. 7.

2.4.2. Losses and FSPL

This term is represented by the sum of all thelosses due to cables, connectors and other, in-cluding the FSPL (Free Space Path Loss). Asmentioned above, the distances are such as tointroduce an incredible attenuation of the sig-nal strength in its path from the exoplanet tothe Earth. Here following it is reported as thedistance d (m) and the frequency f (Hz) areparts of the FSPL formula (5):

FS PL = 20 log10(d) + 20 log10( f ) +

+20 log10(4πc

)(dB) (5)

The Free Space Losses results for the aboveconsidered practical example, are in Tab. 8:

At this point, recalling the formula (3), weare able to calculate the signal power Pout atthe receiver antenna output connector with theformula (6), assuming two levels of the trans-mitter power (1MW and 100 MW) for each ex-oplanet (Fig. 6):

Pout = PT X + GT X − LT X − FS PL − Lmisc −−LRX + GRX(dBm)(6)

where

- Pout is the receiver antenna output - PT Xis the transmitted power (dBm) - GT X is thetransmitter antenna gain (dB) - LT X is thetransmitter losses (coax, connectors...) (dB)- FSPL is the Free SPace Path Losses (dB)- Lmisc is the miscellaneous losses (fadingmargin, body loss, polarization mismatch andother losses) (dB) - LRX is the receiver losses(coax, connectors) (dB) - GRX is the receiverantenna gain (dB)

The results are reported in Tab. 9.The very huge path loss makes radio links

very difficult to establish over long distances,unless extremely low noise receivers, verypowerful ET transmitters and very large Tx andRx antennas are available.

We are now able the compute the maxi-mum Operative Range. To evaluate it we referto the Fig. 6 applying the Kraus formula (7).Providing here measurements in meters andwatts, the range is expressed in light years.

RangeLy = 10−16

√PT AeffTXAeffRX(S/N)kTsysλ2∆ f

(7)

where

- Range (Ly)- PT is the transmitted power (W)- AeffXX is the transmitter and receiver antennaeffective area (m2)- k is the Boltzman’s constant (J/K)- Tsys is the system temperature (K)- ∆ f is the frequency resolution (Hz)- λ is the wavelength (m)- (S/N) is the Signal to Noise ratioAs an example, in the Tab. 10 the rangeswith different antennas (Medicina and SRT),different frequencies (1.42 GHz and 8.5 GHz)and transmitter power of 100 MW, are reportedfor a 1 Hz signal.

The distances obtained, however, are smallin comparison to the extension of the wholegalaxy (100,000 light years radius).

2.5. Electromagnetic visibility interval.


Table 7. The gain for different diameter dish antennas.

Antenna Diameter Frequency Efficiency Gain(m) (GHz) % (dB)

Tx 300 1.42 65 71Rx 64 1.42 65 57

Table 8. List of the Free Space Losses in the case of Trappist-1 and Kepler planets.

Planet Distance Frequency Free Space Loss(km) (GHz) (dB)

Trappist-1 40 Ly 3.8E14 1.42 387Kepler-452 1400 Ly 1.3E16 1.42 418

Fig. 6. Radio link for the discussed example.

A civilizationcharacter-ized by anappropriatetechnology, inits evolution isvisible, froman electromag-netic point ofview, from the outer space for a certain time or,at least, this happened during our evolution!One of the first human appeared on earthabout 4 million years ago (Australopitecusafarensis). It is expected that our planet willbe completely cabled with optical fibers by theyear 2150. In this way, the radio visibility of

the earth from space will drastically decrease!Only radio transmissions for communicationwith spacecraft will remain active (Fig. 7).The period in which the humankind hasintensively emitted radio waves will be about200 years. Then a technologically advancedalien civilization would have seen the earthbright, from a radio point of view, only for 1 /20.000 of its evolution time, therefore a verynarrow time slot.

3. Some remarks.

After the analysis of the parameters that haveto be correctly set in order to be successful withthe program, it comes out that it is very difficult


Table 9. Pout in the TX (300 m) and RX (64 m) configuration.

Planet Frequency TX power Free Space Loss Pout(GHz) (MW) (dB) RX connector

(dBm)

Trappist-1 40 Ly 1.42 1 387 -183Trappist-1 40 Ly 1.42 100 387 -163Kepler-452 1400 Ly 1.42 1 418 -214Kepler-452 1400 Ly 1.42 100 418 -194

Table 10. Different range for a 300 m alien dish TX antenna with different radiotelescopes(Medicina and SRT), operating at 1.4 GHz with a 100MW transmission power.

Rx antenna on Alien Tx Frequency Power Tsys Chan RangeEarth antenna(m) (m) (GHz) (MW) (K) (Hz) (Ly)

Par. 32 m Medicina 300 1.42 100.0 60 1 476Par. 32 m Medicina 300 8.5 100.0 40 1 349664 m (SRT) 300 1.42 100.0 60 1 95464 m (SRT) 300 8.5 100.0 120 1 4036

Fig. 7. The planet will be probably completely cabled with optical fibers by the year 2150, then no moreradio activities except to/from spacecrafts.

to set them in the right way. The ”radio silence”we have had until now maybe means that theobservation parameters have been not correctlyset and not because they don’t exist. It could bepossible also that the alien civilization doesn’tuse radio techniques for local communicationsor to make themselves noticed and/or establishinterstellar communications.

Just to give a practical idea about thechances to correctly point the antenna, we con-sider to operate with the SRT antenna as a re-

ceiver and a similar antenna as a transmitter. Ina complete sky survey, it comes out one chancein 3 · 1010 to point our antenna in their direc-tion. So, a lot of work is requested in the fu-ture to increase such a very small probability,maybe changing the searching paradigm or thesearching instrumentation. In the next step ofSETI the new Square Kilometre Array (www.skatelescope.org), with its wide Field ofView pixeled by thousands of beams at asensitivity never reached before, will be ex-

www.skatelescope.org

www.skatelescope.org


ploited. The observation modes still now areSky Survey and/or Targetted. In both cases theradio transmissions from ET are expected to beeither intentional or not, in the analog domainor in the digital one. It is fundamental to knowthe domain of the signals because the detectionalgorithms are completely different from onecase to the other. In the analog domain searcheswill continue for intentional monochromaticsignals by means of high-frequency resolu-tion FFT working in parallel with possibleWavelets, Duffing Oscillators (optimal for non-linear environment) and Stochastic Resonancealgorithms. In the case of an unknown mod-ulated signals, the proposed transform is theKLT (Karhunen Loewe Transform) (Maccone2012) even if it seems to be affected by a poordetection sensitivity. Anyway, various works toincrease it have already been done with excel-lent results (Schilliro 2008; Maccone 2010). Todate, there is no KLT algorithms operating on-line with the observations at the national radiotelescopes (Medicina, Noto, Cagliari).

In the case of alien signals in the digitaldomain, the detection process becomes evenmore complicated. The final match for theidentification of an extraterrestrial radio signalwill be played, most likely, in the digital do-main. The algorithm for extracting an unknowndigital signal confused into the noise, will beone of the most difficult challenges ever facedin digital signals processing!

4. Future strategies.

The Medicina, Noto and SRT antennas,schematically reported in Fig. 8, are intercon-nected via a very wide band optical fibers. Thisnetwork could be considered a national longbaseline interferometer system, as a concept ofa new generation high performances SETI ra-diotelescope. Such a configuration could offerthe cancellation of the local interferences, forevery single antenna, during the real time cor-relation process. In addition, it could be an ex-tremely suitable radio interferometer for a highgain and high spatial resolution SETI candidatesignals confirmation.

It should represent the first large array inEurope performing so accurate SETI activities.

The future of the SETI program willbe based on the introduction of new re-search paradigms involving the use of SignalProcessing strategies completely different fromthose used to date (Tab. 11). As already men-tioned, the greatest effort must be devoted tothe detection of digital signals modulated inan unknown way. Probably an accurate anal-ysis of entropy could give us indications onthe presence or not of a signal buried in thenoise, even if a high sensitivity is not expectedfrom this method. Given the enormous amountof data acquired and analyzed, a way couldprobably be to discard the noise online andstore the remaining data. This aim could beachieved by introducing the ”Dyspoissonism”,a mode to estimate the randomness of a sig-nal (Russell 2017). From this point on, an ex-tremely new and very vast research scenarioopens up. Maybe in the future, we will facethe problem to decode the received signal. Atpresent, this is not planned because an answeris not foreseen. Answering means understandthe message, deciding who should transmit, atwhat title, what to transmit, how to transmit, onbehalf of whom to transmit, who will pay thelarge expenses for the transmission and then,not less important, the enormous power neededto cover the immeasurable distances, the radiowaves flight time, etc. ...

5. Conclusion.

Due to the huge difficulties to set the rightobservation parameters, SETI is an extremelycomplex program in which the probability ofsuccess is presently very low. Its added value isfor the possible technological and data process-ing (algorithms) feed back that could be usefulfor other scientific programs.

In the future, parallel to the search formonochromatic signals underway today, ajoint multidisciplinary approach will be neededamong the various SETI activities around theworld. The goal is to develop both new tech-nologies and design new algorithms to makethe observations more effective than those con-ducted until today, in the frame of new researchparadigms. Care has to be taken for the meth-ods based on entropy (Shannon) and similar


Fig. 8. The national radiotelescopes linked via a wide band optical fiber.

Table 11. Summary of the SETI past and future post processing.

Past Today Future

Algorithms FFT FFT /KLT/Wavelets - Stochastic resonance- Duffing Oscillator

- Agnentropy- New algorithms for digital

signal detection.

Processor DSP FPGA, Multicore CPU FPGA, Multicore CPU,GPU, Array of GPU

methods, within a desirable future (radio / op-tical) SETI program funded by INAF.

Up to now, no intelligent radio signals havebeen received in the course of the SETI ac-tivities but just man-made radio interferences!Continuing to investigate is a must becausemaybe up to now we have searched in the

wrong mode, in the wrong position, at thewrong time, at the wrong frequency, with notsufficient sensitivity and a not suitable signaldemodulation. We have to continue the searchbecause as the English astronomer MartinReese stated, ”the absence of the evidence isnot the evidence of the absence”. Most likely,


life in the Milky Way is a common event but,due to the incommensurable distances amongthe planets, probably every civilization spendsits own existence completely isolated and sus-pended somewhere into the galaxy as. . . .. if itwas really alone!

Acknowledgements. Special thanks are due to bothINAF President Prof. Nicola D’Amico and theScientific Director Prof. Filippo Maria Zerbi, forthe logistic and economic support for the SETI-INAF meeting, held at the INAF Headquarters on24 October, 2017. Thank also to Dr. Andrea Melis(INAF-Cagliari) for the continuous support duringthe organization of the meeting, the porting of sev-eral contributes from Microsoft-Word to LaTeX andthe formatting of the proceedings volume. Thankalso to Nicolo Antonietti for the support on theporting of some contributes from word to LaTex.Thank to the Medicina radiotelescopes colleagues,Dr. Jader Monari and Dr. Germano Bianchi.

References

Backus, P. R., Jordan, J. C., Harper, D. G.1992, Acta Astronautica, 26, 169

Bowyer, S., et al. 1988, The Berkeleypiggyback Seti program: Serendip 2, inBioastronomy - The Next Steps, ed. G. Marx(Kluwer, Dordrecht), IAU Colloquium, 99,363

Cocconi, G. and Morrison, P. 1959, Nature,184, 844

Drake, F. 1961, Physics Today, 14, 140Frenzel Mon, Louis E. 2012, Understanding

Modern Digital Modulation Techniques, inElectronic Design

Garber, S. J. 1999, Journal of BritishInterplanetary Society, 152, 3

Hickish, J., et al. 2016, Journal of AstronomicalInstrumentation, arXiv:1611.01826v1

Leidich, R. 2017, Introduction to EntropyTransforms, http://agnentropy.blogspot.it/

Maccone, C. 2010, Acta Astronautica, 67,1427

Maccone, C. 2012, Mathematical SETI.Statistics, Signal Processing, SpaceMissions (Springer-Verlag, Berlin)

Recommendation ITU-R SM.1600-1 -Technical identification of digital sig-nals, (09/2012), https://www.itu.int/dms pubrec/itu-r/rec/sm/R-REC-SM.1600-1-201209-S!!PDF-E.pdf

Schilliro, F., Pluchino, S., Montebugnoli, S.2008, La KL Transform: considerazionigenerali sulle metodologie di analisi edimpiego nel campo della Radioastronomia,Internal Report IRA/INAF 416/08

Werthimer D., Ng, D., Bowyer S., DonnellyC. 1995, The Berkeley SETI program:SERENDIP III and IV instrumentation, inProgress in the Search for ExtraterrestrialLife, ed. Seth Shostak (ASP, San Francisco),ASP Conf. Ser., 74, 293

http://agnentropy.blogspot.it/

http://agnentropy.blogspot.it/

https://www.itu.int/dms_pubrec/itu-r/rec/sm/R-REC-SM.1600-1-201209-S!!PDF-E.pdf



Documents

Introduction to the SETI program, criticalities and future ...sait.oats.inaf.it/MSAIt890318/PDF/2018MmSAI..89..316M.pdfIntelligence) program whose aim is to look for artiﬁcial radio