The invention belongs to the field of sonar-type acoustic devices that enable grazing-incidence observation of the seafloor. More particularly, the invention relates to sonars mounted on carriers which can be surface ships or submerged “fishes”, self-propelled or towed by a surface boat, moving along a so-called longitudinal direction X. Still more particularly, the invention belongs to the field of sonars that enable observation of a seafloor strip located directly below the carrier along the longitudinal direction. The sonar can be the only imaging system on board of the carrier where it can be coupled with other sonars and in particular lateral sonars. In the latter case, the strip to be observed is located between two lateral areas observed by means of lateral sonars arranged on each of the carrier's flanks.
Lateral sonars does not allow to observe the seafloor under a sight angle lower than 30° relative to the vertical running through the carrier and following a plane perpendicular to longitudinal direction X. Consequently, the seafloor strip located under the carrier's belly cannot be observed during the carrier passage.
To obtain data about this intermediate strip by means of carriers having only lateral sonars, other passages have to be done along navigation trajectories parallel to a first trajectory, so that the strip not observed during the first passage is now “visible” by means of the lateral sonars.
Of course, resorting to other passages to obtain a seafloor image without “holes” requires additional time. And this time has a cost because it corresponds to the operation of the acquisition device, equipment and crew as a whole.
Moreover, a greater accuracy of the final seafloor map comes from the correlation made between different images of the same site. Now, possessing no redundant images, or possessing some but in a fewer number, reduces the interest of subsequent data processing for image correlation. Therefore, the final image, for example a shadows-and-echoes image, has a lower resolution.
Consequently, there is a need for a carrier-borne sonar device enabling sonar data acquisition relating to the strip located under the carrier, strip which is not observed by other sonar means.
Of course, there exist carriers provided with multibeam sounders which are acoustic devices of another type because they enable normal-incidence observation of the seafloor. But the informations collected by theses devices do not correspond to sonar-obtained informations. If sonars mainly enable shadows-and-echoes images to be obtained, sounders mainly enable to obtain topographic or bathymetric images.
Further, it has been proposed to arrange in front of the carrier sonars that take up the technology used by lateral sonars. Such frontal sonars are thus provided with a composite physical antenna consisting of a row of transducers in alignment with the antenna length. The antenna is so arranged on the carrier to be horizontally oriented, with the length thereof coinciding with lateral direction Y of the carrier. Hence, the frontal sonar antenna is arranged perpendicular to the lateral sonar antennas.
Now, it is known that resolution of such a sonar is given by the ratio of working wavelength to physical antenna length. The carrier must be hydrodynamic, so it is easy to arrange antennas with great linear dimensions on the carrier's flanks, but the available space in the width direction of the carrier prevents great-length antennas to be arranged in front of the carrier. Consequently, physical antennas of frontal sonars according to the prior art are small-sized. If physical antennas of lateral sonars can be generally of the order of 2 m long, frontal sonar antennas are rather of the order of 50 cm long.
Having only short antennas for frontal sonars leads to resolution degradation. In particular, the resolution of images obtained with a frontal sonar is much lower than the resolution of those obtained by the lateral sonars of the same carrier.
An order of magnitude is given by a lateral sonar the physical antenna of which has a length of 2 m and an acoustic working frequency of the order of 100 kiloHertz (kHz). Such a sonar enables a resolution of 15 cm at a maximal range of 300 m to be obtained.
It follows that, even if data about the central strip are acquired during the first passage, it will be necessary to repeat parallel passages to map this strip with a resolution close to that of the lateral areas observed by the lateral sonars. Finally, it almost ends up in the case of a carrier having no frontal sonar.
To increase the resolution of a physical antenna having a limited length, if only for reasons of carrier hydrodynamics, it has thus been proposed to increase the sensor number of the antenna. For example, the COSMOS project developed by IFREMER is a frontal sonar. It comprises, for the emission, 125 emitters arranged in an arc of circle and, for the reception, a 70-cm-long antenna consisting of two rows each comprising 32 transducers. Such a frontal sonar operating at 100 kHz enables a resolution of 160 cm at a range of 80 m to be reached.
The drawbacks of such a solution are known. Firstly, each sensor has to be fitted with electronics arranged near the sensor. It follows that the sensor density is limited by the steric space requirement of the associated electronics. The channel number of the antenna is thus limited. Further, the cost of the sonar antenna increases with the number of sensors used. Finally, the data flow to be processed is significant and the acquisition chain has to be particularly efficient.
Therefore, there is a need for a frontal sonar having a resolution identical or anyway close to those of the lateral sonars used, but having a moderate number of transducers.
To that end, an object of the invention is a frontal sonar for observing the seafloor, comprising: means for insonifying a seafloor area divided into M sectors, suitable for performing a characteristic insonification of each sector of the area during an operation recurrence of said sonar; and receiving means comprising a physical antenna consisting of an array of K transducers, the antenna being lacunar and thus ambiguous, so that it defines N image lobes. Each of the characteristically insonified sectors is associated with a single image lobe of the antenna, so that the antenna ambiguity is removed.
Preferably, insonification means comprise as many emission channels as there are sectors to be characteristically insonified.
In the presently contemplated embodiment, the emission channels insonify four sectors respectively, and four useful lobes among the N image lobes of the antenna are associated with the four sectors respectively.
In an embodiment, during a sonar operation recurrence, each of the emission channels emits around an average frequency chosen in a list of average frequencies, each emission channel emitting at a frequency different from that of another emission channel, the average frequency characterising the insonification of a particular sector.
In a variant, during a sonar operation recurrence, each of the emission channels emits around a common average frequency, each of the emission channels insonifying the sector associated therewith at different instants of the recurrence, the emission instant characterising the insonification of a particular sector.
Preferably, the sonar according to the invention allows a resolution lower than 50 cm for a range of 80 m. Still preferably, it allows a resolution of 40 cm for a range of 80 m.
In the presently preferred embodiment, the lacunar antenna is 0.7 m long, comprises 48 transducers and operates at a working frequency of 400 kHz.
Another object of the invention is a carrier provided with a frontal sonar according to the invention.
Preferably, said carrier is submerged and further comprises two lateral sonars, the frontal sonar enabling to obtain a sonar image having a resolution compatible with the resolution of images obtained by the lateral sonars.
Still another object of the invention is a method of observing a seafloor strip by means of a sonar. The method is characterized in that, the sonar being provided with a lacunar antenna consisting of an array of K transducers defining N image lobes, it consists in:
Preferably, during the insonification step, each of the sectors is insonified by means of a sound signal having a characteristic average frequency.
In a variant, during the insonification step, each of the sectors is insonified by means of a sound signal emitted at a characteristic instant of the recurrence.
Still preferably, the channel forming step is performed over K recurrences of the insonification and acquisition steps in order to synthesize an antenna.
The principle of the frontal sonar according to the invention consists in using a receiving antenna that is lacunar, and consequently ambiguous, and removing the ambiguity on the image lobe in which the source of echo is actually located by insonifying each seafloor sector angularly corresponding to an image lobe with a characteristic sound signal.
Advantageously, the limitation on the sensor number is circumvented using a lacunar antenna while removing ambiguity of this antenna thanks to a sectorized emission.
In the presently preferred embodiment, during a sonar operation recurrence, the sound emissions intended to “light” each of the sectors are performed simultaneously but at distinct respective frequencies. The image lobe is thus characterized by a particular frequency. In a variant, the sectorized emissions can be performed at the same frequency but at successive instants. The image lobe is then characterized by the instant of emission of the sound signal. In still another variant, orthogonal acoustic signals are used. Each orthogonal signal then characterizes a single image lobe.
In the contemplated application, as the frontal sonar goal is to complete the map provided by the lateral sonars, tendency would rather be to use a working frequency and a grazing close to those of the lateral sonars. But this cannot be respected if it is desired to approach the required resolution performances. Indeed, the resolution being proportional to λ/L and the length L of the antenna being necessarily reduced the working wavelength λ has to be reduced too, and thus the working frequency f has to be increased. Advantageously, in the antenna according to the invention, the resolution is improved by increasing the working frequency, which can so be raised to 600 kHz, this working frequency having to be compared to the 100 kHz of the lateral sonars. As a result, the frontal sonar range is reduced. Thus, it is necessary to work at a lower incidence.
Advantageously, obtaining two images at different incidences through the frontal sonar and a lateral sonar offers a gain of information. It is also interesting for the registration operation, which is the referencing of an image relative to the other or relative to an absolute reference.
The invention will be better understood and other purposes, details, features and advantages thereof will become more clearly apparent from the description of a particular embodiment of the invention which is given merely by way of illustrative and non-limitative example with reference to the appended drawings. In these drawings:
A carrier 1, self-propelled or towed, is submerged so as to be positioned at an altitude H of about 30 m relative to seafloor 2. Carrier 1 moves substantially along a navigation trajectory parallel to longitudinal axis X thereof. The direction perpendicular to both direction X and direction H is a lateral direction Y.
Two lateral sonars (not shown) arranged on each of the flanks 1a and 1b of carrier 1 enable a progressive observation of the lateral areas, 2a and 2b respectively, during the displacement of the carrier.
Between these areas observed by the lateral sonars, there is a strip of land of width W that is not observed by the lateral sonars which cannot work at a sight angle lower than 30° relative to the vertical of the carrier in lateral direction Y.
So, the matter is to map a seafloor strip located under the carrier and which, when the latter is at an altitude of 30 m relative to the floor, corresponds to a strip having a width W of about 45 m.
Carrier 1 according to the invention is provided with a sonar 10 arranged in front of it. The goal of frontal sonar 10 is to emit a sound wave able to “light” an insonified area E of seafloor 2, and to “listen to” an imaged area I included in insonified area E. During displacement of carrier 1, acquisition of a succession of imaged areas enables gradual image reconstruction of the central strip not observed by the lateral sonars.
Constraints on the antenna are the following. The reasonable maximal span of the antenna arranged along lateral direction Y is 0.7 m. Taking into account the steric hindrance of acquisition electronics, the number of channels available for the reception is of the order of 48. The working frequency has to be lower than 600 kHz because, beyond this maximal frequency, the effects of signal absorption by the body of water become too large.
Most of the lateral resolution dY depends on the receiving antenna. Actually: dY=λ×R/L=λ×R/(Ni×d), where Ni is the number of sensors and d is the inter-sensor pitch. Given these two parameters, the resolution will be entirely determined by the range and the frequency.
At the present time, the lateral resolution actually reached by an experimental frontal sonar made up according to the described embodiment is of 40 cm at a range of 80 m, with a working frequency of 400 kHz.
By reference to
This lacunar antenna 11 enables to observe a source M under an angle θ in the plane XY. It is considered that the wave emitted by source M is a plane incident wave at antenna 11. Acoustic signals respectively received by each of the transducers are thus identical, to within a temporal phase shift. The matter is then, acting on coefficients τ1 to τK, to correct this phase shift to reconstruct the incident plane wave.
It is shown that the ratio of inter-sensor pitch d (d=L/K) to wavelength λ of the chosen working frequency is approximately equal to four. This means that antenna 11 is widely under-sampled. This antenna is thus a lacunar antenna having image lobes in reception.
The fact that the pitch d separating two successive transducers k and k+1 is larger than wavelength λ entails uncertainty on the phase shift value between these two transducers: the phase shift is known modulo π. Thus, for a same signal detected by the lacunar antenna, source M may be located at different angular positions of the field observed. There is thus an ambiguity about the value of angle θ under which an echo coming from source M is “viewed” by lacunar antenna 11.
Therefore, homogeneously insonifying imaged area I gives the graph of
According to the invention, instead of uniformly insonifying the whole area to be imaged I, a sectorized insonification is performed. The insonified area E is firstly divided into M sectors E1-EM, and then each sector Ei is associated with a particular useful image lobe. Considering that four image lobes are enough to observe the area to be imaged, it is chosen to divide the insonified area E into four sectors E1-E4 (M=4).
The insonification means according to the invention comprise as many emission channels as there are sectors to be insonified. In this case, the sonar according to the invention comprises four emission channels consisting of a transducer operating in emission and electronics for operating this emitter.
The emitters used are directional, i.e. they emit in an emission cone of angle θe chosen so as to insonify a single sector of the seafloor (the angular aperture of which corresponds to the area to be imaged, divided by the number of emission sectors).
In the presently preferred embodiment, sectors E1-E4 are insonified at the same instant but by sound signals having different characteristic frequencies. As a result, it is known how to associate an echo of a particular frequency with a sector, thus a lobe of the antenna, i.e. at an interval of values of observation angle θ. Ambiguity of antenna 11 is thus removed.
In another embodiment of the invention, the sectors are insonified successively in time by a sound signal of a given frequency. The temporal correlation between the echo receiving instant and the emission instant enables once again to associate the echo with a sector, thus with a particular image lobe, i.e. an interval of values of observation angle θ, and then to remove ambiguity of antenna 11.
Possibly, in still another embodiment variant of the invention, these two types of sectorized emission can be combined: for example, by performing a synchronised insonification at different frequencies in a first sonar operation recurrence, and a not-synchronized insonification at a given frequency in the next recurrence.
To insonify a sector, a signal is tried to be produced, centred on an average frequency fi and with a spectral width the closest possible of a 6.25-kHz-wide rectangular function. It will be noticed that the insonification frequencies must be lower than 600 kHz because, beyond this frequency, the propagation effects in the body of water are significant while being close to this upper limit, to obtain high resolution. Moreover, the series of different average frequencies chosen must be within an interval of about 50 kHz corresponding to the detection bandwidth of antenna 11.
According to the example shown in
On the other hand, in
Study of array-antenna is well documented and reference is made to the course given at the ENSIETA, entitled “traitement d'antenne” (“antenna processing”), by Valérie MAZAURIC, 2003 (ref. 1335 DE/SCO-S3-RE/HY).
Choosing a lacunar antenna entails consequences on the antenna contrast function which is in principle degraded relative to that of the adjacent antenna. Contrast is the ratio of the signal received from a shadow area on the main lobe (namely, the noise alone) to the contribution of the other interfering lobes (secondary lobes in emission, array lobe in reception). Concerning these questions on the notion of contrast, reference will be made to the book “Sonars et acoustique sous-marine”, Jean-Paul Marage, edited by Thomson-Sintra.
In other words, the contrast optimization parameter is inter-sensor pitch d because the latter defines the angular spacing between two successive lobes.
The criterion retained to optimize contrast of the sonar according to the invention is an inter-sensor pitch d that enables an emission/reception level at maximal misalignment equal to the amplitude of a conventional secondary lobe in emission.
Though the invention has been described above by reference to a particular embodiment, the invention is not limited to this embodiment and includes all technical equivalents to the described means as well as their combinations that are within the scope of the invention.
In particular, the frontal sonar according to the invention can be used for the antenna synthesis, coherent or not coherent. The matter is to correlate the K transducer signals over a plurality of sonar operation recurrences.
Number | Date | Country | Kind |
---|---|---|---|
06 51763 | May 2006 | FR | national |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/FR2007/051273 | 5/15/2007 | WO | 00 | 4/22/2009 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO2007/132125 | 11/22/2007 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
4970700 | Gilmour et al. | Nov 1990 | A |
5177710 | Gilmour et al. | Jan 1993 | A |
5412618 | Gilmour | May 1995 | A |
5596549 | Sheriff | Jan 1997 | A |
5598163 | Cornic et al. | Jan 1997 | A |
20050099887 | Zimmerman et al. | May 2005 | A1 |
20090231957 | Jean et al. | Sep 2009 | A1 |
Number | Date | Country |
---|---|---|
0 568 427 | Nov 1993 | EP |
2 688 894 | Sep 1993 | FR |
WO 2007132125 | Nov 2007 | WO |
Number | Date | Country | |
---|---|---|---|
20090231957 A1 | Sep 2009 | US |