CALIBRATION

BACKGROUND

[0001] This application claims the benefit of priority of U.S. Provisional Application No. 62/909,849 titled “ADAPTIVE ACOUSTIC CHAMBER AND METHOD FOR ACOUSTIC CALIBRATION” filed October 3, 2019, the contents of which are incorporated herein by reference in their entirety.

[0002] The invention relates to the field of acoustic chambers.

[0003] The manufacture of underwater acoustic equipment is a fast-growing industry. This industry supplies both acoustic sources and acoustic receivers for a variety of products. This includes sonar systems, underwater localization instruments, recorders for marine exploration, sensors for measuring acoustic noise, and systems for monitoring operation of underwater vehicles.

[0004] Testing and calibration is a necessary step during the production of any underwater acoustic equipment. For acoustic transmission, calibration is performed by accurately measuring the sound level. For reception, calibration involves measuring the acoustic receiver’s response for reception of signals emitted by an acoustic source. In both cases, the calibration involves transmission and reception at various carrier frequencies and from different incidence angles.

[0005] Following is a brief discussions of the basics of acoustic propagation. Every propagation medium is characterized by its density p and the speed of sound in the medium, denoted as c. Their product Z = pc is called acoustic impedance: it can be interpreted as a measure of how hard it is for the sound to propagate in the medium and is measured in Rayls, where 1 Rayl is equivalent to 1 kg ^. m ^{-2 .} s ^-1 . In general, sound waves propagating through a medium of impedance Z ₁ = p ₁ c ₁ and encountering an interface with a second medium of impedance Z ₂ = p ₂ c ₂ are partly reflected off the interface and partly refracted inside the second medium. The reflection coefficient that determines the fraction of the wave energy reflected back into medium 1 is derived as where θ _i is the angle of incidence of the wavefront on the interface between the first and the second medium, and θ _t is the angle of refraction inside the second medium, both measured with respect to the direction normal to the interface. The coefficient T ₁₂ = 1 — R ₁₂ determines instead the amount of energy transferred to medium 2. The value — 20log ₁₀ R ₁₂ (in dB) is called echo reduction, and conveys the capability of the material to attenuate echoes originating from the interface.

[0006] Every propagation medium can absorb the energy of the acoustic wave in different ways. Consider a second interface between the first medium and a third medium. The fraction of energy transferred to medium 3 can be computed as T ₁₂ T ₂₃ /A ₂ , where A ₂ conveys the amount of attenuation incurred by the sound wave as it propagates through medium 2. The value — 20log ₁₀ T ₂₃ (in dB), is also called insertion loss, and expresses the capability of medium 2 to retain acoustic energy instead of transmitting it to medium 3.

[0007] The foregoing examples of the related art and limitations related therewith are intended to be illustrative and not exclusive. Other limitations of the related art will become apparent to those of skill in the art upon a reading of the specification and a study of the figures.

SUMMARY

[0008] The following embodiments and aspects thereof are described and illustrated in conjunction with systems, tools and methods which are meant to be exemplary and illustrative, not limiting in scope.

[0009] There is provided, in accordance with an embodiment, an acoustic chamber comprising: an enclosure that is internally coated with a flexible coating, the flexible coating comprising one or more phono-absorbing materials; an actuator configured to exert pressure on and change the geometry of the flexible coating; and a computerized controller configured to operate the actuator, such that acoustic reflection properties inside the enclosure change as a function of the changing geometry of the flexible coating. [0010] In some embodiments, the actuator comprises: an inflatable-deflatable reservoir functionally associated with the flexible coating, wherein inflation and deflation of the reservoir change the geometry of the flexible coating.

[0011] In some embodiments, the actuator comprises: a camshaft functionally associated with the flexible coating, wherein rotation of the camshaft causes one or more cams of the camshaft to push the flexible coating, thereby changing the geometry of the flexible coating.

[0012] In some embodiments, the enclosure is configured to be filled with a fluid. In some embodiments, the fluid is a liquid. In some embodiments, the liquid is water.

[0013] There is provided, in accordance with another embodiment, a method for calibrating acoustic transmission equipment using the acoustic chamber of any one of the above embodiments. The method comprises: positioning an acoustic transmitter and an acoustic receiver inside the enclosure; transmitting an acoustic signal from the acoustic transmitter to the acoustic receiver; utilizing the computerized controller to calculate, based on the acoustic signal as received by the acoustic receiver, a required change to the geometry of the flexible coating; utilizing the computerized controller to operate the actuator so as to affect the required change to the geometry of the flexible coating; and calibrating at least one of the acoustic transmitter and the acoustic receiver.

[0014] In some embodiments, the method further comprises repeating the transmission, calculation, and operation of the actuator iteratively, to estimate one or more stable acoustic paths between the acoustic transmitter and the acoustic receiver, wherein the calibrating is based on the estimated stable acoustic paths.

[0015] In some embodiments, the acoustic signal is of different wavelengths in at least some of the repetitions.

[0016] In addition to the exemplary aspects and embodiments described above, further aspects and embodiments will become apparent by reference to the figures and by study of the following detailed description. BRIEF DESCRIPTION OF THE FIGURES

[0017] Exemplary embodiments are illustrated in referenced figures. Dimensions of components and features shown in the figures are generally chosen for convenience and clarity of presentation and are not necessarily shown to scale. The figures are listed below.

[0018] Fig. 1 shows channel impulse responses for three frequency bands taken in an acoustic tank.

[0019] Fig. 2 shows a wall pattern of the present disclosure in two intermediate configurations: (A) nearly open and (B) halfway closed.

[0020] Fig. 3 shows an experimental prototype of an adaptive acoustic tank which includes a single adjustable plane whose shape is automatically adjusted by a controller.

[0021] Fig. 4A shows a unit structure of right and left interconnected rhombi plates.

[0022] Fig. 4B shows a neoprene sheet configured to be applied to the unit of Fig. 4A.

[0023] Figs. 5A-5B are illustrations of the acoustic chamber with the absorption plane: fully open and fully closed.

[0024] Fig. 5C is a view of the shape illustrating the height of the shape H and the piston extension Z.

[0025] Fig. 6A is a detailed dimensions of rhombi plates and their relative position.

[0026] Fig. 6B shows the height of the shape (H) as it relates to the motion of the motor (Z) described with reference to Fig. 5C.

[0027] Fig. 7 shows channel impulse responses for two configurations of the phonoabsorption plane: stretched by 20% (panel A) and by 60% (panel B).

[0028] Figs. 8A-8C show convergence of the optimization algorithm for adjusting the absorption panel. Signal emitted at a frequency range of 20-30kHz (panels A) and 35-40kHz (panels B).

[0029] Fig. 9 shows channel impulse responses concatenated as a function of optimization call number for frequency ranges 20-30 kHz and 35-40 kHz. [0030] Figs. 10A-10C show channel impulse responses concatenated in a waterfall form for three different frequency ranges.

DETAILED DESCRIPTION

[0031] Disclosed herein is an adaptable acoustic chamber and a method for calibrating acoustic transmission equipment in the chamber. Advantageously, the adaptable acoustic chamber can change its acoustic reflection properties, enabling the calibration of acoustic transmission equipment under different conditions, at different wavelengths, etc.

[0032] In an exemplary embodiment, the acoustic chamber is an enclosure that is internally coated with a flexible coating made of one or more phono-absorbing materials. An actuator is also included in the acoustic chamber or be associated with it, and configured to exert pressure on and change the geometry of the flexible coating. Finally, a computerized controller is also included in the acoustic chamber or is associated with it, the controller configured to operate the actuator such that acoustic reflection properties inside the enclosure change as a function of the changing geometry of the flexible coating.

[0033] The actuator may include, for example, an inflatable-deflatable reservoir that is functionally associated with the flexible coating. For example, the reservoir may be in contact with the flexible coating. Inflation and deflation of the reservoir change the geometry of the flexible coating.

[0034] One example of an actuator is a camshaft functionally. Rotation of the camshaft causes one or more cams of the camshaft to push the flexible coating, thereby changing the geometry of the flexible coating. Other suitable actuators to apply force to the flexible coating, by pushing and/or pulling the coating, will become apparent to those of skill in the art.

[0035] The enclosure may be filled with one or more different fluids, such that the acoustic chamber is capable of calibrating equipment intended to be used underwater, in the atmosphere, or in any other different medium. To calibrate underwater equipment, the enclosure may be filled with a fluid, such as water. To calibrate atmospheric equipment, the enclosure may be filled with a gas, such as air. [0036] Further disclosed herein is a method for calibrating acoustic transmission equipment using the acoustic chamber. The method may include the positioning of an acoustic transmitter and an acoustic receiver inside the enclosure. Then, an acoustic signal is transmitted from the acoustic transmitter to the acoustic receiver. The computerized controller is utilized to calculate, based on the acoustic signal as received by the acoustic receiver, a required change to the geometry of the flexible coating. The computerized controller then operates the actuator so as to affect the required change to the geometry of the flexible coating. Finally, at least one of the acoustic transmitter and the acoustic receiver are calibrated.

[0037] The steps of acoustic signal transmission, calculation of changes to the geometry of the flexible coating, and operation of the actuator, may be repeated iteratively, to estimate one or more stable acoustic paths between the acoustic transmitter and the acoustic receiver. The calibration is therefore based on the estimated stable acoustic paths. The iterations may include: setting a certain structural configuration of the chamber (i.e., by operating the actuator to set a certain shape of the enclosure); calculating a ratio between (a) the power of the direct path between the transmitter and the receiver and (b) the power of all stable multipaths; and repeating the two previous steps until convergence is reached.

[0038] When repeating these steps, the acoustic signal may be of different wavelengths, different bandwidths, and or different durations.

[0039] Typically, the calibration is performed by setting an acoustic emitter (projector) besides an the acoustic receiver (hydrophone) in an acoustic chamber. To allow accurate calibration, the measurement of the acoustic sound level must be performed only for the direct pass from the projector to the receiver. Otherwise, sound reflected from the chamber’s walls, floor, or surface, would be superimpose thereby distorting the received signal.

[0040] To allow this, past solutions performed the calibration in a very large chamber on the order of tens of square meters, such as the one employed by the NATO CMRE (Centre for Maritime Research and Experimentation) facility. This way, short acoustic transmissions traveling in the direct path can be easily separated from other reflections. However, having a larger acoustic chamber is a very costly operation which requires both large space, and expensive maintenance operation. Other past options were, for example, the system described in [19] where tests were performed in a wave flume, which albeit being smaller than a pool, still requires a large facility for its installation. The works described in [17] and [6] were both tested in tanks of different sizes but suffer from multiple reflections. A different approach was taken in [14], where the authors tested acoustic transmission in air by employing a specifically constructed emulator, where phono-absorbing material allows to obtain the same attenuation as a 2-km free-space underwater link within a space of 4 cm. The solution does not rely on actual acoustic transmission and can thus represent only an intermediate step towards accurate calibration.

[0041] With reference to the choice of a phono-absorbing coating material for the present invention, it is noted that there are four propagation media involved: 1) fluid inside the enclosure (tank), such as water; 2) the coating material; 3) the tank walls; 4) fluid outside the enclosure, such as air. There are therefore three interfaces that acoustic waves interact with. The clear water typically used for calibrating tests has an impedance Z _l = 1.48 MRayl; the tank walls are made of high-density polyethylene, which has an impedance Z ₃ = 2.3 MRayl, or of glass walls with an impedance of Z ₃ = 10 MRayl; finally, air has an impedance Z ₄ = 429 Rayl. Of course, water and air are in direct contact at the water surface. Hence, to achieve a good reflection reduction, one needs at least to apply a coating material to the tank walls, and optionally to the water surface/air interface. This material should have the following characteristics: it should be waterproof and exhibit an impedance matching the one of the water and possibly being not too different from the impedance of the tank walls; it should provide good sound absorption properties, high echo reduction and small insertion loss; it should be low cost and possibly off-the-shelf; and it should be thin enough not to take up too much space in the tank.

[0042] In [12], incorporated herein by reference in its entirety, one of the inventors participated in performing a survey of the properties of different phono-absorbing material, and organized them according to their advantage and disadvantage. This list is given in Tables 1-3 below.

[0043] The coating material used in acoustic chambers is designed according to the properties of the emitted acoustic signals. Specifically, the carrier frequency of the signals and the expected emitted sound level. These properties determine the type of the coating material, its thickness, and its shape. This means that the configuration of the acoustic chamber should fit the equipment tested. Naturally, this makes the calibration effort extremely challenging, and companies either use different acoustic chambers to test different acoustic products, or change the coating material of the chamber. Instead, the adaptive acoustic chamber of the present invention can adaptively change its characteristics according to the acoustic signal emitted.

[0044] Based on [12], a low-cost acoustic tank may be constructed. A plastic tank may be used, coated it with a 7mm layer of neoprene. The acoustic properties of this tank may be tested by transmitting acoustic communication signals at a frequency of 18-34kHz using acoustic modems. The measured communication integrity for each multipath arrival may be tested over time as a measure for the tank's quality of reflection mitigation. This is because the higher the integrity, the stronger the path power.

[0045] Experiments have shown that that even for a small tank, the neoprene coating yields a good reflection absorption capability. The dynamics and vibration of an orientation motion platform of a sphere actuated by omnidirectional wheels were analyzed. The work included a vibration and sensitivity analysis. This analysis served as a design tool for the construction of a mobile platform with unlimited rotational motion, as in embodiments of the present invention.

[0046] The flexible coating in the chamber of embodiments of the invention may include one or more of the phono-absorbing materials discussed in [12].

Table 1: General properties of different classes of phono-absorbing materials. Table 2: Advantages and disadvantages of specific phono-absorbing porous and visco-elastic materials. Table 3: Advantages and disadvantages of specific phono-absorbing composite materials.

* Indicates that the corresponding characteristics are as indicated in Table 3.

[0047] Embodiments of the present invention may include an acoustic chamber whose phono-absorbing material is shaped to allow reflection reduction for various carrier frequencies. The acoustic chamber may produce a reduction of at least 10 dB in level of reflecting signal.

[0048] Embodiments further include an algorithm that measures the acoustic attenuation level from the sides of the acoustic chamber, as well as a closed-loop control mechanism that adaptively changing the properties of the acoustic chamber based on the emitted acoustic signals to maximize the absorption properties of the acoustic chamber. [0049] Present embodiments are based on the notion that the performance of an acoustic chamber depends not only on the selection of a proper phono-absorbing material, but also on its geometry. Different wave frequencies require different geometries for optimal reflection reduction of the acoustic chamber. Thus, the ability to easily change the geometry of the phono-acoustic material for different tests is key. The adaptive acoustic chamber is designed such that the shape of the phono-acoustic material can be controlled externally. Since the phono-acoustic material is elastic, applying force or pressure on it can change its geometry. In various embodiments of the invention, different techniques may be used to apply the aforementioned pressure. A) Inserting air tubes that can be inflated using air pressure, thus, inflating the tubes increases their volume, which, in turn, changes the shape of the phono-acoustic material. B) Inserting eccentors (such as in a camshaft mechanism) in strategic areas of the phono-acoustic material, and by rotating them, achieve the desired shape for the phono-acoustic material. The desired pressure will be determined based on measurement of the wave frequency of the transmitter inside the chamber, and active control can be achieved through repeated transmission, measurement, and feedback.

[0050] The channel impulse response from the transmitter to the receiver inside the acoustic chamber highly depends on the locations of the transmitter and receiver. Inside a small chamber, even a movement of a few tens of centimeters can change the response. Multiple path arrivals and fast time variations may be observed. To manage the expected strong multipath inside the tank, the mechanism to evaluate the acoustic power attenuation in the chamber involves the identification of the different path arrivals from the transmitter to the receiver. This way, it is possible to evaluate from which side of the tank power is reflected well, and where additional mitigation is needed. To separate between path arrival, chirp-modulated signals of high bandwidth W and long duration T may be transmitted, and a matched filter during processing may be used. Since the processing gain of a matched filter is WT, the signal-to-noise-ratio (SNR) at the output of the filter will be high. Then, for a channel impulse response h(t) and a transmitted signal s(t), it may be can approximated: where * in the convolution operation, and S(t) is the Kronecker delta function. The result h(t) is an estimate of the channel h(t). [0051] The power attenuation may be measured only for stable path arrivals. Otherwise, the control system would not converge. Obtaining multiple measurements of path classification may be performed. To that end, all estimations may be stacked in a single time-delay image, and identify stable paths by detecting lines of reflections in time. Finally, the power attenuation may be measured over only these identified stable paths.

[0052] The adaptive chamber of the invention has the potential to reduce operational costs and operational complexity for testing and calibrating of acoustic transmitters and receivers.

[0053] In some embodiments, the present disclosure may provide for a novel design of an acoustic tank whose absorption characteristics are made adjustable. In some embodiments, this is achieved by changing the shape of the absorption plates with a single degree of freedom. Performing channel estimation between a projector and hydrophone for a required frequency range, the adjustment of the plates involves an iterative genetic -based optimization that finds the best Pareto front to maximize reflection mitigation. Results from an experimental setup show a fast convergence of the tank to the best reflection suppression for frequency ranges of 20-30kHz and 35-40kHz. The result is an acoustic tank that adapts to various tested frequency ranges and setup of sensors, and can thus improve productivity of acoustic sensor calibration and testing in an efficient manner.

[0054] As detailed above, underwater acoustic tanks are a necessity for the testing of any underwater acoustic equipment, both for calibration procedures and for factory acceptance tests. In underwater acoustic research, tests in acoustic tanks are also used to validate algorithm performance.

[0055] Almost every provider of acoustic equipment, and any sizable research institute performing research and design of acoustic systems, holds at least one acoustic tank. The design of these tanks varies much — from deep large pools with instruments allowing the exact positioning of equipment in water as in the NATO CMRE facility, small plastic chambers, and moderate sized tanks with automatic shifting of equipment. These tanks are designed per operation. For example, for measuring the directional sensitivity of acoustic hydrophones by transmitting from a calibrated projector and carefully turning the tested hydrophones around their axis, full scale experimentation by placing whole systems in the pool and measuring performance over time, or for operational pre-run checks before going out to sea.

[0056] The operations within the acoustic tank include measuring signal distortion levels, evaluating source levels of transmitter, estimating receiver sensitivity, and emission and recording of signals for analysis. For these operations, very much like a quiet room for audio recordings, a good acoustic tank would mitigate reflections from its sides, floor, and surface such that, ideally, only the direct path between a receiver and transmitter exists. Otherwise, due to the tank’s small dimensions, reflections may distort the signal making calibration attempts non-feasible. Besides building a large acoustic pool, which is an extremely costly facility, current solutions to reduce the reflection level include coating the acoustic tank with a phono-absorption material and designing the tank structure to include sharp edges that may capture acoustic reflections by inducing large incidence angles. The choice of the solution trades off between absorption capability, maintenance of material, and ease of implementation over the acoustic tank.

[0057] Different phono-absorption materials can be divided among (i) porous material that trap the signal in pores, but whose absorption properties degrade over time; (2) high-density visco-elastic materials that hold good impedance match to water, but may be sensitive to hydrocarbons and are thus hard to maintain, and (iii) composite materials of high absorption characteristics which may be hard to implement over glass-made or plastic acoustic tanks. The shape of the phono-absorption material determines greatly the reflection attenuation level. Common structures are triangle shapes, tilted planes, or crushed rocks. The attenuation performance is measured by evaluating the reflection within the channel impulse response function. In particular, the length of the channel, the number of channel taps, and the attenuation level of the direct path and its separation level from the nearest multipath.

[0058] A common disadvantage of all methods is the static design of the acoustic tank. Specifically, a certain solution, e.g., a phono-absorption material or a design of absorption planes glued over the tank’s sides, may allow a low reflection level for a certain frequency range and position of the acoustic projector and receiver for low reverberation level. However, testing over different carrier frequencies or in different locations within the tank may suffer from high reflection distortion. An example for such variation in reflection level is shown in Fig. 1, where a channel impulse response estimated for a 2 X 1 X 1 [m ³ ] glass tank is shown for three different frequency bands, taken in an acoustic tank for the same setup: 20kHz-23kHz (panel A), 23kHz-26kHz (panel B), and 26kHz-29kHz (panel C). For this reason, companies may hold different tanks for different configurations, and the testing requires a tedious procedure for positioning the projectors and hydrophones within the tank. A much better solution would be an adaptive acoustic tank whose shape can be altered for the case use.

[0059] Accordingly, in some embodiments, the present disclosure provides for a novel design of an adaptive acoustic tank which is robust to different frequency bands and sensor displacement. In some embodiments, the sides of the present tank are covered with absorption planes made of interconnected plates made of phono-absorption material. However, different from other acoustic tanks, the structure of these planes is adjustable. In particular, the planes attached to the tank’s side are made of separate-interlinked plates that can shrink or stretch to form triangles of different depths. Controlling the shape of the planes may be one or more actuators, e.g., a piston operated by a controller board. For a given setup of a projector and hydrophone within the acoustic tank, the controller may be configured to automatically perform an optimization process that finds the best setup of the plates to obtain a minimal reflection distortion. This involves changing plates configuration while minimizing the delay spread and number of taps for the estimated channel impulse response. Using an optimization algorithm, the system may converge to a global minimum after only a few configuration changes. The result is a fully automatic system that configures the acoustic tank to best fit a given set of sensors and frequency examined. To the best of our knowledge, this is a first attempt to make an acoustic tank that automatically adapts its structure to minimize the reflection level.

[0060] Accordingly, in some embodiments, the present disclosure provides for a novel design of an adaptive acoustic tank whose structure can be efficiently adjustable, as well as an efficient process to automatically configure the shape of the tank to match a given sensor position and carrier frequency used.

[0061] The present inventors have conducted tests using an experimental prototype shown in Fig. 3 of an adaptive acoustic tank, which includes a single adjustable plane whose shape is automatically adjusted by a controller. A pair of projector and hydrophone is used to evaluate the channel impulse response for any adjustment made to the structure of the plane. The results show that, while only a single plane is used, the channel impulse response changes significantly for a modification in the plane’s shape. A fast convergence to a minimum of the delay spread and number of taps is reached at roughly 30 minutes of operation.

Overview

[0062] Calibrations and tests of acoustic sensors are best performed in large pools. To reduce costs of handling such a large facility, tests can be performed in either wave flume or smaller test tanks, at the cost of severe reflection distortion. Alternatively, tests are also performed in air, which, due to the much slower propagation speed, allows separating the direct path from other multipath in smaller volumes. However, for calibration in water, the majority of solutions include tanks covered with phono-absorption material.

[0063] Phono-absorption materials are characterized by their impedance, waterproof capability, and dimensions to reach a certain acoustic absorption capability. With regards to the former, the impedance levels are compared with the acoustic-water impedance, 1.48 MRayl, and the acoustic-tank walls impedance, 10 MRayl for glass.

[0064] Tables 1-3 above include the results of a survey previously conducted by the present inventors, of the properties of different phono-absorbing material. It was found that neoprene rubber, whose acoustic impedance is 2.24 MRayl and whose absorption response is relatively flat in frequency, may represent a satisfactory choice of phono-absorption material.

[0065] Besides the properties of the phono-absorption material, the structure of the absorption planes is also crucial. The structure is required to be multifaceted, where the orientation of each face may be controlled. Since actuating each vertex separately is impractical (for a single wall of the prototype alone there are 80 faces and 158 vertices), a structure with minimal degrees of freedom is required.

[0066] The pattern presented in Fig. 2 is a single degree of freedom structure that allows compact folding of a thin walled sheet using strategically located and oriented creases. For the present application, a rigid, waterproof and phono-absorbent construction is required, e.g., a thick wall controlled by a single linear motor to align a phono-absorbent multifaceted wall to the optimal orientation for minimizing reflections underwater. The entire wall may be referred to as the plane, wherein each face of the structure is a plate. Hence, a plane holds many adjustable plates.

Basics of Underwater Acoustic Propagation

[0067] In relation to acoustic propagation, materials are characterized by acoustic impedance, which reflects on the possibility of sound to propagate in the material. Measured by units of Rayls, where 1 Rayl is equivalent to 1 kg· m ^~2 · s ^-1 , the impedance is defined by the product pc, with p representing material density and c the speed of sound in the medium. When sound propagates through two mediums, it partly penetrates the medium and partly reflects back. The reflection level is determined by the acoustic impedance of the two mediums, Z ₁ and Z ₂ , by the incidence angle, θ _i , and by the angle of refraction, θ _t . The reflection coefficient is set by where both θ _i and θ _t are measured with respect to the direction normal to the interface. In logarithmic scale, the term — 20log ₁₀ R is the echo reduction, also referred to as the target loss, and is an important feature of phono-absorption materials as it reflects the material’s acoustic absorption capability.

[0068] The term T = 1 — R determines the ratio of energy propagated from the first medium to the second. This latter term is used for calculating the capability of an acoustic tank to mitigate reflections. Specifically, set between the water (medium 1) to the tank wall (medium 3), the absorption panel (medium 2) determines the sound’s intensity transferred from medium 1 to medium 3 and reflected back. Denote Ti _j the ratio of energy that transfers from medium i to medium j, this intensity level is set by the product T _1,2 T _1,3 / A ₂ , where A ₂ is the energy loss occurs while the sound passes through medium 2. In turn, the term — 20log ₁₀ T _2,3 is referred to as the insertion loss and is associated with the capability of medium 2 to absorb reflections. Theory of Shape Modifications

[0069] The wall is treated as a mechanism comprising links, which are rigid bodies connected through joints, known as kinematic couples. An important property of a mechanism is its mobility, which is the number of its degrees of freedom. The number of degrees of freedom of a mechanism is the minimum number of independent parameters required to precisely describe the position of all points of the mechanism. This, in turn, translates to the number of parameters needed to control in order to get the mechanism to the required shape. Clearly, the lower the mobility, the simpler the shape modification.

[0070] The folding pattern used herein is similar to foldable structures with pre- determined creases to fold and expand the structure. However, in the present disclosure, a rigid non-foldable material may be used, and thus, instead of one large foldable plane with creases, it comprises multiple rigid rhombi plates with hinges that replace the creases. The adaptive mechanism is a one degree of freedom mechanism. The derivation of the expression for the mobility of this mechanism treats each link as a rigid body in space, thus each has six degrees of freedom, and each kinematic couple (hinge) acts as a constraint which reduces the mobility. The equation for evaluating the mobility of such a mechanism is: m = e — 3v + f , (11) where m is the mobility of the mechanism, v is the number of non-boundary vertices, e is the number of non-boundary edges and f is the number of non-boundary faces. This determines the number of actuators required to fully control the entire mechanism. In the present case, for a mechanism of a X b faces, there is v = (a — 1)(b — 1), e = a(b — 1) + b(a — 1), f = (a — 2 )(b — 2). thus, m = a(b — 1) + b(a — 1) — 3 (a — 1 )(b — 1) + (a — 2 )(b — 2) = 1 (12) so m = 1 independent of the size of the structure, and the simplest way to actuate it is using a linear actuator. Pulling on one side of the mechanism flattens the faces and pushing on it brings them out. Design of the Adaptive Acoustic Tank

[0071] The operation of the adaptive acoustic tank is divided into a setup phase and operational phase. The setup phase comes to determine the shape of the absorption planes, while the operational phase is when the tank is used for testing of acoustic equipment of acoustic applications. The setup phase is performed iteratively by probing the acoustic channel. The process includes the same projector and hydrophone pair and the same frequency band that are aimed to be used in its operation phase. Similarly, the projector and hydrophone are stationed at the same location within the tank as will be set during the operational phase. In the ith iteration, the projector emits short acoustic signals, which are recorded by the hydrophone and analyzed by a controller for channel estimation. An optimization procedure than determines the change needed in the i + 1th iteration for the stretching factor of the planes, i.e., the bending angle of the plate elements in each of the planes with respect to the tank’s wall. Once this change is performed, the iterative process continues until convergence is reached.

[0072] The solution reached is the best setup of the adaptive tank in terms of reflection mitigation. Yet, since the acoustic tank is likely small and thus its reflection pattern is highly spatially and frequency dependent, it is noted that the setup reached only applies to the given locations of the projector and hydrophone pair and for the frequency band test. That is, any change in location of frequency used should follow a new adaptation procedure. An illustration of this process is shown in Fig. 3. In the following, details the design of the absorption planes and the control process are described.

Design of Adaptive Absorption Plane

[0073] The design of the wall is modular. Two basic rhombi plates are designed such that the entire structure is composed of a repeating right-left pattern connected by hinges, as shown in Fig. 4A. All parts may be printed using ABS. Neoprene rubber sheets (Fig. 10B) are then glued over the printed pieces towards the final shape. The entire wall is composed of 10 X 8 rhombi. The top and bottom end rhombi are connected using end units that allow both translation and rotation of the edges. The bottom end unit is a passive unit attached to the bottom of the tank, and the top unit is connected to the linear actuator moving it in the vertical plane, allowing the structure to expand, contract and rotate to the desired orientation.

Control Algorithm

[0074] The control operation involves channel estimation that feeds an optimization algorithm. These channel measures serve for the calculation of the optimization cost function, because of interest is the rejection of reflections, channel delay spread and channel tap number may be selected. The former is a measure of the length of the channel, and the latter reflects the density of the channel. Minimizing both measures thus reflects how close the channel is to a unit channel.

Channel Estimation

[0075] Denote the channel impulse response by h(t). The delay spread is defined by and it is customary to use its root-mean-square value

[0076] In turn, the number of channel path is the number of significant reflections within the channel impulse response. Denote Th a threshold over the power of each received reflection such that only h(t) > Th is considered a significant reflection. Also denote W the bandwidth of the emitted signal used for the channel estimation, such that the two-sided width of each observable reflection peak in h(t) is 2/W. Then, for the sampled impulse response, h[n], n = 0, ..., N and for the sampled bandwidth 2Fs/W, the number of significant reflections is calculated by where d(·) stands for the Kronecker delta function. The condition in (15) aims to define individual significant reflections within h(t), that are separable by the observable width of each reflection. [0077] Due to the small dimensions of the acoustic tank, the reflection pattern within the tank is expected to be complex, and thus a full channel estimation of h(t), e.g., using, e.g., a matching pursuit algorithm, may be highly challenging. However, since only the channel measures are of interest, an accurate channel estimation is not needed. The procedure thus involves a rather simple matched filter to integrate the received signal with the emitted one. Specifically, the normalized matched filter may be selected where s(t), 0 < t < q, being the signal duration, t the matched filter’s running time index, q being the signal duration, t the matched filter’s running time index, and r(t) are the emitted and received signals, respectively. Threshold Th can then be chosen formally by [8] where B(·) is the regularized incomplete beta function, and P _fa is a target false alarm rate. To yield a good estimation of D _RMS and p, a compressed signal s(t) with a large Wq product is preferred, such as a chirp.

Optimization Procedure

[0078] The optimization determines the best configuration of the absorption plane such that both D _RMS from (14) and R from (15) are minimized. The algorithm works iteratively, where for each given input values R and D _RMS , the output of the optimization corresponds to the next stretching ratio of the plane to be tested, with the goal of reaching the global minima for R and D _RMS in a small number of iterations.

[0079] While D _RMS and R are both minimized, they are not interlinked. That is, a channel may have a long delay spread by only a few channel taps. Hence, the optimization should be multi-objective to find the best Pareto front. For this, a genetic algorithm (GA) may be selected, with a population of 10 units and up to 10 generations with a total of 110 function calls. This algorithm showed better performance than a simulated annealing implementation. Still, the present approach is not confined to GA and other solutions can be applied. Prototype Implementation

[0080] The prototype system includes a single adjustable absorption plane that is controlled by an actuator stationed above the tank to push or pull 80 absorption plates. The actuator is operated by a single relay, such that the stretching ratio of the absorption plane is set by the time of actuator operation. In accordance, the operator adapted by the optimization algorithm is the time of operation of the actuator as a proxy for the extension distance of the absorption plane. To avoid feedback from the absorption plate, this operation time is measured by the time elapsed from the initial configuration of the plane, i.e., when it is fully stretched. Consequently, between optimization iterations, the plane is returned to this initial position. Positioning may be determined using, e.g., position sensors and/or based on actuator motion measurements.

[0081] Figs. 5A-5B are illustrations of the acoustic chamber with the absorption plane: fully open (5A) and fully closed (5B). Fig. 5C is a view of the shape illustrating the height of the shape H and the piston extension Z.

[0082] Denote Z as the travel distance of the actuator, where

Z _max = 250 [mm], and H as the height of the shape. The relation between the motion of the linear actuator and shape height is where a and b are the sides of the parallelogram (in the case of a rhombus, they are equal), and g is the parallelogram’s angle as detailed in Fig. 6A. This relation is depicted in Fig. 6B. Thus, the height range of the structure is between 50 — 95 [mm]. Comparing measurements performed before the plane was installed and for the plane being fully stretched shows that the plane can absorb at a ratio of roughly 5 dB, flat across the frequency range of 20-30 kHz. Considering that the prototype has only one plane installed and the tank is small, it may be observed that this is a reasonable absorption capability.

[0083] The projector and hydrophone pair used in the prototype are the Geospectrum M18C-2.5 and M36-V35-900 connected to a power and pre-amplifiers, respectively. As illustrated in Fig.9, the input and output of both sensors are connected to the analog output and input of a National Instruments USB-6221 A/D controlled by a controller, e.g., a computer. Through two I/O outputs, this device is also connected to the actuator to pull or push the absorption plate. To reduce noise in the tank due to water shifting, the channel estimation takes place 5s after the actuator operation finishes. The full iterative cycle including actuator operation, channel estimation, and completion of an optimization step is made fully automatic, and takes up to 30s for a full stretching of the absorption plane.

Experimental Results

[0084] The present inventors have conducted as experiment using as the transmitted signal, s(t), a linear chirp signal of 1 ms duration and at various frequency ranges. A chirp we selected because of its narrow auto-correlation and high processing gain that allows separation of multipath. Fig. 7 shows channel impulse responses for two configurations of the phonoabsorption plane, e.g., stretched by 20% (panel A) and by 60% (panel B). Signal transmitted is chirp of 1 msec long at 20-30 kHz frequency band. A significant difference between the two responses implying that a change in the stretching level of the plane is alternate. The response in Fig. 7 panel A is much less dense in term of reflections implying that this stretching level is more suitable for the given setup (signal emitted and positions of projector and hydrophone).

[0085] Figs. 8A-8B show convergence of the optimization algorithm for adjusting the absorption panel. Signal emitted at a frequency range of 20-30kHz (panels A) and 35-40kHz (panels B). Results in Figs. 8A-8B show the estimated tap number and delay spread of the channel impulse response as a function of the optimization function call number, respectively. The figures clearly show how the GA is capable to escape from local minima while exploring the optional range of stretching ratios efficiently with convergence reached after only 100 trials. Recall each optimization iteration cycle takes up to 30 s for each iteration, the overall time until convergence is less than 50 min, and in practice roughly 30 min. A line shows a decrease trend for the two cost parameters. Since only a single absorption plane is used, the decrease for the 20-30kHz range is not dramatic. Still, it clearly demonstrates the benefit of the procedure. However, for the 35-40kHz range, a very sharp decrease may be observed over time, which reflects the availability of a more distinct optimum point for the higher frequency range.

[0086] The convergence of the procedure is demonstrated in Fig. 8C, where the stretching level of the absorption plane is shown as a function of the optimization call number for the above test configuration. As in, e.g., Fig. 4A, the exploration feature of the GA optimization is also visible with a large variation in the stretching value at the beginning reflects on the insensitivity of the algorithm to the initial configuration of the absorption plane. A similar trend is shown in Fig. 9, showing channel impulse responses concatenated as a function of optimization call number for frequency range: 20-30 kHz (panel A), and 35-40 kHz (panel B), where the channel impulse responses throughout the convergence procedure are concatenated by the optimization call number. This form of result demonstrates the change in the impulse response during the optimization procedure. A clear convergence may be observed towards the end of the procedure showing that, in term of reflections, the channel impulse response after convergence is much less dense than at the beginning of the procedure.

[0087] Figs. 10A-10C show channel impulse responses concatenated in a waterfall form for three different frequency ranges. Results for the adaptation of the acoustic tank for three different frequency ranges, 20-23kHz (10A), 23-26kHz (10B), 26-29kHz (IOC) are shown. The results show the channel impulse response during the optimization procedure. A different channel response may be observed for each of the three frequency bands, where that of the lower range seems the less dense in term of multipath. This result motivates the need for different adaptation of the acoustic tank for different configuration setup. Results of all three adaptation procedures show convergence to a less dense response at the end of the adaptation than in any given instance during the procedure. The best reflection reduction is observed for 20-23 kHz, where due to the large delay spread and high density of the channel response, more room for improvement exists.

[0088] The present invention may be a system, a method, and/or a computer program product. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention. [0089] The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non- exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire. Rather, the computer readable storage medium is a non-transient (i.e., not-volatile) medium.

[0090] Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.

[0091] Computer readable program instructions for carrying out operations of the present invention may be assembler instructions, instmction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, such as Java, Smalltalk, C++ C, Python, Julia, MATLAB, or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field- programmable gate arrays (FPGA), programmable logic arrays (PLA), or microprocessor development boards such as Raspberry Pi or Arduino may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present invention.

[0092] Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.

[0093] These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks. [0094] The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

[0095] The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.

[0096] The descriptions of the various embodiments of the present invention have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the embodiments, the practical application or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein. REFERENCES

[1] Acoustic Polymers Ltd. Kilohertz acoustic tile. Last time accessed: July 2016.

[2] F. A. Barge. Underwater acoustic absorption characteristics of composites of wood, rubber, and steel. Technical Report TM 52-78, Naval Sea Systems Command, May 1978.

[3] L. Bj0rn0 and M. Kjeldgaard. A wide frequency band anechoic water tank. Acta Acustica united with Acustica, 32(2): 103-109, February 1975.

[4] Rodger N. Capps. Elastomeric materials for acustical applications. Technical Report A216 872, Underwater Sound Ref. Detach. Naval Research Lab., September 1989.

[5] Charles L. Darner. An underwater sound absorber for an anechoic tank. Technical Report 31, Underwater Sound Reference Lab., September 1953.

[6] E. Demirors, G. Sklivanitis, T. Melodia, S.N. Batalama, and D.A. Pados. Software- defined underwater acoustic networks: Toward a high-rate real-time reconfigurable modem. IEEE Commun. Mag., 53(11): 64—71 , November 2015.

[7] Andrzej Dobrucki, Bronislaw Zoltogorski, Piotr Pruchnicki, and Romuald Bolejko. Sound-absorbing and insulating enclosures for ultrasonic range. Archives of acoustics, 35(2): 157-164, May 2010.

[8] Pradeep Kumar Garul and Tapan Kumar Chaki. Acoustic & mechanical properties of neoprene rubber for encapsulation of underwater transducers. Inti. J. of Scientific Engineering and Technology, 1(5):231-237, November 2012.

[9] Inti. Inst, of Synthetic Rubber Producers. Synthetic rubber. Last time accessed: July 2016.

[10] Heng Jiang, Yu-Ren Wang, Mi-Lin Zhang, Yan-Ping Hu, Ding Lan, Qun-Li Wu, and Huan-Tong Lu. Wide -band underwater acoustic absorption based on locally resonant unit and interpenetrating network structure. Chinese Physics B, 19(2):026202-l-6, February 2010.

[11] Zhihong Liu and Meiping Sheng. Study on characteristics of sound absorption of underwater visco-elastic coated compound structures. Modern Applied Science, 3( 1 ): 32 — 41 , January 2009. [12] Francesca Meneghello, Filippo Campagnaro, Roee Diamant, Paolo Casari, and Michele Zorzi. Design and evaluation of a low-cost acoustic chamber for underwater networking experiments. In Proceedings of the 11th ACM International Conference on Underwater Networks & Systems, page 1. ACM, 2016.

[13] E. Meyer. Construction of an anechoic underwater sound measuring tank. Technical Report AD-758 095, Navy Underwater Sound Laboratory, January 1965.

[14] Ramanagouda Odugoudar and Ivor Nissen. Ad-hoc network emulation framework for underwater communication applications. In Proc. ACM WUWNet, Woods Hole, MA, September 2010.

[15] Precision Acoustics. Apltile SF5048. Last time accessed: July 2016.

[16] Thomas Ramotowski and Kirk Jenne. NUWC XP-1 polyurethane-urea: a new, “acoustically transparent” encapsulant for underwater transducers and hydrophones. In Proc. MTS/IEEE OCEANS, San Diego, CA, September 2003.

[17] S. Shankar and M. Chitre. Tuning an underwater communication link. In Proc. MTS/IEEE OCEANS, Bergen, Norway, June 2013.

[18] K. Nagakura H. Nakagawa S. Ogata, T. Tanaka and K. Takashima. The experimental investigation on the sound absorption coefficient of the sound absorptive materials in scale model experiment for railway noise. In Proc. Inter-Noise, Nice, Trance.

[19] Giovanni Toso, Ivano Calabrese, Paolo Casari, and Michele Zorzi. RECORDS: a remote control framework for underwater networks. In Proc. IEEE/IFIP Med-Hoc-Net, Piran, Slovenia, June 2014.

[20] Yuren Wang, Heng Jiang, Meng Chen, and Yafei Peng. A wide band underwater strong acoustic absorbing materials. In Proc. ICEM, Porto, Portugal, July 2012.

[21] Jichuan Zhang and Zhaohong Xue. Study on under-water sound absorption properties of eucommia ulmoides gum and its blends. Polymer Bulletin, 67(3):511-525, July 2011.

[22] CMRE facilities & equipment. Last time accessed: June 2016.