Adaptive sound field control system based on metamaterial sandwich and its leakage prevention method

Abstract

This invention discloses an adaptive sound field control system based on a metamaterial interlayer and its leakage prevention method, comprising a main wall structure, an acoustic metamaterial interlayer, an acoustic sensing unit, a feedback sensor array, a sound field execution array, and a control processing module. This invention addresses the problems of existing technologies, such as the lack of specificity in room-level sound masking, difficulty in identifying leakage paths within walls, the separation of active and passive sound control methods, and the lack of directional control capabilities in existing wall-based voice privacy solutions. By constructing an adjustable sound field within the wall interlayer, it achieves path identification, directional cancellation, dynamic disturbance, and closed-loop feedback adjustment of leaked voice, thereby enhancing the information security protection capabilities of building spaces.

Description

Technical Field

[0001] This invention relates to the field of architectural acoustics, and more particularly to an adaptive sound field control system based on metamaterial interlayer and its leakage prevention method, and a wall voice leakage protection system based on acoustic metamaterials and active noise control technology and its leakage prevention method. Background Technology

[0002] With the increasing number of conference rooms, office spaces, and other venues with high requirements for voice privacy, how to suppress the propagation of voice information through walls, gaps, and vibration coupling paths of structural components has become a concern in the fields of architectural acoustics and information security. Existing building sound insulation technologies are generally divided into passive sound insulation and active sound control. Passive sound insulation mainly relies on the building envelope materials, interlayers, cavities, and structural design to achieve sound transmission attenuation; active sound control, on the other hand, modulates the target sound field through sensing, signal processing, and secondary sound sources. Current research indicates that while active noise control has potential in the built environment, there are still research gaps in system integration, application boundaries, and engineering implementation in actual building envelope scenarios.

[0003] To improve voice privacy, existing technologies have proposed room-level sound masking schemes. For example, US7460675B2 discloses an automatically adjusting sound masking system and method, which uses a white noise generator and a masking filter to determine the spectrum and level of the target masking noise based on the room's acoustic response and ambient noise, thereby emitting automatically adjusted masking noise within the room. The main technical idea of ​​this type of scheme is to generate target masking noise based on the overall acoustic conditions of the room to reduce the speech intelligibility at the listening location. This technology can improve speech privacy in open office or room scenarios, but its control object is mainly the room-level background masking sound field, rather than the identification, modeling, and targeted intervention of specific leakage propagation paths inside walls.

[0004] Furthermore, existing technologies have proposed active voice privacy solutions related to walls or partitions. For example, WO2018170045A1 discloses a speech privacy system and acoustic wall assembly. This solution utilizes a microphone to receive sound signals and a sound masking circuit drives a speaker to generate active reverberation or speech intelligibility disruption signals to reduce speech intelligibility through walls. This patent also discloses that the microphone, speaker, and related circuitry can be placed inside the wall, near the wall surface, or in a double-walled cavity, and that interference signals are generated dynamically. Thus, this type of solution has evolved from simple room-level white noise masking to an active voice disruption path integrated with the wall structure.

[0005] However, the aforementioned proactive voice privacy solutions related to walls primarily focus on reducing external auditory recognition through active reverberation, masking, or speech intelligibility disruption. The disclosed key points are the sound pickup, processing, and sound perturbation itself. No explicit disclosures have been made regarding the spatial distribution perception of leakage paths within the wall, the joint modeling of wall vibration and interlayer sound pressure, the directional control of the sound field construction along the target leakage path, or the technical approach of using specific artificial structures within the wall interlayer to enhance controllability.

[0006] On the other hand, acoustic metamaterials, as a technological approach that achieves anomalous acoustic parameters or specific wave propagation characteristics through artificial microstructure design, can reflect, focus, bend, absorb, or block sound waves through bandgap effects, local resonance effects, and anisotropic propagation modulation. Related research indicates that acoustic metamaterials have demonstrated the ability to manipulate sound fields in unconventional ways, but their transformation from laboratory structures to scalable and engineering-deployable devices still faces challenges. Therefore, acoustic metamaterials provide a new structural basis for sound propagation modulation in walls, but existing publications focus more on passive sound control capabilities at the material or device level, and have not naturally derived a closed-loop system solution for preventing voice information leakage and working in conjunction with real-time sensing and active control.

[0007] In summary, the existing technology has the following problems:

[0008] 1) Room-level sound masking solutions mainly target the overall spatial background masking and lack specific control over leakage paths inside the walls;

[0009] 2) Active speech disturbance schemes combined with walls mainly rely on reverberation or intelligibility destruction, and lack a joint perception and path modeling mechanism based on wall vibration response and interlayer sound pressure distribution;

[0010] 3) Although acoustic metamaterials-related technologies can improve the sound control capability in specific frequency bands, existing disclosures mainly focus on passive structure design and lack systematic application solutions that combine with distributed sensing, array execution and feedback adaptive control.

[0011] 4) Existing technologies have not yet developed a voice information leakage prevention solution that combines acoustic metamaterial structure, leakage path identification, directional sound field execution, and feedback adjustment of external residual sound pressure / voice intelligibility within the wall interlayer space.

[0012] Therefore, it is necessary to develop new technologies that can be used to construct adjustable acoustic structures within the wall cavity and combine them with intelligent algorithms to achieve real-time reconstruction and dynamic adjustment of the sound field, thereby enabling proactive control and spatial directional isolation of voice leakage and improving the information security protection capabilities of building spaces. Summary of the Invention

[0013] Purpose of the invention: To address the shortcomings and defects of existing technologies, this invention provides an adaptive sound field control system based on metamaterial interlayers and its leakage prevention method. It solves the problems of existing technologies, such as the lack of specificity in room-level sound masking, difficulty in identifying leakage paths within walls, separation of active and passive sound control methods, and the lack of directional control capabilities in existing wall-mounted voice privacy solutions. By constructing an adjustable sound field within the wall interlayer, it achieves path identification, directional cancellation, dynamic disturbance, and closed-loop feedback adjustment of leaked voice, thereby enhancing the information security protection capabilities of building spaces.

[0014] Technical solution: The adaptive sound field control system based on metamaterial sandwich of the present invention is characterized by comprising:

[0015] The main wall structure includes interior side wall panels and exterior side wall panels;

[0016] Acoustic metamaterial sandwich: disposed between the indoor side wall panel and the outdoor side wall panel, comprising several periodically arranged subwavelength resonant units. The resonant unit array is used to modulate at least one of the following: bandgap modulation, local resonance modulation, and propagation impedance modulation, within the 300Hz to 3400Hz speech frequency band of the leaked sound waves passing through the wall. Acoustic metamaterials refer to materials composed of artificially designed periodic structural units, whose unit size is less than 1 / 2 of the wavelength of the sound wave in the medium, and which generate anomalous equivalent acoustic parameters through local resonance or Bragg scattering.

[0017] Acoustic sensing unit: installed in the main structure of the wall and / or the acoustic metamaterial interlayer, including a vibration sensor array installed on the wall surface or inside the wall and a microphone array installed in the acoustic metamaterial interlayer, to collect the wall vibration response signal and the sound pressure signal in the interlayer, respectively;

[0018] Feedback sensor array: Located on the outside of the wall, used to collect residual sound pressure signals on the outside of the wall and transmit them to the control processing module;

[0019] Sound field execution array: Set within the acoustic metamaterial interlayer and distributed in an array along a preset candidate control region, it includes several independently driveable sound wave emitting units. The sound wave emitting unit array is used to form a directional control sound field targeting the target leakage path within the acoustic metamaterial interlayer according to the control parameters output by the control processing module, so as to cancel or disrupt the leaked speech.

[0020] Control and processing module: Electrically connected to the vibration sensor array, microphone array, sound field execution array, and feedback sensor array, used for: preprocessing and feature extraction of vibration response signals and sound pressure signals; establishing a leaked speech propagation path model based on the joint information of vibration response signals and sound pressure signals; determining the target control area according to the propagation path model, and calculating at least one of the output frequency, amplitude, phase, or time delay parameters of each transmitting unit of the sound field execution array; and updating the output parameters according to the signals collected by the feedback sensor array.

[0021] The system consists of a closed-loop control system that includes sound field perception, path modeling, control solution, array execution, and feedback correction.

[0022] The resonant unit is one or a combination of at least two of the following: a Helmholtz resonant unit, a diaphragm resonant unit, a labyrinthine channel unit, or a local resonant unit; the resonant unit, through a combination of different cavity volumes, neck dimensions, diaphragm thicknesses, channel lengths, mass block parameters, or arrangement periods, enables the acoustic metamaterial interlayer to form multi-band propagation modulation capabilities within the 300Hz–3400Hz speech frequency band.

[0023] The propagation path model is based on the spatial distribution of vibration response signal and sound pressure signal. It is established through joint inversion, transfer function estimation or data-driven modeling to characterize the propagation direction, propagation intensity and leakage location of leaked speech in the corresponding areas inside the wall, the interlayer area and outside the wall.

[0024] The control processing module employs one or more of beamforming algorithms, adaptive filtering algorithms, and model predictive control algorithms to calculate the output parameters of each transmitting unit in the sound field execution array. The control processing module is configured to switch between the following two operating modes:

[0025] The cancellation mode is used to generate control sound waves that are out of phase with at least some frequency bands of the leaked speech to reduce the residual sound pressure outside the wall; the disturbance mode is used to generate control sound waves that are uncorrelated with the leaked speech in the time, frequency, or phase domains to reduce the intelligibility of the speech outside the wall.

[0026] The control processing module switches between cancellation mode and disturbance mode based on at least one of the following: residual sound pressure threshold outside the wall, speech intelligibility threshold, speech duration threshold, or sensitive speech triggering conditions.

[0027] The acoustic wave emitting unit is one or more of a miniature loudspeaker, piezoelectric transducer, electromagnetic exciter, or structural vibration exciter, and each acoustic wave emitting unit is controlled by an independent drive channel.

[0028] The control processing module includes a signal acquisition sub-board, a main control processor, a memory, a power management unit, a power drive interface, and a communication interface; the main control processor is an ARM processor, a DSP processor, an FPGA processor, or a heterogeneous processing platform composed of ARM and FPGA.

[0029] The present invention provides a leakage prevention method based on metamaterial interlayers, characterized by comprising the following steps:

[0030] 1) The vibration response signal of the wall is collected by a vibration sensor array, and the sound pressure signal in the acoustic metamaterial interlayer is collected by a microphone array;

[0031] 2) Denoise, frame, window, and extract spectral features from vibration response signals and sound pressure signals;

[0032] 3) Based on the spatial distribution of vibration response signal and sound pressure signal, a propagation path model characterizing the direction, intensity and location of leaked speech propagation is established through joint inversion or transfer function estimation, and the target leakage path and target control area are determined.

[0033] 4) Based on the propagation path model, calculate at least one of the following parameters for each transmitting unit of the sound field execution array: output frequency, amplitude, phase, or time delay.

[0034] 5) The control sound field execution array forms a directional control sound field targeting the leakage path within the acoustic metamaterial interlayer to cancel or disrupt the leaking speech.

[0035] 6) Obtain the residual sound pressure outside the wall, speech intelligibility, or a combination of both as feedback quantities, and update the output parameters based on the feedback quantities until the preset suppression index is reached.

[0036] In step 3), the propagation path model is obtained through joint inversion of the spatial distribution of vibration response and the spatial distribution of interlayer sound pressure, transfer function estimation, or data-driven modeling. In step 4), beamforming algorithm is used to determine the phase delay and amplitude weight of each transmitting unit so as to minimize the residual sound pressure in the target control area, minimize speech intelligibility, or minimize both simultaneously.

[0037] In step 5), in the cancellation mode, a control sound wave with the opposite phase to the frequency band corresponding to the leaked speech is output, and in the disturbance mode, a multi-frequency uncorrelated control sound wave or a phase disturbance control sound wave is output; in step 6), the preset suppression index includes the sound pressure level reduction threshold outside the wall, the speech intelligibility reduction threshold, or a combination of the two.

[0038] Beneficial Effects: Compared with existing technologies, this invention has the following significant advantages: 1) By setting an acoustic metamaterial interlayer composed of multiple periodic resonant units inside the main wall structure, the wall possesses propagation modulation capabilities within the speech frequency band, thereby improving the basic suppression capability of leaked speech. 2) Through the combined arrangement of vibration sensors and microphone arrays, spatial localization of the leakage path is achieved, thereby improving the targeting of active control. 3) By setting a distributed sound field execution array within the acoustic metamaterial interlayer, and combining it with beamforming, adaptive filtering, or model predictive control algorithms, a directional control sound field can be formed along the target leakage path to reduce residual sound pressure and speech intelligibility outside the wall. 4) By using the residual sound pressure and / or speech intelligibility outside the wall as feedback quantities to perform closed-loop updates to the control parameters, the system can dynamically adjust its working state according to changes in speaker position, speech spectrum, and environmental noise. 5) By setting a dual working mechanism of cancellation mode and disturbance mode, differentiated control can be achieved according to different confidentiality requirements and environmental conditions. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the wall structure of the present invention;

[0040] Figure 2 This is a schematic diagram of the acoustic metamaterial sandwich structure of the present invention;

[0041] Figure 3 This is a schematic diagram of the structure of the vibration sensor array and the sound field execution array of the present invention;

[0042] Figure 4 This is a system structure block diagram of the present invention;

[0043] Figure 5 This is a flowchart of the method of the present invention;

[0044] Figure 6 This is a schematic diagram of the sound field control principle of the present invention;

[0045] In the diagram, 1 is the indoor side wall panel; 2 is the acoustic metamaterial interlayer; 3 is the outdoor side wall panel; 4 is the vibration sensor array; 5 is the resonant unit array; 6 is the control processing module; 7 is the sound field execution array; 8 is the indoor voice sound wave; 9 is the control sound wave; 21 is the resonant unit array; 22 is the resonant cavity; 23 is the neck channel; 24 is the interlayer partition; 25 is the unit side wall; 31 is the vibration sensor array; 32 is the microphone array; 33 is the sound wave emitting unit array; 34 is the leakage area; 35 is the target control area; 36 is the control beam direction; and 37 is the sound propagation path. Detailed Implementation

[0046] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0047] Example 1:

[0048] The adaptive spatial sound field control system based on acoustic metamaterial interlayer in this embodiment is installed inside the building wall structure. It is used to actively control the propagation process of indoor speech sound waves from the interior to the exterior of the wall, thereby reducing the intensity of speech leakage and speech intelligibility.

[0049] 1. Main wall structure:

[0050] like Figure 1 As shown, the structure includes an indoor side wall panel 1, an outdoor side wall panel 3, and an acoustic metamaterial interlayer 2 disposed between the two. The total wall thickness is preferably 150mm to 300mm, and the thickness of the acoustic metamaterial interlayer 2 is preferably 40mm to 80mm. The indoor side wall panel 1 can be made of gypsum board, composite sound insulation board, or concrete board, and the outdoor side wall panel 3 can be made of concrete wall, brick wall structure, or composite enclosure panel.

[0051] 2. Acoustic metamaterial sandwich 2:

[0052] It consists of an array 5 of periodically arranged resonant units. Each resonant unit can be one or more combinations of Helmholtz resonant structure, diaphragm resonant structure, labyrinth channel structure or local resonant structure.

[0053] When using a Helmholtz resonant unit, the volume of the resonant cavity 22 is preferably 20 cm³ to 80 cm³, the diameter of the neck channel 23 is preferably 5 mm to 12 mm, the length of the neck channel 23 is preferably 10 mm to 30 mm, and the arrangement period of the resonant units is preferably 30 mm to 60 mm. By changing the volume of the resonant cavity and the neck size, the resonant frequency can cover the 300 Hz to 3400 Hz speech frequency band, thereby creating a bandgap effect or local resonance effect in the interlayer within the speech frequency band. An interlayer partition 24 is provided between the resonant cavity 22 and the unit sidewall 25.

[0054] In a preferred embodiment, the cavity shell material of the resonant unit can be selected from ABS engineering plastic, polycarbonate, nylon composite material, or aluminum alloy. When a diaphragm resonant unit is used, its diaphragm material can be selected from PET film, TPU film, or silicone rubber film, and the film thickness is preferably 0.03mm to 0.30mm; when a local resonant unit is used, its local mass block can be made of steel, copper, or tungsten material and connected to the substrate through an elastic connector.

[0055] In one embodiment, the acoustic metamaterial interlayer 2 exhibits differentiated propagation modulation capabilities along the wall thickness direction. The resonant unit near the indoor side adopts a higher equivalent acoustic impedance design, while the resonant unit near the outdoor side adopts a lower equivalent acoustic impedance design. Through asymmetric cavity structure, gradually varying neck size, or partition parameter arrangement, the guidance and control capabilities of leaked speech after entering the interlayer are improved.

[0056] 3. Acoustic sensing unit:

[0057] like Figure 3 As shown, it includes a vibration sensor array 5 and a microphone array 32.

[0058] The vibration sensor array 5 is set on the wall surface or the key vibration area of ​​the wall. It can be a piezoelectric vibration sensor or a MEMS accelerometer. The sensor spacing is preferably 200mm to 500mm. It is used to detect the vibration response of the wall structure.

[0059] The microphone array 32 is disposed inside the acoustic metamaterial interlayer 2, preferably using MEMS microphones, and the array spacing is preferably 50mm to 100mm, for collecting the sound pressure distribution inside the interlayer.

[0060] The sampling data from the vibration sensor array 5 and the microphone array 32 are synchronously transmitted to the control processing module 6 to form spatial sound field sampling data for path recognition and control solution.

[0061] Figure 3 In the diagram, 34 represents the leakage area, 36 represents the control beam direction, and 37 represents the sound propagation path.

[0062] 4. Feedback sensor array:

[0063] A feedback sensor array is positioned on the outer side of the wall, preferably 0.5m to 2m from the wall surface, to collect residual sound pressure signals from the outer side of the wall and assess speech intelligibility. The feedback sensor array includes one or more microphones, preferably MEMS microphones of the same model as those used in the interlayer, to achieve signal consistency calibration. The feedback sensor array transmits the collected residual sound pressure signals to the control processing module for adaptive updates of closed-loop control parameters.

[0064] 5. Control and processing module 6:

[0065] like Figure 4 As shown, the control processing module 6 is located in the equipment mounting cavity inside the wall or in the mezzanine area, and includes a signal acquisition sub-board, a main control processor, a memory, a power management unit, a power drive interface, and a communication interface. The main control processor can be an ARM processor, a DSP processor, an FPGA processor, or a heterogeneous processing platform composed of ARM and FPGA.

[0066] In one embodiment, the main control processor has a clock speed of no less than 1.5 GHz, a RAM of no less than 2 GB, and a non-volatile storage capacity of no less than 16 GB. The signal acquisition sub-board includes at least 8 vibration sensor input channels and 16 microphone input channels, with an analog-to-digital conversion accuracy preferably between 16 bits and 24 bits, and a sampling frequency preferably between 8 kHz and 48 kHz. The control processing module includes at least 16 independent control output channels for outputting control signals with adjustable amplitude, phase, and time delay to multiple acoustic wave transmitting units.

[0067] The control processing module 6 can perform the following processing flow: (1) Denoise, frame, window and spectrum analysis of the signals collected by the vibration sensor array 5 and the microphone array 32; (2) Establish a leakage speech propagation path model based on the wall vibration response distribution and the interlayer sound pressure distribution; (3) Determine the target leakage path and target control area according to the propagation path model; (4) Calculate the output parameters of each sound wave emitting unit using beamforming algorithm, adaptive filtering algorithm or model predictive control algorithm; (5) Update the control parameters according to the feedback of the residual sound pressure signal collected by the feedback sensor array, the calculated speech intelligibility or the combination of the two.

[0068] In a preferred embodiment, the single control refresh cycle of the control processing module 6 is 10ms to 50ms, and the end-to-end control delay is no more than 20ms.

[0069] 6. Sound field execution array:

[0070] The sound field execution array consists of multiple sound wave emitting units. The sound wave emitting unit array 33 is embedded in the acoustic metamaterial sandwich layer 2 and distributed in an array along the preset candidate control region. The sound wave emitting units can be one or more of the following: miniature loudspeakers, piezoelectric transducers, electromagnetic exciters, or structural vibration exciters.

[0071] In this embodiment, the preferred spacing between the sound wave emitting units is 50mm to 120mm, the preferred transmission frequency range is 200Hz to 5000Hz, and the preferred maximum output sound pressure level is 80dB to 110dB. Each emitting unit is driven by an independent power amplification module. The control processing module 6 controls the amplitude, phase, and time delay of the sound waves output by each emitting unit, so that multiple sound waves are superimposed in the interlayer space to form a target directional control sound field.

[0072] 7. Working Mode:

[0073] The system includes a cancellation mode and a disturbance mode.

[0074] In the cancellation mode, when the system detects a stable voice signal, the control processing module 6 calculates the spectrum and phase information of the leaked voice and controls the sound field execution array 7 to output a control signal that is in phase with at least some of its frequency bands, so as to reduce the residual sound pressure in the area corresponding to the target leakage path.

[0075] In the scrambling mode, when the system detects continuous speech, highly sensitive information communication, or speech intelligibility outside the wall that is higher than a preset threshold, the system switches to the scrambling mode and outputs multi-frequency uncorrelated signals, phase scrambling signals, or frequency domain uncorrelated control signals through the sound field execution array 7, so that the speech received outside the wall is distorted, blurred, or difficult to identify.

[0076] The two modes can be automatically switched based on the residual sound pressure threshold outside the wall, the speech intelligibility threshold, the speech duration threshold, or the sensitive speech triggering conditions.

[0077] Example 2:

[0078] Intelligent adaptive sound field control methods: such as Figure 5 As shown, it includes the following steps:

[0079] Step S1: Collect the vibration response signal of the wall through a vibration sensor array, and collect the sound pressure signal in the acoustic metamaterial interlayer through a microphone array;

[0080] Step S2: Denoise, frame, window, and extract spectral features from the vibration response signal and the sound pressure signal to obtain the main frequency components and energy distribution of the speech.

[0081] Step S3: Establish a leakage voice propagation path model based on the wall vibration distribution and interlayer sound pressure distribution, and identify the target leakage path and target control area;

[0082] In a preferred embodiment, the propagation path model is established using a joint inversion algorithm. Let the vibration response signal acquired by the vibration sensor array be v(r_v, t), and the sound pressure signal acquired by the microphone array be p(r_p, t), where r_v and r_p are the sensor position vectors, respectively. Signal features within the speech frequency band are extracted through time-frequency analysis, and a joint vibration-sound pressure observation matrix is ​​established.

[0083] Y = [V^T, P^T]^T = H·S + N

[0084] Where V and P are the frequency domain representations of the vibration response and sound pressure signal, respectively, H is the propagation path transfer function matrix, S is the sound source intensity vector, and N is the noise matrix. The leakage path parameters are estimated by minimizing the following cost function:

[0085] J(θ) = ||Y - H(θ)·S||^2 + λ·R(θ)

[0086] Where θ represents the path parameters to be estimated (including propagation direction, attenuation coefficient, and leakage location), λ is the regularization coefficient, and R(θ) is the prior constraint term. An iterative optimization algorithm (such as L-BFGS or gradient descent) is used to solve the above minimization problem to obtain the optimal path parameter estimates.

[0087] In practical applications, the initial value of the transfer function matrix H(θ) can be determined by using a standard sound source during the system calibration phase, or an adaptive algorithm (such as FxLMS or RLS algorithm) can be used during the operation phase to simultaneously update the sound source estimate Ŝ and the path parameter θ.

[0088] Step S4: Calculate the output amplitude, phase, and / or time delay parameters of each acoustic wave emitting unit using beamforming algorithm, adaptive filtering algorithm, or model predictive control algorithm;

[0089] Step S5: Control the sound field execution array to output control sound waves, forming a directional control sound field for the target leakage path in the acoustic metamaterial interlayer, and canceling or disrupting the leaked speech.

[0090] Step S6: The system dynamically adjusts the control parameters based on feedback from the residual sound pressure outside the wall, speech intelligibility, or a combination of both, to achieve closed-loop adaptive control.

[0091] In a preferred embodiment, the speech intelligibility is obtained by calculating the correlation coefficient between the external microphone signal and the reference speech signal or by using a standardized STI (Speech Transmission Index) algorithm. The STI algorithm is based on the modulation transfer function (MTF) and analyzes the intensity modulation depth of 7 octave bands (125Hz to 8kHz) and 14 modulation frequencies (0.63Hz to 12.5Hz) to obtain an intelligibility index between 0 and 1, where a lower STI value indicates poor speech intelligibility and a higher STI value indicates clear and intelligible speech.

[0092] Example 3:

[0093] Specific parameter configurations and control effects for conference room applications:

[0094] Based on the above system configuration, the synergistic effect of joint vibration-sound pressure sensing, leakage path orientation identification, and adaptive sound field modulation can effectively suppress the leakage propagation of indoor speech to the outside of the wall. In the cancellation mode, the system outputs inverse control sound waves, which is expected to significantly reduce the residual sound pressure level in the target area outside the wall; in the disturbance mode, the system outputs uncorrelated control sound waves, which is expected to effectively reduce the intelligibility of speech outside the wall, making the leaked speech difficult to identify and understand.

[0095] This system can target and suppress leaked voice at different frequencies and intensities by adjusting the parameters of the resonant unit, the density of the sensor array, and the weight of the control algorithm, based on the acoustic environment, wall structure characteristics, and confidentiality requirements of the actual conference room, thereby improving the voice information security protection capability of the building space.

[0096] It should be noted that the aforementioned "significantly reduced" and "effectively reduced" technical effects are qualitative effects expected to be achieved based on the physical principles of active noise control theory and beamforming algorithms. The specific noise reduction magnitude and the degree of improvement in speech intelligibility depend on the actual acoustic environment of the conference room, the original sound insulation performance of the walls, the calibration accuracy of the sensor array, and the convergence state of the control algorithm. These factors need to be determined through system debugging and parameter optimization after on-site deployment.

[0097] In a conference room wall application scenario, the wall thickness is 200mm, the acoustic metamaterial interlayer thickness is 60mm, the resonant unit design frequencies include 500Hz, 1000Hz, 2000Hz, and 3000Hz, the number of microphones is 16, the number of vibration sensors is 8, the number of sound wave emitting units is 24, and the control algorithm refresh cycle is 50ms. In this embodiment, the control parameters can be dynamically updated based on the residual sound pressure outside the wall and changes in speech intelligibility to achieve continuous suppression of leaked speech. It should be noted that the resonant unit can be not only a Helmholtz resonant structure, but also a diaphragm resonant structure, a labyrinthine channel structure, or a local resonant structure; the sound wave emitting unit can not only be a miniature loudspeaker, but also a piezoelectric transducer, an electromagnetic exciter, or a structural vibration exciter; the control processing module can not only use an embedded processor, but also an FPGA, DSP, or a heterogeneous processing platform. All equivalent substitutions made under the concept of this invention in terms of structural form, material selection, control parameter solution method, and array arrangement should be included within the protection scope of this invention.

Claims

1. An adaptive sound field control system based on metamaterial sandwich, characterized in that: include: The main structure of the wall includes the interior side wall panel (1) and the exterior side wall panel (3); Acoustic metamaterial interlayer (2): disposed between the indoor side wall panel (1) and the outdoor side wall panel (3), including several periodically arranged subwavelength resonant units, and the resonant unit array (5) is used to perform at least one of bandgap modulation, local resonance modulation and propagation impedance modulation on the leakage sound waves passing through the wall in the speech band. Acoustic sensing unit: set in the main structure of the wall and / or the acoustic metamaterial interlayer (2), including a vibration sensor array (4) set on the surface of the wall or inside the wall and a microphone array (32) set in the acoustic metamaterial interlayer (2), respectively collecting the wall vibration response signal and the sound pressure signal in the interlayer; Feedback sensor array: Located on the outside of the wall, used to collect residual sound pressure signals on the outside of the wall and transmit them to the control processing module (6). Sound field execution array (7): Set in the acoustic metamaterial interlayer (2) and distributed in an array along the preset candidate control area, including several independently driveable sound wave emitting units. The sound wave emitting unit array (33) is used to form a directional control sound field for the target leakage path in the acoustic metamaterial interlayer (2) according to the control parameters output by the control processing module (6) in order to cancel or disturb the leakage voice. Control processing module (6): electrically connected to vibration sensor array (4), microphone array (32), sound field execution array (7) and feedback sensor array, used for: preprocessing and feature extraction of vibration response signal and sound pressure signal; establishing a leaked speech propagation path model based on the joint information of vibration response signal and sound pressure signal; determining the target control area (35) according to the propagation path model, and calculating at least one of the output frequency, amplitude, phase or time delay parameters of each transmitting unit of sound field execution array (7); and updating the output parameters according to the signal collected by feedback sensor array; The system consists of a closed-loop control system that includes sound field perception, path modeling, control solution, array execution, and feedback correction.

2. The adaptive sound field control system based on metamaterial sandwich as described in claim 1, characterized in that: The resonant unit is one or a combination of at least two of the following: Helmholtz resonant unit, diaphragm resonant unit, labyrinth channel unit, or local resonant unit; the resonant unit forms a multi-band propagation modulation capability in the 300Hz to 3400Hz speech frequency band by combining different cavity volumes, neck dimensions, diaphragm thicknesses, channel lengths, mass block parameters, or arrangement periods.

3. The adaptive sound field control system based on metamaterial sandwich as described in claim 1, characterized in that: The propagation path model is based on the spatial distribution of vibration response signal and sound pressure signal. It is established through joint inversion, transfer function estimation or data-driven modeling to characterize the propagation direction, propagation intensity and leakage location of leaked speech in the corresponding areas inside the wall, the interlayer area and outside the wall.

4. The adaptive sound field control system based on metamaterial sandwich as described in claim 1, characterized in that: The control processing module (6) uses one or more of beamforming algorithms, adaptive filtering algorithms, and model predictive control algorithms to calculate the output parameters of each transmitting unit of the sound field execution array (7); the control processing module (6) is configured to switch between the following two operating modes: The cancellation mode is used to generate control sound waves that are out of phase with at least some frequency bands of the leaked speech to reduce the residual sound pressure outside the wall; the disturbance mode is used to generate control sound waves that are uncorrelated with the leaked speech in the time, frequency, or phase domains to reduce the intelligibility of the speech outside the wall.

5. The adaptive sound field control system based on metamaterial sandwich as described in claim 4, characterized in that: The control processing module (6) switches between cancellation mode and disturbance mode based on at least one of the following: residual sound pressure threshold outside the wall, speech intelligibility threshold, speech duration threshold, or sensitive speech triggering condition.

6. The adaptive sound field control system based on metamaterial sandwich as described in claim 1, characterized in that: The acoustic wave emitting unit is one or more of a miniature loudspeaker, piezoelectric transducer, electromagnetic exciter, or structural vibration exciter, and each acoustic wave emitting unit is controlled by an independent drive channel.

7. The adaptive sound field control system based on metamaterial sandwich as described in claim 1, characterized in that: The control processing module (6) includes a signal acquisition sub-board, a main control processor, a memory, a power management unit, a power drive interface, and a communication interface; the main control processor is an ARM processor, a DSP processor, an FPGA processor, or a heterogeneous processing platform composed of ARM and FPGA.

8. A leakage prevention method using the adaptive sound field control system according to any one of claims 1-7, characterized in that: Includes the following steps: 1) The vibration response signal of the wall is collected by the vibration sensor array (4), and the sound pressure signal in the acoustic metamaterial interlayer (2) is collected by the microphone array (32); 2) Denoise, frame, window, and extract spectral features from vibration response signals and sound pressure signals; 3) Based on the spatial distribution of vibration response signal and the spatial distribution of sound pressure signal, a propagation path model characterizing the direction, intensity and location of leaked speech propagation is established through joint inversion or transfer function estimation, and the target leakage path and target control area are determined (35). 4) Based on the propagation path model, calculate at least one of the output frequency, amplitude, phase, or time delay parameters of each transmitting unit of the sound field execution array (7); 5) The control sound field execution array (7) forms a directional control sound field for the target leakage path within the acoustic metamaterial interlayer (2) to cancel or disrupt the leakage speech. 6) Obtain the residual sound pressure outside the wall, speech intelligibility, or a combination of both as feedback quantities, and update the output parameters based on the feedback quantities until the preset suppression index is reached.

9. The leak-proof method based on metamaterial interlayer according to claim 8, characterized in that: In step 3), the propagation path model is obtained through joint inversion of the spatial distribution of vibration response and the spatial distribution of interlayer sound pressure, transfer function estimation, or data-driven modeling. In step 4), beamforming algorithm is used to determine the phase delay and amplitude weight of each transmitting unit so as to minimize the residual sound pressure in the target control area, minimize speech intelligibility, or minimize both simultaneously.

10. The leak-proof method based on metamaterial interlayer according to claim 8, characterized in that: In step 5), in the cancellation mode, a control sound wave (9) with the opposite phase to the frequency band corresponding to the leaked speech is output, and in the disturbance mode, a multi-frequency uncorrelated control sound wave or a phase disturbance control sound wave (9) is output; in step 6), the preset suppression index includes the sound pressure level reduction threshold outside the wall, the speech intelligibility reduction threshold, or a combination of the two.

Images