Vehicle-mounted multimedia audio noise reduction method
Through multi-source sensor collaborative detection and spatial sound field management, rest areas and non-rest areas are dynamically divided, solving the problems of noise source aliasing, insufficient passenger status perception, and equipment signal misjudgment in in-vehicle audio noise reduction technology, achieving precise noise reduction and energy consumption optimization.
Patent Information
- Application Number
- CN202510996292.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-18
- Publication Date
- 2025-09-19
AI Technical Summary
Existing in-vehicle audio noise reduction technology has difficulty distinguishing between different noise sources, resulting in suppression failure, inability to dynamically perceive the physiological state of passengers, misjudging the signals of in-vehicle equipment as noise, and serious energy waste.
It adopts multi-source sensor collaborative detection, obtains noise characteristics through engine compartment vibration and door wind pressure sensors, combines passenger status recognition and equipment status perception, dynamically divides rest areas and non-rest areas, and uses spatial sound field management to achieve precise noise reduction.
It achieves accurate classification of engine noise, door resonance and wind noise, reduces power consumption in the rest area, maintains device voice clarity, improves engine noise elimination rate, and avoids misjudgment and energy waste.
Smart Images
Figure CN120673735A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of audio noise reduction, and more particularly to a vehicle-mounted multimedia audio noise reduction method. Background Art
[0002] In-car audio noise reduction technology is a key component in enhancing the modern driving experience. Its core lies in dynamically canceling out in-car noise through the synergy of acoustic engineering and electronic algorithms. This technology relies primarily on active noise cancellation systems, the principles of which can be traced back to Dr. Bose's acoustic cancellation theory in the 1970s: A microphone collects ambient noise in real time, a processor generates a counter-sound wave, and the sound is released through the car's speakers to neutralize the original noise.
[0003] In the field of in-vehicle audio noise reduction, traditional technologies mainly rely on microphone arrays to collect ambient noise and achieve noise reduction by generating anti-phase sound waves. However, with the increasing intelligence of cars and the upgrading of user demands, existing technologies have exposed the following shortcomings: Noise source aliasing causes suppression failure Current noise reduction systems struggle to distinguish between noise sources with different physical characteristics, such as engine vibration, wind noise, and mechanical resonance. Because engine compartment vibration, door resonance, and rearview mirror wind noise overlap in the frequency domain, traditional broadband noise reduction algorithms cannot accurately separate these noise components.
[0004] Existing systems lack the ability to dynamically perceive a passenger's physiological state. When a passenger falls asleep, the continuous, inverse sound waves produced by active noise cancellation can cause auditory fatigue and even dizziness. Furthermore, traditional, uniform, global noise cancellation methods fail to differentiate between rest areas and active areas: excessive noise cancellation signals for sleeping passengers waste energy, while also failing to enhance the voices of awake passengers.
[0005] In-car navigation prompts and entertainment music are often misinterpreted as noise by noise reduction systems. This is especially true in multi-zone scenarios, where navigation announcements coexist with conversations between rear-seat passengers. Existing technologies are unable to spatially isolate device signals from noise, resulting in the omission of critical information.
[0006] Therefore, how to provide a vehicle-mounted multimedia audio noise reduction method that reduces rest area noise reduction power consumption, maintains device voice clarity, and improves vehicle engine noise elimination rate is a problem that technicians in this field urgently need to solve. Summary of the Invention
[0007] In light of this, the present invention provides a method for in-vehicle multimedia audio noise reduction to address the issues raised in the background technology section above. This method significantly reduces power consumption for noise reduction in a rest area within a vehicle, maintains device voice clarity, and improves the vehicle's engine noise cancellation rate.
[0008] In order to achieve the above object, the present invention adopts the following technical solutions: A vehicle-mounted multimedia audio noise reduction method, comprising: Step S1: The piezoelectric vibration sensor in the engine compartment collects the 20-500 Hz mechanical vibration spectrum. Simultaneously, the MEMS air pressure sensor in the door cavity acquires 0.5-5 kHz wind pressure fluctuation data. The two types of sensor data are synchronized through a time domain alignment controller and then input into the noise signature analysis module. Step S2: In the noise signature analysis module, a classification decision is made based on the vibration energy distribution: when the energy in the 80-120 Hz frequency band accounts for more than 70% of the total vibration energy, an engine noise flag is generated; when the peak value of the correlation coefficient between the pulse signal greater than 1 kHz and the wind pressure fluctuation signal is greater than 0.9, a wind noise flag is generated; when the door resonance condition is met, a door resonance flag is generated, specifically: The vibration signal is processed by 200Hz high-pass filtering and 500Hz low-pass filtering to extract the energy sequence of the 200-500Hz frequency band; With a sliding window length of 1 second, the normalized correlation coefficient between the energy sequence of the frequency band and the vehicle speed sequence is calculated; When the correlation coefficient is greater than 0.8 and the absolute value of the vehicle speed change rate in the window is less than 3 m / s 2 When , it is determined that the door resonance condition is met; Step S3: The passenger's head position and pitch angle are acquired through the far-infrared thermal imager on the roof. When the pitch angle is greater than 30° and the eyelids are closed for 5 minutes, a rest mode activation instruction is sent to the sound zone control module. Step S4: Receive the device status code of the in-vehicle infotainment system, trigger the central control screen speaker position lock when the status code is navigation broadcast, trigger the door speaker position lock when the status code is music playback, and transmit the coordinates to the spatial sound field management module; Step S5: In response to the noise identification, the processing units are driven in sequence under the control of the noise reduction processing scheduling module: the engine noise identification activates the anti-phase cancellation unit, the door resonance identification activates the notch filter unit, and the wind noise identification activates the adaptive tracking unit; Step S6: receiving a rest mode activation instruction, and performing partition processing in the sound zone control module: generating an active noise reduction shutdown instruction for the rest area, and outputting voice enhancement beam parameters for the non-rest area; Step S7: generating a spatial sound field reserved area centered on the locked position based on the coordinate data of the spatial sound field management module, and returning the noise reduction gain in the area to zero; Step S8: Using the domain-specific output actuator, the headrest in the rest area outputs psychoacoustic masking noise, and the ceiling in the non-rest area outputs a composite noise reduction signal.
[0009] The present invention realizes the accurate classification of engine noise, door resonance and wind noise through the collaborative detection and time domain alignment of multi-source sensors; dynamically divides the noise reduction strategies of rest areas and non-rest areas by combining passenger status recognition and equipment status perception; utilizes the spatial sound field reserved area to protect the output of navigation / music equipment from noise reduction interference; and finally realizes adaptive noise reduction of the entire vehicle domain through the partitioned output of rest area masking noise and non-rest area composite noise reduction. Preferably, in the above-mentioned vehicle-mounted multimedia audio noise reduction method, the noise identification generation in step S2 forms a closed loop with the noise reduction control through a vehicle parameter linkage mechanism, including: The engine noise identification driver calls the engine speed data from the OBD-II interface and inputs the speed data into the inverting amplitude mapping unit, so that the output amplitude of the inverting cancellation unit is dynamically adjusted according to the preset speed-gain curve; The door resonance indicator triggers the reading of the body roll angle sensor data. When the data is determined by the cornering state identifier to be a cornering condition, a depth increase instruction is automatically sent to the notch filter unit; The wind noise indicator activates the rearview mirror angle tracking loop, allowing the center frequency control module of the adaptive tracking unit to follow the changes in the deflection angle of the electronic rearview mirror in real time.
[0010] The noise signature and vehicle dynamic parameters form a closed-loop linkage, optimizing noise reduction parameters in real time. The engine noise reduction amplitude increases with the increase of engine speed, solving the problem of sudden noise increase under acceleration conditions. The door resonance suppression depth is enhanced when cornering to offset the abnormal noise caused by body roll. The wind noise frequency is dynamically tracked when the rearview mirror is deflected to avoid missing wind noise when changing lanes at high speed, thereby improving the overall real-time performance and scene adaptability of noise reduction.
[0011] Preferably, in the above-mentioned in-vehicle multimedia audio noise reduction method, the generation of the rest mode activation instruction in step S3 realizes a state response closed loop through a biometric analysis chain, including: The raw data collected by the thermal imager is input into the head posture recognition unit for processing, and the pitch angle parameters and eye opening and closing status are output; When the output pitch angle parameter is greater than 30° and the eye state is closed, activating the timing unit to start accumulating the duration; When the accumulated time in the timing unit exceeds the 5-minute threshold, a Rest Mode activation command is generated and transmitted to the audio zone control module, triggering the shutdown of the active noise reduction circuit in the corresponding audio zone. Multi-level verification and timing based on head posture and eye state ensure that Rest Mode is only activated when the passenger is actually asleep, avoiding false triggering. Disabling the active noise reduction circuit in the corresponding zone eliminates the ear pressure discomfort caused by traditional noise reduction systems while avoiding energy waste, significantly improving passenger sleep comfort and system energy efficiency.
[0012] Preferably, in the above-mentioned vehicle-mounted multimedia audio noise reduction method, the speaker position locking operation in step S4 is driven by a device status response mechanism and includes: The navigation broadcast status code triggers the central control screen speaker positioning program, which calculates the coordinates of a spherical sound field reserved area with a radius of 15 cm and centered on the central control screen, and transmits the calculation results to the spatial coordinate database; The music playing status code triggers the door speaker positioning program, which calculates the coordinates of a cylindrical sound field reservation area with a radius of 25 cm, centered on the door speaker diaphragm, and transmits the calculation results to the spatial coordinate database; The spatial coordinate database responds to the calling request of the spatial sound field management module and outputs the coordinate set of the sound field reserved area.
[0013] The device status code (navigation / music) directly triggers the generation of a spatial sound field reservation area for the corresponding speaker, ensuring that navigation voice is not mistakenly eliminated by the noise reduction system, maintaining command clarity; the core frequency band of music playback is fully preserved to avoid sound quality degradation; and the spatial coordinate database centrally manages coordinate sets, improving processing efficiency and effectively resolving conflicts between vehicle device output and the noise reduction system.
[0014] Preferably, in the above-mentioned vehicle-mounted multimedia audio noise reduction method, the noise reduction processing in step S5 implements resource allocation control through a time-sharing scheduling mechanism, including: The engine noise indicator triggers the processing resource allocator to exclusively allocate resources within a 0-5ms period, driving the anti-phase cancellation unit to generate a cancellation signal and output it to the driver's footwell area speaker; The door resonance indicator triggers resource switching after the engine noise processing is completed, driving the notch filter unit to generate a filtered signal within a 5-10ms period and output it to the floor speaker; After pre-processing, the wind noise indicator triggers resource reallocation, driving the adaptive tracking unit to generate a notch signal within a 10-15ms period and output it to the ceiling array. The footwell speakers refer to those installed in the footrest area of the cockpit.
[0015] The time-sharing scheduling mechanism ensures the sequential execution of the three types of noise processing to avoid conflicts in multi-task parallelism; exclusive resource allocation ensures real-time suppression of critical noise; and spatial division of labor deployment matches the physical location of the noise source to achieve efficient collaborative processing of multiple noise sources.
[0016] Preferably, in the above-mentioned in-vehicle multimedia audio noise reduction method, the deployment of the anti-phase cancellation signal is optimized and executed by a spatial load balancing system, including: Based on the distribution of engine noise energy in the vehicle interior, the load distributor allocates the driver's seat engine noise processing task to the left driver's seat footwell area speaker; The load distributor distributes the task of handling the passenger seat engine noise to the right driver's footwell area speaker; The power control module receives the noise intensity detection data of each area in real time and dynamically adjusts the output power ratio of each speaker group accordingly.
[0017] The spatial load balancing system allocates processing tasks to the corresponding speaker groups based on the noise distribution map, and dynamically adjusts the power based on the noise intensity, so that the engine noise in the driver's seat and the front passenger seat can be independently suppressed to avoid crosstalk. At the same time, the speaker power is distributed on demand to reduce overall energy consumption. The noise in each area is balanced and offset to improve the overall noise reduction uniformity, effectively solving the problems of uneven speaker load and local excessive noise in traditional solutions.
[0018] Preferably, in the above-mentioned in-vehicle multimedia audio noise reduction method, the human voice enhancement beam in step S6 is precisely controlled by a physical constraint optimization mechanism, including: The beamforming controller receives passenger ear coordinate data and constrains the beam spreading angle to be greater than 25° based on the physical layout parameters of the ceiling array; The pointing calibration unit calculates the beam axis angle in real time according to the ear coordinate data, allowing a spatial tolerance range of ±10 cm; The sound pressure feedback loop monitors the changes in ambient volume in real time and controls the beam sound pressure level to maintain at +3dB of the ambient volume.
[0019] The present invention adopts a physical constraint optimization mechanism with a minimum diffusion angle of 25°, a tolerance of ±10cm, and closed-loop sound pressure control to ensure that the voice enhancement beam achieves optimal directionality within the physical limitations of the on-board speakers. The tolerance range also avoids voice interruptions caused by slight head movements of passengers. At the same time, the sound pressure level is adaptively adjusted with the ambient noise to maintain clarity, thereby significantly improving the voice communication quality for passengers in non-rest areas.
[0020] Preferably, in the above-mentioned vehicle-mounted multimedia audio noise reduction method, the spatial sound field reserved area in step S7 is optimized by an intelligent gradual control system, including: The boundary expander is based on the boundary of the sound field reserved area and expands outward by 10cm to form a transition zone space area; The gradient controller performs noise reduction intensity control in the transition zone space, making it decrease linearly from 100% at the boundary of the retention area to 0% at the outer edge of the transition zone; The emergency response module monitors changes in navigation volume in real time. When a sudden increase of more than 6dB is detected, the reserved area radius is triggered to expand by 5cm instantly.
[0021] The intelligent gradient control system can eliminate auditory discomfort caused by sudden changes in the sound field at the protection boundary. It instantly expands the protection zone when the volume suddenly increases, preventing the navigation prompt sound from being mistakenly suppressed by the noise reduction system. The smooth transition ensures natural switching of the sound field, solving the problems of sound field jumps at the protection boundary and sudden volume leakage in traditional solutions.
[0022] Preferably, in the above-mentioned vehicle-mounted multimedia audio noise reduction method, the psychoacoustic masking noise in step S8 is generated by a real-time spectrum matching system, including: The ambient noise collected by the headrest microphone is input into the fast spectrum analysis unit for processing; Based on the 1 / 3 octave spectrum data updated every 200ms by the unit, the masking generator drives the generation of a masking signal with a spectrum level difference of less than 3dB from the ambient noise; The masking signal is transmitted to the headrest speakers in the rest area for playback after quality verification.
[0023] The real-time spectrum matching system ensures that the masking sound is highly similar to the ambient noise, utilizing the auditory masking effect to maximize noise reduction perception. The quality verification process avoids distorted signal output, while low-latency refresh adapts to dynamic noise environments, providing passengers in the rest area with a comfortable noise reduction experience without ear pressure.
[0024] Preferably, in the above-mentioned vehicle-mounted multimedia audio noise reduction method, the continuity of the psychoacoustic masking noise is ensured by a dynamic response system, including: The energy mutation detector monitors the environmental noise energy changes in real time and triggers the spectrum reconstruction process when the change exceeds 5dB; The emergency switch is activated during spectrum reconstruction, calling the best matching template in the pre-stored masking template library to maintain the output; Once the stability monitoring unit confirms that the ambient energy is stable, it triggers the system to switch back to real-time analysis and generation mode. The dynamic response system addresses issues such as masking failure caused by sudden noise changes and output interruptions during the reconfiguration phase, preventing auditory discomfort caused by frequent reconfiguration. This ensures continuous noise reduction and prevents sudden noise interruptions from disrupting passengers' sleep.
[0025] It can be seen from the above technical solution that compared with the existing technology, the present invention discloses a method for vehicle-mounted multimedia audio noise reduction. The present invention fundamentally solves the contradiction between vehicle-mounted environmental noise suppression and useful signal protection by constructing a dual mechanism of multimodal sensing collaboration and dynamic spatial partitioning control.
[0026] This invention integrates three key technical components—noise source physical property identification, human-machine state perception, and spatial isolation of device signals—into a closed-loop system, enabling precise control of the entire vehicle's acoustic environment. Specifically, through the collaborative analysis of vibration spectra and wind pressure fluctuation data, the system accurately separates mixed noise components, such as engine, door, and wind noise, and achieves targeted suppression of these noise types through a time-sharing scheduling mechanism. Furthermore, it utilizes biometric recognition to dynamically divide sound zones, implementing voice enhancement beam control in active areas while ensuring auditory comfort in rest areas.
[0027] Furthermore, this invention completely avoids the useful signal damage caused by traditional frequency-domain filtering by establishing a spatial sound field preservation zone centered on the device's speaker. The interaction of these three technical elements enables the vehicle to maintain a dynamic balance between noise reduction depth and signal fidelity under complex operating conditions such as high-speed lane changes, rapid acceleration, and open windows.
[0028] This invention uses the vehicle bus to acquire dynamic parameters such as engine speed and vehicle posture in real time, enabling adaptive matching of noise reduction strategies with driving conditions. This significantly improves the real-time and stability of noise tracking. Ultimately, this system optimizes the entire process, from noise source isolation and state perception to spatial sound field control. This reduces system energy consumption while comprehensively enhancing the acoustic experience for drivers and passengers. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0030] Figure 1 The accompanying drawing is a flow chart of a vehicle-mounted multimedia audio noise reduction method according to the present invention. DETAILED DESCRIPTION
[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.
[0032] An embodiment of the present invention discloses a method for reducing noise in a vehicle-mounted multimedia audio system, comprising: Step S1: The piezoelectric vibration sensor in the engine compartment collects the 20-500 Hz mechanical vibration spectrum. Simultaneously, the MEMS air pressure sensor in the door cavity acquires 0.5-5 kHz wind pressure fluctuation data. The two types of sensor data are synchronized through a time domain alignment controller and then input into the noise signature analysis module. Step S2: In the noise signature analysis module, a classification decision is made based on the vibration energy distribution: when the energy in the 80-120 Hz frequency band accounts for more than 70% of the total vibration energy, an engine noise flag is generated; when the peak value of the correlation coefficient between the pulse signal greater than 1 kHz and the wind pressure fluctuation signal is greater than 0.9, a wind noise flag is generated; when the door resonance condition is met, a door resonance flag is generated, specifically: The vibration signal is processed by 200Hz high-pass filtering and 500Hz low-pass filtering to extract the energy sequence of the 200-500Hz frequency band; With a sliding window length of 1 second, the normalized correlation coefficient between the energy sequence of the frequency band and the vehicle speed sequence is calculated; When the correlation coefficient is greater than 0.8 and the absolute value of the vehicle speed change rate in the window is less than 3m / s 2 When , it is determined that the door resonance condition is met; Step S3: The passenger's head position and pitch angle are acquired through the far-infrared thermal imager on the roof. When the pitch angle is greater than 30° and the eyelids are closed for 5 minutes, a rest mode activation instruction is sent to the sound zone control module. Step S4: Receive the device status code of the in-vehicle infotainment system, trigger the central control screen speaker position lock when the status code is navigation broadcast, trigger the door speaker position lock when the status code is music playback, and transmit the coordinates to the spatial sound field management module; Step S5: In response to the noise identification, the processing units are driven in sequence under the control of the noise reduction processing scheduling module: the engine noise identification activates the anti-phase cancellation unit, the door resonance identification activates the notch filter unit, and the wind noise identification activates the adaptive tracking unit; Step S6: receiving a rest mode activation instruction, and performing partition processing in the sound zone control module: generating an active noise reduction shutdown instruction for the rest area, and outputting voice enhancement beam parameters for the non-rest area; Step S7: generating a spatial sound field reserved area centered on the locked position based on the coordinate data of the spatial sound field management module, and returning the noise reduction gain in the area to zero; Step S8: Using the domain-specific output actuator, the headrest in the rest area outputs psychoacoustic masking noise, and the ceiling in the non-rest area outputs a composite noise reduction signal.
[0033] The present invention realizes the accurate classification of engine noise, door resonance and wind noise through the collaborative detection and time domain alignment of multi-source sensors; dynamically divides the noise reduction strategies of rest areas and non-rest areas by combining passenger status recognition and equipment status perception; utilizes the spatial sound field reserved area to protect the output of navigation / music equipment from noise reduction interference; and finally realizes adaptive noise reduction of the entire vehicle domain through the partitioned output of rest area masking noise and non-rest area composite noise reduction. To further optimize the above technical solution, the noise signature generation in step S2 forms a closed loop with the noise reduction control through a vehicle parameter linkage mechanism, including: The engine noise identification driver calls the engine speed data from the OBD-II interface and inputs the speed data into the inverting amplitude mapping unit, so that the output amplitude of the inverting cancellation unit is dynamically adjusted according to the preset speed-gain curve; The door resonance indicator triggers the reading of the body roll angle sensor data. When the data is determined by the cornering state identifier to be a cornering condition, a depth increase instruction is automatically sent to the notch filter unit; The wind noise indicator activates the rearview mirror angle tracking loop, allowing the center frequency control module of the adaptive tracking unit to follow the changes in the deflection angle of the electronic rearview mirror in real time.
[0034] The noise signature and vehicle dynamic parameters form a closed-loop linkage, optimizing noise reduction parameters in real time. The engine noise reduction amplitude increases with the increase of engine speed, solving the problem of sudden noise increase under acceleration conditions. The door resonance suppression depth is enhanced when cornering to offset the abnormal noise caused by body roll. The wind noise frequency is dynamically tracked when the rearview mirror is deflected to avoid missing wind noise when changing lanes at high speed, thereby improving the overall real-time performance and scene adaptability of noise reduction.
[0035] To further optimize the above technical solution, the rest mode activation instruction generation in step S3 implements a state response closed loop through a biometric analysis chain, including: The raw data collected by the thermal imager is input into the head posture recognition unit for processing, and the pitch angle parameters and eye opening and closing status are output; When the output pitch angle parameter is greater than 30° and the eye state is closed, the activation timing unit starts to accumulate the duration; When the accumulated time in the timing unit exceeds the 5-minute threshold, a Rest Mode activation command is generated and transmitted to the audio zone control module, triggering the shutdown of the active noise reduction circuit in the corresponding audio zone. Multi-level verification and timing based on head posture and eye state ensure that Rest Mode is only activated when the passenger is actually asleep, avoiding false triggering. Disabling the active noise reduction circuit in the corresponding zone eliminates the ear pressure discomfort caused by traditional noise reduction systems while avoiding energy waste, significantly improving passenger sleep comfort and system energy efficiency.
[0036] To further optimize the above technical solution, the speaker position locking operation in step S4 is driven by the device status response mechanism and includes: The navigation broadcast status code triggers the central control screen speaker positioning program, which calculates the coordinates of a spherical sound field reserved area with a radius of 15 cm and centered on the central control screen, and transmits the calculation results to the spatial coordinate database; The music playing status code triggers the door speaker positioning program, which calculates the coordinates of a cylindrical sound field reservation area with a radius of 25 cm, centered on the door speaker diaphragm, and transmits the calculation results to the spatial coordinate database; The spatial coordinate database responds to the calling request of the spatial sound field management module and outputs the coordinate set of the sound field reserved area.
[0037] The device status code (navigation / music) directly triggers the generation of a spatial sound field reservation area for the corresponding speaker, ensuring that navigation voice is not mistakenly eliminated by the noise reduction system, maintaining command clarity; the core frequency band of music playback is fully preserved to avoid sound quality degradation; and the spatial coordinate database centrally manages coordinate sets, improving processing efficiency and effectively resolving conflicts between vehicle device output and the noise reduction system.
[0038] In order to further optimize the above technical solution, the noise reduction process in step S5 implements resource allocation control through a time-sharing scheduling mechanism, including: The engine noise indicator triggers the processing resource allocator to exclusively allocate resources within a 0-5ms period, driving the anti-phase cancellation unit to generate a cancellation signal and output it to the driver's footwell area speaker; The door resonance indicator triggers resource switching after the engine noise processing is completed, driving the notch filter unit to generate a filtered signal within a 5-10ms period and output it to the floor speaker; After pre-processing, the wind noise indicator triggers resource reallocation, driving the adaptive tracking unit to generate a notch signal within a 10-15ms period and output it to the ceiling array. The footwell speakers refer to those installed in the footrest area of the cockpit.
[0039] The time-sharing scheduling mechanism ensures the sequential execution of the three types of noise processing to avoid conflicts in multi-task parallelism; exclusive resource allocation ensures real-time suppression of critical noise; and spatial division of labor deployment matches the physical location of the noise source to achieve efficient collaborative processing of multiple noise sources.
[0040] To further optimize the above technical solution, the deployment of the anti-phase cancellation signal is optimized through a spatial load balancing system, including: Based on the distribution of engine noise energy in the vehicle interior, the load distributor allocates the driver's seat engine noise processing task to the left driver's seat footwell area speaker; The load distributor distributes the task of handling the passenger seat engine noise to the right driver's footwell area speaker; The power control module receives the noise intensity detection data of each area in real time and dynamically adjusts the output power ratio of each speaker group accordingly.
[0041] The spatial load balancing system allocates processing tasks to the corresponding speaker groups based on the noise distribution map, and dynamically adjusts the power based on the noise intensity, so that the engine noise in the driver's seat and the front passenger seat can be independently suppressed to avoid crosstalk. At the same time, the speaker power is distributed on demand to reduce overall energy consumption. The noise in each area is balanced and offset to improve the overall noise reduction uniformity, effectively solving the problems of uneven speaker load and local excessive noise in traditional solutions.
[0042] To further optimize the above technical solution, the voice enhancement beam in step S6 is precisely controlled through a physical constraint optimization mechanism, including: The beamforming controller receives passenger ear coordinate data and constrains the beam spreading angle to be greater than 25° based on the physical layout parameters of the ceiling array; The pointing calibration unit calculates the beam axis angle in real time based on the ear coordinate data, allowing a spatial tolerance range of ±10cm; The sound pressure feedback loop monitors the changes in ambient volume in real time and controls the beam sound pressure level to maintain at +3dB of the ambient volume.
[0043] The present invention adopts a physical constraint optimization mechanism with a minimum diffusion angle of 25°, a tolerance of ±10cm, and closed-loop sound pressure control to ensure that the voice enhancement beam achieves optimal directionality within the physical limitations of the on-board speakers. The tolerance range also avoids voice interruptions caused by slight head movements of passengers. At the same time, the sound pressure level is adaptively adjusted with the ambient noise to maintain clarity, thereby significantly improving the voice communication quality for passengers in non-rest areas.
[0044] In order to further optimize the above technical solution, the spatial sound field reserved area in step S7 is optimized by an intelligent gradual control system, including: The boundary expander is based on the boundary of the sound field reserved area and expands outward by 10cm to form a transition zone space area; The gradient controller performs noise reduction intensity control in the transition zone space, making it decrease linearly from 100% at the boundary of the retention area to 0% at the outer edge of the transition zone; The emergency response module monitors changes in navigation volume in real time. When a sudden increase of more than 6dB is detected, the reserved area radius is triggered to expand by 5cm instantly.
[0045] The intelligent gradient control system can eliminate auditory discomfort caused by sudden changes in the sound field at the protection boundary. It instantly expands the protection zone when the volume suddenly increases, preventing the navigation prompt sound from being mistakenly suppressed by the noise reduction system. The smooth transition ensures natural switching of the sound field, solving the problems of sound field jumps at the protection boundary and sudden volume leakage in traditional solutions.
[0046] In order to further optimize the above technical solution, the psychoacoustic masking noise in step S8 is generated by a real-time spectrum matching system and includes: The ambient noise collected by the headrest microphone is input into the fast spectrum analysis unit for processing; Based on the 1 / 3 octave spectrum data updated every 200ms by the unit, the masking generator drives the generation of a masking signal with a spectrum level difference of less than 3dB from the ambient noise; After quality verification, the masking signal is transmitted to the headrest speakers in the rest area for playback.
[0047] The real-time spectrum matching system ensures that the masking sound is highly similar to the ambient noise, utilizing the auditory masking effect to maximize noise reduction perception. The quality verification process avoids distorted signal output, while low-latency refresh adapts to dynamic noise environments, providing passengers in the rest area with a comfortable noise reduction experience without ear pressure.
[0048] To further optimize the above technical solution, the continuity of the psychoacoustic masking noise is ensured by a dynamic response system, including: The energy mutation detector monitors the environmental noise energy changes in real time and triggers the spectrum reconstruction process when the change exceeds 5dB; The emergency switch is activated during spectrum reconstruction, calling the best matching template in the pre-stored masking template library to maintain the output; Once the stability monitoring unit confirms that the ambient energy is stable, it triggers the system to switch back to real-time analysis and generation mode. The dynamic response system addresses issues such as masking failure caused by sudden noise changes and output interruptions during the reconfiguration phase, preventing auditory discomfort caused by frequent reconfiguration. This ensures continuous noise reduction and prevents sudden noise interruptions from disrupting passengers' sleep.
[0049] Technical principle: The core principle of this invention is to build a closed-loop interactive system between multimodal sensor data and vehicle dynamic status, and achieve precise noise reduction through a three-level collaborative mechanism of physical noise source separation, human-machine state space mapping, and device signal space isolation.
[0050] Specifically, the physical property differences between vibration sensors and air pressure sensors are first used to separate independent identifiers of noise sources such as engine, door, and wind noise from the vibration spectrum energy distribution and wind pressure fluctuation characteristics; at the same time, the passenger activity status is dynamically divided through biometric features captured by the thermal imager to generate differentiated noise reduction instructions; in this process, the engine speed, body inclination and other data transmitted in real time by the vehicle bus continuously calibrate the noise reduction parameters to form an adaptive match with the driving conditions.
[0051] To protect useful signals, the system analyzes device status codes and locks the physical location of the speakers, generating a three-dimensional sound field reservation zone within the digital signal processing layer, rendering noise reduction ineffective within this area. Ultimately, a time-sharing scheduling mechanism prioritizes various noise suppression commands to directional speaker arrays, including those in the footwell, floor, and ceiling. Simultaneously, the system combines biometric state-based zoning to output masking sound waves and enhancement beams, achieving dynamic balance control of the entire vehicle's acoustic environment. The entire system, through real-time data exchange among the noise classification module, spatial sound field management module, and zoning control module, forms a closed-loop control chain from noise perception to spatial execution, ensuring the balance between noise reduction effectiveness and auditory comfort.
[0052] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.
[0053] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for reducing noise in vehicle-mounted multimedia audio, characterized in that: include: Step S1: The piezoelectric vibration sensor in the engine compartment collects the 20-500 Hz mechanical vibration spectrum. Simultaneously, the MEMS air pressure sensor in the door cavity acquires 0.5-5 kHz wind pressure fluctuation data. The two types of sensor data are synchronized through a time domain alignment controller and then input into the noise signature analysis module. Step S2: In the noise signature analysis module, a classification decision is made based on the vibration energy distribution: when the energy in the 80-120 Hz frequency band accounts for more than 70% of the total vibration energy, an engine noise flag is generated; when the door resonance condition is met, a door resonance flag is generated; and when the peak value of the correlation coefficient between the pulse signal greater than 1 kHz and the wind pressure fluctuation signal is greater than 0.9, a wind noise flag is generated. Step S3: The passenger's head position and pitch angle are acquired through the far-infrared thermal imager on the roof. When the pitch angle is greater than 30° and the eyelids are closed for 5 minutes, a rest mode activation instruction is sent to the sound zone control module. Step S4: Receive the device status code of the in-vehicle infotainment system, trigger the central control screen speaker position lock when the status code is navigation broadcast, trigger the door speaker position lock when the status code is music playback, and transmit the coordinates to the spatial sound field management module; Step S5: In response to the noise identification, the processing units are driven in sequence under the control of the noise reduction processing scheduling module: the engine noise identification activates the anti-phase cancellation unit, the door resonance identification activates the notch filter unit, and the wind noise identification activates the adaptive tracking unit; Step S6: receiving a rest mode activation instruction, and performing partition processing in the sound zone control module: generating an active noise reduction shutdown instruction for the rest area, and outputting voice enhancement beam parameters for the non-rest area; Step S7: generating a spatial sound field reserved area centered on the locked position based on the coordinate data of the spatial sound field management module, and returning the noise reduction gain in the area to zero; Step S8: Using the domain-specific output actuator, the headrest in the rest area outputs psychoacoustic masking noise, and the ceiling in the non-rest area outputs a composite noise reduction signal.
2. The vehicle-mounted multimedia audio noise reduction method according to claim 1, characterized in that: The noise identification generation in step S2 forms a closed loop with the noise reduction control through the vehicle parameter linkage mechanism, including: The engine noise identification driver calls the engine speed data from the OBD-II interface and inputs the speed data into the inverting amplitude mapping unit, so that the output amplitude of the inverting cancellation unit is dynamically adjusted according to the preset speed-gain curve; The door resonance indicator triggers the reading of the body roll angle sensor data. When the data is determined by the cornering state identifier to be a cornering condition, a depth increase instruction is automatically sent to the notch filter unit; The wind noise indicator activates the rearview mirror angle tracking loop, allowing the center frequency control module of the adaptive tracking unit to follow the changes in the deflection angle of the electronic rearview mirror in real time.
3. The vehicle-mounted multimedia audio noise reduction method according to claim 1, characterized in that: The generation of the rest mode activation instruction in step S3 realizes a state response closed loop through a biometric analysis chain, including: The raw data collected by the thermal imager is input into the head posture recognition unit for processing, and the pitch angle parameters and eye opening and closing status are output; When the output pitch angle parameter is greater than 30° and the eye state is closed, activating the timing unit to start accumulating the duration; When the accumulated time of the timing unit exceeds the 5-minute threshold, a rest mode activation instruction is generated and transmitted to the sound zone control module, triggering the operation of shutting down the active noise reduction circuit of the corresponding sound zone.
4. The vehicle-mounted multimedia audio noise reduction method according to claim 1, characterized in that: The speaker position locking operation in step S4 is driven by the device status response mechanism and includes: The navigation broadcast status code triggers the central control screen speaker positioning program, which calculates the coordinates of a spherical sound field reserved area with a radius of 15 cm and centered on the central control screen, and transmits the calculation results to the spatial coordinate database; The music playing status code triggers the door speaker positioning program, which calculates the coordinates of a cylindrical sound field reservation area with a radius of 25 cm, centered on the door speaker diaphragm, and transmits the calculation results to the spatial coordinate database; The spatial coordinate database responds to the calling request of the spatial sound field management module and outputs the coordinate set of the sound field reserved area.
5. The vehicle-mounted multimedia audio noise reduction method according to claim 1, characterized in that: The noise reduction process in step S5 implements resource allocation control through a time-sharing scheduling mechanism, including: The engine noise indicator triggers the processing resource allocator to exclusively allocate resources within a 0-5ms period, driving the anti-phase cancellation unit to generate a cancellation signal and output it to the driver's footwell area speaker; The door resonance indicator triggers resource switching after the engine noise processing is completed, driving the notch filter unit to generate a filtered signal within a 5-10ms period and output it to the floor speaker; The wind noise indicator triggers resource reallocation after the pre-processing is completed, driving the adaptive tracking unit to generate a notch signal within a 10-15ms period and output it to the ceiling array.
6. The vehicle-mounted multimedia audio noise reduction method according to claim 5, characterized in that: The deployment of the anti-phase cancellation signals is optimized through a spatial load balancing system, including: Based on the distribution of engine noise energy in the vehicle interior, the load distributor allocates the driver's seat engine noise processing task to the left driver's seat footwell area speaker; The load distributor distributes the task of handling the passenger seat engine noise to the right driver's footwell area speaker; The power control module receives the noise intensity detection data of each area in real time and dynamically adjusts the output power ratio of each speaker group accordingly.
7. The vehicle-mounted multimedia audio noise reduction method according to claim 1, characterized in that: The vocal enhancement beam in step S6 is precisely controlled through a physical constraint optimization mechanism, including: The beamforming controller receives passenger ear coordinate data and constrains the beam spreading angle to be greater than 25° based on the physical layout parameters of the ceiling array; The pointing calibration unit calculates the beam axis angle in real time according to the ear coordinate data, allowing a spatial tolerance range of ±10 cm; The sound pressure feedback loop monitors the changes in ambient volume in real time and controls the beam sound pressure level to maintain at +3dB of the ambient volume.
8. The vehicle-mounted multimedia audio noise reduction method according to claim 1, characterized in that: The spatial sound field reserved area in step S7 is optimized by using an intelligent gradual change control system, including: The boundary expander is based on the boundary of the sound field reserved area and expands outward by 10cm to form a transition zone space area; The gradient controller performs noise reduction intensity control in the transition zone space, making it decrease linearly from 100% at the boundary of the retention area to 0% at the outer edge of the transition zone; The emergency response module monitors changes in navigation volume in real time. When a sudden increase of more than 6dB is detected, the reserved area radius is triggered to expand by 5cm instantly.
9. The vehicle-mounted multimedia audio noise reduction method according to claim 1, characterized in that: The psychoacoustic masking noise in step S8 is generated by a real-time spectrum matching system and includes: The ambient noise collected by the headrest microphone is input into the fast spectrum analysis unit for processing; Based on the 1 / 3 octave spectrum data updated every 200ms by the unit, the masking generator drives the generation of a masking signal with a spectrum level difference of less than 3dB from the ambient noise; The masking signal is transmitted to the headrest speakers in the rest area for playback after quality verification.
10. The vehicle-mounted multimedia audio noise reduction method according to claim 9, characterized in that: The continuity of psychoacoustic masking noise is ensured by a dynamic response system, including: The energy mutation detector monitors the environmental noise energy changes in real time and triggers the spectrum reconstruction process when the change exceeds 5dB; The emergency switch is activated during spectrum reconstruction, calling the best matching template in the pre-stored masking template library to maintain the output; After the stability monitoring unit confirms that the environmental energy is stable, it triggers the system to switch back to the real-time analysis and generation mode.