Multi-source Bluetooth audio stream arbitration system based on vehicle state perception and anti-interference method

Abstract

The invention provides a multi-source Bluetooth audio stream arbitration system based on vehicle state perception and an anti-interference method, and relates to the technical field of vehicle-mounted communication. The system comprises a vehicle state and environment sensing module, a multi-source Bluetooth communication management module and a central arbitration processing unit. The method comprises the following steps: collecting vehicle kinetic parameters, monitoring electrical characteristics of an execution part by using an LC oscillation detection unit, and constructing a dynamic electromagnetic interference potential energy model to predict a future interference trend; when the predicted value exceeds the standard, preemptive link reconstruction is triggered before interference occurs, the type of an underlying data packet is forcibly adjusted to be a high-anti-interference single-slot packet, and an anti-interference frequency hopping table is locked; meanwhile, the audio stream weight is calculated in combination with the vehicle speed and the safety index, and dynamic sorting and output arbitration are carried out on multi-source audios. According to the invention, spectral bandwidth replacement is carried out by using a wind noise masking effect, and joint optimization of active anti-interference of a physical layer and intelligent scheduling of an application layer is realized by cooperating with user interaction feedback.

Description

Technical Field

[0001] This invention relates to the field of vehicle communication technology, and in particular to a multi-source Bluetooth audio stream arbitration system and anti-interference method based on vehicle status awareness. Background Technology

[0002] With the increasing sophistication of automotive smart cockpit functions, in-vehicle Bluetooth technology has become a crucial link connecting mobile terminals and vehicle systems, widely used in scenarios such as hands-free calling, media playback, and navigation voice transmission. In the complex driving environment, the vehicle interior contains numerous inductive load components, such as window lift motors, wiper motors, and seat adjustment motors. These components generate broadband radio frequency noise during frequent starts, stops, or reversals, easily causing electromagnetic interference to Bluetooth communication links operating in similar frequency bands. Existing in-vehicle Bluetooth communication solutions typically rely on the protocol stack's own retransmission mechanisms or adaptive frequency hopping technology to handle data loss. These mechanisms are primarily based on statistical analysis and responses to errors, triggering channel adjustment only after detecting an increase in the bit error rate or a link timeout. This delayed processing often fails to guarantee the continuity of the audio stream when facing sudden, strong interference.

[0003] Furthermore, drivers' audio information needs vary significantly under different driving conditions. During high-load conditions such as high-speed driving or emergency maneuvers, drivers' need for environmental awareness increases. If low-priority entertainment audio continues to occupy channel resources, it may affect the transmission efficiency of critical navigation warning information and even distract the driver. Current audio management logic mostly adopts fixed priority strategies, rarely incorporating the vehicle's real-time dynamic state, electrical interference characteristics, and the driver's psychological load into the arbitration process. This makes it difficult for audio stream scheduling strategies to balance communication stability and driving safety when dealing with extreme physical conditions or sudden changes in the electromagnetic environment. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-source Bluetooth audio stream arbitration system and anti-interference method based on vehicle state perception, so as to solve the problems pointed out in the background art.

[0005] In a first aspect, the present invention provides a multi-source Bluetooth audio stream arbitration system based on vehicle state awareness, the system comprising:

[0006] The vehicle status and environment perception module is used to collect the vehicle's dynamic parameters in real time through the vehicle's CAN bus interface, and simultaneously monitor the electrical characteristic data of the drive circuit of the vehicle body's actuators.

[0007] The multi-source Bluetooth communication management module is equipped with a Bluetooth baseband controller that supports multi-link concurrency. It is used to establish and maintain Bluetooth audio communication links with multiple external terminal devices and parse the protocol attributes of the audio stream.

[0008] The central arbitration processing unit is connected to the vehicle status and environment perception module and the multi-source Bluetooth communication management module respectively.

[0009] The central arbitration processing unit is configured to execute the following control logic:

[0010] A dynamic electromagnetic interference potential energy model is constructed, and the trend value of electromagnetic interference intensity within a future preset time window is calculated based on the electrical characteristic data.

[0011] When the electromagnetic interference intensity trend value exceeds the preset safety threshold, a preemptive link reconfiguration command is generated. Before the electromagnetic interference actually occurs, the Bluetooth baseband controller is controlled to forcibly adjust the underlying transmission data packet type of the Bluetooth audio communication link in order to improve the anti-interference and error correction redundancy of the link.

[0012] Simultaneously, the multi-source audio streams are dynamically sorted and arbitrated based on weights by combining the dynamic parameters and the electromagnetic interference intensity trend value.

[0013] Optionally, the central arbitration processing unit is pre-installed with a dynamic priority calculation engine, which calculates the real-time arbitration weight of each audio stream using the following formula. :

[0014] ;

[0015] in, The basic type score for audio streams, The current vehicle speed is one of the dynamic parameters. The preset reference velocity constant, The preset velocity offset constant, For the current vehicle safety index, , , These are preset weighting coefficients; when a certain audio stream's... When the value is lower than the adjudication threshold, the system suspends the audio stream and records the breakpoint timestamp at the time of suspension.

[0016] Optionally, the multi-source Bluetooth communication management module is configured with an audio type fingerprint database to identify whether the accessed audio stream belongs to navigation warning, real-time call, or media entertainment.

[0017] When the central arbitration processing unit determines that the current driving state is under high load, it locks the physical channel of the Bluetooth audio communication link, allowing only navigation warning audio to pass through, and sends a flow control pause command to media entertainment devices.

[0018] Optionally, the system may also include a user interaction feedback module;

[0019] When the output arbitration results in an audio source switch, the user interaction feedback module controls the vehicle ambient light to perform a breathing flash at a specific frequency. The frequency of the flash is positively correlated with the current electromagnetic interference intensity trend value to provide visual status feedback.

[0020] Optionally, the multi-source Bluetooth communication management module has a disconnection protection mechanism;

[0021] During the execution of the preemptive link reconfiguration instruction, if the connection of a non-playing Bluetooth device times out, the system simulates sending a virtual heartbeat maintenance packet to prevent the non-playing device from disconnecting its physical connection due to timeout, ensuring seamless switching during arbitration recovery.

[0022] Optionally, the vehicle body actuator includes a window lift motor; the vehicle status and environment perception module includes an LC vibration detection unit connected to the window lift motor;

[0023] The electrical characteristic data includes the rate of change of the oscillation frequency output by the LC oscillation detection unit;

[0024] The central arbitration processing unit identifies instantaneous abrupt changes in the rate of change of the oscillation frequency as precursor signals of electromagnetic interference. The greater the magnitude of the rate of change, the higher the predicted trend value of the electromagnetic interference intensity.

[0025] Optionally, the underlying transmission data packet types include high-throughput multi-slot packets and high-interference-resistance single-slot packets;

[0026] The specific execution logic of the preemptive link reconfiguration instruction is as follows:

[0027] The Bluetooth baseband controller is controlled to pause the transmission of the high-throughput multi-slot packets and forcibly switch to the high-interference-resistant single-slot packets with 2 / 3 ratio forward error correction coding;

[0028] At the same time, the anti-interference frequency hopping table is locked based on the electromagnetic interference intensity trend value, and frequency points that are predicted to be affected by motor commutation spark interference are actively eliminated.

[0029] Optionally, the central arbitration processing unit further includes a spectrum bandwidth replacement submodule;

[0030] The spectrum bandwidth replacement submodule is used to: construct a wind noise spectrum model based on the vehicle speed in the dynamic parameters, and identify invalid frequency bands in the audio stream that are physically masked by wind noise;

[0031] When switching to the high-instability single-time-slot packet causes a decrease in bandwidth, a high-pass filter is used in the preprocessing stage of audio encoding to suppress the signal amplitude of the invalid frequency band, thereby reducing the data entropy after encoding; or the bit allocation pool of the Bluetooth encoder is dynamically adjusted to reduce the number of quantization bits in the invalid frequency band, and the saved air interface bandwidth resources are used to transmit high-priority audio stream data to achieve joint optimization of the source channel.

[0032] Optionally, the system is also equipped with a Bluetooth tire pressure monitoring interface;

[0033] When the system detects that the vehicle-mounted Bluetooth tire pressure monitoring sensor is sending a data packet, the central arbitration processing unit inserts a silent time slot in the Bluetooth audio communication link. The duration of the silent time slot is synchronized with the data transmission cycle of the tire pressure monitoring sensor to eliminate co-channel interference.

[0034] In a second aspect, the present invention provides a multi-source Bluetooth audio stream arbitration system and anti-interference method based on vehicle state awareness, characterized in that, based on the system described in any one of the first aspects, the method includes the following steps:

[0035] Step S1: Real-time acquisition of vehicle CAN bus dynamic parameters, and monitoring of the electrical characteristics of the actuators using the LC oscillation detection unit;

[0036] Step S2: Using the dynamic electromagnetic interference potential energy model, predict the interference intensity within the next millisecond window based on the frequency change rate of electrical characteristics;

[0037] Step S3: If the predicted interference intensity does not exceed the limit, calculate the priority of each audio stream using the weighted formula and perform regular arbitration;

[0038] Step S4: If the interference intensity is predicted to exceed the limit, immediately trigger preemptive link reconstruction and switch Bluetooth transmission to high error correction, single time slot mode;

[0039] Step S5: Simultaneously calculate the wind noise masking effect, trim invalid frequency band data, and replace the bandwidth to ensure error correction transmission of critical audio data.

[0040] The present invention has achieved the following beneficial effects:

[0041] This invention achieves proactive defense against in-vehicle electromagnetic interference at the physical layer by constructing a vehicle state and environment perception system. The system utilizes an LC oscillation detection unit to capture the microscopic electrical characteristics of the vehicle's actuators. This allows it to predict interference trends and trigger preemptive link reconfiguration before electromagnetic interference actually deteriorates the communication link. It switches data packets to a highly interference-resistant single-timeslot mode and locks the anti-interference frequency hopping table, significantly improving communication stability under strong interference environments and avoiding the latency and bandwidth waste caused by traditional passive retransmission mechanisms.

[0042] This invention introduces a dynamic arbitration mechanism based on vehicle dynamics parameters and safety indices, which can intelligently adjust the priority weights of various audio streams under high-load conditions such as high-speed driving or emergency avoidance. The system calculates the audio stream weights based on real-time vehicle speed and vehicle safety status, automatically suppressing entertainment audio and prioritizing the transmission of navigation warning information. This achieves intelligent scheduling in accordance with ergonomic principles, ensuring the lossless and timely delivery of critical interactive commands in complex driving scenarios.

[0043] This invention utilizes the wind noise masking effect for spectrum bandwidth replacement. When the physical bandwidth decreases due to anti-interference mode, it actively eliminates invalid frequency band data that is imperceptible to the human ear, using the saved air interface resources to enhance the error correction redundancy of key signals. This ensures connection stability while maintaining subjective auditory clarity. Simultaneously, the integrated user interaction feedback module and disconnection protection mechanism provide intuitive status prompts through light effects and prevent inactive devices from unexpectedly dropping out, further optimizing the user experience in complex driving environments.

[0044] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0045] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0046] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0047] Figure 1 This is a schematic diagram of a multi-source Bluetooth audio stream arbitration system based on vehicle state perception in an embodiment of the present invention;

[0048] Figure 2 This is a flowchart of the system anti-interference method in an embodiment of the present invention. Detailed Implementation

[0049] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0050] Example 1:

[0051] See the instruction manual appendix Figure 1This invention provides a multi-source Bluetooth audio stream arbitration system based on vehicle state awareness, aiming to solve the problems of lag and passivity in existing in-vehicle wireless communication systems when facing interference from complex electromagnetic environments and concurrent multitasking within vehicles. In actual deployment, this system is mainly integrated into the vehicle's intelligent cockpit domain controller or a separate central gateway unit, utilizing its powerful edge computing capabilities to process high-frequency physical signals and protocol stack data.

[0052] The system provided in this embodiment of the invention includes four modules: a vehicle status and environment perception module, a multi-source Bluetooth communication management module, a central arbitration processing unit, and a user interaction feedback module. These modules communicate with each other with low latency via an onboard gigabit Ethernet or high-speed CAN-FD bus, while sharing the power supply network of the onboard power management system to ensure stable operation throughout the vehicle's entire lifecycle.

[0053] In this embodiment, the core function of the vehicle status and environment perception module is to construct an active physical environment detection network. This module is configured to directly interact with the vehicle's chassis powertrain domain and body control domain at the low-level signal level. Specifically, the module integrates a high-performance 32-bit microcontroller, which is physically connected to the vehicle's chassis powertrain network segment (P-CAN) and body comfort network segment (B-CAN) via a CAN transceiver with opto-isolation protection. In this embodiment, the vehicle status and environment perception module is configured to capture and parse specific ID messages on the bus in real time at a sampling frequency of no less than 50Hz, thereby obtaining the vehicle's dynamic parameters. Dynamic parameters are fundamental variables for assessing driver cognitive load and vehicle physical state. The system calculates high-precision real-time vehicle speed and longitudinal acceleration by analyzing raw pulse data from wheel speed sensors sent by the Anti-lock Braking System (ABS) or Electronic Stability Program (ESP); it calculates the vehicle's steering intention by analyzing the absolute value and angular velocity of the steering wheel angle sent by the Steering Angle Sensor (SAS); and it assesses the vehicle's stability by analyzing the yaw rate and lateral acceleration sent by the Inertial Measurement Unit (IMU). These dynamic parameters are strictly aligned on the time axis, forming a multi-dimensional feature vector describing the vehicle's current physical motion state. For example, when longitudinal acceleration is detected to be less than -0.5g and the ABS flag is activated, the system physically determines that the vehicle is in an emergency braking state, providing a physical basis for subsequent audio priority arbitration.

[0054] More importantly, considering the complexity of the in-vehicle electromagnetic environment, especially the potential impact of broadband radio frequency interference generated by high-power inductive loads during operation on the Bluetooth communication frequency band, the vehicle status and environment perception module also integrates an electrical characteristic monitoring circuit to synchronously monitor the electrical characteristic data of the drive circuits of the vehicle body actuators. These vehicle body actuators specifically include components with frequent start-stop characteristics and high operating current, such as window lift motors, electric sunroof drive motors, wiper motors, and electric seat adjustment motors. When these DC brushed motors are operating, the moment the commutator and carbon brushes separate, a high-frequency electrical spark is generated, radiating broadband radio frequency noise outwards. To accurately detect the precursors of electromagnetic interference generated by these actuators during commutation or sudden load changes, this embodiment does not use conventional Hall current sensors for simple current amplitude measurement in the drive power supply circuits of the vehicle body actuators. Instead, a high-sensitivity LC oscillation detection unit is connected in parallel.

[0055] The LC oscillation detection unit is a high-frequency signal capture circuit based on the principle of parallel resonance. Its circuit topology includes a high-frequency magnetic core inductor L and a low ESR (equivalent series resistance) ceramic capacitor C, which together form a parallel resonant circuit. In the specific circuit parameter design, considering that the center frequency of the Bluetooth band is approximately 2.44GHz, a high-quality microstrip line inductor or wire-wound inductor is selected, with an inductance value preferably between 1.0nH and 3.0nH. Combined with a picofarad-level high-precision ceramic capacitor, the resonant frequency of this circuit is specifically tuned to the high-energy frequency band generated by the motor commutation spark. When the carbon brushes inside the motor experience microscopic contact separation from the commutator contacts, the resulting arc excites the LC circuit to generate high-frequency damped oscillations. The high-speed analog-to-digital converter (ADC) inside the vehicle status and environment perception module performs envelope detection sampling of this oscillation signal at a sampling rate of up to 200kHz. To overcome the physical limitation of single envelope detection in obtaining carrier frequency information, the electrical characteristic monitoring circuit actually adopts a dual-path parallel architecture: the first path is the aforementioned envelope detection path, connected to an ADC to obtain the oscillation amplitude; the second path is the frequency monitoring path, where the output signal of the LC oscillation detection unit is simultaneously connected to a high-speed comparator or Schmitt trigger to shape the analog oscillation signal into a TTL level square wave. After the frequency is reduced by a hardware prescaler, it is input to the timer input capture pin of the microcontroller. By calculating the time difference between adjacent pulse edges, the microcontroller accurately calculates the real-time frequency and rate of change of the oscillation signal without relying on the ADC sampling rate. .

[0056] The electrical characteristic data specifically includes the rate of change (df / dt) of the oscillation frequency output by the LC oscillation detection unit and the instantaneous value of the oscillation amplitude. In the control logic of this embodiment, the central arbitration processing unit identifies the instantaneous abrupt change in the rate of change of the oscillation frequency as a precursor signal of electromagnetic interference. Physics principles show that in the initial stage of motor spark discharge, as the plasma channel forms and extinguishes, the equivalent impedance of the circuit changes drastically, causing the LC oscillation frequency to drift rapidly. The larger the rate of change, the stronger the arc energy and the more unstable the spectrum, resulting in a higher predicted trend value of electromagnetic interference intensity. This detection method based on physical layer microscopic electrical characteristics has a significant time lead compared to the post-hoc statistical method based on Bluetooth bit error rate (BER) in existing technologies. It can typically provide a warning 10 to 50 milliseconds in advance, thus providing sufficient time for the system to perform preemptive link reconfiguration.

[0057] Closely collaborating with the aforementioned sensing module is the multi-source Bluetooth communication management module, which serves as the system's wireless communication hub. At its core is an automotive-grade SoC chip supporting the Bluetooth 5.3 dual-mode (BR / EDR+BLE) standard. This module is equipped with a Bluetooth baseband controller supporting multi-link concurrency, capable of maintaining connections for at least three master and slave devices simultaneously: for example, connecting as the master device to the driver's smartphone for hands-free calling, simultaneously connecting to the passenger's tablet for A2DP media playback, and maintaining a BLE connection with the rear passenger's wearable device. At the software protocol stack level, this module has deeply extended the standard Bluetooth Host Controller Interface (HCI), enabling it to expose underlying link state information to the central arbitration processing unit and accept underlying configuration commands. Specifically, this module has the ability to deeply analyze the attributes of audio stream protocols. It can not only identify audio encoding formats (such as SBC, AAC, LDAC), but also accurately identify the service type of the access audio stream by analyzing the packet header information and quality of service (QoS) parameters of the Logical Link Control and Adaptation Protocol (L2CAP) layer, as well as the establishment requests of synchronous connection (SCO / eSCO) and asynchronous connection (ACL).

[0058] To support intelligent priority arbitration, the multi-source Bluetooth communication management module is equipped with an audio type fingerprint database. This database stores a matching template based on data stream characteristics to identify whether the incoming audio stream belongs to navigation alerts, real-time calls, or media entertainment. For example, navigation alert audio is characterized by strong bursts, short packet lengths, and transmission often completed shortly after connection establishment, or as mono audio data in the HFP protocol. Real-time call audio is characterized by strictly isochronous bidirectional data streams, typically occupying SCO or eSCO links, and strict restrictions on retransmission mechanisms. Media entertainment audio is characterized by continuous high-throughput data streams, occupying ACL links, and high buffer occupancy. By comparing the current data stream with the fingerprint database characteristics in real time, the module can determine with extremely high accuracy whether the current audio is navigation broadcast, teleconference, or music playback, thus providing a classification basis for subsequent priority arbitration. Specifically, for high-load operating conditions, when the central arbitration processing unit determines based on dynamic parameters that the vehicle is currently in a high-load driving state (e.g., detecting a steering wheel angle change rate exceeding a threshold, accompanied by significant lateral acceleration, indicating that the vehicle is performing a sharp lane change or evasive maneuver), the system will enter an absolute focus mode. At this time, the control logic will directly lock the physical channel of the Bluetooth audio communication link, implement a whitelist mechanism, allowing only navigation warning audio data packets to pass through, and send a flow control pause command to media entertainment devices, forcing the source device to pause transmission and ensuring that critical navigation or alarm commands can exclusively occupy all air interface bandwidth.

[0059] The central arbitration processing unit is connected to the vehicle status and environment perception module and the multi-source Bluetooth communication management module. Internally, it runs a real-time operating system and deploys a dynamic electromagnetic interference potential energy model and a dynamic priority calculation engine. This unit achieves proactive defense against electromagnetic interference by constructing the dynamic electromagnetic interference potential energy model. Unlike traditional threshold-triggered simple logic, this model introduces the concept of potential energy to describe the accumulation and other processes of electromagnetic interference. It should be noted that the electromagnetic interference potential energy defined in this invention is not the concept of energy in classical physics, but rather refers to a virtual state index characterizing the degree of accumulation and outbreak trend of interference risk over time. The system calculates the trend value of electromagnetic interference intensity within a preset time window based on electrical characteristic data (mainly the rate of change of oscillation frequency and its first derivative). Specifically, the model uses a sliding time window integration method. (The specific algorithm implementation is repeated here.) Electromagnetic interference potential energy index As shown in the following formula:

[0060] ;

[0061] in, The rate of change of the LC oscillation frequency , The amplitude of the fluctuation. The length of the sliding time window (e.g., 50ms). and These are the normalized weighting coefficients. In this embodiment, and The adaptive historical extreme value reciprocal method is used to determine this. The system maintains the maximum absolute value of the rate of change of frequency monitored during historical operating cycles in non-volatile memory. With the maximum amplitude of oscillation .set up ,set up For example, if statistics are collected... , Then dynamically set for , for This process ensures that the two indices with vastly different physical dimensions in the formula are mapped to the dimensionless interval [0,1], guaranteeing the mathematical validity of the potential energy superposition. Specifically, and Internally, it includes a dimension conversion factor and an amplitude scaling factor, used to convert the rate of change of frequencies that are completely different in physical units. (unit: Hz / s) and oscillation amplitude (Unit V) is mapped to a dimensionless value of the same order of magnitude (e.g., normalized to a closed interval of 0 to 1). This process eliminates the difference in physical dimensions and prevents issues arising from significant differences in their absolute values ​​(e.g., for level for This leads to the problem that small numerical terms are ignored in integration operations. This represents the current system sampling time; Let be the integral variable, representing the time window during the sliding time. The integral operation is used to calculate the historical instant within the range; Over a period of time, the rate of change of oscillation frequency With oscillation amplitude The cumulative electromagnetic interference energy trend due to the combined effects. This integral value reflects the degree of recent interference energy accumulation, effectively filtering out occasional spike noise and accurately characterizing the continuous interference trend during motor commutation. When the motor is in the initial startup phase, the frequency change rate R increases exponentially, and the future trend value calculated by the model will rise rapidly. When the electromagnetic interference intensity trend value exceeds the preset safety threshold, the system determines that the physical channel will face severe signal-to-noise ratio degradation. At this time, the central arbitration processing unit no longer waits for the Bluetooth baseband to report an increase in packet loss rate, but immediately generates a preemptive link reconstruction command. Before the electromagnetic interference actually occurs, it controls the Bluetooth baseband controller to forcibly adjust the underlying transmission data packet type of the Bluetooth audio communication link to improve the link's anti-interference and error correction redundancy.

[0062] In response to the preemptive link reconfiguration command, the underlying transmission data packet type adjustment performed by the system is the key processing logic for physical layer anti-interference in this invention. In the Bluetooth baseband specification, the data packet type determines the balance between transmission efficiency and anti-interference capability. Typically, to ensure high audio quality, the system tends to use high-throughput multi-slot packets, such as 3-DH5 or 2-DH5. These packets, after being sent at the beginning of one slot, continuously occupy the next four slots, carrying a large amount of data. Their advantage is a high payload ratio, but their disadvantage is that even a microsecond-level pulse interference during transmission will cause the entire long packet to fail verification, leading to retransmission, significantly wasting bandwidth and causing delays. When an interference warning is triggered, the system forcibly switches the packet type to a high-immunity single-slot packet, such as the DM1 (Data-Medium Rate) packet. The DM1 packet occupies only one slot, and its payload data is forcibly augmented with a 2 / 3 ratio of forward error correction (FEC). This means that every 15 bits of encoded data contains 10 information bits and 5 parity bits, capable of correcting random bit errors during transmission. Although the theoretical throughput of DM1 packets is much lower than that of 3-DH5, its stability is extremely strong in strong interference environments. Combined with the FEC mechanism, the receiver can automatically repair damaged data without requesting retransmission, thus ensuring the continuity of the audio stream and avoiding popping and stuttering. Simultaneously, based on the trend value of electromagnetic interference intensity, the central arbitration processing unit also locks the anti-interference frequency hopping table. Traditional adaptive frequency hopping is based on post-event statistics, meaning that a frequency point is only marked as a bad channel after a large number of packet losses. This system, however, is based on pre-event prediction. Based on the spectral characteristics of motor interference (usually broadband noise, but with the strongest energy at certain harmonic points), the model predicts the affected frequency bands and actively removes these frequency points from the frequency hopping sequence, forcing the Bluetooth RF front-end to switch to a relatively clean frequency band for transmission.

[0063] After resolving the transmission reliability issue, the central arbitration processing unit is pre-installed with a dynamic priority calculation engine. This engine uses a formula to calculate the real-time arbitration weight of each audio stream. The specific expression of this formula is as follows: In this formula, The base score for the audio stream is a static score assigned based on the essential attributes of the audio content. In this embodiment, navigation warning audio, which involves driving route guidance and safety tips, is assigned the highest score; real-time call audio, which involves interpersonal interaction, is assigned a medium score; and media entertainment audio, which is considered non-essential, is assigned the lowest score. The current vehicle speed is a dynamic parameter that incorporates the impact of driving load on audio demand. In the formula... The term employs an inverse proportional decay model. (Introduction) (For example, setting it to 10km / h) is to avoid calculation overflow caused by the denominator being zero when the vehicle is stationary; introducing (For example, setting it to 120km / h) is to eliminate The physical dimensions are transformed into dimensionless coefficients between 0 and 1, thus enabling them to be correlated with... and Perform a valid weighted summation. The faster the vehicle speed, the closer the driver's ability to process environmental information approaches its limit; at this point, the tolerance for entertainment audio decreases, while the demand for warning audio increases. Therefore, as vehicle speed increases, this value decreases, affecting the overall weight of all audio streams. Suppressed, but due to Due to the differences, entertainment-related audio is more likely to fall below the adjudication threshold and be suspended. This is the current vehicle safety index, provided by the vehicle's active safety systems (such as ADAS), and its value is typically normalized to 0 to 1. When the vehicle is driving smoothly and safely, this value is 1; when a collision risk, lane departure, or vehicle instability is detected, this value drops sharply or even becomes negative. This feature ensures that in emergency situations, regardless of other parameters, the arbitration result will prioritize quiet entertainment and prominent warnings. Meanwhile, , and All three must maintain a consistent unit of speed measurement (e.g., km / h) to ensure that the intermediate term calculation result is a dimensionless coefficient, and that the formula contains... and The relationship between them is a division operation. , , These are preset weighting coefficients, which are calibration values ​​derived from a large number of real vehicle road tests and ergonomic experiments. They are used to balance the weights of each factor and ensure that the system can make arbitration decisions that conform to human intuition under various operating conditions.

[0064] As a specific preferred embodiment, the following is set (Basic score weighting) (Speed ​​weight) (Security weight); Setting , .

[0065] Assuming the current content is media entertainment audio ( The vehicle was traveling at 120 km / h and was safe. ),but If this value is lower than a preset threshold (e.g., 30), the system will suspend the operation.

[0066] If it is navigation audio ( Under the same working conditions The value is far above the threshold, ensuring smooth transmission.

[0067] When a certain audio stream When the value falls below the system's set adjudication threshold, the system suspends the audio stream and records the timestamp of the suspension point. When operating conditions improve (e.g., vehicle speed decreases or interference disappears), the system will suspend the audio stream. Once the value recovers, the system will automatically resume playback from the recorded timestamp, achieving seamless continuation of playback.

[0068] To further optimize the user experience, this embodiment also introduces a user interaction feedback module. When output arbitration causes an audio source switch, or when a strong anti-interference strategy is implemented due to detected strong interference, the user interaction feedback module takes over control of the ambient lighting in the vehicle. The system controls the ambient lighting to perform a breathing flash at a specific frequency. The flashing frequency is positively correlated with the current electromagnetic interference intensity trend value; that is, the stronger the interference, the deeper the system intervention, and the more rapid the light breathing. This non-verbal visual feedback effectively conveys the system's operating status information without distracting the driver, enhancing the user's sense of trust and security, while reducing the user's anxiety about brief fluctuations or switching in audio quality, providing visual status feedback.

[0069] When performing preemptive link reconstruction (switching to DM1 packets), the physical bandwidth drops significantly, potentially failing to meet the transmission requirements of high-bitrate audio. To address this issue, the central arbitration processing unit also includes a spectrum bandwidth replacement submodule. When a vehicle is in motion, wind noise and tire noise are primarily concentrated in the low-frequency band, and the sound pressure level increases significantly with vehicle speed. When ambient noise is sufficiently high, the human ear's sensitivity to audio signals in the same frequency band decreases dramatically, sometimes to the point of being inaudible. The spectrum bandwidth replacement submodule first constructs a wind noise spectrum model based on the vehicle speed in the dynamic parameters. This model pre-stores the in-vehicle noise frequency response curves at different vehicle speeds. The system compares the spectrum of the current audio stream with the wind noise model in real time, identifying invalid frequency bands in the audio stream that are physically masked by wind noise. The wind noise spectrum model contains a pre-defined linear cutoff frequency mapping formula: .in The starting speed for wind noise (e.g., 60 km / h). This is the slope coefficient (e.g., 4.0 Hz / (km / h)). When the vehicle speed... hour, ;when hour, The system controls the pre-stage high-pass filter of the audio encoder accordingly to filter out low-frequency components below 200Hz. When switching to high-immunity single-time-slot packets causes a decrease in bandwidth, the system proactively discards audio payload data in invalid frequency bands at the audio coding layer (e.g., by performing dynamic high-pass filtering) and uses the saved air interface bandwidth resources to transmit high-priority audio stream data (such as vocal frequencies or high-frequency details), or to increase the redundancy of error correction codes. This achieves joint optimization of the source and channel, maintaining audio clarity and coherence in subjective listening perception despite a reduction in objective physical bandwidth.

[0070] Furthermore, to prevent inactive devices from unexpectedly disconnecting during resource-constrained periods, the multi-source Bluetooth communication management module incorporates a disconnection protection mechanism. During the execution of a preemptive link reconfiguration command, if a non-playing Bluetooth device times out, the system simulates sending virtual heartbeat maintenance packets to prevent non-playing devices from disconnecting their physical connection due to timeout, ensuring seamless handover during arbitration recovery. These heartbeat packets are typically zero-payload POLL or NULL packets, their sole purpose being to reset the peer device's watchdog timer and maintain the physical link's dormant state. Specifically, the virtual heartbeat maintenance packets are implemented using the Logical Link Control and Adaptation Protocol (L2CAP). The system periodically sends L2CAP_ECHO_REQ commands (Ping packets) to non-playing devices. This command does not carry upper-layer application data but can forcibly trigger link activity in the underlying baseband, thereby resetting the receiving device's link monitoring timer and effectively preventing connection timeouts due to prolonged inactivity.

[0071] To address potential co-channel interference from Bluetooth tire pressure monitoring systems (TPMS) in modern vehicles, the system also includes a Bluetooth TPMS coordination interface. Considering that TPMS sensors typically operate in the 2.4GHz band and periodically transmit data packets, when the system detects a data packet transmission from the vehicle's Bluetooth TPMS sensor, the central arbitration processing unit inserts a silence slot into the Bluetooth audio communication link. The length of the silence slot is strictly synchronized with the data transmission period and duration of the TPMS sensor. This silence slot insertion is achieved by sending the HCI_Hold_Mode command through the Host Controller Interface (HCI). When a TPMS synchronization signal is detected, the central arbitration processing unit forces the audio ACL link into hold mode and sets the hold time to cover the TPMS data transmission window (e.g., 2.5ms). During this period, the Bluetooth RF transceiver is temporarily turned off, thus physically eliminating the risk of co-channel interference.

[0072] Example 2:

[0073] See the instruction manual appendix Figure 2 Based on the hardware system architecture detailed in Embodiment 1, this embodiment of the invention further provides a multi-source Bluetooth audio stream arbitration system and anti-interference method based on vehicle state perception. This method relies on the hardware architecture described in Embodiment 1, and by executing firmware instructions stored in non-volatile memory, it completes the perception of the vehicle's physical state and the reallocation of wireless communication resources within a closed-loop control cycle through periodic real-time task scheduling. Specifically, it includes the following steps:

[0074] Step S1: Real-time acquisition of vehicle CAN bus dynamic parameters, and monitoring of the electrical characteristics of the actuators using the LC oscillation detection unit.

[0075] The core of this step lies in establishing a digital twin state space that is strictly synchronized with the real physical world. The central arbitration processing unit first initializes the onboard direct memory access (DMA) controller and establishes two independent high-speed data channels: one for capturing CAN messages for vehicle dynamic parameters and the other for sampling electrical characteristic data using an ADC.

[0076] For acquiring dynamic parameters, the system employs an event-triggered message filtering mechanism. The system's CAN controller is configured to receive only specific ID segments, such as chassis-critical messages, through a hardware filter. Upon interrupt triggering, the DMA controller immediately moves the message payload to a circular buffer. Subsequently, a parsing algorithm is used. For vehicle speed signals, the system not only reads the average speed displayed on the instrument panel but also directly reads the raw pulse count differences from the four wheel speed sensors, calculating the vehicle's longitudinal micro-slip rate using a differential algorithm. This minute slip rate parameter is crucial for determining whether the vehicle is in a state of extreme grip and is also essential for subsequent calculations of the safety index. This is a crucial input. For acceleration signals, the system performs coordinate system transformation on the raw triaxial acceleration data acquired from the IMU, removes the gravitational acceleration component, and performs low-pass filtering to remove high-frequency noise caused by road bumps, thereby extracting the acceleration and deceleration intentions purely caused by the driver's operation.

[0077] For acquiring electrical characteristics, the system initiates a high-frequency ADC to continuously sample the output of the LC oscillation detection unit. The sampling frequency is set to 200kHz, sufficient to capture microsecond-level arc discharge characteristics. The DSP (Digital Signal Processing) core performs real-time wavelet transform on the sampled stream. Compared to Fourier transform, wavelet transform has better localization characteristics in the time-frequency domain, enabling precise capture of the non-stationary abrupt signal generated at the instant of motor commutation spark. The system focuses on monitoring the rate of change of the oscillation frequency, df / dt. Before the motor has fully established a stable electromagnetic field, i.e., during the dead time when the current is just beginning to rise, the LC circuit is affected by weak mutual inductance coupling, resulting in a characteristic frequency drift. The system defines this slight drift characteristic as the zero-time feature, used to mark the starting point T0 of the interference event. To eliminate potential random thermal noise interference during sampling, the system also introduces a moving average filtering algorithm to smooth the five continuously acquired sampling points, ensuring that only genuine oscillation signals with a certain energy persistence are identified as valid electrical characteristic data. It is worth noting that, in order to ensure the consistency of dynamic parameters and electrical characteristic data on the time axis, the system adopts a global timestamp synchronization technology based on hardware timers. Every time a CAN message frame is acquired or an LC interrupt is triggered, the system adds an RTC timestamp with microsecond precision to ensure that subsequent multi-source data fusion is performed under the same time slice, avoiding causal reversal caused by time misalignment.

[0078] Step S2: Using the dynamic electromagnetic interference potential energy model, predict the interference intensity within a future millisecond-level window based on the frequency change rate of electrical characteristics.

[0079] After acquiring the basic physical data, the algorithm enters the core prediction stage. The dynamic electromagnetic interference potential energy model proposed in this invention is a set of nonlinear differential equations constructed based on the physical laws of electromagnetic field energy accumulation.

[0080] The model first defines a virtual state variable, namely the electromagnetic interference potential energy PEMI(t) at time t. The growth rate of this potential energy is proportional to the detected electrical characteristic amplitude A(t) and the rate of frequency change R(t), and is modulated by the current Bluetooth channel background noise Nfloor. Specifically, to ensure the physical validity of the calculation, the system first performs preprocessing on the acquired raw electrical characteristic data. Using a max-min normalization algorithm, based on the extreme boundary of historical statistics, the real-time rate of frequency change is... With amplitude This is converted into a dimensionless scalar. Subsequently, in the model, when the vehicle motor (such as the window motor) starts, the energy accumulation process in its armature winding follows the inductor charging formula. However, the intensity of the externally radiated interference does not directly correspond to the current I, but rather to the rate of change of the current di / dt. Therefore, the model uses the detected LC oscillation characteristics to invert the di / dt of the motor circuit, and combines it with the pre-stored parasitic parameter matrix of the wiring harness distribution for this vehicle model to calculate the estimated interference voltage coupled to the Bluetooth antenna port.

[0081] This model introduces an adaptive time window prediction algorithm. The system sets a prediction window ΔT (e.g., 50ms) extending forward from the current time t. Within this window, the model calculates the total potential energy of future disturbances by integrating based on the current potential energy growth slope. If the detected frequency change rate shows an exponential upward trend, it indicates that the motor is in a stage of rapid change in commutation current, at which point the generated electrical spark energy is the largest and the frequency band is the widest.

[0082] To quantify this prediction, the system calculates a normalized dimensionless electromagnetic interference intensity trend value, Itrend. This value reflects not only the amplitude of the interference but also its spectral width. If the model predicts that the motor is about to enter a region of frequent commutation (such as repeated reversals triggered by the anti-pinch logic of a car window), Itrend is given a very high weight. When Itrend exceeds a preset safety threshold, Ith, the system determines that the channel is about to become unavailable, thus triggering subsequent preemptive operations. It is worth noting that this threshold, Ith, is not fixed but dynamically adjusted based on the current Bluetooth Received Signal Strength Indication (RSSI). If the current Bluetooth device is very close and the signal is extremely strong, the system will appropriately increase Ith to reduce unnecessary interference; conversely, if the signal is weak, the system will decrease Ith, becoming more sensitive and aggressive to ensure connection stability under weak signal conditions.

[0083] Step S3: If the predicted interference intensity does not exceed the limit, calculate the priority of each audio stream using the weighted formula and perform regular arbitration.

[0084] While predicting the channel environment, the system's parallel processing also assesses the value of meeting the needs of upper-layer applications.

[0085] The system retrieves the weighted calculation formula stored in memory. Real-time calculations are performed. To make the formula more practical, this embodiment quantifies and defines each parameter. For the basic type score... The system uses 8-bit integers for quantization. Navigation warnings (such as ADAS alarm sounds and navigation voice) are set to 255 (0xFF) and assigned the highest score; real-time communication (HFP protocol voice) is set to 128 (0x80) and assigned a medium score; media entertainment (A2DP music and audiobooks) is set to 64 (0x40) and assigned the lowest score. For vehicle speed factors... To prevent the denominator from being zero, a smoothing constant is used for correction in the actual calculation. When the vehicle speed reaches the high-speed range, this term is adjusted using a weighting coefficient. The adjustment significantly reduces the overall weight, simulating the psychological contraction effect of drivers under high pressure. That is, at high speeds, the human brain's ability to process unnecessary information decreases, thus the system actively suppresses the weight of entertainment information. Regarding the safety index... This index is derived from multiple safety flags on the vehicle's CAN bus. It is high during normal driving, but drops sharply if ESP intervenes or AEB (Automatic Emergency Braking) warning is triggered.

[0086] The system refreshes each audio stream every 10ms. Value. For example, in the current scenario: the driver is cruising at 100 km / h, listening to loud rock music, when suddenly the car in front brakes sharply, triggering AEB. The calculation result will affect the weight of the rock music. The system's survival threshold was instantly breached. At this point, the scheduler not only sends a pause command but also performs a suspension operation. That is, the system saves a snapshot of the current decoder state in memory, including the current playback timestamp and the pointer position of the decoding buffer, ensuring millisecond-level accuracy during future resumption, rather than rebuffering. This fine-grained state management ensures that the psychological discontinuity experienced by the user is minimized during the preemption and resumption of the audio stream.

[0087] Step S4: If the interference intensity is predicted to exceed the standard, immediately trigger preemptive link reconstruction and switch Bluetooth transmission to high error correction, single time slot mode.

[0088] Once the interference warning in step S2 is triggered, and step S3 identifies the high-priority links that need protection, this step will immediately execute preemptive link reconfiguration.

[0089] First, the central arbitration processing unit sends a vendor-defined mandatory packet type instruction via the HCI interface. This instruction has the highest execution priority. The instruction parameters explicitly specify that only high-immunity single-slot packets (such as DM1 packets) are allowed, and force FEC encoding to be enabled. Upon receiving this instruction, the Bluetooth baseband immediately stops its current packet scheduling strategy at the next slot boundary.

[0090] Specifically, if a 3-DH5 long packet (occupying 5 time slots) is currently being sent, the normal logic is to wait for the packet to finish sending. However, under the preemptive logic of this system, if the interference trend value is extremely high, the baseband may even forcibly truncate the current long packet transmission. Although this will result in the loss of the current packet, it avoids the inevitable errors of the long packet under strong interference and the subsequent long retransmission, thus making time available for subsequent highly interference-resistant short packets.

[0091] Simultaneously, the system performs dynamic pruning of the frequency hopping sequence. Standard Bluetooth adaptive frequency hopping requires negotiation between the master and slave devices, which is time-consuming. This system adopts a unilateral active avoidance strategy. Since the interference source (such as the window motor) is inside the vehicle, the arbitration unit can accurately sense the center frequency of the interference. The system directly modifies the frequency synthesizer control word of the local baseband. When generating the next frequency hopping point, if the calculated frequency point falls within the motor interference band, the hardware logic will automatically skip that frequency point and select the next available frequency point in the sequence. This process does not require negotiation with the peer mobile phone. Although it may cause the peer mobile phone to lose signal in that time slot (manifested as a normal packet loss), it avoids invalid transmission on dirty channels and protects the RF front-end from being saturated by high-power noise floor. In addition, to prevent inactive devices from accidentally disconnecting when resources are scarce, the system sends virtual heartbeat maintenance packets to these devices. These packets do not carry a payload and are only used to reset the link monitoring timer.

[0092] Step S5: Simultaneously calculate the wind noise masking effect, trim invalid frequency band data, and replace the bandwidth to ensure error correction transmission of critical audio data.

[0093] After step S4 is executed, the bandwidth of the physical channel is compressed to its limit (the theoretical rate of the DM1 packet is only about 1 Mbps, and the payload is low). In order to transmit high-quality audio, the audio data must be processed. This step performs spectrum bandwidth permutation.

[0094] The system activates the wind noise masking model within the DSP. This model receives the vehicle speed signal in real time and calculates the current background noise power spectral density inside the vehicle. For example, at 120 km / h, the low-frequency (<300 Hz) background noise inside the vehicle can reach over 70 dB. Due to the masking effect of the human ear, components in the audio signal below 300 Hz with amplitudes not significantly higher than 70 dB are completely inaudible and invalid information.

[0095] The adaptive filter bank within the DSP dynamically adjusts the parameters of the audio encoder based on the calculated masking curve. For SBC or AAC encoding, the system forcibly relocates bit resources from the bit allocation pool from low-frequency subbands to mid-to-high-frequency subbands (such as the 2kHz-5kHz vocal-sensitive area). Furthermore, the adaptive filter bank within the DSP preprocesses and filters the PCM audio stream before it enters the Bluetooth encoder based on the calculated masking curve. By significantly attenuating the low-frequency signal energy masked by wind noise, the encoder (such as SBC or AAC) automatically reduces the number of bits allocated to the low-frequency subbands during psychoacoustic model analysis. Thus, while maintaining or reducing the total bit rate, the saved bit resources are automatically redistributed by the encoder to the mid-to-high-frequency subbands (i.e., the vocal-sensitive area), indirectly achieving sound quality optimization within limited bandwidth.

[0096] Users will not subjectively feel the lack of bass while driving at high speeds (because it is already covered by wind noise), but they will clearly feel that the mid-to-high frequency navigation voice or call voice becomes clearer and sharper, effectively avoiding distortion or interruption.

[0097] Example 3:

[0098] To more intuitively illustrate the robustness of this invention in complex real-world vehicle environments, the following provides the system behavior under several typical extreme scenarios:

[0099] Scenario 1: Electromagnetic surge and call protection when the anti-pinch function of the car window is triggered.

[0100] In this scenario, a rear passenger attempts to close the window, but the window encounters an object, triggering the anti-pinch function. The window motor undergoes a rapid electrical process involving stall, current surge, and reverse rotation within a very short time, resulting in an extremely high rate of current change (di / dt) and generating extremely strong electric arc energy.

[0101] In the initial few milliseconds of motor stall, the oscillation frequency of the LC detection unit experiences a severe Doppler shift. The Itrend calculated by the model increases dramatically. At this time, the system is in a Bluetooth teleconference (real-time call, medium priority).

[0102] According to traditional logic, strong interference would cause continuous packet loss on the SCO link, resulting in a harsh "buzzing" sound or interrupted words during calls. However, this system immediately triggers preemptive reconstruction the moment it detects stalling characteristics (even before the window controller performs a reversing action). The system not only switches to DM1 packets but also further activates the packet loss concealment algorithm. Since it predicts that the channel may be completely unavailable within the next 50ms, the system performs waveform stretching and predictive padding on the voice signal at the DSP level.

[0103] Simultaneously, the user interaction feedback module is activated. The ambient lighting on the door armrests and dashboard is taken over by the system. Based on the electromagnetic interference intensity trend value, the ambient lighting instantly switches from a soft cool tone to a warm orange, and flashes at a high frequency (e.g., 5Hz). This flashing frequency is positively correlated with the oscillation frequency of the motor current. When the interference actually occurs, the physical link is indeed interrupted for 40ms. However, during these 40ms, the user hears a fitted speech synthesized by the DSP based on the preceding speech features. Although the timbre is slightly mechanical, the semantics are completely coherent without any popping sounds. Visually, the rapidly flashing ambient lighting conveys to the user that the system is trying its best to resist the interference. After the motor reverses and stabilizes, the interference disappears, the link is restored instantly, and the light returns to a constant cool tone. This process completes a closed loop from underlying communication assurance to upper-level user perception, greatly reducing the user's anxiety about intermittent sound quality fluctuations.

[0104] Scenario 2: Wiper interference and navigation priority during heavy rain.

[0105] In this scenario, the vehicle is traveling at 110 km / h in heavy rain, with the windshield wiper motor running at its highest setting (high speed), generating periodic broadband electromagnetic interference. Simultaneously, the driver is playing high-bitrate symphonic music via Bluetooth and has activated navigation on their mobile phone.

[0106] Each time the wiper motor reverses direction, the LC oscillation detection unit in the vehicle status and environment perception module detects the frequency change 20ms in advance. The central arbitration processing unit identifies this as a periodic interference source and establishes a pulse interference suppression schedule synchronized with the wiper movement.

[0107] Within each tiny time window (approximately 30ms) of wiper interference, the system forcibly switches the Bluetooth packet to DM1 and pauses the low-frequency data transmission of the symphony (bandwidth substitution). This is because the symphony is then identified as... Lowest, and Due to slippery roads caused by rain (ESP frequently adjusts), the system automatically reduces the music volume and narrows the sound field.

[0108] Suddenly, an accident occurred ahead, and the driver sharply swerved to avoid it. At this moment, the lateral acceleration sensor reading surged, and the system determined that the vehicle had entered a high-load driving state. Instantly, the central arbitration processing unit issued the highest priority instruction: suspend all media entertainment streams. The symphony immediately stopped (recording the breakpoint). The physical channel was completely locked to the navigation warning audio. Just then, the navigation software broadcast, "Accident ahead, please slow down." This voice data packet, using the highest anti-interference mode of DM1+FEC, reliably transmitted to the vehicle's infotainment system and played back despite the electromagnetic interference from the wiper motor and the vehicle's violent dynamics. The driver clearly heard the instruction and completed the avoidance maneuver. After the vehicle's posture stabilized (approximately 5 seconds later), the system detected... The music resumes automatically from the point of interruption, fading back in without any manual intervention from the driver.

[0109] Scenario 3: Quiet rest mode in the parking lot in conjunction with TPMS.

[0110] In this scenario, the vehicle is parked in a quiet underground garage with the engine off. The driver is resting inside, listening to high-fidelity lossless Bluetooth music. The background noise inside the car is extremely quiet, and the wind noise masking effect is ineffective.

[0111] At this point, all the vehicle's ECUs enter low-power sleep mode, but the Bluetooth tire pressure monitoring system (TPMS) is configured to wake up every 5 minutes to send a status packet. The TPMS sensor uses the same 2.4GHz frequency band as the Bluetooth audio.

[0112] When the system learns through the coordination interface that TPMS is about to send data (e.g., upon receiving a pre-wake-up interrupt from the TPMS receiver), it executes a silent insertion strategy to avoid music stuttering caused by even a tiny packet loss (easily noticeable in a quiet environment). The system calculates that TPMS takes only 2ms to send a data packet. Therefore, the system precisely inserts 2ms of silent data into the Bluetooth audio buffer queue, or fine-tunes the audio sampling rate (slowing down the playback speed by 0.1% through a resampling algorithm, imperceptible to the human ear), thus accumulating 2ms of idle time.

[0113] Within those 2ms, the Bluetooth audio link paused transmission, resulting in an extremely clean air interface. The TPMS sensor successfully transmitted packets, and the receiver successfully received them. Subsequently, the audio link resumed full-speed transmission. Thanks to the smooth buffering strategy, the music heard by the driver was fluid and seamless, without the driver realizing that a frequency band resource contention and release had occurred behind the scenes. This processing method demonstrates the system's refined control capabilities under low-load, high-quality audio demand scenarios.

[0114] Furthermore, in terms of hardware selection, the vehicle status and environment perception module of this invention employs a high-performance microcontroller based on the ARM Cortex-M7 core, with a main frequency of up to 600MHz, to support complex wavelet transform and potential energy model calculations. The ADC module uses a successive approximation (SAR) ADC independent of the MCU core, featuring differential input functionality to cancel common-mode noise. The LC oscillation detection circuit is encapsulated in an independent shield to prevent itself from becoming an interference source, and is connected to the detection point via a microstrip line on a Teflon substrate to ensure high-frequency characteristics.

[0115] In terms of software architecture, the system is developed based on a real-time operating system (RTOS) compliant with the OSEK / VDX standard. The logic code of the central arbitration processing unit is encapsulated as a highest-priority urgent task, with its scheduling cycle strictly locked within 5ms. The Bluetooth protocol stack has been deeply customized, stripping away unnecessary application layer profiles and retaining the core L2CAP, HCI, and LinkManagerProtocol (LMP) layers. To improve data throughput efficiency, the system implements a zero-copy mechanism in the underlying UART / USB interface driver, meaning that data is transferred directly from the receive buffer to the transmit buffer via a DMA linked list without CPU handling, greatly reducing CPU load and ensuring system stability under multi-task concurrency.

[0116] Furthermore, the system's code implementation follows the MISRA-C 2012 programming standard, does not use dynamic memory allocation (malloc / free), and all data structures are managed using static memory pools, effectively avoiding the risk of system crashes caused by memory fragmentation and meeting the requirements of automotive functional safety ISO 26262 ASIL-B level.

[0117] Example 4:

[0118] This invention also provides a computer-readable storage medium storing a computer program. The storage medium can be a non-volatile memory, such as flash memory, hard disk drive (HDD), or solid-state drive (SSD), or a read-only memory (ROM). When executed by one or more processors, the computer program can implement all the steps of the multi-source Bluetooth audio stream arbitration system and anti-interference method based on vehicle state awareness described in Embodiment 2 above. Specifically, the computer program is organized into a collection of multiple functional modules, including but not limited to: a low-level hardware driver module (for operating the CAN controller, ADC module, and GPIO), a middleware layer signal processing module (for implementing Kalman filtering, FFT transformation, and interference potential energy model calculation), a protocol stack layer Bluetooth management module (for processing HCI commands, L2CAP packet splitting, and SCO link management), and an application layer business logic module (for performing priority arbitration, user interaction feedback, and audio stream routing). This program code is highly optimized, written in C / C++, and follows automotive software development standards to ensure deterministic and secure code execution. In addition, the storage medium pre-stores the system's default configuration file, which includes basic score tables for various audio types, threshold tables for vehicle dynamics parameters, and a spectral template for wind noise masking effects. During system operation, these parameters can be dynamically updated in specific areas of the storage medium based on machine learning algorithms. For example, as user habits accumulate, the system will automatically adjust... , , The weighting coefficients make the arbitration results more in line with the psychological expectations of specific drivers.

[0119] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A multi-source Bluetooth audio stream arbitration system based on vehicle state awareness, characterized in that, The system includes: The vehicle status and environment perception module is used to collect the vehicle's dynamic parameters in real time through the vehicle's CAN bus interface, and simultaneously monitor the electrical characteristic data of the drive circuit of the vehicle body's actuators. The multi-source Bluetooth communication management module is equipped with a Bluetooth baseband controller that supports multi-link concurrency. It is used to establish and maintain Bluetooth audio communication links with multiple external terminal devices and parse the protocol attributes of the audio stream. The central arbitration processing unit is connected to the vehicle status and environment perception module and the multi-source Bluetooth communication management module respectively. The central arbitration processing unit is configured to execute the following control logic: A dynamic electromagnetic interference potential energy model is constructed, and the trend value of electromagnetic interference intensity within a future preset time window is calculated based on the electrical characteristic data. When the electromagnetic interference intensity trend value exceeds the preset safety threshold, a preemptive link reconfiguration command is generated. Before the electromagnetic interference actually occurs, the Bluetooth baseband controller is controlled to forcibly adjust the underlying transmission data packet type of the Bluetooth audio communication link in order to improve the anti-interference and error correction redundancy of the link. Simultaneously, the multi-source audio streams are dynamically sorted and arbitrated based on weights by combining the dynamic parameters and the electromagnetic interference intensity trend value.

2. The multi-source Bluetooth audio stream arbitration system based on vehicle state awareness according to claim 1, characterized in that, The central arbitration processing unit is pre-installed with a dynamic priority calculation engine, which calculates the real-time arbitration weight of each audio stream using the following formula. : ； in, The basic type score for audio streams, The current vehicle speed is one of the dynamic parameters. The preset reference velocity constant, The preset velocity offset constant, For the current vehicle safety index, , , These are preset weighting coefficients; when a certain audio stream's... When the value is lower than the adjudication threshold, the system suspends the audio stream and records the breakpoint timestamp at the time of suspension.

3. The multi-source Bluetooth audio stream arbitration system based on vehicle state awareness according to claim 1, characterized in that, The multi-source Bluetooth communication management module is equipped with an audio type fingerprint database to identify whether the accessed audio stream belongs to navigation warning, real-time call, or media entertainment. When the central arbitration processing unit determines that the current driving state is under high load, it locks the physical channel of the Bluetooth audio communication link, allowing only navigation warning audio to pass through, and sends a flow control pause command to media entertainment devices.

4. The multi-source Bluetooth audio stream arbitration system based on vehicle state awareness according to claim 1, characterized in that, The system also includes a user interaction feedback module; When the output arbitration results in an audio source switch, the user interaction feedback module controls the vehicle ambient light to perform a breathing flash at a specific frequency. The frequency of the flash is positively correlated with the current electromagnetic interference intensity trend value to provide visual status feedback.

5. A multi-source Bluetooth audio stream arbitration system based on vehicle state awareness according to claim 1, characterized in that, The multi-source Bluetooth communication management module has a disconnection protection mechanism; During the execution of the preemptive link reconfiguration instruction, if the connection of a non-playing Bluetooth device times out, the system simulates sending a virtual heartbeat maintenance packet to prevent the non-playing device from disconnecting its physical connection due to timeout, ensuring seamless switching during arbitration recovery.

6. The multi-source Bluetooth audio stream arbitration system based on vehicle state awareness according to claim 1, characterized in that, The vehicle body actuators include a window lift motor; the vehicle status and environment perception module includes an LC vibration detection unit connected to the window lift motor. The electrical characteristic data includes the rate of change of the oscillation frequency output by the LC oscillation detection unit; The central arbitration processing unit identifies instantaneous abrupt changes in the rate of change of the oscillation frequency as precursor signals of electromagnetic interference. The greater the magnitude of the rate of change, the higher the predicted trend value of the electromagnetic interference intensity.

7. A multi-source Bluetooth audio stream arbitration system based on vehicle state awareness according to claim 1, characterized in that, The underlying transmission data packet types include high-throughput multi-slot packets and high-interference-resistant single-slot packets; The specific execution logic of the preemptive link reconfiguration instruction is as follows: The Bluetooth baseband controller is controlled to pause the transmission of the high-throughput multi-slot packets and forcibly switch to the high-interference-resistant single-slot packets with 2 / 3 ratio forward error correction coding; At the same time, the anti-interference frequency hopping table is locked based on the electromagnetic interference intensity trend value, and frequency points that are predicted to be affected by motor commutation spark interference are actively eliminated.

8. A multi-source Bluetooth audio stream arbitration system based on vehicle state awareness according to claim 7, characterized in that, The central arbitration processing unit also includes a spectrum bandwidth replacement submodule; The spectrum bandwidth replacement submodule is used to: construct a wind noise spectrum model based on the vehicle speed in the dynamic parameters, and identify invalid frequency bands in the audio stream that are physically masked by wind noise; When switching to the high-invalidity single-slot packet causes a decrease in bandwidth, a high-pass filter is used in the preprocessing stage of audio encoding to suppress the signal amplitude of the invalid frequency band, thereby reducing the entropy of the encoded data. Alternatively, the bit allocation pool of the Bluetooth encoder can be dynamically adjusted to reduce the number of quantization bits in invalid frequency bands, and the saved air interface bandwidth resources can be used to transmit high-priority audio stream data to achieve joint optimization of the source channel.

9. A multi-source Bluetooth audio stream arbitration system based on vehicle state awareness according to claim 1, characterized in that, The system is also equipped with a Bluetooth tire pressure monitoring interface; When the system detects that the vehicle-mounted Bluetooth tire pressure monitoring sensor is sending a data packet, the central arbitration processing unit inserts a silent time slot in the Bluetooth audio communication link. The duration of the silent time slot is synchronized with the data transmission cycle of the tire pressure monitoring sensor to eliminate co-channel interference.

10. A multi-source Bluetooth audio stream arbitration system and anti-interference method based on vehicle state awareness, characterized in that, Based on the system according to any one of claims 1-9, the method includes the following steps: Step S1: Real-time acquisition of vehicle CAN bus dynamic parameters, and monitoring of the electrical characteristics of the actuators using the LC oscillation detection unit; Step S2: Using the dynamic electromagnetic interference potential energy model, predict the interference intensity within the next millisecond window based on the frequency change rate of electrical characteristics; Step S3: If the predicted interference intensity does not exceed the limit, calculate the priority of each audio stream using the weighted formula and perform regular arbitration; Step S4: If the interference intensity is predicted to exceed the limit, immediately trigger preemptive link reconstruction and switch Bluetooth transmission to high error correction, single time slot mode; Step S5: Simultaneously calculate the wind noise masking effect, trim invalid frequency band data, and replace the bandwidth to ensure error correction transmission of critical audio data.

Images