A star flash wireless audio transmission system and a networking control method thereof

Abstract

The application discloses a star flash wireless audio transmission system and a networking control method thereof, and belongs to the technical field of audio transmission, and comprises a double-clock source integrated module, a random check module between devices, a clock compensation module and a synchronization state monitoring module.The double-clock source integrated module is used for generating a synchronization master clock and an auxiliary clock and integrating a bidirectional synchronization protocol, the bidirectional clock protocol adopts a bidirectional timestamp chain technology.The random check module between devices is used for performing random check between different slave devices based on a lightweight check protocol and generating a check result.The clock compensation module is used for compensating synchronization failure caused by long-term error and short-term error of a local device in a star flash wireless audio transmission process.The synchronization state monitoring module is used for monitoring the compensation process and a clock signal after compensation is completed.In the technical scheme implementation process of the application, through effective cooperation of a dynamic check and compensation mechanism, accurate synchronization of clocks of various devices is ensured, so that audio transmission quality is significantly improved, and the demand for sound field consistency in a long-time and multi-device scene is met.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of audio transmission, in particular to a star flash wireless audio transmission system and a networking control method thereof. BACKGROUND

[0002] Audio is an important information transmission medium and is widely used in daily life and work. The transmission of audio usually relies on wired connection or wireless connection technology to achieve the transmission of audio signals.

[0003] At present, with the development of portable audio transmission, wireless connection technology has gradually become the mainstream, and wired transmission is only used in some specific occasions. Wireless connection technology needs to rely on wireless communication, such as Bluetooth A2DP, LDAC, Wi-Fi direct connection, star flash, etc., and is widely used in pure audio transmission scenarios, such as professional recording studio monitoring, stadium sound amplification, stage return, car audio system, etc., and has the advantages of wireless portability, strong protocol compatibility, etc.

[0004] However, in the prior art, the above technical solutions still have some pain points. For example, the existing scheme involves audio transmission mainly through periodic broadcast of synchronization packets by the master device, such as the AVDTP synchronization instruction of Bluetooth. This scheme needs to set up multiple levels of flow process synchronization clocks for receiving, calculating, and adjusting the device. Due to the delay of wireless communication and the delay of the master device processing, a gradually accumulated time delay error is generated, which cannot meet the requirements of long-time, multi-device sound field consistency. Moreover, for some pure audio transmission vertical scenes, there will also be accumulated errors between multiple devices after long-time running, resulting in audio synchronization error, affecting the sound quality and user experience.

[0005] Therefore, it is necessary to provide a star flash wireless audio transmission system and a networking control method thereof to solve the above problems.

[0006] It should be noted that the above information disclosed in the background section of the present application is only used to understand the background technology of the concept of the present application, and therefore, it can contain information that does not constitute prior art. SUMMARY

[0007] Based on the above problems existing in the prior art, the present application solves the problem of providing a star flash wireless audio transmission system and a networking control method thereof. Through a dynamic verification and compensation mechanism, the clock synchronization state between devices is monitored and adjusted in real time, the accumulated time delay error is effectively reduced, and the stability and consistency of multi-device audio transmission are ensured.

[0008] The technical scheme adopted by the present application to solve its technical problems is: a star flash wireless audio transmission system, the system comprising:

[0009] The double clock source integrated module is used for generating a synchronous master clock and an auxiliary clock, and integrating a bidirectional clock protocol, the bidirectional clock protocol adopts a bidirectional timestamp chain technology, and realizes offset calibration between nodes through a bidirectional synchronization protocol, the nodes include at least one master device and a plurality of slave devices;

[0010] The inter-device random check module is used for random checking between different slave devices based on a lightweight check protocol, and generating a check result, and fine-tuning the synchronous master clock or the auxiliary clock according to the generated check result, wherein the lightweight check protocol is a short message based on UDP;

[0011] The clock compensation module is used for compensating for synchronization failure caused by long-term error and short-term error of the local device in the star flash wireless audio transmission process, wherein the long-term error is caused by the inherent crystal oscillator aging or environmental temperature change of the local device, and the short-term error is caused by short-term noise, and an adaptive multi-source fusion compensation mechanism is established for clock compensation;

[0012] The synchronization state monitoring module is used for monitoring the compensation process and the clock signal after compensation, and is linked with the inter-device random check process.

[0013] In the implementation process of the technical scheme of the present application, through the effective cooperation of the dynamic check and compensation mechanism, the precise synchronization of the clocks of each device is ensured, thereby significantly improving the audio transmission quality and meeting the sound field consistency requirements in long-time and multi-device scenarios.

[0014] Further, the bidirectional synchronization protocol realizes the calibration of the clocks between nodes, including the following steps:

[0015] Firstly, the master device broadcasts a synchronization start packet to all slave devices, the synchronization start packet containing a global start time of the master clock and a master clock frequency;

[0016] Then, after the slave device receives the synchronization start packet broadcast from the master device, the local receiving time is recorded based on the auxiliary clock, and the initial offset is calculated according to the global start time, the initial offset being equal to the sum of the local receiving time minus the global start time and the transmission delay;

[0017] After the slave device calculates the initial offset, the initial offset is fed back to the master device, and the master device adjusts the synchronization start packet according to the feedback initial offset and rebroadcasts the adjusted synchronization information.

[0018] Further, the random check among different slave devices and generating a check result specifically includes the following steps: setting a fixed check period and randomly selecting a group of slave devices participating in the check, in each check period, each group of slave devices exchanges the current synchronization offset; calculating the absolute difference of the synchronization offset of each group of devices, and fine-tuning the master clock or the auxiliary clock according to the absolute difference; after fine-tuning, the synchronization state is confirmed, the adjusted synchronization offset is recorded, and the check log is updated.

[0019] Further, according to the ratio of the absolute difference to the preset threshold, it is judged whether to adjust the master clock or the auxiliary clock, when the ratio of the absolute difference to the preset threshold is less than one, the auxiliary clock is adjusted preferentially, otherwise, the master clock is directly adjusted, triggering a full-link re-adjustment.

[0020] Further, the adaptive multi-source fusion compensation mechanism includes the following steps:

[0021] A clock error model is established by multi-source data, and model parameters are dynamically adjusted in combination with historical data and real-time monitoring data; the clock error model is trained offline and online respectively, the offline training is performed by using historical data, and the online training is performed by using current measured data; the clock error model is optimized based on offline training and adjusted through online training; a predicted clock error value within a fixed interval is output by the clock error model, the clock error value includes a long-term error component and a short-term error component; a double-closed-loop compensation strategy is established, and the clock is coarsely adjusted and finely adjusted according to the clock error value, wherein the coarse adjustment directly adjusts the control voltage of the voltage-controlled oscillator of the phase-locked loop in the local crystal oscillator, and the fine adjustment generates a compensation signal through a low-noise digital frequency synthesizer.

[0022] Further, the clock error model adopts a lightweight LSTM neural network structure.

[0023] Further, when the predicted clock error value is positive, the control voltage is reduced and the phase-locked loop output frequency is reduced; when the phase jitter value is detected to be positive, a negative leading compensation signal is generated.

[0024] Further, the synchronization state monitoring generates a check data packet at a fixed interval, and sends the check data packet to the adjacent device, evaluates the synchronization effect by comparing the received check data packet with the local clock, if the deviation exceeds the preset threshold, a new round of compensation adjustment is triggered, and the clock synchronization precision is ensured to be maintained within the set range.

[0025] A networking control method of a star flash wireless audio transmission system, comprising:

[0026] Generate a synchronization master clock and an auxiliary clock, and integrate a bidirectional clock protocol, which adopts a bidirectional timestamp chain technology and realizes offset calibration between nodes through a bidirectional synchronization protocol, the nodes including at least one master device and multiple slave devices;

[0027] Randomly check between different slave devices based on a lightweight check protocol, and generate a check result, and fine-tune the synchronization master clock or the auxiliary clock according to the generated check result, wherein the lightweight check protocol is a short message based on UDP;

[0028] Compensate for the synchronization failure caused by long-term error and short-term error of the local device in the star flash wireless audio transmission process, wherein the long-term error is caused by the aging of the inherent crystal oscillator of the local device or the change of the environmental temperature, and the short-term error is caused by short-term noise, and an adaptive multi-source fusion compensation mechanism is established for clock compensation;

[0029] Monitor the compensation process and the clock signal after compensation, and link with the random check process between devices, generate check data packets at fixed intervals, and send the check data packets to adjacent devices, evaluate the synchronization effect by comparing the received check data packets with the local clock, and if the deviation exceeds the preset threshold, trigger a new round of compensation adjustment, and ensure that the clock synchronization accuracy is always maintained within the set range.

[0030] The beneficial effects of the present application are: the star flash wireless audio transmission system and the networking control method thereof provided by the present application ensure the accurate synchronization of the clocks of each device through the effective cooperation of the dynamic check and compensation mechanism, thereby significantly improving the audio transmission quality, meeting the sound field consistency demand in long time and multi-device scenarios, reducing the audio delay and distortion caused by clock deviation, ensuring the audio synchronous playback between multiple devices, and improving the user experience.

[0031] In addition to the purposes, features and advantages described above, the present application has other purposes, features and advantages. The present application will be further described below with reference to the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0032] The drawings accompanying the specification of the present application form a part of the present application, the illustrative embodiments of the present application and their descriptions serve to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0033] Figure 1 The drawings accompanying the specification of the present application form a part of the present application, the illustrative embodiments of the present application and their descriptions serve to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0034] Figure 2 The drawings accompanying the specification of the present application form a part of the present application, the illustrative embodiments of the present application and their descriptions serve to explain the present application, and do not constitute an improper limitation on the present application. In the drawings: DETAILED DESCRIPTION

[0035] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0036] In order for those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the scope of protection of the present application.

[0037] Embodiment one: as shown, the present application provides a star flash wireless audio transmission system, which is applied to a star flash-based wireless audio transmission scene, realizes stable transmission of audio data between master devices and slave devices through decentralized and cumulative error-free clock synchronization, reduces delay while reducing audio loss and interference, and the system comprises: Figure 1 A dual-clock source integration module is used to generate a synchronization master clock and an auxiliary clock, and integrate a bidirectional clock protocol. The bidirectional clock protocol adopts a bidirectional timestamp chain technology and realizes offset calibration between nodes through a bidirectional synchronization protocol. The nodes include at least one master device and a plurality of slave devices.

[0038] Unlike traditional centralized broadcasting relying on master devices, the present system adopts a distributed node coordination mechanism, each node synchronizes the clock independently, and each node adopts a dual-clock source to generate a synchronization master clock and an auxiliary clock through a dual-clock source integration module. The synchronization master clock is realized based on a temperature-compensated crystal oscillator or a voltage-controlled crystal oscillator. The temperature-compensated crystal oscillator is a high-precision clock source that can effectively offset the frequency deviation caused by temperature changes and is used for calibration of local time reference to ensure the clock synchronization accuracy between nodes. The auxiliary clock is realized based on a MEMS oscillator (programmable silicon oscillator) and is used for temporary timing in the synchronization process to provide a flexible time adjustment mechanism and ensure high-precision synchronization in complex environments.

[0039]

[0040] ​Meanwhile, the dual-clock source integrated module is also integrated with a bidirectional synchronization protocol, and the clock calibration between nodes is realized through the bidirectional synchronization protocol. The bidirectional synchronization protocol adopts a bidirectional timestamp chain technology. The bidirectional timestamp chain is a chain timestamp structure, supports forward time sequence tracing and reverse time correlation verification, and also supports bidirectional data tracing. For example, the integrity and correlation of the timestamp can be verified forward or backward at any node. The bidirectional synchronization protocol not only differs from the traditional timestamp service, but also differs from the unidirectional extension timestamp chain in the blockchain technology. The bidirectional timestamp chain technology enhances the reliability and security of time synchronization. Specifically, the clock calibration between nodes realized by the bidirectional synchronization protocol includes the following steps.

[0041] Firstly, the master device broadcasts a synchronization start packet to all slave devices. The synchronization start packet contains the global start time of the master clock and the master clock frequency.

[0042] In the star flash wireless audio transmission process, there are at least one master device and multiple slave devices. The prior art is to broadcast a clock signal by the master device, and the slave device passively receives. This way is not only easy to be disturbed, but also has low synchronization accuracy. To this end, in the above steps, the broadcast function of the master device is first used to broadcast a synchronization start packet instead of a clock signal. The synchronization start packet contains the global start time of the master clock and the master clock frequency. The unit of the master clock frequency is hertz (Hz). The accurate time reference and frequency information received by the slave device are ensured, so that more accurate clock synchronization is realized.

[0043] Then, after receiving the synchronization start packet broadcast from the master device, the slave device records the local receiving time based on the auxiliary clock, and calculates the initial offset according to the global start time.

[0044] During the broadcast process of the master device and the slave device, whether it is broadcasting a clock signal or other information, there will be a certain transmission delay. This transmission delay is reflected in the numerical value as the initial offset of the timestamp. In the traditional method, the initial offset is difficult to quantify, which limits the synchronization accuracy. In the present application, after the slave device records the local receiving time based on the auxiliary clock, the initial offset caused by the transmission delay can be calculated in combination with the global start time of the master clock. Specifically, the calculation formula of the initial offset is: the global start time of the master clock is T0, the local receiving time recorded based on the auxiliary clock is T1, and the initial offset ΔT = T1 - (T0 + d). When the link of the bidirectional timestamp chain is symmetrical, d = (T1 - T0) / 2, that is, the offset is the same as the transmission delay of the bidirectional timestamp chain. Therefore, the initial offset is equal to the local receiving time minus the sum of the global start time and the transmission delay. By calculating the initial offset, the clock can be calibrated through the initial offset, thereby effectively reducing the error caused by the transmission delay.

[0045] After the initial offset is calculated by the slave device, the initial offset is fed back to the master device, and the master device adjusts the synchronization start packet according to the fed-back initial offset and rebroadcasts the adjusted synchronization information;

[0046] After the initial offset in the link is obtained, the slave device feeds back the initial offset to the master device, the master device synthesizes the initial offsets of the slave devices, optimizes the synchronization strategy, fine-tunes the synchronization start packet, and rebroadcasts again, so that the time references of all slave devices are consistent, and the clock is calibrated based on the new synchronization start packet and rebroadcasted, so that the audio transmission process between the master and slave devices has a dynamic synchronization adjustment mechanism rather than a static fixed mode, and the time deviation is corrected in real time during the audio transmission process. In a multi-channel scene and complex sound mixing, this method can also effectively avoid the problem of uncoordinated channels;

[0047] An inter-device random check module is configured to perform random checking between different slave devices based on a lightweight check protocol and generate a check result, and fine-tune the synchronization master clock or the auxiliary clock according to the generated check result, wherein the lightweight check protocol is a short message based on UDP.

[0048] In the process of audio transmission, in addition to the clock synchronization between the master device and the slave device, the cooperative work between other slave devices also needs to be considered. Traditional synchronization usually only involves one-way calibration from the master device to the slave device, and cannot detect the clock deviation between the slave devices. In order to facilitate the cooperation between devices, an inter-device random check module is introduced to perform random checking between different devices through a lightweight check protocol without a complex configuration process, and fine-tune the synchronization master clock or the auxiliary clock according to the check result. The lightweight check protocol is a short message transmission based on UDP. The transmission protocol based on UDP short message has the characteristics of low delay and high efficiency, and can avoid the redundant overhead of traditional check methods, without the need to identify the identity information of the check parties, simplify the check process, and not need to consider the order of the transmission parties. Therefore, it is suitable for random checking between different devices. Specifically, the process of short message transmission based on UDP can refer to the prior art. Since this protocol is only responsible for transmitting check data and does not involve data integrity verification, it will not be described in detail in this embodiment.

[0049] The random checking between different slave devices and the generation of the check result specifically include the following steps:

[0050] A fixed check period is set, and a group of slave devices participating in the check are randomly selected. In each check period, the slave devices in each group exchange the current synchronization offset.

[0051] Before the random check from the device, first need to set a fixed check period, according to this check period, the group of slave devices participating in the check is randomly selected, to ensure that each check covers different devices, in this embodiment, the check period can be set to 100 milliseconds, that is, the check process is performed once every 100 milliseconds, and during the check process, each group of slave devices exchanges the current synchronization offset, the calculation method of the synchronization offset is consistent with the calculation method of the initial offset in the foregoing process, therefore, it is not described here;

[0052] The absolute difference of the synchronization offset of each group of devices is calculated, and the master clock or the auxiliary clock is fine-tuned according to the absolute difference;

[0053] After obtaining the synchronization offset of each group of slave devices, the absolute difference is calculated, under ideal conditions, the absolute difference should tend to zero, that is, the clock deviation between each slave device is approximately equal, and only partial clock drift exists, but no obvious cross-device error exists, when the absolute difference exceeds the preset threshold, the master clock or the auxiliary clock is adjusted to correct the clock deviation, whether the master clock or the auxiliary clock is adjusted depends on the ratio of the absolute difference to the preset threshold, if the ratio is less than one, the auxiliary clock is preferentially adjusted to reduce the impact on the stability of the master clock; if the ratio is greater than one, the master clock is directly adjusted, triggering a full-link re-adjustment, the master device repeats the operation of broadcasting the synchronization start packet in the foregoing process until all slave devices complete synchronization and the absolute difference returns to the preset threshold, thereby considering that the synchronization is completed;

[0054] After fine-tuning, the synchronization state is confirmed, the adjusted synchronization offset is recorded, and the check log is updated;

[0055] In each synchronization process, the synchronization state needs to be confirmed, and each adjustment process needs to be recorded to establish a check log of each device, to ensure that each adjustment has a traceable record, and to monitor the device running state to prevent clock oscillation caused by frequent adjustment, and to ensure the overall stability and synchronization accuracy of the system.

[0056] The clock compensation module is used for compensating for the synchronization failure caused by the long-term error and the short-term error of the local device in the star flash wireless audio transmission process, wherein the long-term error is caused by the aging of the inherent crystal oscillator of the local device or the change of the environmental temperature, and the short-term error is caused by short-term noise, and an adaptive multi-source fusion compensation mechanism is established for clock compensation, specifically, the adaptive multi-source fusion compensation mechanism includes the following steps:

[0057] A clock error model is established through multi-source data, and model parameters are dynamically adjusted in combination with historical data and real-time monitoring data;

[0058] Due to the complexity of long-term errors and short-term errors, multi-source data needs to be combined for analysis, wherein the multi-source data includes multi-dimensional data such as different audio test environments, audio device running states, audio transmission quality, and audio transmission data in the current environment;

[0059] In order to more accurately identify and compensate for errors, a clock error model is first established, which adopts a lightweight LSTM neural network structure and can not only process time series data but also effectively capture long-term dependencies and output predicted clock error values. After the clock error model is established, historical data and real-time monitoring data need to be aligned for model parameter adjustment, which includes but is not limited to input data preprocessing, model training, and parameter optimization, to reduce the error of the model output;

[0060] The clock error model is trained offline and online respectively. Offline training uses historical data, and online training uses current measured data. Based on offline training, online training is used for adjustment to optimize the clock error model;

[0061] In the traditional clock synchronization scheme, offline training often fails to cover all scenarios. Although online training can respond to environmental changes in real time, the training data is limited and lacks adjustment benchmarks, resulting in low optimization efficiency of the model. Therefore, by combining the advantages of offline and online training, the method of offline training as the basis and dynamic adjustment through online training is adopted, which not only guarantees the generalization ability of the model but also improves its adaptability to real-time environmental changes, thereby significantly improving the accuracy of clock compensation and the stability of system operation. The historical data used in offline training includes multi-dimensional data such as different audio test environments, audio device running states, and audio transmission quality. Online training real-time collects audio transmission data in the current environment to optimize the model;

[0062] The clock error model outputs a predicted clock error value within a fixed interval, which includes a long-term error component and a short-term error component;

[0063] After the clock error model is optimized, the predicted clock error value output by the model is used to accurately distinguish between long-term and short-term error components, thereby facilitating subsequent compensation;

[0064] The long-term error component is caused by factors such as hardware aging (for example, crystal oscillator frequency drift), slow environmental changes, and the like, and is specifically manifested as a low-frequency cumulative trend, while the short-term error is usually caused by high-frequency interference, such as power supply noise, sudden temperature changes, and the like, and is specifically manifested as a high-frequency irregular deviation, so that the long-term and short-term error components can be distinguished by a residual-based adaptive decomposition method combined with the residual characteristics of the model prediction error, the extracted data is first used as a long-term error component by using a moving average method, and then the remaining data is checked for statistical characteristics, since the short-term error has irregularity, no regular trend should appear in the checking process, such as regular noise, regular change, and the like, when the above regular change appears in the checking process of the remaining data, the long-term error component is re-extracted, and the checking process is repeated until the long-term and short-term error components are completely distinguished;

[0065] In addition, a time window-based statistical decomposition method can also be used to distinguish the long-term and short-term error components, which can refer to the prior art, and will not be described in detail in the present embodiment;

[0066] Different compensation strategies are usually needed for different error components, for example, the long-term error can be compensated by adjusting the crystal oscillator frequency, while the short-term error can be fine-tuned by real-time adjustment of the clock phase to ensure that the system can still maintain high-precision synchronization in complex environments;

[0067] A double-loop compensation strategy is established to coarsely and finely adjust the clock according to the clock error value, wherein the coarse adjustment directly adjusts the control voltage of the voltage-controlled oscillator of the local phase-locked loop, and the fine adjustment generates a compensation signal through a low-noise digital frequency synthesizer;

[0068] When dealing with complex environmental changes, the traditional single-loop compensation strategy is difficult to balance accuracy and response speed, so in the present embodiment, a double-loop compensation strategy is used, the coarse adjustment directly adjusts the control voltage of the voltage-controlled oscillator of the local phase-locked loop, thereby directly changing the clock frequency, for example, if the predicted clock error value is positive, the control voltage is reduced and the phase-locked loop output frequency is reduced;

[0069] Fine adjustment generates a compensation signal through a low-noise digital frequency synthesizer and injects it into the clock generation circuit to cancel out phase jitter. For example, if a phase jitter of +5 nanoseconds is detected, a -5 nanosecond lead compensation signal is generated to ensure accurate clock phase alignment. Through the synergistic effect of dual closed loops, it can quickly respond to environmental changes while maintaining high-precision synchronization. Clock errors are compensated for using different methods, including: first, adjusting the clock frequency based on the predicted long-term error component to stabilize the clock reference; second, fine-tuning the clock phase in real time based on the short-term error component to eliminate instantaneous jitter; and finally, combining coarse and fine adjustment to ensure that the system maintains high-precision synchronization even in dynamic environments.

[0070] The synchronization status monitoring module is used to monitor the compensation process and the clock signal after compensation, and to link with the random verification process between devices. It generates verification data packets at fixed intervals and sends the verification data packets to adjacent devices. By comparing the received verification data packets with the local clock, the synchronization effect is evaluated. If the deviation is found to exceed the preset threshold, a new round of compensation adjustment is triggered to ensure that the clock synchronization accuracy is always maintained within the set range.

[0071] After compensation, clock synchronization requires further monitoring of the synchronization status, including monitoring the clock signal during and after the compensation process, to ensure that the compensation effect meets expectations. Since a complete inter-device random verification framework is already in place, this framework can be directly used for continuous monitoring of the synchronization status. Specifically, verification data packets are generated according to a preset period and sent to adjacent slave devices. Upon receiving the packets, the slave devices compare them with their local clocks. If the deviation exceeds a threshold, feedback is sent to the master device, which adjusts its compensation strategy accordingly, forming a closed-loop control to ensure stable synchronization accuracy. Simultaneously, the results of each verification are recorded to optimize subsequent clock error models. By accumulating data, the prediction algorithm is continuously corrected to improve compensation accuracy.

[0072] Example 2: Figure 2 As shown, this application provides a networking method for a StarFlash wireless audio transmission system. This method is used to achieve efficient and stable networking connections in the StarFlash wireless audio transmission system. The method includes the following steps:

[0073] Step 101: Generate a synchronous master clock and auxiliary clock, and integrate a bidirectional clock protocol. The bidirectional clock protocol adopts bidirectional timestamp chain technology and realizes offset calibration between nodes through a bidirectional synchronization protocol. The node includes at least one master device and multiple slave devices.

[0074] Step 102: Perform random verification between different slave devices based on a lightweight verification protocol and generate verification results. Fine-tune the synchronization master clock or auxiliary clock according to the generated verification results. The lightweight verification protocol is a short message based on UDP.

[0075] Step 103: compensate for the synchronization failure caused by the long-term error and short-term error of the local device in the star flash wireless audio transmission process, wherein the long-term error is caused by the aging of the inherent crystal oscillator of the local device or the change of the environmental temperature, and the short-term error is caused by the short-term noise, and an adaptive multi-source fusion compensation mechanism is established to compensate the clock;

[0076] Step 104: monitor the compensation process and the compensated clock signal, and link with the inter-device random checking process, generate checking data packets at fixed intervals, and send the checking data packets to adjacent devices, evaluate the synchronization effect by comparing the received checking data packets with the local clock, and if the deviation exceeds the preset threshold, trigger a new round of compensation adjustment to ensure that the clock synchronization accuracy is always maintained within the set range.

[0077] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Those skilled in the art can make various changes and modifications to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A star flash wireless audio transmission system characterized by: The system comprises: A dual-clock source integration module for generating a synchronization master clock and an auxiliary clock, and integrating a bidirectional clock protocol, which adopts a bidirectional timestamp chain technology and realizes offset calibration between nodes through a bidirectional synchronization protocol, the nodes including at least one master device and a plurality of slave devices; An inter-device random check module for performing random check between different slave devices based on a lightweight check protocol, and generating a check result, and fine-tuning the synchronization master clock or the auxiliary clock according to the generated check result, wherein the lightweight check protocol is a short message based on UDP, including the following steps: Setting a fixed check period and randomly selecting a group of slave devices participating in the check, in each check period, each group of slave devices exchanges the current synchronization offset; calculating the absolute difference of the synchronization offset of each group of devices, and fine-tuning the master clock or the auxiliary clock according to the absolute difference; determining whether to adjust the master clock or the auxiliary clock according to the ratio of the absolute difference to the preset threshold value, when the ratio of the absolute difference to the preset threshold value is less than one, the auxiliary clock is adjusted preferentially, otherwise, the master clock is directly adjusted, triggering a full-link re-adjustment; After fine-tuning, the synchronization state is confirmed, the adjusted synchronization offset is recorded, and the check log is updated; A clock compensation module for compensating for synchronization failure caused by long-term error and short-term error of the local device in the star flash wireless audio transmission process, wherein the long-term error is caused by the aging of the inherent crystal oscillator of the local device or the change of the environmental temperature, and the short-term error is caused by short-term noise, and an adaptive multi-source fusion compensation mechanism is established for clock compensation; A synchronization state monitoring module for monitoring the compensation process and the compensated clock signal, and linking with the inter-device random check process.

2. A star flash wireless audio transmission system according to claim 1, characterized in that: The bidirectional synchronization protocol realizes the calibration of the clock between nodes, including the following steps: First, the master device broadcasts a synchronization start packet to all slave devices, the synchronization start packet containing the global start time of the master clock and the master clock frequency; Then, after the slave device receives the synchronization start packet broadcast from the master device, the local receiving time is recorded based on the auxiliary clock, and the initial offset is calculated according to the global start time, which is equal to the sum of the local receiving time minus the global start time and the transmission delay; After the slave device calculates the initial offset, the initial offset is fed back to the master device, and the master device adjusts the synchronization start packet according to the feedback initial offset and rebroadcasts the adjusted synchronization information.

3. A star flash wireless audio transmission system according to claim 1, characterized in that: The adaptive multi-source fusion compensation mechanism includes the following steps: The clock error model is established by multi-source data, and the model parameters are dynamically adjusted combined with historical data and real-time monitoring data; the clock error model is trained offline and online respectively, the offline training is performed by using historical data, and the online training is performed by using current measured data, the offline training is taken as a basis, and the clock error model is adjusted and optimized through online training; the clock error value in a fixed interval is output by the clock error model, and the clock error value includes a long-term error component and a short-term error component; a double closed loop compensation strategy is established, and the clock is coarsely adjusted and finely adjusted according to the clock error value, wherein the coarse adjustment directly adjusts the control voltage of the voltage controlled oscillator of the phase-locked loop in the local crystal oscillator, and the fine adjustment generates a compensation signal through a low-noise digital frequency synthesizer.

4. A star flash wireless audio transmission system according to claim 3, wherein: The clock error model adopts a lightweight LSTM neural network structure.

5. A star flash wireless audio transmission system according to claim 3, wherein: When the predicted clock error value is positive, the control voltage is reduced, and the phase-locked loop output frequency is reduced; when the detected phase jitter value is positive, a negative leading compensation signal is generated.

6. A star flash wireless audio transmission system according to claim 1, characterized in that: The synchronization state monitoring generates a check data packet at a fixed interval, and sends the check data packet to the adjacent equipment, and evaluates the synchronization effect by comparing the received check data packet with the local clock, if the deviation exceeds the preset threshold, a new round of compensation adjustment is triggered, and the clock synchronization precision is ensured to be maintained within the set range.

7. A networking control method of a star flash wireless audio transmission system, characterized in that: It comprises: A synchronization master clock and an auxiliary clock are generated, and a bidirectional clock protocol is integrated, the bidirectional clock protocol adopts a bidirectional timestamp chain technology, and offset calibration between nodes is realized through a bidirectional synchronization protocol, the nodes include at least one master device and a plurality of slave devices; Random checking is performed between different slave devices based on a lightweight checking protocol, and a checking result is generated, and the synchronization master clock or the auxiliary clock is fine-tuned according to the generated checking result, wherein the lightweight checking protocol is a short message based on UDP, comprising the following steps: A fixed checking period is set, and a group of slave devices participating in checking are randomly selected, and in each checking period, each group of slave devices exchanges the current synchronization offset; the absolute difference value of the synchronization offset of each group of devices is calculated, and the master clock or the auxiliary clock is fine-tuned according to the absolute difference value; whether the master clock or the auxiliary clock is adjusted is determined according to the ratio of the absolute difference value to the preset threshold, when the ratio of the absolute difference value to the preset threshold is less than one, the auxiliary clock is adjusted preferentially, otherwise, the master clock is directly adjusted, and full-link re-adjustment is triggered; After fine-tuning, the synchronization state is confirmed, the adjusted synchronization offset is recorded, and the checking log is updated; The synchronization failure caused by long-term error and short-term error of the local device in the star flash wireless audio transmission process is compensated, the long-term error is caused by aging of the inherent crystal oscillator of the local device or change of the environmental temperature, and the short-term error is caused by short-term noise, and an adaptive multi-source fusion compensation mechanism is established for clock compensation. The compensation process and the compensated clock signal are monitored, and are linked with the random check process between devices, check data packets are generated at fixed intervals, and the check data packets are sent to adjacent devices. By comparing the received check data packets with the local clock, the synchronization effect is evaluated. If the deviation exceeds the preset threshold, a new round of compensation adjustment is triggered to ensure that the clock synchronization accuracy is always maintained within the set range.

8. A networking control method of a star flash wireless audio transmission system, characterized in that: The application is applied to a star flash wireless audio transmission system in any one of claims 1 to 6.

Images