A fragmentation transmission method for low-power-consumption industrial audio acquisition terminal

Abstract

This application relates to a segmented transmission method for low-power industrial audio acquisition terminals. The method includes: the terminal device sending a start identifier message to a server; segmenting the acquired audio data according to preset segmentation conditions to generate multiple data segments and attaching segmentation identifier information; transmitting the data segments to the server via a cellular communication network; the server establishing a data transmission session and allocating buffer space; storing the data segments in the buffer space according to the segment sequence identifier; the terminal device sending an end identifier message; and the server performing integrity checks and merging the data segments after the checks pass to generate a complete audio data sequence. This application reduces the instantaneous power consumption on the terminal side and improves the reliability of audio data transmission through the segmented transmission mechanism.

Description

Technical Field

[0001] This application relates to the field of Internet of Things (IoT) communication technology, and in particular to a segmented transmission method for low-power industrial audio acquisition terminals. Background Technology

[0002] With the increasing demand for condition monitoring and intelligent operation and maintenance of industrial equipment, audio signals are gradually being introduced for equipment operating status analysis and anomaly detection. Compared with traditional monitoring parameters such as temperature and pressure, audio signals have the characteristics of high sampling frequency, large data volume, and strong continuity, which places higher demands on the bandwidth capacity and transmission stability of wireless communication networks. In practical applications, audio acquisition devices are usually deployed in industrial sites. Due to limitations in installation location and power supply conditions, they often use battery power and wireless communication methods to interact with backend systems.

[0003] Currently, wireless communication solutions in the field of industrial field data acquisition mainly include fieldbus networks based on industrial wireless protocols and remote communication solutions based on cellular communication networks. Industrial wireless protocol solutions achieve communication through centralized network management and a multi-hop ad hoc network architecture, offering high reliability when transmitting low-speed, small-volume data such as temperature and pressure. However, these networks have limited bandwidth resources, and communication scheduling primarily relies on periodic time slot allocation. When used to transmit continuous, large-volume audio signals, they are prone to data queuing and increased transmission latency. Furthermore, they depend on field gateways for data aggregation and forwarding, resulting in high network deployment and maintenance costs.

[0004] Cellular network-based solutions offer advantages such as wide coverage and no need for on-site networking; terminal devices can achieve remote data transmission simply by connecting to the public network via a cellular communication module. However, in battery-powered low-power devices, existing cellular network transmission solutions largely rely on the terminal side to complete the transmission control of the entire data. When packet loss, retransmission, or network interruption occurs, the terminal needs to repeatedly send large amounts of data, leading to a surge in instantaneous power consumption, which is detrimental to the long-term operation of battery-powered devices. Furthermore, existing solutions lack mechanisms for server-side fragmentation and reassembly, buffer management, and timeout detection for audio data after transmission, which can easily lead to decreased data transmission reliability in scenarios involving continuous large data transfers. Therefore, how to achieve stable uploading and reliable reassembly of audio data under cellular network conditions without increasing terminal complexity and power consumption is a pressing technical problem that needs to be solved. Summary of the Invention

[0005] In order to achieve stable uploading and reliable reconstruction of audio data under cellular communication network conditions without increasing terminal complexity and power consumption, this application provides a segmented transmission method for low-power industrial audio acquisition terminals.

[0006] This application provides a segmented transmission method for low-power industrial audio acquisition terminals, which adopts the following technical solution: A segmented transmission method for low-power industrial audio acquisition terminals includes: S1. The terminal device sends a start identifier message to the server, the start identifier message carrying the device identifier information and the data collection task identifier information of the terminal device; S2. The terminal device performs segmentation processing on the acquired audio data according to preset segmentation conditions to generate multiple data segments, and adds segmentation identification information to each data segment. The segmentation identification information includes the acquisition task identification information and the segment sequence identification. S3. The terminal device sends the multiple data fragments to the server via a cellular communication network; S4. The server establishes a data transmission session based on the device identification information and the collection task identification information in the start identifier message, and determines a corresponding cache space for the data transmission session; S5. The server receives the plurality of data segments and stores each data segment into the cache space corresponding to the data transmission session according to the segment order identifier carried in each data segment; S6. After all the data segments have been sent, the terminal device sends an end identifier message to the server, the end identifier message carrying expected quantity information; S7. The server performs integrity checks on the data segments received in the data transmission session based on the expected quantity information in the end identifier message; S8. In response to the integrity check being passed, the server merges the data segments in the cache space according to the segment order identifier to generate a complete audio data sequence to restore the audio data.

[0007] By adopting the above technical solution, the terminal device segments the audio data and sends it piece by piece during the audio data acquisition process, eliminating the need to cache complete audio data on the terminal side, thus reducing the storage resource consumption and instantaneous power consumption of the terminal device. Simultaneously, by centralizing complex logic such as data reconstruction and integrity detection on the server side, the computational load on the terminal side is reduced. The server manages data transmission sessions and allocates cache space to perform ordered storage and integrity verification of received data segments, enabling reliable audio data reconstruction even under conditions of fluctuating cellular network bandwidth.

[0008] Optionally, the preset segmentation conditions are preset time length or preset data volume, and each data segment corresponds to a continuous time interval of audio data.

[0009] By adopting the above technical solution, using time length or data volume as the segmentation condition, the size of each data segment is controllable and corresponds one-to-one with the timeline of the audio data. This makes it easier for the server to restore the complete audio signal in chronological order during reconstruction, while avoiding the problem of excessive retransmission costs due to excessively large single data segments causing transmission failures.

[0010] Optionally, step S2 includes the following sub-steps: S21. The terminal device continuously samples the audio signal according to preset sampling parameters to generate the audio data; S22. During the acquisition of the audio data, the terminal device performs real-time segmentation processing on the audio data according to the preset segmentation conditions to generate the multiple data segments; S23. The terminal device adds the fragment identification information to each of the data segments.

[0011] By adopting the above technical solution, the three actions of sampling, segmentation, and labeling are chained together and executed synchronously during the acquisition process. This allows the terminal device to complete segmentation and labeling encapsulation while the audio signal is being acquired, avoiding the additional buffering and processing overhead required for unified segmentation after acquisition.

[0012] Optionally, the terminal device does not cache the complete audio data during the execution of S22. Whenever the amount of audio data accumulated during the continuous sampling process meets the preset segmentation condition, the terminal device generates a data segment and executes S3.

[0013] By adopting the above technical solution, the terminal device adopts an instantaneous streaming segmentation mode, which sends each data segment immediately after it is generated, without waiting for the complete audio data acquisition to be completed. This reduces the peak storage requirements of the terminal device from the amount of complete audio data to the amount of data of a single data segment, and reduces the additional power consumption caused by large-capacity cache read and write under battery power conditions.

[0014] Optionally, step S4 includes the following sub-steps: S41. The server determines whether a corresponding data transmission session already exists based on the device identification information and the data collection task identification information; S42. In response to the absence of a corresponding data transmission session, the server creates a new data transmission session and allocates the cache space for the newly created data transmission session; S43. In response to the existence of a corresponding data transmission session, the server obtains the existing data transmission session.

[0015] By adopting the above technical solution, the server executes deduplication logic before creating a session, which avoids the repeated creation of sessions and allocation of cache space when the terminal device resends the start identifier message due to network jitter. This ensures that data fragments of the same collection task are always managed in the same data transmission session, thus guaranteeing data consistency.

[0016] Optionally, the server records the creation time and active time of the data transmission session when it is created, and updates the active time of the data transmission session each time a data fragment is received.

[0017] By adopting the above technical solution, recording and updating the active time provides a time reference for subsequent timeout detection, enabling the server to distinguish between sessions in normal transmission and abnormal sessions that have not received new data segments for a long time due to network interruption or terminal failure.

[0018] Optionally, the method further includes: In response to the failure of the integrity check, the server continues to wait for the arrival of subsequent data segments for a preset waiting time, and re-executes the integrity check each time a new data segment is received; In response to the expiration of the preset waiting time and the failure of the integrity check, the server terminates the data transmission session and releases the cache space corresponding to the data transmission session.

[0019] By adopting the above technical solution, a certain waiting window is provided when the integrity check fails to accommodate the delayed arrival of data segments due to network latency, thus avoiding misjudgment of transmission failure due to instantaneous network fluctuations. At the same time, by setting an upper limit constraint on the waiting time, abnormal sessions that have been incomplete for a long time can be terminated in a timely manner and cache space can be released, preventing server resources from being continuously occupied by invalid sessions.

[0020] Optionally, in step S7, the server compares the number of data segments received in the data transmission session with the expected number indicated by the expected number information by periodically detecting the data segments received. If the number of data segments received is equal to the expected number, the server determines that the integrity check has passed.

[0021] By adopting the above technical solution and using a quantity comparison method for integrity detection, the calculation logic is simple. The server does not need to verify the content of each data segment to determine whether all data segments have been received, thus reducing the computational overhead of the server in the integrity detection process. The periodic detection method eliminates the need for the server to trigger a detection immediately for each arriving data segment, reducing the detection frequency.

[0022] Optionally, the server can simultaneously receive data segments from multiple terminal devices and establish an independent data transmission session for each acquisition task of each terminal device, thereby isolating and managing data segments from different terminal devices and different data transmission sessions.

[0023] By adopting the above technical solution, and by establishing an independent data transmission session for each acquisition task of each terminal device, data isolation is achieved in multi-terminal concurrent transmission scenarios. This ensures that the transmission abnormality of any terminal device or acquisition task will not affect the data reception and reassembly of other terminal devices or acquisition tasks, thereby improving the stability and scalability of the system in multi-device access scenarios.

[0024] Optionally, in step S8, before merging the data segments in the cache space, the server performs an order check on the received data segments in the cache space according to the segment order identifier carried in each data segment; In response to the result of the order verification indicating that the order of the data segments in the cache space is inconsistent with the order indicated by the segment order identifier, the server rearranges the data segments in the cache space according to the segment order identifier, and performs the merging after the rearrangement is completed.

[0025] By adopting the above technical solution, since the arrival order of data segments may be inconsistent with the sending order due to network routing changes, congestion control, etc. during the transmission process of cellular communication networks, the addition of the sequence verification and rearrangement steps before merging allows the server to tolerate out-of-order data segments. Without requiring the terminal device to retransmit, the data segments can be restored to the correct order before merging, avoiding audio data sequence misalignment caused by out-of-order delivery.

[0026] In summary, this application includes at least one of the following beneficial technical effects: 1. By introducing an audio data fragmentation and transmission mechanism on the terminal side and centralizing complex logic such as data reassembly, integrity detection, and exception handling on the server side, the terminal device does not need to cache complete audio data or perform complex processing operations, which reduces the instantaneous power consumption and computing load on the terminal side and is suitable for low-power battery-powered device scenarios.

[0027] 2. By establishing an independent data transmission session for each terminal device and each acquisition task, and combining active time recording and timeout termination mechanisms, data isolation management and resource reclamation of abnormal sessions are achieved in multi-terminal concurrent transmission scenarios, thereby improving the stability and scalability of the system in multi-device access scenarios.

[0028] 3. By performing sequence checksum and rearranging before merging data segments on the server side, the system can tolerate out-of-order arrival of data segments during cellular communication network transmission, ensuring the correctness of the merged audio data sequence without increasing the processing burden on the terminal side. Attached Figure Description

[0029] Figure 1 A flowchart illustrating the data fragmentation transmission process of a terminal device in a fragmentation transmission method for a low-power industrial audio acquisition terminal according to an embodiment of the present invention is shown.

[0030] Figure 2 A flowchart illustrating the server-side data reception and reassembly process of a segmented transmission method for low-power industrial audio acquisition terminals according to an embodiment of the present invention is shown.

[0031] Figure 3 A schematic diagram of a segmented transmission system for a low-power industrial audio acquisition terminal is shown in one embodiment of the present invention. Detailed Implementation

[0032] The present application will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the scope of the application.

[0033] Before detailing the embodiments of this application, some terms will be explained first.

[0034] An audio acquisition terminal is a device used to acquire audio signals and send audio data to a server via wireless communication. This terminal includes a low-power microcontroller, an audio acquisition module, and a wireless communication module. As an example, an audio acquisition terminal could be a wireless sensor for monitoring safety valve leaks or a wireless audio analysis sensor installed in an industrial site. It operates on battery power and is limited by the device's power supply conditions, processing performance, and storage capacity.

[0035] A data transmission session refers to a data receiving and management unit established by the server for a specific data acquisition task of a terminal device. It is used to isolate and manage all data segments of that acquisition task. Each data transmission session is uniquely identified by a combination of device identification information and acquisition task identification information. That is, different acquisition tasks of the same terminal device correspond to different data transmission sessions, and acquisition tasks of different terminal devices also correspond to different data transmission sessions.

[0036] Segment identification information refers to the identification data attached to each data segment, including acquisition task identification information and segment sequence identification. Acquisition task identification information is used to indicate the acquisition task to which the data segment belongs, and segment sequence identification is used to indicate the position of the data segment in the complete audio data, such as segment 1, segment 2, etc.

[0037] Preset segmentation conditions refer to the division rules used by the terminal device when segmenting audio data. Preset segmentation conditions can be preset time lengths, that is, segmenting audio data according to fixed time intervals; or preset data volume, that is, segmenting audio data according to a fixed number of bytes.

[0038] The following describes an embodiment of this application using a specific industrial application scenario. In a chemical plant, multiple safety valves require leak monitoring. Audio acquisition terminals are installed on the safety valves at different locations in the plant. For example, a safety valve leak monitoring sensor numbered DEV-001 is installed on valve V-101, a wireless audio analysis sensor numbered DEV-002 is installed on pump body P-201, and a safety valve leak monitoring sensor numbered DEV-003 is installed on valve V-305. All three terminal devices operate on battery power and transmit the collected audio data to a remote server for centralized analysis via a cellular communication network.

[0039] Reference Figure 1 and Figure 2 This application's method for segmented transmission of low-power industrial audio acquisition terminals includes a data segmentation and transmission process on the terminal device side and a data reception and reassembly process on the server side. The following description uses a single acquisition task of DEV-001 on valve V-101 as the main thread to illustrate each step of the method.

[0040] When the trigger on DEV-001 initiates data acquisition, DEV-001 sends a start identifier message to the server. This message carries the device identifier information of DEV-001 and the acquisition task identifier information for this acquisition task. For example, the device identifier information is the unique number of the terminal device, DEV-001, and the acquisition task identifier information is the acquisition task number, TASK-20260204-001. The purpose of the start identifier message is to notify the server that a new data transmission is about to begin, enabling the server to prepare receiving resources in advance for subsequent data segments.

[0041] After sending the start flag message, the DEV-001 continuously samples the audio signal at valve V-101 according to preset sampling parameters, generating audio data. For example, the DEV-001 samples the audio signal at a sampling rate of 16kHz and a quantization precision of 16bit, continuously collecting audio data for 10 seconds, generating a total audio data volume of approximately 320KB.

[0042] Because audio data is relatively large, the DEV-001, limited by its storage capacity and processing performance under battery power, does not cache the complete audio data all at once. Instead, it performs real-time segmentation processing on the audio data according to preset segmentation conditions. These preset segmentation conditions have the following two implementation methods.

[0043] In the first implementation, the preset segmentation condition is a preset time length. Taking DEV-001 as an example, the preset time length is 1 second. DEV-001 samples at a 16kHz sampling rate and 16-bit quantization precision, generating 16000 × 2 = 32000 bytes of data per second, or approximately 32KB. DEV-001 generates a data segment after collecting 1 second of audio data. 10 seconds of audio data are divided into 10 data segments. Each data segment corresponds to a continuous time interval of the audio data. The first data segment corresponds to the audio signal from second 0 to second 1, the second data segment corresponds to the audio signal from second 1 to second 2, and so on. The characteristic of this implementation is that the time length covered by each data segment is fixed, and the server can directly splice them according to the timeline during reconstruction.

[0044] In the second implementation, the preset segmentation condition is a preset data volume. Taking DEV-002 as an example, DEV-002 samples the audio signal at pump P-201 at an 8kHz sampling rate and 16-bit quantization precision for 15 seconds, generating a total audio data volume of 8000 × 2 × 15 = 240,000 bytes, or approximately 240KB. The preset data volume is 48KB, so DEV-002 generates a data segment every 48KB of accumulated audio data, dividing the 240KB audio data into 5 data segments. The characteristic of this implementation is that the data volume of each segment is fixed, maintaining a consistent transmission load for each data segment in the cellular communication network across terminal devices with different sampling parameters.

[0045] It should be understood that both of the above implementation methods enable each data segment to correspond to a continuous time interval of audio data; the only difference lies in the criteria for segmentation. In actual deployment, suitable segmentation conditions can be selected based on the sampling parameters of the terminal device and the bandwidth conditions of the cellular communication network; no restrictions are imposed here.

[0046] In traditional data transmission schemes, terminal devices typically read all audio data from their memory and send it to the server all at once after acquiring all the audio data. This method requires the terminal device's memory to be able to hold the complete audio data, and the terminal device needs to read data from the memory in batches during the transmission phase, resulting in high instantaneous power consumption. For terminal devices operating on batteries, there are constraints in terms of both storage capacity and instantaneous power consumption.

[0047] In the embodiments of this application, the terminal device performs real-time segmentation processing on the audio data during the acquisition process. Specifically, the terminal device continuously samples the audio signal according to preset sampling parameters to generate audio data. During the acquisition process, it performs real-time segmentation processing on the audio data according to preset segmentation conditions to generate multiple data segments, and adds segmentation identification information to each data segment. That is, the three actions of sampling, segmentation, and identification addition are executed synchronously during the acquisition process, and the terminal device can complete segmentation and identification encapsulation at the same time as the audio signal is acquired.

[0048] Taking DEV-001 as an example, DEV-001 uses a 1-second segmentation condition, continuously sampling the audio signal at valve V-101 starting from second 0. When the sampling time reaches second 1, DEV-001 generates the first data segment from the approximately 32KB of audio data accumulated from second 0 to second 1, and adds segmentation identification information {task_id: TASK-20260204-001, seq: 1} to the first data segment. At this time, DEV-001 does not stop sampling, but continues to sample the audio signal. When the sampling time reaches second 2, DEV-001 generates the second data segment from the approximately 32KB of audio data accumulated from second 1 to second, and adds segmentation identification information {task_id: TASK-20260204-001, seq: 2}. This process continues until sampling ends.

[0049] During the above process, DEV-001 does not cache complete audio data. Whenever the accumulated audio data during continuous sampling meets the preset segmentation conditions, DEV-001 generates a data segment and sends it to the server via the cellular communication network. That is, DEV-001 adopts an on-demand streaming segmentation mode, sending each generated data segment immediately without waiting for the complete audio data acquisition. For example, DEV-001 sends the first data segment to the server immediately after generating it in the first second, and the second data segment is generated and sent to the server immediately after generating it in the second second. In this way, DEV-001's peak storage requirement is reduced from 320KB to the order of 32KB, reducing the additional power consumption caused by large-capacity cache read / write operations under battery power conditions.

[0050] After all data segments have been sent, DEV-001 sends an end-of-life flag message to the server. This flag message carries the expected number of data segments to be sent in this acquisition task. For example, if DEV-001 acquires 10 seconds of audio data, dividing it into 1-second segments, generating 10 data segments, then the expected number carried in the end-of-life flag message would be 10.

[0051] On the server side, after receiving the start identifier message from DEV-001, the server enters the data transmission session establishment process based on the device identifier information and data acquisition task identifier information carried in the message.

[0052] In the first scenario, the server has not previously received any messages related to TASK-20260204-001 for DEV-001. The server maintains a session index table with the device identifier and the acquisition task identifier as a composite key. The server queries the session index table based on the device identifier DEV-001 and the acquisition task identifier TASK-20260204-001, and determines that no corresponding data transmission session exists. The server then creates a new data transmission session and allocates cache space for it. When creating a data transmission session, the server records the session's creation time and active time; for example, the creation time is 2026-02-04 10:00:00, and the initial active time is also 2026-02-04 10:00:00.

[0053] In the second scenario, due to network jitter in weak signal environments, DEV-001 failed to receive a timely network layer response from the server after sending the start identifier message. This caused DEV-001's communication module to automatically retransmit the start identifier message. Upon receiving the second start identifier message, the server queries the session index table based on the device identifier information DEV-001 and the acquisition task identifier information TASK-20260204-001 to confirm that a corresponding data transmission session already exists. The server retrieves the existing data transmission session without creating a new one or allocating duplicate cache space. Through this deduplication mechanism, data fragments from the same acquisition task are always managed within the same data transmission session.

[0054] In the third scenario, DEV-001 and DEV-002 each triggered their respective data acquisition tasks within the same time period. DEV-001 sent a start identifier message to the server, carrying device identifier information DEV-001 and data acquisition task identifier information TASK-20260204-001. DEV-002 sent a start identifier message to the server, carrying device identifier information DEV-002 and data acquisition task identifier information TASK-20260204-002. The server created independent data transmission sessions for TASK-20260204-001 of DEV-001 and TASK-20260204-002 of DEV-002, and allocated buffer space for each. The two data transmission sessions are indexed with different composite keys in the session index table on the server side. The data fragments sent by DEV-001 are stored in the cache space of the TASK-20260204-001 session, and the data fragments sent by DEV-002 are stored in the cache space of the TASK-20260204-002 session.

[0055] After a data transmission session is established, the server receives data segments sent by the terminal device and stores each data segment in the corresponding session's cache space according to the segment sequence identifier carried in each segment. Taking TASK-20260204-001 of DEV-001 as an example, the server receives a data segment with segment identifier information {task_id: TASK-20260204-001, seq: 1} at 10:00:01, stores this data segment in the cache space at the position corresponding to seq=1, and updates the active time of the data transmission session to 2026-02-04 10:00:01. The server receives a data segment with seq=2 at 10:00:02, stores it in the position corresponding to seq=2, and updates the active time to 2026-02-04 10:00:02. Each time a new data segment is received, the server updates the active time of the data transmission session so that the active time always reflects the moment when the session last received a data segment.

[0056] In some embodiments, the server simultaneously receives data fragments from multiple terminal devices and establishes an independent data transmission session for each acquisition task of each terminal device, thereby isolating and managing data fragments in different terminal devices and different data transmission sessions.

[0057] For example, DEV-001 collects audio data at valve V-101, while DEV-002 collects audio data at pump body P-201 and DEV-003 collects audio data at valve V-305 within the same time period. The server establishes three independent data transmission sessions for DEV-001's TASK-20260204-001, DEV-002's TASK-20260204-002, and DEV-003's TASK-20260204-003, respectively. The buffer spaces for these three sessions are independent of each other. During the data reception phase, data fragments from DEV-001 are only stored in the buffer space of TASK-20260204-001, data fragments from DEV-002 are only stored in the buffer space of TASK-20260204-002, and data fragments from DEV-003 are only stored in the buffer space of TASK-20260204-003.

[0058] With this isolation management method, even if DEV-003 experiences transmission abnormalities due to cellular communication network signal interruption, the transmission abnormality of DEV-003 only affects the data reception status of the TASK-20260204-003 session. The data transmission sessions of DEV-001 and DEV-002 are not affected and can still complete the reception and reassembly of data segments normally.

[0059] In some embodiments, multiple data collection tasks triggered by the same terminal device at different times also establish independent data transmission sessions. For example, DEV-001 triggers TASK-20260204-001 at 10:00 and TASK-20260204-004 at 10:30. The server establishes separate data transmission sessions for these two data collection tasks, and the data segments of TASK-20260204-001 and TASK-20260204-004 are stored in their respective independent cache spaces.

[0060] When the server receives the end-of-transmission flag message from the terminal device, it initiates the integrity check process. Because cellular communication networks may experience delays in data transmission due to network congestion or signal fluctuations, the server's buffer may not have received all data segments by the time the end-of-transmission flag message arrives. Therefore, the server uses periodic checks to determine whether all data segments have arrived, rather than performing a one-time integrity check immediately upon receiving the end-of-transmission flag message.

[0061] Specifically, the server compares the number of data segments received in the data transmission session with the expected number indicated by the expected number information carried in the end-of-transmission flag message. When the number of received data segments equals the expected number, the integrity check is considered passed. For example, the expected number carried in the end-of-transmission flag message of DEV-001 (TASK-20260204-001) is 10. The server performs a count comparison every second. In the first check, 9 data segments have been received in the buffer space, which is not equal to 10, and the integrity check fails. In the next check cycle, 10 data segments have been received in the buffer space, which equals 10, and the integrity check is considered passed.

[0062] The following three scenarios illustrate the three situations: integrity check passes, integrity check passes after waiting, and integrity check terminates due to timeout.

[0063] Scenario 1, taking TASK-20260204-001 of DEV-001 as an example. DEV-001 collects 10 seconds of audio data, dividing it into 1-second segments to generate 10 data fragments, with an expected quantity of 10. Since the cellular signal in the valve V-101 area where DEV-001 is located is good, all 10 data fragments have been received by the server and stored in the buffer space before the end-of-life flag message arrives. After receiving the end-of-life flag message, the server performs its first periodic check. The number of received data fragments is 10, equal to the expected quantity of 10. The integrity check passes, and the server directly enters the merging process.

[0064] Scenario 2, taking DEV-002's TASK-20260204-002 as an example. DEV-002 collects 15 seconds of audio data, dividing it into 5 data segments of 48KB each, with an expected quantity of 5. Intermittent signal obstruction exists in the pump body P-201 area where DEV-002 is located, causing the 4th data segment to experience a significant network delay during transmission. After receiving the end-of-transmission flag message, the server performs its first periodic check. The number of received data segments is 4, which is not equal to the expected 5, and the integrity check fails. The server continues to wait for subsequent data segments to arrive within a preset waiting time of 30 seconds. At the 12th second, the 4th data segment arrives at the server and is stored in the buffer. In the next detection cycle, the server performs an integrity check. The number of received data segments is 5, which is equal to the expected 5, and the integrity check passes. The server then enters the merging process.

[0065] Scenario 3, taking TASK-20260204-003 of DEV-003 as an example. DEV-003 collects 10 seconds of audio data, generating 10 data segments with a 1-second segmentation condition, with an expected quantity of 10. During transmission, the cellular communication network in the valve V-305 area where DEV-003 is located experienced an interruption, and the 8th and 9th data segments were lost. After the network was restored, DEV-003 only successfully sent the 10th data segment and the end marker message. After receiving the end marker message, the server performs an integrity check. The number of received data segments is 8, which is not equal to the expected quantity of 10, and the integrity check fails. The server continues to wait for a preset waiting time of 30 seconds, but since the 8th and 9th data segments were lost during the network interruption, the server still only receives 8 data segments after 30 seconds. The server terminates the data transmission session corresponding to TASK-20260204-003 and releases the cache space occupied by the session to prevent abnormal sessions from continuously occupying server resources.

[0066] After the integrity check passes, the server merges the data segments in the cache space according to the segment sequence identifier to generate a complete audio data sequence to restore the audio data. Taking TASK-20260204-001 of DEV-001 as an example, the cache space stores 10 data segments from seq=1 to seq=10. The server concatenates the audio data content of each data segment in the order of seq=1, seq=2, seq=3...seq=10 to generate a complete 320KB audio data sequence. This audio data sequence corresponds one-to-one with the original audio signal collected by DEV-001 at valve V-101 on the time axis, restoring 10 seconds of continuous audio signal.

[0067] After merging, the server will hand over the complete audio data sequence to subsequent business modules for storage or analysis. For example, the business module can perform spectral analysis on the audio data sequence to determine whether valve V-101 exhibits leakage characteristics.

[0068] The following extended embodiment further illustrates the server-side processing in a scenario of concurrent transmission across multiple terminal devices.

[0069] In this embodiment, the chemical plant needs to simultaneously monitor the audio of valve V-101, pump body P-201, and valve V-305 during maintenance. DEV-001, DEV-002, and DEV-003 trigger their respective data acquisition tasks within the same time window. DEV-001 acquires 10 seconds of audio data and generates 10 data segments (expected quantity 10) with a 1-second segmentation condition; DEV-002 acquires 15 seconds of audio data and generates 5 data segments (expected quantity 5) with a 48KB segmentation condition; and DEV-003 acquires 10 seconds of audio data and generates 10 data segments (expected quantity 10) with a 1-second segmentation condition.

[0070] The server establishes three independent data transmission sessions for TASK-20260204-001 (DEV-001), TASK-20260204-002 (DEV-002), and TASK-20260204-003 (DEV-003). Data segments sent by the three terminal devices arrive at the server via interleaving through the cellular communication network. For example, the server might receive seq=1 from DEV-001 at 10:00:01, seq=1 from DEV-002 at 10:00:01.2, and seq=1 from DEV-003 at 10:00:01.5. Based on the acquisition task identifier information carried in each data segment, the server routes the data segments to the corresponding data transmission session's buffer space.

[0071] All 10 data segments of DEV-001 arrived before 10:00:10, and the end-of-transmission message for DEV-001 arrived at 10:00:11. The server performed an integrity check on TASK-20260204-001, and after passing the check, merged them to generate a complete 320KB audio data sequence. All 5 data segments of DEV-002 arrived before 10:00:16 (the 4th data segment arrived at 10:00:14 due to network latency), and the end-of-transmission message for DEV-002 arrived at 10:00:17. The server confirmed that the integrity check of TASK-20260204-002 passed in the second detection cycle and merged them to generate a complete 240KB audio data sequence. During the transmission of DEV-003, the 8th and 9th data segments were lost. After a preset waiting time of 30 seconds, the server terminated TASK-20260204-003 and released the buffer space.

[0072] During the above process, the transmission anomaly of DEV-003 only resulted in the termination of the TASK-20260204-003 session. The data transmission sessions of DEV-001 and DEV-002 were unaffected and successfully completed the reassembly of data fragments. The lifecycles of the three data transmission sessions on the server side are independent of each other; the creation, detection, merging, or termination of any one session does not affect the state of the other sessions.

[0073] Reference Figure 3 This application also provides a segmented transmission system for low-power industrial audio acquisition terminals. The system includes a terminal device and a server.

[0074] The terminal device includes an audio sensor module, a data acquisition buffer module, and a cellular communication module. The audio sensor module is configured to acquire audio signals, such as picking up the acoustic signal at valve V-101 using a built-in microphone sensor. The data acquisition buffer module is configured to digitize the audio signal output from the audio sensor module according to preset sampling parameters to generate audio data, perform real-time segmentation processing on the audio data according to preset segmentation conditions to generate multiple data segments, and attach segmentation identification information to each data segment. The buffer capacity of the data acquisition buffer module only needs to hold the data volume of a single data segment, not the complete audio data. The cellular communication module is configured to send a start identifier message, multiple data segments, and an end identifier message to the server.

[0075] The server-side includes a message access module and a segmented data processing module. The message access module is configured to receive start identifier messages, data segments, and end identifier messages from one or more terminal devices, and route the messages to the segmented data processing module based on the device identifier and acquisition task identifier information carried in the messages. The segmented data processing module is configured to establish a data transmission session and allocate buffer space based on the start identifier message, store each data segment in the corresponding buffer space according to the segment order identifier, record and update the creation time and active time of the data transmission session, perform integrity checks on the received data segments using a periodic detection method based on the expected quantity information, and merge the data segments according to the segment order identifier to generate a complete audio data sequence after the integrity check passes. If the integrity check fails within a preset waiting time, the segmented data processing module terminates the corresponding data transmission session and releases the buffer space.

[0076] The implementation principle of the segmented transmission method for low-power industrial audio acquisition terminals in this application is as follows: By introducing an audio data segmentation and transmission mechanism on the terminal side, continuously acquired audio data is segmented in real time and sent to the server piece by piece, enabling the terminal device to complete data upload without caching complete audio data. On the server side, by establishing an independent data transmission session for each acquisition task of each terminal device, and combining active time recording, periodic integrity detection, and timeout termination mechanisms, isolated management of concurrent data transmission from multiple terminals and reliable reassembly of audio data are achieved. This reduces power consumption and processing pressure on the terminal side while improving the reliability and integrity of audio data transmission under cellular communication network conditions.

[0077] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0078] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A segmented transmission method for low-power industrial audio acquisition terminals, characterized in that, include: S1. The terminal device sends a start identifier message to the server, the start identifier message carrying the device identifier information and the data collection task identifier information of the terminal device; S2. The terminal device performs segmentation processing on the acquired audio data according to preset segmentation conditions to generate multiple data segments, and adds segmentation identification information to each data segment. The segmentation identification information includes the acquisition task identification information and the segment sequence identification. S3. The terminal device sends the plurality of data fragments to the server via a cellular communication network; S4. The server establishes a data transmission session based on the device identification information and the data collection task identification information in the start identification message, and determines a corresponding cache space for the data transmission session; S5. The server receives the plurality of data segments and stores each data segment into the cache space corresponding to the data transmission session according to the segment order identifier carried in each data segment; S6. After all the data segments have been sent, the terminal device sends an end identifier message to the server, the end identifier message carrying expected quantity information; S7. The server performs integrity checks on the data segments received in the data transmission session based on the expected quantity information in the end identifier message; S8. In response to the integrity check being passed, the server merges the data segments in the cache space according to the segment order identifier to generate a complete audio data sequence to restore the audio data.

2. The segmented transmission method for low-power industrial audio acquisition terminals according to claim 1, characterized in that, The preset segmentation condition is a preset time length or a preset data volume, and each data segment corresponds to a continuous time interval of the audio data.

3. The segmented transmission method for low-power industrial audio acquisition terminals according to claim 1, characterized in that, S2 includes the following sub-steps: S21. The terminal device continuously samples the audio signal according to preset sampling parameters to generate the audio data; S22. During the acquisition of the audio data, the terminal device performs real-time segmentation processing on the audio data according to the preset segmentation conditions to generate the multiple data segments; S23. The terminal device adds the fragment identification information to each of the data segments.

4. The segmented transmission method for low-power industrial audio acquisition terminals according to claim 3, characterized in that, During the execution of S22, the terminal device does not cache the complete audio data. Whenever the amount of audio data accumulated during the continuous sampling process meets the preset segmentation condition, the terminal device generates a data segment and executes S3.

5. The segmented transmission method for low-power industrial audio acquisition terminals according to claim 1, characterized in that, S4 includes the following sub-steps: S41. The server determines whether a corresponding data transmission session already exists based on the device identification information and the data collection task identification information; S42. In response to the absence of a corresponding data transmission session, the server creates a new data transmission session and allocates the cache space for the newly created data transmission session; S43. In response to the existence of a corresponding data transmission session, the server obtains the existing data transmission session.

6. The segmented transmission method for low-power industrial audio acquisition terminals according to claim 5, characterized in that, The server records the creation time and active time of the data transmission session when it creates the data transmission session, and updates the active time of the data transmission session each time it receives the data fragment.

7. The segmented transmission method for low-power industrial audio acquisition terminals according to claim 6, characterized in that, The method further includes: In response to the failure of the integrity check, the server continues to wait for the arrival of subsequent data segments for a preset waiting time, and re-executes the integrity check each time a new data segment is received; In response to the expiration of the preset waiting time and the failure of the integrity check, the server terminates the data transmission session and releases the cache space corresponding to the data transmission session.

8. The segmented transmission method for low-power industrial audio acquisition terminals according to claim 1, characterized in that, In step S7, the server compares the number of data segments received in the data transmission session with the expected number indicated by the expected number information through periodic detection. In response to the number of data segments received being equal to the expected number, the server determines that the integrity detection has passed.

9. The segmented transmission method for low-power industrial audio acquisition terminals according to claim 1, characterized in that, The server simultaneously receives data segments from multiple terminal devices and establishes an independent data transmission session for each acquisition task of each terminal device, thereby isolating and managing the data segments in different terminal devices and different data transmission sessions.

10. The segmented transmission method for low-power industrial audio acquisition terminals according to claim 1, characterized in that, In step S8, before merging the data segments in the cache space, the server performs an order check on the received data segments in the cache space according to the segment order identifier carried in each data segment. In response to the result of the order verification indicating that the order of the data segments in the cache space is inconsistent with the order indicated by the segment order identifier, the server rearranges the data segments in the cache space according to the segment order identifier, and performs the merging after the rearrangement is completed.

Images