Method and system for identifying abnormal signal of internal communication bus of vending machine

Abstract

The invention discloses a vending machine internal communication bus abnormal signal identification method and system, and the method comprises the steps: synchronously collecting a power supply signal and a bus signal through employing a dual-channel ADC, extracting the noise components of the two signals, carrying out the source synchronization analysis, quantifying the data, and calculating the correlation score between the signals, therefore, the causal relationship generated by the abnormity is deconstructed on the physical signal level. Through the mode, accurate source tracing of an abnormal signal source is realized, and environment interference generated by power supply fluctuation through electromagnetic coupling and endogenous faults generated by a communication link are effectively distinguished, so that the operation stability and the operation and maintenance efficiency of the vending machine in a complex electromagnetic environment are improved.

Description

Technical Field

[0001] This application relates to the field of anomaly identification, and more specifically, to a method and system for identifying abnormal signals on the internal communication bus of a vending machine. Background Technology

[0002] As vending machines evolve towards greater intelligence and financialization, they are no longer limited to simple mechanical dispensing; instead, they integrate complex payment systems and multi-dimensional data interaction functions. Their internal communication bus, serving as a crucial neural hub connecting the main control unit, payment module, and actuators, carries massive amounts of control commands and transaction data. Developing an efficient and accurate communication bus anomaly signal identification solution is of paramount importance for ensuring the stable 24 / 7 operation of vending machines, reducing maintenance costs, and guaranteeing the security of transaction data.

[0003] However, existing vending machine communication monitoring technologies primarily determine communication status by verifying frame errors, timeouts, or CRC checksums in the communication protocol. This traditional monitoring method reveals significant technical shortcomings when facing increasingly complex on-site environments. Specifically, monitoring at the logic layer alone cannot address subtle differences introduced at the physical layer. The system cannot effectively distinguish between routine interference caused by equipment aging and environmental noise, and bus anomalies caused by unauthorized intrusion, based on the physical layer characteristics of the signals. This lack of awareness of physical layer characteristics makes it difficult for the system to identify the true nature of abnormal signals. This not only fails to provide accurate preventative maintenance for aging equipment in the existing market but also leaves undetectable financial security risks in new markets.

[0004] Therefore, an optimized method for identifying abnormal signals on the internal communication bus of vending machines is needed. Summary of the Invention

[0005] To address the aforementioned technical problems, this application provides a method and system for identifying abnormal signals on the internal communication bus of a vending machine.

[0006] According to one aspect of this application, a method for identifying abnormal signals on the internal communication bus of a vending machine is provided, comprising: In response to receiving an error trigger signal, the power supply voltage and bus signal are sampled at high speed using a dual-channel ADC to obtain the bus signal waveform and the power supply voltage waveform. Noise components are extracted from the bus signal waveform and power supply voltage waveform to obtain the bus noise component and power supply noise component; Noise source synchronization analysis is performed on bus noise components and power supply noise components to obtain signal source correlation scores; The interference source causality is determined by comparing the signal source correlation score and the correlation threshold to obtain the interference source label; Differentiated fault handling and response based on interference source labels are used to obtain system actions.

[0007] According to another aspect of this application, a system for identifying abnormal signals on the internal communication bus of a vending machine is provided, comprising: The dual-channel high-speed synchronous sampling module is used to respond to the received error trigger signal by performing high-speed synchronous sampling of the power supply voltage and bus signal through a dual-channel ADC to obtain the bus signal waveform and the power supply voltage waveform. The noise component extraction module is used to extract noise components from the bus signal waveform and the power supply voltage waveform to obtain the bus noise component and the power supply noise component. The noise source synchronization analysis module is used to perform noise source synchronization analysis on bus noise components and power supply noise components to obtain signal source correlation scores. The interference source causality determination module is used to determine the interference source causality based on the comparison between the signal source correlation score and the correlation threshold in order to obtain the interference source label. The differentiated fault handling and response module is used to perform differentiated fault handling and response based on interference source labels in order to obtain system actions.

[0008] Compared with existing technologies, this application provides a method and system for identifying abnormal signals on the internal communication bus of a vending machine. It utilizes a dual-channel ADC to synchronously acquire power and bus signals, extracts the noise components of both, and performs source synchronization analysis to quantify the data and calculate the correlation score between signals. This allows for the deconstruction of the causal relationship of the abnormality at the physical signal level. In this way, precise source tracing of abnormal signals is achieved, effectively distinguishing between environmental interference caused by power fluctuations via electromagnetic coupling and intrinsic faults in the communication link itself. This improves the operational stability and maintenance efficiency of the vending machine in complex electromagnetic environments. Attached Figure Description

[0009] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0010] Figure 1 This is a flowchart of a method for identifying abnormal signals on the internal communication bus of a vending machine according to an embodiment of this application; Figure 2 This is a data flow diagram illustrating the method for identifying abnormal signals on the internal communication bus of a vending machine according to an embodiment of this application; Figure 3 This is a block diagram of a vending machine internal communication bus abnormal signal identification system according to an embodiment of this application. Detailed Implementation

[0011] Hereinafter, exemplary embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.

[0012] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not specifically singular and may include plural forms. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0013] While this application makes various references to certain modules of the systems according to embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The modules described are merely illustrative, and different aspects of the systems and methods may use different modules.

[0014] Flowcharts are used in this application to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0015] The technical solution of this application proposes a method for identifying abnormal signals of the internal communication bus of a vending machine. Figure 1 This is a flowchart of a method for identifying abnormal signals on the internal communication bus of a vending machine according to an embodiment of this application. Figure 2 This is a system architecture diagram of a method for identifying abnormal signals on the internal communication bus of a vending machine according to an embodiment of this application. Figure 1 and Figure 2As shown, the method for identifying abnormal signals of the internal communication bus of a vending machine according to an embodiment of this application includes the following steps: S1, in response to receiving an error trigger signal, performing high-speed synchronous sampling of the power supply voltage and bus signal using a dual-channel ADC to obtain the bus signal waveform and the power supply voltage waveform; S2, extracting noise components from the bus signal waveform and the power supply voltage waveform to obtain the bus noise component and the power supply noise component; S3, performing noise source synchronization analysis on the bus noise component and the power supply noise component to obtain a signal source correlation score; S4, determining the causal relationship of the interference source based on the comparison between the signal source correlation score and the correlation threshold to obtain an interference source label; S5, performing differentiated fault handling and response based on the interference source label to obtain system action.

[0016] Specifically, in S1, in response to receiving an error trigger signal, dual-channel ADC is used to perform high-speed synchronous sampling of the power supply voltage and bus signal to obtain the bus signal waveform and power supply voltage waveform. It should be understood that the electromagnetic environment inside a vending machine is extremely complex. Power supply voltage fluctuations generated when high-power components such as the compressor start can interfere with the communication bus through direct coupling or differential coupling (capacitive coupling). To accurately distinguish whether an anomaly on the bus originates from an inherent fault in the communication link itself or from external interference in the power supply system, the system must acquire the original physical layer signal at the moment the anomaly occurs. In the technical solution of this application, dual-channel high-speed synchronous sampling is used to capture the true physical state of the bus signal and power supply voltage at the moment the anomaly occurs, thereby providing necessary data support for subsequent identification of whether the "instantaneous rate of change of the power supply noise waveform" has a differential coupling effect on the signal line.

[0017] In practice, the system first enters a continuous monitoring state, awaiting an error trigger signal. This error trigger signal, acting as a hardware interrupt or a high-priority software event, signifies the system's switch from normal communication monitoring mode to high-precision fault diagnosis data acquisition mode. In response to this signal, the system immediately activates the configured dual-channel analog-to-digital converter (ADC); then, it performs high-speed synchronous sampling of dual signals through the system's dual-channel high-speed synchronous sampling module. The core hardware of this module is a dual-channel ADC, whose two input channels are connected to the vending machine's internal communication bus signal line and power supply voltage sampling point, respectively. During this process, the two channels are physically connected to the vending machine's power supply voltage line (e.g., a 24V power line) and bus signal line (e.g., the MDB bus data line), respectively. High-speed synchronous sampling refers to the simultaneous quantization and conversion of the analog signals connected to each channel under the same clock pulse, ensuring that the two acquired data points are strictly aligned in time and have no phase difference. Then, through continuous sampling by the dual-channel ADC, the system converts the analog bus signal and power supply voltage into discrete digital sequences. This process lasts for a predetermined time window, such as several milliseconds, to cover a period before and after the occurrence of the error trigger signal, ensuring that the complete waveform of the abnormal event is captured. Through this operation, the system ultimately outputs two digital waveform data containing time-series information: one is the bus signal waveform representing the fluctuation of the physical level of the communication line, and the other is the power supply voltage waveform representing the voltage fluctuation of the power supply line. These two waveform data will serve as the core input for subsequent signal processing.

[0018] Specifically, in step S2, noise components are extracted from the bus signal waveform and power supply voltage waveform to obtain bus noise components and power supply noise components. It should be understood that the original waveform contains both normal communication levels or DC power supply components, as well as transient high-frequency noise caused by electromagnetic interference, etc. The amplitude of these abnormal noises is often smaller than that of normal signals, and their morphological characteristics are easily overwhelmed by the main signal. Directly analyzing the original bus signal waveform and power supply voltage waveform is inefficient and inaccurate. Therefore, in the technical solution of this application, noise components are extracted from the bus signal waveform and power supply voltage waveform to separate the high-frequency noise components representing potential anomalies from the original waveform, thereby amplifying the abnormal features and improving the sensitivity and accuracy of subsequent correlation analysis. This preprocessing effectively separates the target signal (noise) from the background signal (normal operating level).

[0019] In practice, firstly, high-frequency transient components are extracted from the bus signal waveform and power supply voltage waveform using a first-order IIR filter to obtain the filtered bus signal sequence and the filtered power supply voltage sequence. The first-order IIR (Infinite Impulse Response) filter is a computationally efficient digital high-pass filter suitable for real-time processing. Its function is to attenuate low-frequency and DC components in the signal while allowing high-frequency components to pass. Specifically, the transfer function of a first-order IIR filter can typically be expressed as a difference equation, which can efficiently filter out low-frequency and DC components in the signal, thereby preserving or even highlighting the high-frequency transient noise of interest. Specifically, this filter recursively calculates the input waveform sequence, allowing high-frequency components to pass while suppressing low-frequency components, ultimately outputting a sequence mainly containing rapidly changing components, i.e., the filtered sequence. Through this filtering operation, the slowly changing DC level and low-frequency communication fundamental frequency in the original waveform are significantly suppressed, while the rapidly changing high-frequency glitches and oscillations caused by interference are preserved and output, thus obtaining the filtered bus signal sequence and the filtered power supply voltage sequence mainly containing high-frequency transient components.

[0020] Next, amplitude normalization based on the maximum absolute value is performed on the filtered bus signal sequence and the filtered power supply voltage sequence to obtain the bus noise component and the power supply noise component. That is, amplitude normalization based on the maximum absolute value eliminates potential amplitude differences between different time points and different signal channels, unifying all noise components to the same scale, thereby ensuring the fairness and accuracy of subsequent correlation comparisons. Specifically, by calculating the maximum absolute value of the data in the sequence as a benchmark, the amplitude of the entire sequence is linearly mapped to a unified numerical range (such as between -1 and 1 or 0 and 1), thus obtaining the final bus noise component and power supply noise component. This normalization process eliminates the influence of dimensions and absolute amplitude, allowing the subsequent calculation of the two-domain correlation coefficient to focus on the similarity of waveform shape and trend, rather than the magnitude of the values.

[0021] Specifically, S3 involves performing noise source synchronization analysis on the bus noise component and the power supply noise component to obtain a signal source correlation score. It should be understood that the traditional noise source identification mechanism based on cross-correlation functions suffers from a technical bottleneck rooted in an overly idealized physical model assumption: that the noise waveform on the communication bus should be directly similar in shape to the noise waveform on the power supply. However, this assumption has significant flaws in the complex electromagnetic environment of a vending machine. The actual physical coupling path is far more complex than direct reproduction, especially neglecting a crucial special relationship—differential coupling effect. Inside the vending machine, the power supply voltage fluctuates drastically when high-power components such as the compressor start up. This interference is not always directly transmitted to the signal bus as a voltage disturbance of the same form. More commonly, the inherent parasitic capacitance between circuit board traces plays a key role, causing the noise coupled to the signal line to be essentially closer to the instantaneous rate of change of the power supply noise waveform, i.e., its first derivative. A rapidly changing power supply notch, even for a very short duration, can induce a sharp interference spike on the signal line through this capacitive coupling path. Therefore, the original cross-correlation algorithm can only measure the morphological similarity between waveforms, but cannot effectively capture the intrinsic relationship between the rate of change of bus noise and power supply noise. When physical coupling is dominated by capacitance, the recognition sensitivity of the algorithm will drop significantly, and it is very easy to make false judgments, wrongly judging the interference that should be attributed to the power supply as uncorrelated, thus losing the reliability of the technical solution.

[0022] To address the aforementioned technical deficiencies, this application proposes an optimization mechanism that integrates two physical models—direct coupling and differential coupling—to accurately match the actual physical coupling path.

[0023] In practice, firstly, the transient rate of change of power supply noise is extracted from the power supply noise components to obtain a power supply noise rate of change sequence. It should be understood that the differential coupling effect is the core physical cause of misjudgment; therefore, the rate of change of power supply noise is first quantified as an independent feature. Specifically, by performing a first-order difference operation on the normalized power supply voltage noise sequence, its first derivative is approximated, generating a new sequence that can characterize the degree of abrupt change in power supply disturbance. This process is expressed by the formula:

[0024] in, This represents the normalized power supply noise component at time point n. This represents the sample value at the previous moment, and the output is... This refers to the sample value of the power supply noise change rate sequence at time point n. This step accurately simulates the physical process by which power supply noise induces interference spikes on the signal bus through parasitic capacitance at the algorithm level. It successfully separates and quantifies the key features that can cause differential coupling effects from the original power supply noise waveform, providing core data input for subsequent two-domain model analysis.

[0025] Next, dual-domain correlation coefficients are calculated for the power supply noise component, the power supply noise rate of change sequence, and the bus noise component to obtain the direct correlation peak and the differential correlation peak. It should be understood that a single signal similarity metric cannot accurately describe the actual electromagnetic interference coupling mechanism inside the vending machine. Specifically, the original cross-correlation algorithm can only measure the morphological similarity between waveforms, but cannot effectively capture the intrinsic relationship between bus noise and the power supply noise rate of change. When physical coupling is predominantly capacitive, the algorithm's recognition sensitivity drops significantly, easily leading to missed detections. Therefore, in the technical solution of this application, dual-domain correlation coefficients are calculated for the power supply noise component, the power supply noise rate of change sequence, and the bus noise component to perform independent and parallel quantitative evaluation of the two possible main coupling paths. Considering that the actual physical coupling path is far more complex than direct reproduction, it may be direct coupling, differential coupling, or a mixture of both. During execution, the system will calculate the peak values ​​of two sets of cross-correlation functions in parallel: one set is used to evaluate the morphological similarity between the bus noise and the power supply noise itself, and the other set is used to evaluate the trend synchronization between the bus noise and the power supply noise rate of change extracted in the previous step. This process is expressed by the formula:

[0026] in, It is a bus noise component. It is the time offset. The peak value represents the direct correlation. It is the peak value of differential correlation. This is the power supply noise component. This is a sequence of power supply noise change rates. This parallel, dual-channel computation provides a complete and unbiased data foundation for correlation score calculation, thus overcoming the blind spots in the original scheme.

[0027] Furthermore, a correlation score for the signal source is generated based on the direct correlation peak and the differential correlation peak. It should be understood that real circuit coupling effects often involve both direct and differential coupling modes, and a single peak indicator cannot comprehensively and fairly reflect the complex physical reality. If judgment is based solely on the direct correlation peak, it will lead to numerous missed detections in scenarios dominated by capacitive coupling; conversely, if only the differential correlation peak is used, it may result in misjudgments in scenarios dominated by resistive direct coupling. Therefore, to allow the algorithm model to approximate physical reality infinitely and possess universality adaptable to different hardware designs, the technical solution of this application organically integrates these two independent indicators to overcome the limitations of a single model. Specifically, by weighted summation of the direct correlation peak and the differential correlation peak, a hybrid correlation score that comprehensively reflects the two physical coupling paths is constructed. This process is expressed by the formula:

[0028] in, The correlation score of the signal source. It is a physical coupling coefficient with a value between 0 and 1, used to characterize the inherent electromagnetic characteristics of the current vending machine hardware design. It can be obtained through mass production line calibration or subsequent adaptive optimization using machine learning. The peak value represents the direct correlation. This represents the peak value of the differential correlation. This step establishes a configurable mathematical model capable of adaptively matching the coupling characteristics of different hardware circuits. When When the value approaches 1, it indicates that the coupling of the device is dominated by direct (resistive) coupling; when it approaches 0, it indicates that it is dominated by differential (capacitive) coupling. In this way, the accuracy of abnormal signal identification is significantly improved, especially the false negative rate in specific coupling scenarios is greatly reduced, and more accurate fault diagnosis directions are provided for operation and maintenance work.

[0029] In summary, this optimized mechanism successfully overcomes the blind spots in identification caused by the oversimplification of the physical model in the original solution. It can accurately and reliably distinguish between power coupling noise caused by the startup of high-power components inside the vending machine and communication anomalies caused by problems with the communication link itself (such as equipment damage or poor line contact). Ultimately, it significantly improves the accuracy of anomaly signal identification, especially by drastically reducing the false negative rate in scenarios where capacitive coupling is dominant. Furthermore, by introducing a configurable physical coupling coefficient, this mechanism has universality and adaptability for vending machines of different models and hardware layouts. This directly translates into higher system operational stability, effectively avoiding equipment service interruptions or unnecessary system restarts caused by incorrectly identifying power interference as communication faults. This improves the operational efficiency and user experience of the vending machine and provides more accurate fault diagnosis directions for maintenance work.

[0030] Specifically, in step S4, the interference source causality is determined based on a comparison between the signal source correlation score and the correlation threshold to obtain the interference source label. That is, by determining the causality of the interference source, the system accurately and reliably distinguishes between power coupling noise caused by the startup of high-power components inside the vending machine and communication anomalies truly caused by problems with the communication link itself (such as equipment damage or poor line contact). Through this determination process, the system can significantly improve the accuracy of anomaly signal identification, especially effectively avoiding equipment service interruptions or unnecessary system restarts caused by incorrectly identifying power interference as communication failures. This improves the vending machine's operational efficiency and user experience, and provides more accurate fault diagnosis directions for maintenance work.

[0031] In practice, the first step involves a correlation quantification comparison of the signal source correlation score and the correlation threshold to obtain a correlation Boolean flag. This step aims to determine whether the correlation between the currently observed bus anomaly and power fluctuation is sufficient to constitute a causal relationship, thereby generating an intermediate state variable, namely the correlation Boolean flag. Specifically, the calculated signal source correlation score is compared with a pre-set correlation threshold stored in the system, and based on the comparison result, one of two predefined labels is selected for output. This determination process is defined by a clear decision logic. Specifically, the system reads the signal source correlation score and compares it numerically with a fixed, preset correlation threshold. The correlation threshold is an empirical value between 0 and 1, or a value calibrated through extensive experiments. The determination rule is as follows: if the signal source correlation score is greater than or equal to the correlation threshold, it is determined to be power coupling interference, and a corresponding first label is generated; conversely, if the signal source correlation score is less than the correlation threshold, it is determined to be an intrinsic fault in the bus link, and a corresponding second label is generated.

[0032] Then, interference source labels are assigned to the correlation Boolean flags to obtain the interference source labels. This is a classification process based on logical branches: in response to the correlation Boolean flag being true, the system logically confirms that the current anomaly is caused by power fluctuations, and therefore determines the interference source label as power coupling; conversely, in response to the correlation Boolean flag being false, the system determines that the current anomaly is not caused by the power supply, but is an independent problem of the communication bus itself, and therefore determines the interference source label as communication endogenous.

[0033] Specifically, S5 involves differentiated fault handling and response based on interference source tags to obtain system action. It should be understood that faults with different root causes require drastically different response strategies; a single approach cannot effectively solve the problem and may even exacerbate the fault. Using the same handling logic for power-coupled interference and bus link-inherent faults will lead to serious risks of misoperation. For example, when power-coupled interference is identified, the correct strategy is to adopt tolerant measures such as soft degradation operation and adding retry mechanisms, because this type of interference is usually transient and recoverable. However, if it is misidentified as power interference and a tolerant strategy is adopted, when it is actually a bus link hardware fault, it will lead to serious consequences such as continuous communication interruption and data loss. Conversely, if an aggressive strategy of immediate shutdown or alarm is adopted for power-coupled interference, frequent false alarms will affect the normal operation of the vending machine. Therefore, differentiated responses based on interference source tags are necessary. In the technical solution of this application, by distinguishing the nature of the interference source, it is effective to avoid equipment service interruptions or unnecessary system restarts caused by incorrectly identifying power interference as a communication fault. This differentiated approach can significantly improve the operational efficiency and user experience of vending machines, while also providing more accurate fault diagnosis directions for maintenance work, ensuring that the fault information stored in the logs truly reflects the device status, rather than false alarms contaminated by external noise.

[0034] In practice, the first step involves tag parsing, policy mapping, and routing selection for the interference source tags to obtain a protocol identifier. This protocol identifier includes ignore protocols and standard recovery protocols. During this process, the system receives the interference source tags from the preceding steps as input and performs tag parsing, policy mapping, and routing selection on them. Internally, this process manifests as a table lookup or state machine transition operation. The system searches for the corresponding processing logic in a predefined policy library based on the tag's specific content (e.g., "power coupling" or "communication-inherent") to determine a protocol identifier. This protocol identifier includes two types: ignore protocols and standard recovery protocols. Specifically, the ignore protocol is a specialized processing mechanism for non-fatal external interference (such as power noise), instructing the system to filter out the anomaly to maintain business continuity. The standard recovery protocol, on the other hand, addresses genuine internal communication failures (such as disconnections or damage), instructing the system to take proactive intervention measures such as resetting or error reporting to restore functionality.

[0035] Then, based on the protocol identifier, system actions are generated. Specifically, the corresponding instruction set is executed according to the protocol identifier determined in the previous step: if the identifier points to the ignore protocol, the system action may be to only log a low-level log in the background indicating that external interference has been detected, while keeping the current communication session and business process uninterrupted; if the identifier points to the standard recovery protocol, the system action may include immediately suspending the current service, resetting the communication bus transceiver, sending a retransmission request, or triggering a serious hardware fault alarm.

[0036] Taking the solution in this application as an example, assuming that in the preceding steps, the system identifies that the current communication error is caused by the compressor starting, and therefore outputs a power coupling interference source label. In this step, the system reads this label, and through its internal policy mapping table, finds that the processing policy corresponding to power coupling is to ignore the protocol. Therefore, the system decides not to send a communication fault interruption request to the host computer, nor to reset the bus, but only to write a timestamped record in the log file of the underlying driver: "Power-side transient interference detected, automatically shielded". Conversely, if in another scenario, the system outputs an intrinsic communication label (e.g., signal attenuation due to line aging), the system will map to the standard recovery protocol and then execute the system action: immediately stop the current vending process, display "Equipment under maintenance" on the screen, and attempt to reset the communication module three times. If this still fails, a hardware repair work order will be sent to the backend server via the network. This differentiated response ensures that the vending machine will not stop working due to a normal compressor start, while also not missing a real line fault.

[0037] In summary, the method for identifying abnormal signals on the internal communication bus of a vending machine according to the embodiments of this application is explained. It utilizes a dual-channel ADC to synchronously acquire power and bus signals, extracts the noise components of both, and performs source synchronization analysis to quantify the data and calculate the correlation score between the signals, thereby deconstructing the causal relationship of the abnormality at the physical signal level. In this way, accurate tracing of the source of abnormal signals is achieved, effectively distinguishing between environmental interference caused by power fluctuations via electromagnetic coupling and intrinsic faults generated by the communication link itself, thereby improving the operational stability and maintenance efficiency of the vending machine in complex electromagnetic environments.

[0038] Furthermore, a system for identifying abnormal signals in the internal communication bus of a vending machine is also provided.

[0039] Figure 3 This is a block diagram of a vending machine internal communication bus abnormality signal identification system according to an embodiment of this application. Figure 3 As shown, the vending machine internal communication bus abnormal signal identification system 300 according to an embodiment of this application includes: a dual-channel high-speed synchronous sampling module 310, used to perform dual-channel high-speed synchronous sampling of power supply voltage and bus signal through a dual-channel ADC to obtain bus signal waveform and power supply voltage waveform in response to receiving an error trigger signal; a noise component extraction module 320, used to extract noise components from the bus signal waveform and power supply voltage waveform to obtain bus noise component and power supply noise component; a noise source synchronization analysis module 330, used to perform noise source synchronization analysis on the bus noise component and power supply noise component to obtain signal source correlation score; an interference source causality determination module 340, used to determine interference source causality based on the comparison between signal source correlation score and correlation threshold to obtain interference source label; and a differentiated fault handling and response module 350, used to perform differentiated fault handling and response based on interference source label to obtain system action.

[0040] As described above, the vending machine internal communication bus abnormality signal identification system 300 according to the embodiments of this application can be implemented in various wireless terminals, such as servers with vending machine internal communication bus abnormality signal identification algorithms. In one possible implementation, the vending machine internal communication bus abnormality signal identification system 300 according to the embodiments of this application can be integrated into the wireless terminal as a software module and / or hardware module. For example, the vending machine internal communication bus abnormality signal identification system 300 can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the vending machine internal communication bus abnormality signal identification system 300 can also be one of many hardware modules of the wireless terminal.

[0041] Alternatively, in another example, the identification system 300 for abnormal signals of the vending machine's internal communication bus can be a separate device from the wireless terminal, and the identification system 300 for abnormal signals of the vending machine's internal communication bus can be connected to the wireless terminal via wired and / or wireless networks, and transmit interactive information in accordance with an agreed data format.

Claims

1. A method for identifying abnormal signals on the internal communication bus of a vending machine, characterized in that, include: In response to receiving an error trigger signal, the power supply voltage and bus signal are sampled at high speed using a dual-channel ADC to obtain the bus signal waveform and the power supply voltage waveform. Noise components are extracted from the bus signal waveform and power supply voltage waveform to obtain the bus noise component and power supply noise component; Noise source synchronization analysis is performed on bus noise components and power supply noise components to obtain signal source correlation scores; The interference source causality is determined by comparing the signal source correlation score and the correlation threshold to obtain the interference source label; Differentiated fault handling and response based on interference source labels are used to obtain system actions.

2. The method for identifying abnormal signals of the internal communication bus of a vending machine according to claim 1, characterized in that, Noise components are extracted from the bus signal waveform and power supply voltage waveform to obtain bus noise components and power supply noise components, including: High-frequency transient components are extracted from the bus signal waveform and the power supply voltage waveform based on a first-order IIR filter to obtain the filtered bus signal sequence and the filtered power supply voltage sequence. The bus signal filtered sequence and the power supply voltage filtered sequence are normalized based on the maximum absolute value to obtain the bus noise component and the power supply noise component.

3. The method for identifying abnormal signals of the internal communication bus of a vending machine according to claim 1, characterized in that, Noise source synchronization analysis is performed on the bus noise component and power supply noise component to obtain the signal source correlation score, including: The transient rate of change of power supply noise is extracted from the power supply noise components to obtain the power supply noise rate of change sequence; The direct correlation peak and the differential correlation peak are obtained by calculating the dual-domain correlation coefficients of the power supply noise component, the power supply noise rate of change sequence and the bus noise component. The correlation score of the signal source is generated based on the direct correlation peak and the differential correlation peak.

4. The method for identifying abnormal signals of the internal communication bus of a vending machine according to claim 3, characterized in that, The direct correlation peak and differential correlation peak are obtained by calculating the dual-domain correlation coefficients of the power supply noise component, the power supply noise rate of change sequence, and the bus noise component. This includes calculating the dual-domain correlation coefficients of the power supply noise component, the power supply noise rate of change sequence, and the bus noise component using the following formula: in, It is a bus noise component. It is the time offset. The peak value represents the direct correlation. It is the peak value of differential correlation. This is the power supply noise component. This is a sequence of power supply noise change rates.

5. The method for identifying abnormal signals of the internal communication bus of a vending machine according to claim 4, characterized in that, The signal source correlation score is generated based on the direct correlation peak and the differential correlation peak, including: calculating the signal source correlation score using the following formula: in, The correlation score of the signal source. It is a physical coupling coefficient with a value between 0 and 1. The peak value represents the direct correlation. This represents the peak value of differential correlation.

6. The method for identifying abnormal signals of the internal communication bus of a vending machine according to claim 1, characterized in that, The interference source causality is determined by comparing the signal source correlation score and the correlation threshold to obtain the interference source label, including: The correlation score and correlation threshold of the signal source are compared quantitatively to obtain a Boolean correlation flag. Interference source labels are assigned to the relevant Boolean flags to obtain the interference source labels.

7. The method for identifying abnormal signals of the internal communication bus of a vending machine according to claim 6, characterized in that, To obtain the interference source label, the Boolean flag of the correlation is assigned an interference source label, including: If the correlation Boolean flag is true, the interference source is identified as power coupling. If the correlation Boolean flag is false, the interference source is identified as communication-endogenous.

8. The method for identifying abnormal signals of the internal communication bus of a vending machine according to claim 1, characterized in that, Differentiated fault handling and response based on interference source labels to obtain system actions, including: The interference source labels are parsed, policy mapped, and routed to obtain protocol identifiers, which include ignored protocols and standard recovery protocols; and Generate system actions based on protocol identifiers.

9. A system for identifying abnormal signals on the internal communication bus of a vending machine, characterized in that, include: The dual-channel high-speed synchronous sampling module is used to respond to the received error trigger signal by performing high-speed synchronous sampling of the power supply voltage and bus signal through a dual-channel ADC to obtain the bus signal waveform and the power supply voltage waveform. The noise component extraction module is used to extract noise components from the bus signal waveform and the power supply voltage waveform to obtain the bus noise component and the power supply noise component. The noise source synchronization analysis module is used to perform noise source synchronization analysis on bus noise components and power supply noise components to obtain signal source correlation scores. The interference source causality determination module is used to determine the interference source causality based on the comparison between the signal source correlation score and the correlation threshold in order to obtain the interference source label. The differentiated fault handling and response module is used to perform differentiated fault handling and response based on interference source labels in order to obtain system actions.

Images