Wireless electrocardiograph monitor based on Internet of Things and data sharing method

Abstract

The invention relates to the technical field of remote monitoring, in particular to a wireless electrocardiograph monitor based on the Internet of Things and a data sharing method. A lead signal sequence is collected, window variance and adjacent difference slope are evaluated to exceed a threshold value to generate a trigger instruction, a low-noise channel packaging data packet is called to monitor time delay and interrupt to generate a parameter set, and the parameter set is sent to a server; and extracting interrupt over-limit interface detection impedance to trigger port switching to generate a connection state, and counting a link stability period comparison threshold value to restart a channel. By calculating the time window variance and the differential slope mean value of the electrocardiosignal in real time, comparing with a preset threshold value, accurately identifying abnormity, reducing abrupt change detection delay, dynamically screening low-interference channels based on channel noise, optimizing a transmission path in combination with data priority classification and transmission interruption statistics, reducing the influence of environmental fluctuation on the waveform, and improving the detection accuracy of the electrocardiosignal. Physical lead switching is triggered through wire harness impedance detection, a double-link redundancy architecture is constructed, and the risk of single-link failure is avoided.

Description

Technical Field

[0001] The present invention relates to the field of remote monitoring technology, and in particular to a wireless electrocardiogram monitor and a data sharing method based on the Internet of Things. Background Art

[0002] The field of remote monitoring technology includes the real-time collection, dynamic analysis and cross-regional sharing of physiological data through information transmission and processing technology. Based on medical monitoring equipment, combined with the Internet of Things communication protocol and data exchange standards, a multi-node collection network for patients' physiological parameters is constructed, and a data transmission channel is established through a low-power wide area network or cellular network. Data integration is completed according to the medical information system interface specifications, and ultimately a full-process monitoring system covering pre-hospital emergency care, in-hospital diagnosis and treatment, and out-of-hospital follow-up is formed. Its core lies in eliminating the technical limitations of traditional monitoring equipment in terms of spatial coverage, data transmission timeliness, and multi-terminal collaboration capabilities, and ensuring the continuous acquisition and standardized flow of vital signs data in different medical scenarios.

[0003] The IoT-based wireless ECG monitor refers to an ECG monitoring terminal built by integrating a wearable sensor unit with a narrowband IoT communication module. A two-way data link is established between the monitoring terminal and a cloud database based on the Medical IoT platform. The message queue telemetry transmission protocol is used to compress and encapsulate ECG waveform data and transmit it asynchronously. Data access permissions between medical institutions are set based on an open authorization mechanism. This covers the miniaturization of ECG signal acquisition devices, the compatibility configuration of multi-protocol communication gateways, integrity verification of data packet transmission processes, and security authentication strategies for shared data access interfaces. The technical implementation is achieved through optimizing the bioelectric sensor layout, adapting to wireless communication standards in different frequency bands, deploying data verification code generation rules, and configuring a role-based access control model.

[0004] In the existing technology, abnormal physiological signal detection relies on a single threshold or dimension, and does not integrate dynamic time series feature analysis. It is easy to miss poor lead contact or electromyographic interference, which reduces data reliability. Data transmission uses a fixed channel allocation mechanism, which cannot adjust the transmission path according to real-time noise fluctuations. Electromagnetic interference from multiple devices can easily cause high-priority data to be lost, resulting in the loss of waveform features. The physical lead and wireless transmission modes operate independently and lack the ability to dynamically switch autonomously based on transmission quality. When the wireless module fails due to power or hardware failure, the system cannot seamlessly switch to the physical cable, resulting in interruption of monitoring continuity. For example, in an ambulance scenario, traditional equipment causes wireless transmission interruption due to multi-band interference and there is no physical link to automatically take over, delaying the emergency team from obtaining the patient's heart rhythm data. The above defects are caused by the single signal processing dimension, the lack of dynamic channel management and insufficient redundancy of the transmission medium, which restrict the reliability of the monitoring system. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a wireless ECG monitor and data sharing method based on the Internet of Things.

[0006] In order to achieve the above-mentioned object, the present invention adopts the following technical solution: a wireless ECG monitor based on the Internet of Things, the ECG monitor comprising:

[0007] The lead signal mutation detection module collects the physiological signal sequence of the RR interval continuously output by the lead patch, evaluates whether the variance value of the current window sequence point and the mean of the differential slope of the adjacent sequence points exceed the threshold, and generates a lead signal mutation trigger instruction;

[0008] A wireless channel adaptive encapsulation module, based on the lead signal mutation trigger instruction, calls a low-noise wireless channel to encapsulate data packets of different transmission rates, monitors the transmission delay and the number of interruptions, and generates a channel transmission quality parameter set;

[0009] A physical lead connection switching module extracts the channel transmission quality parameter, collects the transmission interruption times of the wiring harness interface exceeding the interruption threshold, detects the impedance value of the wiring harness interface, triggers the physical port data flow switching, and generates a physical connection activation state;

[0010] A dual-state link backtracking detection module, based on the physical connection activation state, counts the number of consecutive stable periods of the physical link, compares the stability threshold with the noise backtracking threshold, restarts the wireless channel, and generates a wireless channel restart instruction;

[0011] The monitoring terminal data synchronization module integrates the lead signal mutation trigger instruction, the channel transmission quality parameter set, the physical connection activation status, encapsulates the ECG waveform and link status field, and generates a terminal synchronization data stream.

[0012] As a further solution of the present invention, the lead signal mutation trigger instruction includes an abnormal timestamp, a variance over-limit flag, and a slope over-limit flag; the channel transmission quality parameter set includes an available channel priority list, a data frame type identifier, a delay distribution statistical value, and an interruption frequency statistical value; the physical connection activation status includes an interface impedance qualification flag, a physical link enable instruction, and a wireless component shutdown status mark; the wireless channel restart instruction includes a transmission cycle compliance flag, a channel noise qualification flag, and a wireless component restart trigger mark; the terminal synchronous data stream includes a waveform encapsulation protocol version, an abnormal event coding table, a channel mapping relationship index, and a link status change time series.

[0013] As a further solution of the present invention, the lead signal mutation detection module includes:

[0014] The signal acquisition and feature extraction submodule collects the output signal of the lead patch physiological electrical signal sensing chip, identifies the R wave apex, extracts the continuous RR interval time series, intercepts the data segment according to the preset time window, and generates a time domain segmented sequence;

[0015] The dynamic variance analysis submodule calls the time domain segmentation sequence and calculates the set of time differences between adjacent RR intervals in the window based on the formula:

[0016]

[0017] Obtain the RR interval fluctuation intensity within the window and generate the variance value;

[0018] Among them, σ 2 Represents the variance of RR interval within the window, t i represents the i-th time difference, μ represents the average time difference in the window, and n represents the total number of time differences in the window;

[0019] The mutation determination submodule calls the variance value and calculates the mean set of the absolute values ​​of the adjacent time differences within the window using the formula:

[0020]

[0021] Obtain the differential change rate, generate the slope mean, compare the variance value with the preset mutation variance threshold, and compare the slope mean with the preset mutation slope threshold. When both indicators exceed the threshold, generate a lead signal mutation trigger instruction;

[0022] Among them, K represents the mean of the difference slope, d j represents the absolute value of the jth time difference, Δt represents the sampling interval, and j is the time difference index value.

[0023] As a further solution of the present invention, the wireless channel adaptive encapsulation module includes:

[0024] The channel screening and optimization submodule, based on the lead signal mutation trigger instruction, calls the channel interference noise list of the wireless transmission component, arranges the channel numbers in ascending order by noise value, eliminates channels with noise values ​​higher than the dynamic interference threshold, and generates an available channel queue;

[0025] The data framing adaptation submodule calls the available channel queue to divide the ECG waveform data into high-priority data frames and regular data frames according to a preset frame length, allocates redundant check bits to the high-priority frames based on the frame length difference, and generates a priority data frame set;

[0026] The transmission quality assessment submodule calls the priority data frame set, monitors the confirmation signal delay returned by the receiving end, counts the number of interruptions caused by channel switching during the transmission of each frame, calculates the product of the delay mean and the interruption frequency, and generates a channel transmission quality parameter set.

[0027] As a further solution of the present invention, the physical lead connection switching module includes:

[0028] The interruption monitoring and impedance detection submodule monitors the number of transmission interruptions of the channel transmission quality parameter concentration. When the number exceeds a preset interruption threshold, the impedance value of the lead patch harness interface is detected to generate a real-time impedance value.

[0029] The connection stability evaluation submodule calls the real-time impedance value based on the formula:

[0030]

[0031] Calculate the connection stability index, compare the result with the dynamic stability threshold, and generate a stability judgment mark;

[0032] Among them, S represents the connection stability index, N represents the number of current transmission interruptions, and Z τ represents the preset connection threshold, Z represents the real-time impedance value, ΔZ represents the standard deviation of the three most recent impedance detection values, and α represents the harness aging correction factor;

[0033] The physical port activation submodule calls the stability determination flag. When the determination flag is abnormal, the power supply of the wireless transmitting component is turned off, the data flow of the physical lead wire transmission port is activated, and the physical connection activation state is generated.

[0034] As a further solution of the present invention, the dual-state link backtracking detection module includes:

[0035] a transmission cycle monitoring submodule, which counts the number of continuous and uninterrupted transmission cycles of the physical lead wire based on the activation state of the physical connection, compares the number of cycles with a preset stability threshold, and generates a cycle compliance flag;

[0036] The composite noise evaluation submodule calls the noise value sequence in the channel transmission quality parameter based on the formula:

[0037]

[0038] Calculate the noise retrospective assessment coefficient, compare the result with the retrospective threshold, and generate a noise compliance mark;

[0039] Among them, Q represents the noise retrospective evaluation coefficient, N β Represents the preset backtracking threshold, n k represents the kth noise value, t krepresents the time attenuation factor corresponding to the noise value, represents the noise mean, γ represents the noise fluctuation correction, and λ represents the time attenuation coefficient;

[0040] The wireless restart decision submodule calls the period compliance flag and the noise compliance flag, and when both flags are true, restarts the power supply of the wireless transmitting component and generates a wireless channel restart instruction.

[0041] As a further solution of the present invention, the monitoring terminal data synchronization module includes:

[0042] The data source integration submodule calls the abnormal timestamp in the lead signal mutation trigger instruction, counts the number of interruptions of the channel transmission quality parameter set, extracts the switching records of the physical connection activation state, aligns the three types of data according to the time axis and eliminates redundant entries to generate an integrated data set;

[0043] A protocol field encapsulation submodule, based on the integrated data set, adds an abnormality mark field to the original ECG waveform data, matches the link status field according to the channel number, arranges the field order according to the data structure defined by the remote monitoring protocol, and generates an encapsulated data frame;

[0044] The terminal synchronization generation submodule calls the encapsulated data frame, checks the consistency of the field length with the protocol specification, truncates and fills the excess field with zeros, adds a data check code and a timestamp watermark, and generates a terminal synchronization data stream.

[0045] A method for sharing data of a wireless ECG monitor based on the Internet of Things is provided. The method is performed based on the above-mentioned wireless ECG monitor based on the Internet of Things and comprises the following steps:

[0046] S1: Collect the physiological signal sequence of the RR interval continuously output by the lead patch, evaluate whether the variance value of the current window sequence point and the differential slope mean of the adjacent sequence points exceed the threshold, and generate a lead signal mutation trigger instruction;

[0047] S2: Based on the lead signal mutation trigger instruction, a low-noise wireless channel is called to encapsulate data packets with different transmission rates, the transmission delay and the number of interruptions are monitored, and a channel transmission quality parameter set is generated;

[0048] S3: extracting the wiring harness interface whose transmission interruption times exceed the interruption threshold in the channel transmission quality parameter set, detecting the impedance value of the wiring harness interface, triggering the physical port data flow switching, and generating a physical connection activation state;

[0049] S4: Based on the physical connection activation state, counting the number of consecutive stable periods of the physical link, comparing the stability threshold with the noise backtracking threshold, restarting the wireless channel, and generating a wireless channel restart instruction;

[0050] S5: Integrate the lead signal mutation trigger instruction, channel transmission quality parameter set, and physical connection activation status, encapsulate the ECG waveform and link status field, and generate a terminal synchronization data stream.

[0051] Compared with the prior art, the advantages and positive effects of the present invention are:

[0052] In the present invention, by calculating the variance of the time window of the ECG signal and the mean of the differential slope in real time and comparing it with the preset threshold, accurate anomaly identification is achieved, and the delay in detecting signal mutations is reduced. Low-interference channels are dynamically screened based on channel noise, and the transmission path is optimized by combining data priority classification and transmission interruption statistics to reduce the impact of environmental fluctuations on waveform integrity. The physical lead connection switching is triggered by harness impedance detection to build a dual-link redundant architecture to avoid the risk of single-link failure. The physical link stability period and channel noise backtracking threshold are combined to determine the intelligent recovery of wireless transmission and ensure operational continuity in complex environments. The abnormal timestamp, channel parameters and link status are integrated to generate a structured data stream, enhance the multi-dimensional traceability analysis capability, and improve the accuracy of remote diagnosis and treatment decision-making. This solution systematically solves the problems of signal response lag, weak anti-interference and insufficient link reliability in traditional monitoring through timing feature analysis, channel adaptive optimization, media redundancy switching and data fusion encapsulation, and achieves the simultaneous improvement of vital sign data acquisition quality and transmission stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a system flow chart of the present invention;

[0054] Figure 2 This is an acquisition flow chart of the lead signal mutation detection module of the present invention;

[0055] Figure 3 This is a flowchart of obtaining the wireless channel adaptive encapsulation module of the present invention;

[0056] Figure 4 This is a flow chart of obtaining the physical lead connection switching module of the present invention;

[0057] Figure 5 This is a flowchart of obtaining the dual-state link backtracking detection module of the present invention;

[0058] Figure 6 This is an acquisition flow chart of the monitoring terminal data synchronization module of the present invention. DETAILED DESCRIPTION

[0059] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0060] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0061] In the embodiments of the present invention, the terms "image" and "picture" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same. The terms "of," "corresponding," and "corresponding" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same.

[0062] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0063] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0064] See also Figure 1 The present invention provides a technical solution: a wireless ECG monitor based on the Internet of Things, the ECG monitor comprising:

[0065] The lead signal mutation detection module collects the output of the physiological electrical signal sensing chip of the lead patch, extracts the continuous RR interval sequence, intercepts the data segment according to the preset time window, calculates the variance value within the window based on the time difference between the sequence points, calculates the differential slope by the mean of the absolute value of the difference between adjacent points, compares the variance value with the preset mutation variance threshold, and compares the mean of the differential slope with the preset mutation slope threshold. If both comparison results exceed the threshold, a lead signal mutation trigger instruction is generated;

[0066] The wireless channel adaptive encapsulation module, based on the trigger instruction of the lead signal mutation, calls the channel interference noise list of the wireless transmission component, selects the available channel numbers in ascending order of noise value, divides the ECG waveform data into high-priority data frames and regular data frames according to the preset frame length, monitors the delay of the confirmation signal returned by the receiving end, counts the number of transmission interruptions of each frame, and generates a channel transmission quality parameter set;

[0067] The physical lead connection switching module detects the impedance value of the lead patch's harness interface when the number of transmission interruptions of the channel transmission quality parameter exceeds a preset interruption threshold, compares the impedance value with a preset connection threshold, and if the impedance value is lower than the threshold, turns off the power supply of the wireless transmitter component, activates the data flow of the physical lead wire transmission port, and generates a physical connection activation state;

[0068] The dual-state link backtracking detection module counts the number of continuous, uninterrupted transmission cycles of the physical lead line based on the activation state of the physical connection, compares the number of cycles with a preset stability threshold, and compares the noise value in the channel transmission quality parameter with the backtracking threshold. If the number of cycles meets the standard and the noise value is lower than the backtracking threshold, the wireless transmitter power is restarted and a wireless channel restart instruction is generated;

[0069] The monitoring terminal data synchronization module integrates the abnormal timestamp in the lead signal mutation trigger instruction, the number of interruptions in the channel transmission quality parameter set, and the switching record of the physical connection activation state, and encapsulates the original ECG waveform, abnormal mark, channel number and link status field according to the remote monitoring protocol to generate the terminal synchronization data stream.

[0070] The lead signal mutation trigger instruction includes the abnormal timestamp, variance exceeding limit flag, and slope exceeding limit flag. The channel transmission quality parameter set includes the available channel priority list, data frame type identifier, delay distribution statistics, and interruption frequency statistics. The physical connection activation status includes the interface impedance qualification flag, physical link enable instruction, and wireless component shutdown status mark. The wireless channel restart instruction includes the transmission cycle compliance flag, channel noise qualification flag, and wireless component restart trigger mark. The terminal synchronization data stream includes the waveform encapsulation protocol version, abnormal event coding table, channel mapping relationship index, and link status change time series.

[0071] See also Figure 2 , the lead signal mutation detection module includes:

[0072] The signal acquisition and feature extraction submodule collects the output signal of the lead patch physiological electrical signal sensing chip, identifies the R wave apex, extracts the continuous RR interval time series, intercepts the data segment according to the preset time window, and generates a time domain segmented sequence;

[0073] The lead patch is attached to the V2 lead position of the patient's chest, and the original electrical signal is collected at a sampling rate of 1000Hz through the AD8232 chip. The input signal is filtered through a second-order Butterworth bandpass filter (0.5Hz-40Hz) to eliminate 50Hz power frequency interference and respiratory baseline drift. The filtered signal is identified by the moving window peak detection algorithm: the dynamic detection threshold is set to the sliding average of the amplitudes of the first five R waves (the initial threshold is set to 0.4mV). When the voltage of three consecutive sampling points exceeds the threshold and the slope difference of adjacent points is greater than 8mV, the R wave is detected. / ms is marked as the R wave vertex. If the interval between two detected R waves is less than 200ms, it is determined to be an artifact and eliminated. For example, the measured R wave timestamp sequence is [1.235s, 1.842s, 2.517s, 3.204s], and the calculated RR interval sequence is [0.607s, 0.675s, 0.687s]. The data segment is intercepted according to the 10-second window. When the window contains 15 RR intervals, a time domain segmented sequence is generated. If the data in the window is less than 5 intervals, zero padding is triggered to fill it to the standard length.

[0074] Assume that the measured data of chest leads of a patient is:

[0075] Original sampling point voltage sequence: [0.12mV, 0.35mV, 1.02mV (R wave apex), 0.68mV, 0.21mV];

[0076] Dynamic threshold update: If the average amplitude of the first five R waves is 1.1 mV, a missed detection is considered if the sixth R wave amplitude is 0.95 mV (lower than the threshold × 0.8 = 0.88 mV).

[0077] Abnormal elimination: When the interval between adjacent R waves was 0.18 seconds (<0.3 seconds), it was marked as myoelectric interference.

[0078] The dynamic variance analysis submodule calls the time domain segmentation sequence and calculates the set of time differences between adjacent RR intervals in the window based on the formula:

[0079]

[0080] Obtain the RR interval fluctuation intensity within the window and generate the variance value;

[0081] Among them, σ 2 Represents the variance of RR interval within the window, t i represents the i-th time difference, μ represents the average time difference in the window, and n represents the total number of time differences in the window;

[0082] Call the third window data [0.62s, 0.64s, 0.61s, 0.66s, 0.59s] in the time domain segmented sequence, calculate the time difference set ΔT = [0.62-0.64 = -0.02s, 0.64-0.61 = 0.03s, 0.61-0.66 = -0.05s, 0.66-0.59 = 0.07s], take the absolute value to get [0.02, 0.03, 0.05, 0.07]s, calculate the mean μ = (0.02+0.03+0.05+0.07) / 4 = 0.0425s, and find the variance σ 2 =[(0.02-0.0425) 2 +(0.03-0.0425) 2 +(0.05-0.0425) 2 +(0.07-0.0425) 2 ] / (4-1)=(0.0005+0.00015+0.00005+0.00075) / 3=0.000483s 2 , when the variance in the window exceeds the dynamic threshold for three consecutive times (the mean variance in the first 10 minutes of the baseline period is 0.0002s 2 For example, the measured variance sequence for a period of time is [0.00041, 0.00052, 0.00063], and the dynamic threshold is 0.0006s. 2 , the third data triggers an alarm;

[0083] The threshold setting is based on the analysis of resting heart rate variability in 200 healthy adults:

[0084] Normal RR interval variance range: 0.0001-0.0005s 2 (SDNN30-50ms);

[0085] The mean variance of lead loosening abnormal events is 0.0021s 2 (Standard deviation 0.0007s 2 );

[0086] Dynamic threshold = health mean + 3 × health standard deviation = 0.0005 + 3 × 0.0001 = 0.0008s 2 .

[0087] The mutation determination submodule calls the variance value and calculates the mean set of the absolute values ​​of the adjacent time differences within the window using the formula:

[0088]

[0089] Obtain the differential change rate, generate the slope mean, compare the variance value with the preset mutation variance threshold, and compare the slope mean with the preset mutation slope threshold. When both indicators exceed the threshold, generate a lead signal mutation trigger instruction;

[0090] Among them, K represents the mean of the difference slope, d j represents the absolute value of the jth time difference, Δt represents the sampling interval, and j is the time difference index value;

[0091] Assume that a time window contains the following five RR interval time differences (unit: seconds): d j =(0.75, 0.82, 0.68, 0.91, 0.62, total number of time differences n = 5, sampling interval Δt = 0.004s (corresponding to 250Hz resampling frequency), mutation slope threshold K th =5s -1 ;

[0092] Calculate the absolute value of the adjacent time difference (|d j -d j-1 |):

[0093] From j=2 to j=n-2=3 (i.e., calculating the first three adjacent differences):

[0094] j=2: |d2-d1|=|0.82-0.75|=0.07s;

[0095] j=3: |d3-d2|=|0.68-0.82|=0.14s;

[0096] Calculate the absolute value of the subsequent time difference (|d j+1 -d j |):

[0097] Continue to calculate for j=2 and j=3:

[0098] j=2: |d3-d2|=|0.68-0.82|=0.14s;

[0099] j=3: |d4-d3|=|0.91-0.68|=0.23s;

[0100] Find the sum of the absolute values ​​of each term:

[0101] Calculate |d for each j j -d j-1 |+|d j+1 -d j |:

[0102] j=2:0.07+0.14=0.21s;

[0103] j=3:0.14+0.23=0.37s;

[0104] Total: Σ = 0.21 + 0.37 = 0.58 s;

[0105] Substitute into the formula to calculate the slope mean:

[0106]

[0107] The results show that the average change rate of the adjacent time differences in the current window is 36.25 mutations per second (s -1 ), reflecting the dramatic fluctuations in the RR interval. This value directly quantifies the steepness of the signal mutation. Its physical meaning is: if the lead signal is stable (such as healthy heart rate variability), the K value is usually less than 5s -1 When the lead contact is poor or there is noise interference, the sharp jump of adjacent time differences will cause the K value to increase significantly;

[0108] Assume that the preset mutation slope threshold K th =5s -1 Calculated value K = 36.25s -1 >K th , then the signal mutation in the current window is significant, and the lead signal mutation trigger instruction is generated. The result shows that the K value (36.25s -1 ) far exceeds the preset threshold (5s -1 ), indicating that there are non-physiological jumps in the time difference sequence, combined with the output variance value (needs to be calculated separately, assuming it is 0.0007s 2 And exceeds the variance threshold of 0.0003s 2 ), when both conditions are met, the system determines that the lead signal is abnormal. The threshold is set by clinical data analysis: for example, statistics of 200 lead-off events show that the K value is >4.7s -1 (5th percentile), K value range of healthy people at rest: 0.2s -1 ~3.8s -1 , determine K by ROC curve th =5s -1 When the sensitivity is 98% and the specificity is 96%, if the electrode partially falls off, causing the R wave amplitude to drop sharply and the difference between adjacent RR intervals to increase (such as a sudden change from 0.75s to 0.62s), the K value will increase significantly. If the patient's limb movement generates myoelectric noise, the R wave misdetection will cause the time difference to fluctuate violently (such as 0.68s→0.91s→0.62s), and the K value will far exceed the threshold.

[0109] See also Figure 3 , the wireless channel adaptive encapsulation module includes:

[0110] The channel screening and optimization submodule, based on the trigger instruction of the lead signal mutation, calls the channel interference noise list of the wireless transmission component, sorts the channel numbers in ascending order by noise value, eliminates channels with noise values ​​higher than the dynamic interference threshold, and generates an available channel queue;

[0111] Based on the timestamp of the lead signal mutation trigger instruction (for example, 2023-08-15T14:23:17.352), call the 16 channel interference noise lists recorded by the wireless transmission component in the current period (±5 seconds). The noise value unit is dBm, and the list data is [-92, -88, -95, -78, -85, -90, -80, -83, -97, -75, -89, -93, -82, -91, -86, -84]. After sorting in ascending order, the sorting index is [9(-75), 4(-78), 7(-80), 13(-82), 6(-83), 14(-84), 1(-88), 10(-89), 3(-95), 8(-97)] (excluding the channel numbers in the positive sequence with noise values ​​higher than the dynamic interference threshold of -85dBm). Dynamic The interference threshold is set based on historical noise data: the 75th percentile of the noise value within the previous hour is used. (Calculation process: Arrange the noise value sequence [-80, -82, -85, -88, -90, ...] sampled every 5 seconds for the previous hour in ascending order, and take the value at the index position multiplied by 0.75 of the total number.) If the noise value distribution in the current period is skewed (for example, due to sudden interference), the dynamic threshold is adjusted to the average of -85 dBm and the maximum noise value of the three most recent sudden noise events. (For example, if the historical maximum noise value is -78 dBm, the adjusted threshold is (-85 + (-78)) / 2 = -81.5 dBm). This ultimately generates the available channel queue: [Channel 4 (-85 dBm), Channel 7 (-83 dBm), Channel 13 (-82 dBm), Channel 6 (-83 dBm), Channel 14 (-84 dBm)].

[0112] The data framing adaptation submodule calls the available channel queue to split the ECG waveform data into high-priority data frames and regular data frames according to the preset frame length, allocates redundant check bits to the high-priority frames based on the frame length difference, and generates a priority data frame set;

[0113] Channel 4 (bandwidth 2MHz, rate 250kbps) in the available channel queue is called, and the ECG waveform data is divided according to the preset frame length: the high-priority frame (including 0.5 seconds of waveform data before and after the R wave peak) is 128 bytes long, and the regular frame (baseline waveform) is 256 bytes long. The frame length difference triggers the redundant check bit allocation rule. The high-priority frame adds 20% redundant check bits (128×0.2=25.6 bytes, rounded to 26 bytes, total length 154 bytes), and the regular frame adds 5% redundant check bits (256×0.05=12.8 bytes, rounded to 13 bytes, total length 269 bytes). Example data segmentation: For 10 seconds of ECG data (sampling rate 500Hz, total The data volume is 5000 points × 2 bytes = 10,000 bytes), 3 R-wave events are identified, and 3 high-priority frames are generated (each frame covers 0.5 seconds × 500 Hz = 250 points, 3 × 250 × 2 = 1,500 bytes) and 17 normal frames (remaining 8.5 seconds × 500 Hz = 4,250 points, 17 × 250 × 2 = 8,500 bytes). After adding redundancy, the total data volume is 3 × 154 + 17 × 269 = 462 + 4,573 = 5,035 bytes, and a priority data frame set [frame type tag: HP1 (154B), HP2 (154B), HP3 (154B), NP1 (269B) ... NP17 (269B)] is generated.

[0114] The transmission quality assessment submodule calls the priority data frame set, monitors the delay of the confirmation signal returned by the receiving end, counts the number of interruptions caused by channel switching during the transmission of each frame, calculates the product of the average delay and the interruption frequency, and generates a channel transmission quality parameter set;

[0115] Call the high-priority frame HP1 (transmitted on channel 4) in the priority data frame set, monitor the delay sequence of the confirmation signal returned by the receiver [120ms, 150ms, 130ms], calculate the average delay = (120+150+130) / 3 = 133.3ms, count the number of interruptions caused by channel switching during transmission = 2 times (occurring in the second retransmission of HP1 and the first transmission of NP5), calculate the quality parameter = average delay × number of interruptions = 133.3 × 2 = 266.6, and preset the quality threshold = 2 00 (set based on historical transmission data: when the parameter > 200, the channel quality is determined to be substandard). The current parameter 266.6 exceeds the threshold, and the channel transmission quality parameter set [channel 4: 266.6, channel 7: 180 (example value), channel 13: 310 (example value)] is generated. The dynamic threshold update rule is: if the quality parameter exceeds the threshold for three consecutive time periods, the channel is removed from the available queue (for example, if the parameters of channel 4 in time periods T1 to T3 are 266.6, 280, and 310, channel 4 is blocked in time period T4).

[0116] See also Figure 4, the physical lead connection switching module includes:

[0117] The interruption monitoring and impedance detection submodule monitors the number of transmission interruptions of the channel transmission quality parameter. When the number exceeds the preset interruption threshold, it detects the impedance value of the lead patch harness interface and generates a real-time impedance value.

[0118] The channel transmission quality parameter set for the current time period is monitored, and the transmission interruption count parameter is extracted. When this value exceeds the preset interruption threshold (for example, the number of interruptions in the last 5 minutes is ≥3), the impedance detection of the lead patch harness interface is triggered. The contact impedance is measured at a frequency of 1kHz using a four-wire detection method. Three consecutive measurement values ​​[52kΩ, 48kΩ, 50kΩ] are obtained. Outliers that deviate from the mean by ±10% are eliminated (48kΩ is retained within the range of ±5kΩ of the mean of 50kΩ). The real-time impedance value Z = 50kΩ is calculated. The preset interruption threshold is dynamically set according to the lead type: the standard lead (chest lead) threshold is 2 times / minute, and the limb lead threshold is 1 time / minute. When the chest lead interruption count is detected as 4 times (exceeding the threshold by 2 times) between 14:00 and 14:05, the impedance detection process is activated and the real-time impedance value is generated.

[0119] The connection stability evaluation submodule calls the real-time impedance value based on the formula:

[0120]

[0121] Calculate the connection stability index, compare the result with the dynamic stability threshold, and generate a stability judgment mark;

[0122] Among them, S represents the connection stability index, N represents the number of current transmission interruptions, and Z τ represents the preset connection threshold, Z represents the real-time impedance value, ΔZ represents the standard deviation of the three most recent impedance detection values, and α represents the harness aging correction factor (range of 0.05-0.15);

[0123] The real-time impedance value Z = 50 kΩ is called up, combined with the preset connection threshold Zτ = 45 kΩ (according to the lead wire specification: the qualified impedance range is 30-60 kΩ, and the median is 45 kΩ). The standard deviation of the last three impedance detection values ​​ΔZ = √[(52-50) 2 +(48-50) 2 +(50-50)2] / 3=√[(4+4+0) / 3]=√(8 / 3)=1.63kΩ, harness aging correction factor α=0.1 (set according to the cumulative use time of the harness: 0.15 for >500 hours, 0.1 for 200-500 hours, and 0.05 for <200 hours). Substitute into the formula:

[0124]

[0125] Step-by-step calculation:

[0126] Numerator: 4 × 5 = 20;

[0127] Numerator sum: 20 + 5.26 = 25.26;

[0128] Denominator: 45 × 1.1 = 49.5

[0129] Final result: S = 25.26 / 49.5 = 0.51;

[0130] If the dynamic stability threshold is set to 1.2 (according to 500 sets of clinical data statistics: when S>1.2, the probability of lead-off is>95%), the current result 0.51<1.2, and the generated stability determination mark is normal.

[0131] Formula parameter description:

[0132] N=4: Number of transmission interruptions in the current period (obtained from paragraph 1);

[0133] Z τ =45kΩ: Based on the impedance range median provided by the lead wire manufacturer;

[0134] ΔZ = 1.63 kΩ: standard deviation of the three most recent impedance measurements;

[0135] α=0.1: Set according to the cumulative usage time of the wiring harness (200-500 hours);

[0136] The results show that the current connection stability index (S = 0.51) is significantly lower than the dynamic threshold (1.2), indicating that the impedance deviation of the lead harness (Z τ The combined effect of the impedance deviation (-Z = 5kΩ) and the historical fluctuation (ΔZ = 1.63kΩ) is small, and the connection status is within the stable range. τ -Z) and the geometric superposition of historical fluctuations (ΔZ) The number of interruptions (N=4) is introduced as a weight factor. The final generated S value can be directly compared with the preset threshold without further calculation, only logical judgment (S≤S th Normal, otherwise abnormal) can generate a stability judgment mark.

[0137] The physical port activation submodule calls the stability determination flag. When the determination flag is abnormal, the wireless transmission component power is turned off, the physical lead wire transmission port data stream is activated, and the physical connection activation state is generated;

[0138] The stability determination flag (normal) is called, and the power of the wireless transmitting component is maintained on. When the flag is abnormal (for example, S=1.3>1.2), the wireless component is turned off, the physical lead wire transmission port is activated, and the data stream is transmitted using the RS-485 protocol. The baud rate is set to 115200 bps, the parity bit is configured to even parity, and the data frame format is [frame header 0xAA, impedance value 50kΩ (2 bytes), status code 0x01 (normal)]. The physical connection activation status parameters [port ID = COM3, transmission rate = 115200 bps, parity mode = even parity] are generated.

[0139] Assume that in a certain implementation scenario, when poor lead contact occurs, if the number of transmission interruptions between 14:10 and 14:15 is 5 (exceeding the threshold), the measured impedance sequence is [62 kΩ, 58 kΩ, 65 kΩ], Z = 61.7 kΩ, ΔZ = 3.5 kΩ, α = 0.1, and S = (5 × | 45 - 61.7 | + √ (278.89 + 12.25)) / (45 × 1.1) = (5 × 16.7 + 17.03) / 49.5=(83.5+17.03) / 49.5≈2.03>1.2, triggering physical port switching; if the harness ages, the harness has been used for 600 hours, α=0.15, Z=48kΩ, ΔZ=2.1kΩ, S=(3×|45-48|+√(9+4.41)) / (45×1.15)=(9+3.63) / 51.75≈0.245<1.2, maintaining wireless transmission.

[0140] See also Figure 5 , the dual-state link backtracking detection module includes:

[0141] The transmission cycle monitoring submodule counts the number of continuous and uninterrupted transmission cycles of the physical lead wire based on the activation status of the physical connection, compares the number of cycles with the preset stability threshold, and generates a cycle compliance flag;

[0142] Based on the duration record of the physical connection activation state (for example, port COM3 has been working continuously for 35 minutes), the number of uninterrupted transmission times of the physical lead line in the last 10 transmission cycles (each cycle is 5 minutes) is counted. When the number of cycles is ≥ 8, it is determined to be stable. The preset stability threshold is set according to the lead type: chest lead threshold = 7 times / 10 cycles, limb lead threshold = 9 times / 10 cycles. Example data: The number of uninterrupted transmission times of the current chest lead in cycles 1 to 10 is [1,1,1,1,1,1,1, 1,0,1], the statistical valid number of times = 9 times, which exceeds the threshold 7 times, and the cycle compliance flag is generated [status: True, failure cycle index: 9]. The stability threshold is set based on: clinical data show that the average failure cycle rate of chest leads is ≤30% (that is, ≤3 interruptions are allowed in 10 cycles), and that of limb leads is ≤10% (≤1 interruption is allowed). When the limb lead cycle data [1,1,1,1,1,1,1,0,1,1] (failure number = 1) is detected, the True flag is also generated.

[0143] The composite noise evaluation submodule calls the noise value sequence in the channel transmission quality parameter based on the formula:

[0144]

[0145] Calculate the noise retrospective assessment coefficient, compare the result with the retrospective threshold, and generate a noise compliance mark;

[0146] Among them, Q represents the noise retrospective evaluation coefficient, N β Represents the preset backtracking threshold, n k represents the kth noise value, t k represents the time attenuation factor corresponding to the noise value, represents the noise mean, γ represents the noise fluctuation correction, and λ represents the time attenuation coefficient;

[0147] Call the noise value sequence N in the channel transmission quality parameter set i =[-82dBm,-85dBm,-88dBm,-90dBm,-92dBm] (corresponding to timestamps T1-T5), the time attenuation factor is calculated as: γ i =0.9 Δt (Δt=t current -t i , in minutes), where t current =10:00:00, t i The acquisition time of each noise value (for example, the acquisition time of T1 noise value -82dBm is 09:55:00, Δt = 5 minutes, γ1 = 0.9 5 ≈0.590), the noise mean is calculated as: The noise fluctuation correction is calculated as:

[0148]

[0149] Substitute into the noise retrospective evaluation formula:

[0150] Molecular computing:

[0151] -82×0.590=-48.38;

[0152] -85×0.656==-55.76;

[0153] -88×0.729==-64.15;

[0154] -90×0.810==-72.90;

[0155] -92×0.900==-82.80;

[0156] ∑==-48.38-55.76-64.15-72.90-82.80=-323.99;

[0157] Denominator calculation: μ + Δ = -87.4 + 3.56 = -83.84;

[0158] Final result:

[0159] Preset backtracking threshold R th =3.0 (according to the historical noise data distribution: when R>3.0, the channel noise is within a controllable fluctuation range), and a noise compliance flag [status: True, R value: 3.86] is generated. This result shows that the ratio of the sum of the noise energy after weighted attenuation (numerator -323.99) to the composite index of the noise mean and fluctuation (denominator -83.84) is 3.86, which exceeds the preset threshold of 3.0, indicating that the current channel noise fluctuation is within the historical controllable range. The generation of the noise compliance flag directly depends on the logical comparison of the R value and the threshold (3.86>3.0).

[0160] The wireless restart decision submodule calls the cycle compliance flag and the noise compliance flag. When both flags are true, the wireless transmitter power is restarted and a wireless channel restart instruction is generated.

[0161] The cycle compliance flag (True) and the noise compliance flag (True) are called. When both flags are True, a restart command is sent to the wireless transmitter component. After a delay of 500ms, the power is turned on again. The handshake signal response time is tested (for example, the first signal response time after the restart = 120ms ≤ the preset maximum value of 200ms). The wireless channel restart command [Status: Success, Restart Timestamp: 2023-08-20T10:00:05, Response Time: 120ms] is generated. The preset restart condition is: the cycle compliance flag validity period ≤ 15 minutes (for example, if the cycle flag is generated at 10:00:00, the restart must be completed before 10:15:00). In an abnormal scenario, when the noise compliance flag is False (R = 2.8 < 3.0), even if the cycle meets the standard, the restart is prohibited and the error code [0x05: Noise Not Converged] is recorded.

[0162] See also Figure 6 , the monitoring terminal data synchronization module includes:

[0163] The data source integration submodule calls the abnormal timestamp in the lead signal mutation trigger instruction, counts the number of interruptions in the channel transmission quality parameter set, extracts the switching records of the physical connection activation state, aligns the three types of data according to the time axis, removes redundant entries, and generates an integrated data set;

[0164] Call the abnormal timestamp sequence [2023-08-15T14:23:17.352, 2023-08-15T14:25:03.201] in the lead signal mutation trigger instruction, count the number of interruptions of the channel transmission quality parameter set within the corresponding time window (±30 seconds) [4 times, 2 times], extract the switching records of the physical connection activation state [time stamp: 14:23:17.352 (physical port COM3 activation), 14:25:03.201 (wireless restart)], align the three types of data according to the time axis (time difference threshold ±500ms), and remove redundant entries (such as 14:23:17.800 to 14:23:18.20 0) to generate an integrated data set [timestamp: 14:23:17.352, interruption number: 4, switching state: COM3 activated, timestamp: 14:25:03.201, interruption number: 2, switching state: wireless restart]. The time difference threshold is set based on the following: the ECG signal sampling rate of 500Hz corresponds to a sampling interval of 2ms. The redundancy judgment rule is that if the same lead repeatedly triggers an abnormality within 500ms, only the first record is retained. For example, the original data contains timestamps [14:23:17.352, 14:23:17.800, 14:23:18.100], and after elimination, [14:23:17.352] is retained.

[0165] The protocol field encapsulation submodule adds an abnormality mark field to the original ECG waveform data based on the integrated data set, matches the link status field according to the channel number, arranges the field order according to the data structure defined by the remote monitoring protocol, and generates an encapsulated data frame;

[0166] Based on the abnormal timestamp [14:23:17.352] in the integrated data set, an abnormal flag field [field name: Abnormal_Flag, length: 1 byte, value: 0x01 (abnormal)] is added to the original ECG waveform data (500 Hz sampling, 16-bit precision), and the link status field [field name: Channel_Status, length: 2 bytes, value: 0x0401 (channel 4 activated)] is matched according to channel number 4. The field order [frame header 0xA5 (1B), timestamp] is arranged according to the data structure defined in the remote monitoring protocol (RFC-ECG-2023). (8B), Abnormal_Flag (1B), Channel_Status (2B), ECG data (512B)], generate an encapsulated data frame [frame header 0xA5, timestamp 14:23:17.352 (Hex: 0x18F05E30), Abnormal_Flag 0x01, Channel_Status 0x0401, ECG data (0x120x34...)], instance field length verification: the ECG data segment must be strictly 512 bytes (if the original data is 510 bytes, it is padded with zeros 0x00 at the end).

[0167] The terminal synchronization generation submodule calls the encapsulated data frame, verifies the consistency of the field length with the protocol specification, truncates and fills the excess fields with zeros, adds data checksums and timestamp watermarks, and generates the terminal synchronization data stream;

[0168] Call the ECG data segment (512 bytes) in the encapsulated data frame, check the field length (frame header 1B + timestamp 8B + tag 1B + status 2B + data 512B = 524B), and add a data check code (CRC16: polynomial 0x8005, initial value 0xFFFF, calculate the 512-byte CRC value from the frame header to the end of the data segment) to the overrun field (for example, if the abnormal tag field is mistakenly entered as 2 bytes 0x0101, it is truncated to 1 byte 0x01). For example, if the input data is 0xA5...0x00 Generate CRC = 0x3D7A), embed a timestamp watermark (convert the timestamp 14:23:17.352 to the Unix timestamp 1692102197, Hex: 0x64E3F6F5, and insert it at the end of the data stream), generate a terminal synchronization data stream [frame header 0xA5...CRC0x3D7A, watermark 0x64E3F6F5], and perform truncation and zero padding operations as follows: When the ECG data is 515 bytes, truncate it to 512 bytes (discard the last 3 bytes) and pad it with zeros to 512 bytes.

[0169] A method for sharing data of a wireless electrocardiogram monitor based on the Internet of Things comprises the following steps:

[0170] S1: Collect the physiological signal sequence of the RR interval continuously output by the lead patch, evaluate whether the variance value of the current window sequence point and the differential slope mean of the adjacent sequence points exceed the threshold, and generate a lead signal mutation trigger instruction;

[0171] S2: Based on the trigger instruction of the lead signal mutation, the low-noise wireless channel is used to encapsulate data packets with different transmission rates, monitor the transmission delay and interruption times, and generate a channel transmission quality parameter set;

[0172] S3: extract the channel transmission quality parameter, concentrate on the wiring harness interface whose transmission interruption times exceed the interruption threshold, detect the impedance value of the wiring harness interface, trigger the physical port data flow switching, and generate the physical connection activation state;

[0173] S4: Based on the physical connection activation status, count the number of consecutive stable periods of the physical link, compare the stability threshold with the noise backtracking threshold, restart the wireless channel, and generate a wireless channel restart instruction;

[0174] S5: Integrate the lead signal mutation trigger instruction, channel transmission quality parameter set, and physical connection activation status, encapsulate the ECG waveform and link status fields, and generate terminal synchronization data stream.

[0175] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A wireless ECG monitor based on the Internet of Things, characterized in that: The electrocardiogram monitor comprises: The lead signal mutation detection module collects the physiological signal sequence of the RR interval continuously output by the lead patch, evaluates whether the variance value of the current window sequence point and the mean of the differential slope of the adjacent sequence points exceed the threshold, and generates a lead signal mutation trigger instruction; A wireless channel adaptive encapsulation module, based on the lead signal mutation trigger instruction, calls a low-noise wireless channel to encapsulate data packets of different transmission rates, monitors the transmission delay and the number of interruptions, and generates a channel transmission quality parameter set; A physical lead connection switching module extracts the channel transmission quality parameter, collects the transmission interruption times of the wiring harness interface exceeding the interruption threshold, detects the impedance value of the wiring harness interface, triggers the physical port data flow switching, and generates a physical connection activation state; A dual-state link backtracking detection module, based on the physical connection activation state, counts the number of consecutive stable periods of the physical link, compares the stability threshold with the noise backtracking threshold, restarts the wireless channel, and generates a wireless channel restart instruction; The monitoring terminal data synchronization module integrates the lead signal mutation trigger instruction, the channel transmission quality parameter set, the physical connection activation status, encapsulates the ECG waveform and link status field, and generates a terminal synchronization data stream.

2. The wireless ECG monitor based on the Internet of Things according to claim 1, characterized in that: The lead signal mutation trigger instruction includes an abnormal timestamp, a variance excess flag, and a slope excess flag; the channel transmission quality parameter set includes an available channel priority list, a data frame type identifier, a delay distribution statistic, and an interruption frequency statistic; the physical connection activation status includes an interface impedance qualification flag, a physical link enable instruction, and a wireless component shutdown status mark; the wireless channel restart instruction includes a transmission cycle compliance flag, a channel noise qualification flag, and a wireless component restart trigger mark; the terminal synchronization data stream includes a waveform encapsulation protocol version, an abnormal event coding table, a channel mapping relationship index, and a link status change time series.

3. The wireless ECG monitor based on the Internet of Things according to claim 1, characterized in that: The lead signal mutation detection module includes: The signal acquisition and feature extraction submodule collects the output signal of the lead patch physiological electrical signal sensing chip, identifies the R wave apex, extracts the continuous RR interval time series, intercepts the data segment according to the preset time window, and generates a time domain segmented sequence; The dynamic variance analysis submodule calls the time domain segmentation sequence and calculates the set of time differences between adjacent RR intervals in the window based on the formula: Obtain the RR interval fluctuation intensity within the window and generate the variance value; Among them, σ 2 Represents the variance of RR intervals within the window, t i represents the i-th time difference, μ represents the average time difference in the window, and n represents the total number of time differences in the window; The mutation determination submodule calls the variance value and calculates the mean set of the absolute values ​​of the adjacent time differences within the window using the formula: Obtain the differential change rate, generate the slope mean, compare the variance value with the preset mutation variance threshold, and compare the slope mean with the preset mutation slope threshold. When both indicators exceed the threshold, generate a lead signal mutation trigger instruction; Among them, K represents the mean of the difference slope, d j represents the absolute value of the jth time difference, Δt represents the sampling interval, and j is the time difference index value.

4. The wireless ECG monitor based on the Internet of Things according to claim 1, characterized in that: The wireless channel adaptive encapsulation module includes: The channel screening and optimization submodule, based on the lead signal mutation trigger instruction, calls the channel interference noise list of the wireless transmission component, arranges the channel numbers in ascending order by noise value, eliminates channels with noise values ​​higher than the dynamic interference threshold, and generates an available channel queue; The data framing adaptation submodule calls the available channel queue to divide the ECG waveform data into high-priority data frames and regular data frames according to a preset frame length, allocates redundant check bits to the high-priority frames based on the frame length difference, and generates a priority data frame set; The transmission quality assessment submodule calls the priority data frame set, monitors the confirmation signal delay returned by the receiving end, counts the number of interruptions caused by channel switching during the transmission of each frame, calculates the product of the delay mean and the interruption frequency, and generates a channel transmission quality parameter set.

5. The wireless ECG monitor based on the Internet of Things according to claim 1, characterized in that: The physical lead connection switching module includes: The interruption monitoring and impedance detection submodule monitors the number of transmission interruptions of the channel transmission quality parameter concentration. When the number exceeds a preset interruption threshold, the impedance value of the lead patch harness interface is detected to generate a real-time impedance value. The connection stability evaluation submodule calls the real-time impedance value based on the formula: Calculate the connection stability index, compare the result with the dynamic stability threshold, and generate a stability judgment mark; Among them, S represents the connection stability index, N represents the number of current transmission interruptions, and Z τ represents the preset connection threshold, Z represents the real-time impedance value, ΔZ represents the standard deviation of the three most recent impedance detection values, and α represents the harness aging correction factor; The physical port activation submodule calls the stability determination flag. When the determination flag is abnormal, the power supply of the wireless transmitting component is turned off, the data flow of the physical lead wire transmission port is activated, and the physical connection activation state is generated.

6. The wireless ECG monitor based on the Internet of Things according to claim 1, characterized in that: The dual-state link backtracking detection module includes: a transmission cycle monitoring submodule, which counts the number of continuous and uninterrupted transmission cycles of the physical lead wire based on the activation state of the physical connection, compares the number of cycles with a preset stability threshold, and generates a cycle compliance flag; The composite noise evaluation submodule calls the noise value sequence in the channel transmission quality parameter based on the formula: Calculate the noise retrospective assessment coefficient, compare the result with the retrospective threshold, and generate a noise compliance mark; Among them, Q represents the noise retrospective evaluation coefficient, N β Represents the preset backtracking threshold, n k represents the kth noise value, t k represents the time attenuation factor corresponding to the noise value, represents the noise mean, γ represents the noise fluctuation correction, and λ represents the time attenuation coefficient; The wireless restart decision submodule calls the period compliance flag and the noise compliance flag, and when both flags are true, restarts the power supply of the wireless transmitting component and generates a wireless channel restart instruction.

7. The wireless ECG monitor based on the Internet of Things according to claim 1, characterized in that: The monitoring terminal data synchronization module includes: The data source integration submodule calls the abnormal timestamp in the lead signal mutation trigger instruction, counts the number of interruptions of the channel transmission quality parameter set, extracts the switching records of the physical connection activation state, aligns the three types of data according to the time axis and eliminates redundant entries to generate an integrated data set; A protocol field encapsulation submodule, based on the integrated data set, adds an abnormality mark field to the original ECG waveform data, matches the link status field according to the channel number, arranges the field order according to the data structure defined by the remote monitoring protocol, and generates an encapsulated data frame; The terminal synchronization generation submodule calls the encapsulated data frame, checks the consistency of the field length with the protocol specification, truncates and fills the excess field with zeros, adds a data check code and a timestamp watermark, and generates a terminal synchronization data stream.

8. A wireless ECG monitor data sharing method based on the Internet of Things, characterized in that: The method is used for the wireless ECG monitor based on the Internet of Things according to any one of claims 1 to 7, The following steps are involved: S1: Collect the physiological signal sequence of the RR interval continuously output by the lead patch, evaluate whether the variance value of the current window sequence point and the differential slope mean of the adjacent sequence points exceed the threshold, and generate a lead signal mutation trigger instruction; S2: Based on the lead signal mutation trigger instruction, a low-noise wireless channel is called to encapsulate data packets with different transmission rates, the transmission delay and the number of interruptions are monitored, and a channel transmission quality parameter set is generated; S3: extracting the wiring harness interface whose transmission interruption times exceed the interruption threshold in the channel transmission quality parameter set, detecting the impedance value of the wiring harness interface, triggering the physical port data flow switching, and generating a physical connection activation state; S4: Based on the physical connection activation state, counting the number of consecutive stable periods of the physical link, comparing the stability threshold with the noise backtracking threshold, restarting the wireless channel, and generating a wireless channel restart instruction; S5: Integrate the lead signal mutation trigger instruction, channel transmission quality parameter set, and physical connection activation status, encapsulate the ECG waveform and link status field, and generate a terminal synchronization data stream.

Images