RFID phase data processing method for landslide monitoring

Abstract

The invention provides an RFID phase data processing method for landslide monitoring, and relates to the field of landslide monitoring, and the method comprises the steps: collecting the original phase information of a passive RFID tag in a monitoring point through a reader-writer, carrying out the phase unwrapping of the original phase information, and obtaining the real phase information, a data packet fusion compensation missing phase algorithm is adopted to compensate missing of real phase information; the compensated phase is processed through an improved Kalman filtering algorithm based on the phase change rate, and an optimal phase estimation value is obtained; and an RFID phase data fusion algorithm is used to carry out multi-label and multi-antenna data fusion, so that a phase with higher quality is obtained, and earth surface displacement monitoring of a landslide point is realized. The beneficial effects of the invention are that for the landslide monitoring-oriented RFID positioning scheme, through a series of data processing schemes, the interference of the outdoor environment on the positioning scheme is reduced, the positioning precision is improved, and the reliability of the landslide monitoring scheme is further enhanced.

Description

Technical Field

[0001] The present invention relates to the field of landslide monitoring, in particular to ultra-high frequency RFID positioning, and in particular to an RFID phase data processing method for landslide monitoring. Background Art

[0002] Landslide monitoring is to monitor and warn of landslides through various technical means to reduce the losses caused by landslide disasters. The displacement of landslides usually ranges from a few centimeters to a few meters per year, and is irregular in time and space. Sometimes landslides show catastrophic acceleration, so they need to be monitored so that early warnings can be given.

[0003] Currently, landslide monitoring is mainly limited by the high cost of existing solutions and environmental constraints, such as meteorology, rugged terrain or dense vegetation. Among the various remote sensing methods, interferometric synthetic aperture radar is usually used for surface monitoring of large landslides. However, whether it is through satellites or ground stations, there are defects such as poor real-time performance or high cost. Landslide monitoring methods based on GNSS or wireless sensor networks have high real-time performance, but they require power supplies, which greatly increases initial and maintenance costs. The optical-based lidar measurement solution is highly sensitive to the surrounding environment. The laser is easily scattered and reflected by rain, snow, fog, etc., and even absorbed, which affects the accuracy. Considering the harmfulness of landslides and the respective defects of the above methods, there is an urgent need for a low-cost, environmentally adaptable and easy-to-maintain method to better realize landslide monitoring. Summary of the invention

[0004] In order to solve the problems of high cost, weak environmental adaptability and difficult maintenance of the existing technology, the present invention provides an RFID phase data processing method for landslide monitoring, which mainly includes:

[0005] S1: Use the reader to collect the original phase information of the passive RFID tag in the monitoring point, use the path integral method to phase unwrap the original phase information to obtain the real phase information, and use the data group fusion compensation missing phase algorithm to compensate for the missing real phase information;

[0006] S2: The improved Kalman filter algorithm based on the phase change rate is used to process the compensated phase to obtain the optimal phase estimation value;

[0007] S3: Use RFID phase data fusion algorithm to fuse the best phase estimation value to the tag and multi-antenna data to obtain higher quality phase and realize surface displacement monitoring of landslide points.

[0008] A storage device stores instructions and data for implementing an RFID phase data processing method for landslide monitoring.

[0009] An RFID phase data processing device for landslide monitoring includes: a processor and a storage device; the processor loads and executes instructions and data in the storage device to implement an RFID phase data processing method for landslide monitoring.

[0010] The technical solution provided by the present invention has the following beneficial effects: the present invention performs a series of data processing on the phase information collected by the RFID positioning technology applied to landslide monitoring, uses a reader to collect the original phase information of the passive RFID tag in the monitoring point, uses the path integral method to perform phase unwrapping to obtain the real phase information, and proposes a missing phase compensation algorithm on this basis, which solves the phase missing problem and ensures the integrity of the unwrapped phase. The Kalman filter algorithm based on the phase change rate suppresses the noise interference generated by the outdoor environment and improves the smoothness of the phase curve. The phase data fusion algorithm is adopted to reduce the data gap of the phase data and improve the reliability of the phase data, so that the positioning accuracy of the RFID positioning solution for landslide monitoring is further improved, and the monitoring and early warning functions of the landslide monitoring solution are further improved, which has progressive significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0012] Figure 1 It is a flow chart of an RFID phase data processing method for landslide monitoring in an embodiment of the present invention.

[0013] Figure 2 Schematic diagram of landslide monitoring in an embodiment of the present invention.

[0014] Figure 3 is a schematic diagram of a multi-antenna multi-tag method in an embodiment of the present invention.

[0015] Figure 4 It is a schematic diagram of the operation of the hardware device in the embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0017] Example 1

[0018] Please refer to Figure 1 , Figure 1 : is a flow chart of an RFID phase data processing method for landslide monitoring in an embodiment of the present invention, which specifically includes:

[0019] S1: The original phase information of the passive RFID tag in the monitoring point is collected by the reader, and the path integral method is used to phase unwrap the original phase information to obtain the real phase information. The missing phase information is compensated by the data group fusion compensation missing phase algorithm; the data group fusion compensation missing phase algorithm solves the phase loss problem caused by insufficient phase signal data in the signal sampling process, thereby correcting the unwrapped phase after the data gap, restoring it to the real phase curve, and improving the positioning accuracy. The positioning system based on passive radio frequency identification technology (RFID) is easy to install, has low maintenance cost, can provide intensive measurement in time and space, has low sensitivity to obstacles, and has good weather robustness. Therefore, RFID positioning technology should have great development potential in the field of landslide monitoring. And RFID phase information has the advantages of fine granularity and less interference, so positioning technology based on RFID phase information has become a research hotspot in recent years.

[0020] S2: An improved Kalman filter algorithm based on phase change rate is used to process the compensated phase; the improved Kalman filter algorithm based on phase change rate improves the prediction equation in the Kalman filter algorithm based on the phase change rate, and uses the improved Kalman filter algorithm to filter the observed phase to suppress noise interference, thereby obtaining a smoother phase curve.

[0021] S3: Use the RFID phase data fusion algorithm to fuse the processed phase with the tag and multi-antenna data to obtain a higher quality phase and realize the surface displacement monitoring of the landslide point. The RFID phase data fusion algorithm first fuses the dual-antenna winding phase data of a single tag through a complex superposition and averaging operation, and further fuses the dual-antenna fused winding phase data of multiple tags at the same monitoring point through a complex superposition and averaging operation, so that the continuity and robustness of the obtained phase data are improved, thereby improving the positioning accuracy of the monitoring scheme.

[0022] like Figure 1 As shown, the present invention proposes an RFID phase data processing method for landslide monitoring. First, a missing phase compensation algorithm is proposed to solve the phase missing problem. A Kalman filtering algorithm based on phase change rate is proposed to suppress the noise interference generated by the outdoor environment. Then a phase data fusion algorithm is proposed to reduce data gaps and further improve the reliability of phase data.

[0023] Preferred embodiment: Figure 2As shown, the RFID passive tag is bound to the glass fiber rod, and the glass fiber rod is deeply inserted into the mountain soil as a monitoring point. Multiple RFID passive tags are set at each monitoring point. The RFID antenna is bound to a relatively stable location at the top of the landslide and connected to the reader. The phase information of the RFID passive tag collected by the reader is collected by the host computer. After a series of data processing, the tag is positioned, and finally the surface displacement monitoring of the landslide point is realized.

[0024] In the missing phase compensation algorithm, the first thing to mention is the prerequisite for phase unwrapping: the signal acquisition rate satisfies the Nyquist sampling theorem, that is, the acquisition frequency of the phase signal must be higher than twice the highest frequency of the signal. Only when this condition is met can the correctness and accuracy of the obtained unwrapped phase sequence be guaranteed. In the actual signal sampling environment, the phase measurement value of RFID will have a phase missing problem due to insufficient amount of collected phase signal data.

[0025] In the RFID positioning scenario of landslide monitoring, when a landslide occurs, the displacement characteristics of multiple tags often show consistency. Therefore, in the landslide monitoring scenario, for the phase loss problem, it is proposed to group the tag phase data and fuse them into a reference phase sequence to compensate for the missing phase. The approach is: the tags belonging to the same area are grouped together. When a tag undergoes rapid displacement and the amount of phase data is insufficient to disentangle, the phase data of the tag group in the area is merged to synthesize a reference phase sequence. Assume is the phase sequence of label t, by using The average rate of change To obtain the reference phase sequence fused by the tag group

[0026]

[0027] n t is the number of labels in this group, The reference phase sequence The time derivative is obtained by Integrate over time to obtain the reference phase sequence Assuming the path integral method is U, the phase unwrapping process is as follows:

[0028]

[0029] in is the unwrapped phase obtained by guiding the reference phase sequence, is the measured phase where the phase missing label appears.

[0030] In the Kalman filter algorithm section, the first thing to mention is that when positioning in an outdoor environment, the phase data collected by the reader contains a lot of noise, including the multipath effect in the environment and the random errors caused by the thermal noise of the reader's receiving circuit. Because the effective phase signal and the noise may overlap in the spectrum, it is difficult for traditional filtering methods such as low-pass filters to filter out random interference signals. Therefore, the prediction equation in the Kalman filter algorithm is improved based on the phase change rate. Specifically, the prediction equation predicts the phase value at the current moment based on the best estimate of the phase at the previous moment and the appropriate weights of the phase difference values ​​at the previous multiple moments. Therefore, the mathematical expression of the improved prediction equation in the improved Kalman filter algorithm is as follows:

[0031]

[0032] in, is the predicted value of the phase at the i-th moment, is the best estimated value of the phase at the previous n moments, k 1 ,k 2 ,k 3 ,...,k n-1 is the weight of the phase difference. The phase values ​​of the first three moments are selected to participate in preprocessing, and the specific design of the improved Kalman filter algorithm is as follows:

[0033] The process of initializing parameters and setting initial values ​​is:

[0034] (1) Set the first three phase estimates to Set the weight of the phase difference to k 1 , k 2 , initialize the state observation matrix H, process noise covariance matrix Q, and measurement noise covariance matrix R.

[0035] (2) Calculate the predicted value of the phase at the i-th moment

[0036] According to the best phase estimate at time i-1 The best phase estimate at time i-2 The best phase estimate at time i-3 And the corresponding weights, get the phase prediction value at the i-th moment

[0037]

[0038] Furthermore, the phase prediction value of the prior estimate of the covariance at the i-th moment is calculated

[0039]

[0040] Further, calculate the Kalman gain K at the current momenti :

[0041]

[0042] Furthermore, according to the phase prediction value at the i-th moment Phase Observation and the Kalman gain K i Calculate the best phase estimate of the phase at time i

[0043]

[0044] In order to reduce the prediction error of the optimal phase estimate, the phase estimate of the covariance prior estimate at the i-th moment is calculated as follows:

[0045]

[0046] in, Represents the prediction error. When the prediction error is within the error threshold, the best phase estimation value at this time is the final best phase estimation value.

[0047] At this point, the design of the improved Kalman filter algorithm based on the phase change rate has been completed, which can effectively filter out the noise error that interferes with the RFID phase information in the mountain environment.

[0048] In the phase data fusion algorithm part, the first thing that needs to be mentioned is that when using RFID positioning technology to monitor landslides, the tag phase is easily affected by various influencing factors. When there is a large gap in the phase data, the phase unwrapping of this phase data will fail, resulting in a phase missing problem, which will affect the final positioning accuracy.

[0049] like Figure 3 As shown, the phase data fusion method of multiple antennas and multiple tags can be used to obtain better winding phase, with one tag corresponding to two antennas. When two antennas with similar distances measure the phase of the same tag, their phase changes are usually very similar, so a complex superposition and averaging operation can be performed. Similarly, multiple tags with similar distances within a single monitoring point and on the same support usually have very similar phase changes, so a complex superposition and averaging operation can also be performed. The specific implementation is as follows.

[0050] Define a threshold value σ, antenna 1 at t 1 The best phase estimate measured at time is Antenna 2 at t 2 The best phase estimate measured at time is If the two moments satisfy the following relationship:

[0051] |t 1-t 2 |≤σ (9)

[0052] The two phase values ​​are fused by complex superposition and averaging:

[0053]

[0054] in, is the multi-antenna fusion phase, z 1 Indicates that antenna 1 is at t 1 The complex form of the best phase estimate measured at time, z 2 It indicates that antenna 2 is at t 2 The complex form of the best phase estimate measured at time t, and phase fusion is performed in complex form. If there is a time t in the phase data of one of the antennas 1 , while the phase data of the other antenna does not contain the time t that satisfies equation (9) 2 , then directly change t 1 The phase value at the time is added to the fused phase sequence.

[0055] After fusing the dual-antenna phase data of a single tag, the tag phase data from the same monitoring point is further subjected to complex superposition and averaging, and then the fused phase sequence of the monitoring point is phase unwrapped to obtain the tag radial displacement. This process fuses the available phase data of each tag to obtain a displacement index to monitor the radial displacement of the tag.

[0056] Example 2

[0057] An RFID phase data processing device 401 for landslide monitoring, such as Figure 4 As shown, it includes: a processor 402 and a storage device 403; the processor 402 loads and executes instructions and data in the storage device 403 to implement an RFID phase data processing method for landslide monitoring.

[0058] Example 3

[0059] A storage device stores instructions and data for implementing an RFID phase data processing method for landslide monitoring.

[0060] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A RFID phase data processing method for landslide monitoring, characterized in that: include: S1: Use the reader to collect the original phase information of the passive RFID tag in the monitoring point, use the path integral method to phase unwrap the original phase information to obtain the real phase information, and use the data group fusion compensation missing phase algorithm to compensate for the missing real phase information; S2: The improved Kalman filter algorithm based on the phase change rate is used to process the compensated phase to obtain the optimal phase estimation value; S3: Use RFID phase data fusion algorithm to perform multi-tag and multi-antenna data fusion on the optimal phase estimation value to obtain higher quality phase and realize surface displacement monitoring of landslide points.

2. The RFID phase data processing method for landslide monitoring according to claim 1, characterized in that: The specific steps of step S1 are: The tags belonging to the same area are grouped together. When a tag is rapidly displaced and the amount of phase data is insufficient to untangle, the phase data of the tag group in the area is merged to synthesize a reference phase sequence. is the phase sequence of label t, by using The average rate of change To obtain the reference phase sequence fused by the tag group Among them, n t is the number of labels in this group, The reference phase sequence The time derivative is obtained by Integrate over time to obtain the reference phase sequence Assuming the path integral method is U, the phase unwrapping process is as follows: in, is the unwrapped phase obtained by guiding the reference phase sequence, is the measured phase where the phase missing label appears.

3. The RFID phase data processing method for landslide monitoring according to claim 1, characterized in that: In step S2, the improved Kalman filter algorithm based on the phase change rate includes prediction equation improvement, initialization parameter and initial value setting, and Kalman gain calculation. The prediction equation predicts the phase value at the current moment based on the best estimated value of the phase at the previous moment and the appropriate weights of the phase difference values ​​at the previous moments: S2.1: The improved prediction equation is: in, is the predicted value of the phase at the i-th moment, is the best estimated phase value at the previous n moments, k1, k2, k3, ..., k n-1 is the weight of the phase difference; S2.2: The process of initializing parameters and setting initial values ​​is as follows: (1) Set the first three phase estimates to Set the weights of the phase difference values ​​to k1, k2, initialize the state observation matrix H, process noise covariance matrix Q, and measurement noise covariance matrix R; (2) Calculate the predicted value of the phase at the i-th moment According to the best phase estimate at time i-1 The best phase estimate at time i-2 The best phase estimate at time i-3 And the corresponding weights, get the phase prediction value at the i-th moment S2.3: According to the phase prediction value at the i-th moment Calculate the Kalman gain K at the current moment i : According to the phase prediction value at the i-th moment Phase Observation and the Kalman gain K i Get the best phase estimate at time i At this point, the improved Kalman filter algorithm based on phase change rate is designed to filter out noise errors that interfere with RFID phase information in mountain environments.

4. The RFID phase data processing method for landslide monitoring according to claim 3, characterized in that: The calculation formula of the phase prediction value of the prior estimate of the covariance at the i-th moment is as follows: Among them, A is the state transfer matrix from time i-1 to time i.

5. The RFID phase data processing method for landslide monitoring according to claim 1, characterized in that: In step S3, the RFID phase data fusion algorithm fuses the dual-antenna phase data of a single tag by complex superposition and averaging: Define the threshold σ, the best phase estimate of antenna 1 at time t1 is The best phase estimate for antenna 2 at time t2 is If the two moments satisfy the following relationship: |t1-t2|≤σ (8) Then the two phase values ​​are fused by complex superposition and averaging operation: in, is the multi-antenna fusion phase; z1 represents the complex form of the best phase estimate of antenna 1 at time t1, and z2 represents the complex form of the best phase estimate of antenna 2 at time t2; If there is a time t1 in the phase data of one of the antennas, but there is no time t2 satisfying formula (8) in the phase data of the other antenna, the phase value at time t1 is directly added to the fused phase sequence.

6. The RFID phase data processing method for landslide monitoring according to claim 5, characterized in that: In step S3, after the dual-antenna phase data of a single tag is fused, the tag phase data from the same monitoring point is further subjected to a complex superposition and averaging operation, and then the dual-antenna fusion data of multiple tags are fused, and then the fused phase sequence of the monitoring point is phase unwrapped to obtain the radial displacement of the tag; this process fuses the available phase data of each tag to obtain a displacement index in order to monitor the radial displacement of the tag.

7. A storage device, characterized in that: The storage device stores instructions and data for implementing the RFID phase data processing method for landslide monitoring as described in any one of claims 1 to 6.

8. An RFID phase data processing device for landslide monitoring, characterized in that: include: Processor and storage device; the processor loads and executes instructions and data in the storage device to implement the RFID phase data processing method for landslide monitoring as described in any one of claims 1 to 6.

Images