Ultra-lightweight passive wireless signal sensing system and method based on ultra-high frequency RFID technology

Abstract

The present invention discloses an ultra-light passive wireless signal sensing system and method based on ultra-high frequency RFID technology, the system includes: N ultra-light passive wireless signal sensing tags, 1 reader / writer, and 1 host computer; a multi-dimensional wireless signal feature acquisition technology is used to extract environmental wireless signal features, and the signal category in the current environment is identified by an ultra-light multi-dimensional composite signal recognition and sorting method, and the reader / writer and the host computer are used to build a wireless signal sensing system based on RFID technology. The present invention integrates an ultra-light passive wireless signal sensing circuit and a signal recognition and sorting method in a board-level electronic tag, which has the beneficial effects of low cost, low power consumption, and convenient maintenance. The system has a large sensing range, low cost, low power consumption, and easy deployment. The designed multi-dimensional wireless signal feature acquisition circuit and ultra-light multi-dimensional composite signal recognition and sorting method have the beneficial effects of wide recognition range, high recognition accuracy, low resource occupation, and ultra-light weight.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communication, and in particular relates to an ultra-light passive wireless signal sensing system and method based on ultra-high frequency RFID technology. Background Art

[0002] With the rapid development of wireless communication technology, the communication spectrum environment is becoming more and more complex and dense, and various types of wireless signals are also becoming increasingly complex. When conducting wireless communications, it is often necessary to identify the type and modulation method of the other party's wireless signal in order to demodulate and obtain information. Therefore, signal recognition and sorting and parameter estimation have received close attention in recent years. They have been widely used in the civilian field, such as electronic countermeasures, threat analysis, interference identification, spectrum management, etc., especially in the communication confrontation of electronic countermeasures, if you want to obtain the opponent's intelligence by intercepting the opponent's communication signal, wireless signal recognition and sorting technology is indispensable.

[0003] Usually, when the content and information of the modulation are uncertain, the modulation form of the communication signal is identified in a specific way, and some of the corresponding parameters can be solved to provide a basis for the demodulator to successfully select a suitable algorithm, thereby obtaining effective information. This is called modulation identification and signal sorting of wireless signals. To sort wireless signals, it is first necessary to obtain signal information. The common method of existing wireless signal sensing technology is down-conversion combined with orthogonal demodulation, which moves the wireless signal from the high frequency band to the baseband and extracts the amplitude and phase information. However, the down-conversion operation requires a high-frequency and high-precision local oscillator, which will cause high energy loss to the carrier system. For the extracted amplitude and phase information, signal features are generally extracted from them for judgment. There are two main types of commonly used wireless signal recognition and sorting methods. The first type is to obtain time-frequency information through time-frequency transformation to obtain signal features, and then combine classification algorithms to judge signal categories based on signal features. However, time-frequency transformation requires a lot of computing resources and storage space. After time-frequency transformation, the signal modulation method needs to be judged in combination with subsequent classification algorithms. The classification algorithm needs to process a large amount of time-frequency information and perform recognition and classification. The calculation is highly complex and occupies a lot of computing resources. The second type is a signal recognition and sorting method that combines artificial intelligence technology. The characteristics of the signal are autonomously learned through artificial intelligence algorithms such as neural networks for judgment and classification. Although this method can autonomously learn the characteristics of the signal and then perform signal recognition and sorting, it is unknown what signal features the neural network has extracted and how it has learned them. In addition, running the neural network algorithm requires high-performance equipment and high power consumption. If the collected signal samples are uploaded to the cloud server for signal recognition and sorting, although the performance requirements of the terminal equipment are reduced, it will bring huge communication overhead and huge communication pressure.

[0004] In the existing methods, the circuit complexity of down-conversion and orthogonal demodulation is high. The wireless signal identification and sorting equipment has the characteristics of high power consumption, high cost and high complexity. It basically operates in active mode, which puts certain requirements on fixed power supplies or portable batteries and increases the cost of regular maintenance. The existing wireless signal sensing equipment is relatively large and expensive, and it is impossible to achieve large-area wireless sensor network (WSN, Wireless Sensor Networks) layout. The existing methods are not applicable to regional wireless signal monitoring scenarios that require low cost, strong portability and wide coverage, such as invigilation scenarios, confidential units and other scenarios that require a large number of devices for full coverage. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides an ultra-lightweight passive wireless signal sensing system and method based on ultra-high frequency RFID technology, which can realize high-precision, large-range and highly intelligent monitoring of wireless signals.

[0006] The technical solution adopted by the present invention is: an ultra-light passive wireless signal sensing system based on ultra-high frequency RFID technology, comprising: N ultra-light passive wireless signal sensing tags, a reader / writer, and a host computer.

[0007] The ultra-light passive wireless signal sensing tag is used to respond to the command of the reader, monitor the existence of wireless signals in the current environment and their modulation mode, and transmit the collected environmental signal information back to the reader by backscattering or transmit it to the subsequent signal demodulation circuit by wire, and the ultra-light passive wireless signal sensing tag complies with ISO / IEC 18000-6 Type C standard. The reader is controlled by the host computer to send radio frequency signals to the tag group, and the reader runs the anti-collision algorithm of ISO / IEC 18000-6 Type C standard to realize the multi-tag anti-collision function. The host computer is used to control the working state of the reader and display the environmental wireless signal parameter information returned by the tag.

[0008] Furthermore, in the sensing system, the ultra-light passive wireless signal sensing tag includes: an antenna, a matching network circuit, a first power divider circuit, a second power divider circuit, a rectification and energy management circuit, a multi-dimensional wireless signal feature acquisition circuit, a baseband processing unit, a backscattering circuit, and a solar panel.

[0009] The antenna is connected to the input end of the matching network circuit and the output end of the backscattering circuit; the output end of the matching network circuit is connected to the input end of the first power divider circuit; the output end 1 of the first power divider circuit is connected to the input end of the second power divider circuit; the output end 2 of the first power divider circuit is connected to the input end 1 of the rectification and energy management circuit; the output end of the second power divider circuit is connected to the input end of the multi-dimensional wireless signal feature acquisition circuit; the output end of the multi-dimensional wireless signal feature acquisition circuit is connected to the signal input end of the baseband processing unit; the solar cell panel is connected to the input end 2 of the rectification and energy management circuit; the output end of the rectification and energy management circuit is connected to the power input end of the baseband processing unit; the output end of the baseband processing unit is connected to the input end of the backscattering circuit.

[0010] Among them, the antenna, matching network circuit, first power distributor circuit and second power distributor circuit are used to receive and distribute the radio frequency signal and radio frequency energy sent by the reader; the rectification and energy management circuit is used to convert the radio frequency energy into DC power available to the tag, and store the DC power of the radio frequency rectification and the DC power of the solar panel, and intermittently supply the tag for operation; the solar panel is used to convert solar energy into electrical energy for collection by the energy management circuit; the multi-dimensional wireless signal feature acquisition circuit is used to collect signal features of environmental wireless signals; the baseband processing unit is used to parse the reader command and control the tag to work according to the reader command, and execute the signal recognition and sorting method according to the collected environmental wireless signal features; the backscattering circuit sends a signal to the reader by backscattering under the control of the modulation signal; the sensing tag can be connected to the subsequent circuit through a wired interface.

[0011] Furthermore, in the ultra-lightweight passive wireless signal sensing tag, the multi-dimensional wireless signal feature acquisition circuit includes: a first envelope detection circuit, a frequency-variable phase change detection circuit, and a resonant frequency sweeping circuit, which are specifically as follows:

[0012] (1) The first envelope detection circuit adopts a Dickson detection structure;

[0013] (2) The frequency-variable phase-change detection circuit includes: an adjustable RC differential circuit, a second envelope detection circuit and a peak detection circuit; used for wireless signals with frequency and phase changes, including: angle-modulated waves, FSK, LFM, frequency hopping, PSK;

[0014] (3) The resonant frequency sweep circuit is used to extract the approximate spectral characteristics of the wireless signal.

[0015] The present invention also provides an ultra-lightweight passive wireless signal sensing method based on ultra-high frequency RFID technology, and the specific steps are as follows:

[0016] S1. The sensing system starts to work. The reader / writer is controlled by the host computer to send a radio frequency signal to the tag group. The radio frequency signal contains radio frequency energy and commands, charges the sensing tag wirelessly and activates it, and controls the activated tag to work according to the working mode set by the host computer, so that the tag senses the type and parameters of the wireless signal in the environment.

[0017] The wireless charging and activation of the sensing tag is specifically as follows:

[0018] An ultra-lightweight passive wireless signal sensing tag is placed in the environment to be tested. The antenna receives the RF signal in the environment and the solar panel receives the solar energy and converts it into DC power for its own operation and stores it. When the stored DC power is sufficient for the tag to work normally, the tag is powered. When there is a signal of a certain power in the environment, the tag wakes up and activates.

[0019] S2. Determine whether the signal in the current environment is a command sent by the reader / writer. If it is a command sent by the reader / writer or other tags, parse the command and execute it. If it is other wireless signals in the environment, use a multi-dimensional wireless signal feature acquisition circuit to extract signal features, and identify and sort the signal through an ultra-lightweight multi-dimensional composite signal recognition and sorting method.

[0020] S3. Based on the sorting result obtained in step S2, the data is transmitted back to the reader through the backscattering circuit and the antenna, and then the wireless signal sensing of the environment to be tested is continued until the DC power stored this time is exhausted, the tag is powered off and stops working, until the next time the RF signal and solar energy in the environment make the tag powered on and activated again, and the wireless signal of the environment to be tested is monitored.

[0021] Furthermore, in step S1, the process of the reader / writer being controlled by the host computer is specifically as follows:

[0022] A1. The reader initializes and takes inventory of the tags around it, records all the tag IDs and updates their timing timestamps, and records duplicate tags only once;

[0023] A2. Each tag enters the working state, and determines whether the reader receives the tag signal. If so, the tag signal is analyzed and the tag ID and data are recorded. If not, it goes to step A3;

[0024] A3, determine whether the host computer wants to send a command to the tag, if yes, control the reader to send the command through the antenna, if no, go to step A4;

[0025] A4. Determine whether there is any tag that has timed out and not been updated among the currently counted tags, that is, the countdown timestamp is 0. If so, page it according to its tag ID. Then, if there is a signal returned within T time, update the timestamp and record the tag ID and data. If no return data is received from the tag within T time, determine that the tag is faulty, and then repeat step A4 until there is no tag that has timed out and not been updated among the currently counted tags, then return to step A2.

[0026] Furthermore, the working process of the ultra-lightweight passive wireless signal sensing tag is as follows:

[0027] B1. After the tag is powered, the baseband processing unit is initialized and enters a dormant state. When there is a wireless signal of a certain strength in the environment, the baseband processing unit is awakened;

[0028] B2. The baseband processing unit determines whether the signal is sent by a known device according to the result of envelope detection. If so, the baseband processing unit performs command parsing and implements the command to enter step B4. If not, the baseband processing unit determines that the signal is other wireless signals in the current environment and enters step B3.

[0029] B3, extracting signal features from the environmental signal through a multi-dimensional wireless signal feature acquisition circuit, and executing an ultra-lightweight multi-dimensional composite signal recognition and sorting method according to the signal features to obtain a signal recognition and sorting result, and proceeding to step B4;

[0030] B4. Determine the signal strength in the current environment. If the signal strength in the environment is too low, the signal backscattered back to the reader is weak, and the information that needs to be sent will be stored, and the sleep mode will be ended. If the signal strength in the environment is strong, the amplitude of the current environmental signal will be modulated by backscattering and sent to the reader to complete the signal transmission.

[0031] Furthermore, in the multi-dimensional wireless signal feature acquisition circuit, the frequency-variable phase-change detection circuit and the resonant frequency sweeping circuit have the following specific working processes:

[0032] (1) Frequency-variable phase-change detection circuit;

[0033] The input signal is differentiated by an adjustable RC differential circuit. The RC differential circuit selects appropriate resistance and capacitance values ​​to convert the frequency change in the wireless signal into amplitude change. The expression is as follows:

[0034] V in (t) = Acos(2πf c t+φ(t)) (1)

[0035]

[0036] Z total =R 1||Z after (3)

[0037] Among them, V in (t) represents the input time domain signal at time t, A represents the amplitude of the input time domain signal, φ(t) represents the phase-frequency change of the signal, and f c Indicates the carrier frequency, V out (t) represents the output time domain signal amplitude, R 1 Represents the resistance value of the RC differential circuit, Z after Represents the total impedance of the subsequent circuit, Z total It represents the parallel value of the impedance of the RC differential circuit and the impedance of the subsequent circuit, and C represents the capacitance value of the RC differential circuit.

[0038] After differential processing, the signal flows to the second envelope detection circuit or the peak detection circuit. The baseband processing unit switches the frequency-varying detection mode and the phase-varying detection mode by changing the capacitance value of the adjustable capacitor of the frequency-varying phase-varying detection circuit, the peak detection circuit control I / O, and the envelope detection control I / O state, as follows:

[0039] When the peak detection circuit controls the I / O to be set to high voltage, the envelope detection circuit controls the I / O port to be set to high impedance, and the adjustable capacitor value is adjusted so that its RC time constant satisfies equation (4), it works in the frequency variation detection mode, and the signal flows to the envelope detection circuit, and the frequency variation information is extracted through the envelope detection circuit and the ADC.

[0040] When working in frequency-varying detection mode, the time constant of RC is less than the inverse of the modulation signal frequency and greater than the inverse of the carrier signal frequency. The expression is as follows:

[0041]

[0042] Among them, f mod represents the modulation signal frequency, C 1 It represents the capacitance value of C during frequency variation detection.

[0043] The peak value is extracted from the discontinuous point of the phase mutation of the carrier through differential operation, and then the peak value is detected by the passive peak detection circuit. The peak detection part uses the baseband processing unit control pin to initially set the capacitor voltage to 0V, that is, the peak detection circuit controls the I / O to be pre-set to 0V to pull the capacitor voltage to 0V, and then the control pin is set to high impedance, and the envelope detection circuit controls the I / O port to be set to high voltage. The adjustable capacitor value is adjusted so that its RC time constant satisfies formula (5). The unidirectional conduction characteristic of the diode is used to save the peak voltage in the capacitor, and then the voltage amplitude is collected by the ADC, working in the phase change detection mode, and the signal flows to the peak detection circuit.

[0044] When working in the phase change detection mode, the resistance and capacitance values ​​of the phase change detection circuit satisfy equation (5), and the time constant of RC is less than the inverse of the carrier signal frequency, as shown in the following expression:

[0045]

[0046] Among them, C 2 Represents the capacitance value of C during phase change detection.

[0047] (2) Resonant frequency sweep circuit;

[0048] By using the strong frequency selection performance of the high-Q LC resonant circuit, the input signal spectrum is swept, and the expression is as follows:

[0049]

[0050] Z sum (f) = Z LC (f)||Z after (7)

[0051] V out (f) = V in (f)Z sum (f) (8)

[0052] Among them, the capacitor C var represents the adjustable capacitor, L represents the inductance of the LC resonant network, Z LC (f) represents the impedance value of the LC resonant network at each frequency point, Z L represents the inductive impedance of the LC resonant network, Z C Represents the capacitive impedance of the LC resonant network, Z sum (f) represents the total impedance value of the LC resonant network and the subsequent circuit at each frequency point, f represents the input signal frequency, Z after Represents the total impedance of the subsequent circuit, V in (f) represents the amplitude of each frequency point of the input resonant sweep frequency circuit signal, V out (f) represents the amplitude of each frequency point of the output signal of the resonant sweep frequency circuit.

[0053] Among them, the resonant frequency sweeping circuit changes the capacitance value by adjusting the voltage at the control input end of the capacitor, and then changes the resonant frequency point of the LC resonant network by continuously adjusting the capacitance value of the adjustable capacitor, and obtains the average voltage of each frequency point through the subsequent rectifier circuit, and finally obtains the frequency domain information of the wireless signal.

[0054] Furthermore, in step B3, the ultra-lightweight multi-dimensional composite signal identification and sorting method is specifically as follows:

[0055] C1. Determine whether the tag has deployed a trained signal recognition and sorting model. If not, proceed to step C2. If yes, proceed to step C3.

[0056] Among them, the model is trained by a signal source, an ultra-lightweight passive wireless signal sensing tag and a host computer. The host computer controls the signal source to generate various modulated signals. The ultra-lightweight passive wireless signal sensing tag extracts signal features through a multi-dimensional wireless signal feature acquisition circuit and transmits them back to the host computer through a wired interface. The host computer performs model training based on the returned data.

[0057] C2, the ultra-lightweight multi-dimensional composite signal recognition and sorting method is initialized, and the model training is performed. After the training, the model accuracy is judged. When the model accuracy obtained by training meets the requirements, the model is deployed on the baseband processing unit of the ultra-lightweight passive wireless signal sensing tag. If the model accuracy does not meet the standard, the data is collected again for training until it meets the standard. After the model accuracy meets the standard, it returns to step C1 for re-judgment;

[0058] Among them, the signal features collected by the multi-dimensional wireless signal feature acquisition circuit include: the envelope time domain signal collected by the envelope detection circuit, the envelope after signal differentiation collected by the frequency-variable phase change detection circuit and the amplitude peak of the phase change position, and the spectrum information collected by the resonant sweep frequency circuit.

[0059] C3. The tags deployed with the model will be placed in the environment to be tested to sense the wireless signals in the environment to be tested. The tags use the multi-dimensional wireless signal feature acquisition circuit to collect data, extract feature parameters as input, and send them to the trained signal recognition and sorting model. The model first uses the decision tree method for pre-classification, and classifies the signals into no signal, constant envelope, amplitude modulation, amplitude keying, orthogonal amplitude modulation, and orthogonal frequency division multiplexing through the decision tree;

[0060] Among them, the extracted characteristic parameters include: the mean value, discrete coefficient, difference mean value and difference discrete coefficient of the envelope detection time domain signal; the envelope discrete coefficient, difference discrete coefficient, amplitude peak value and envelope mean ratio of the signal acquired by the frequency-variable phase change detection circuit after differentiation; and the 10 characteristic parameters of the signal acquired by the resonant sweeping circuit, including the number of effective frequency points, effective spectrum width and spectrum flatness within the effective spectrum width.

[0061] Furthermore, in step C3, the constant envelope signal is subdivided into two sub-classifications using a support vector machine (SVM) method, as follows:

[0062] The envelope discrete coefficient, difference discrete coefficient, amplitude peak value and envelope mean ratio of the phase change position after differentiation of the signal collected by the frequency-variable phase change detection circuit; the six characteristic parameters of the effective frequency points, effective spectrum width and spectrum flatness within the effective spectrum width of the signal collected by the resonant sweeping circuit are used to sort the five types of signals. Firstly, the input characteristic parameters are dimensionally upgraded by the Gaussian radial basis function kernel function to increase the discrimination, and a five-pair-remainder SVM model is constructed. The six characteristic parameters of the signal that needs to be classified are extracted and calculated using the Gaussian radial basis function kernel function and the five-pair-remainder SVM model to obtain the probability values ​​of the five classification results. The one with the largest comprehensive probability value is the final signal recognition and sorting result.

[0063] The constant envelope signals include: angle modulation wave, linear frequency modulation, frequency hopping, frequency shift keying and phase shift keying.

[0064] Beneficial effects of the present invention: The system of the present invention comprises: N ultra-light passive wireless signal sensing tags, 1 reader / writer, and 1 host computer; the multi-dimensional wireless signal feature acquisition technology is used to extract the environmental wireless signal features, the signal category in the current environment is identified by the ultra-light multi-dimensional composite signal identification and sorting method, and the reader / writer and the host computer are used to build a wireless signal sensing system based on RFID technology. The present invention integrates an ultra-light passive wireless signal sensing circuit and a signal identification and sorting method in a board-level electronic tag, which has the beneficial effects of low cost, low power consumption, and convenient maintenance. The system has the beneficial effects of large sensing range, low cost, low power consumption, and easy deployment. The designed multi-dimensional wireless signal feature acquisition circuit and ultra-light multi-dimensional composite signal identification and sorting method have the beneficial effects of wide recognition range, high recognition accuracy, low resource occupation, and ultra-lightweight. The software and hardware of the system of the present invention have the beneficial effect of large secondary development space, can be applied in many fields, can identify many signal types and modulation modes and extract wireless signal parameters, and are particularly suitable for scenes such as invigilation scenes and confidential units that require a large number of devices for all-round coverage, and have broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a structural diagram of an ultra-light passive wireless signal sensing system based on ultra-high frequency RFID technology of the present invention.

[0066] Figure 2 Schematic diagram of the structure of the ultra-light passive wireless signal sensing tag described in an embodiment of the present invention.

[0067] Figure 3 4 is a structural diagram of a multi-dimensional wireless signal feature acquisition circuit according to an embodiment of the present invention.

[0068] Figure 4 This is a flow chart of the reader / writer being controlled by a host computer in an embodiment of the present invention.

[0069] Figure 5 Schematic diagram of the host computer interface of the ultra-lightweight passive wireless signal sensing system in an embodiment of the present invention.

[0070] Figure 6 This is a working flow diagram of the ultra-lightweight passive wireless signal sensing tag described in an embodiment of the present invention.

[0071] Figure 7 This is a flow chart of the ultra-lightweight multi-dimensional composite signal identification and sorting method in an embodiment of the present invention.

[0072] Figure 8 This is a pre-classification architecture diagram of the ultra-lightweight multi-dimensional composite signal recognition and sorting method in an embodiment of the present invention. DETAILED DESCRIPTION

[0073] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0074] like Figure 1 As shown, an ultra-light passive wireless signal sensing system based on ultra-high frequency RFID technology of the present invention includes: N ultra-light passive wireless signal sensing tags, 1 reader / writer, and 1 host computer.

[0075] The ultra-light passive wireless signal sensing tag is placed in the environment to be tested, and is used to respond to the command of the reader, monitor the existence of wireless signals in the current environment and their modulation methods, and transmit the collected environmental signal information back to the reader by backscattering or transmit it to the subsequent signal demodulation circuit by wire, so that the subsequent circuit can select a suitable method to demodulate the signal. In order to solve the collision problem, the ultra-light passive wireless signal sensing tag follows the ISO / IEC18000-6Type C standard, improves the overall efficiency of the system, and realizes the multi-tag anti-collision function. The reader is controlled by the host computer to send radio frequency signals to the tag group; and the reader runs the anti-collision algorithm of the ISO / IEC 18000-6Type C standard to realize the multi-tag anti-collision function. The host computer is used to control the working status of the reader and display the environmental wireless signal parameter information returned by the tag.

[0076] like Figure 2 As shown, in this embodiment, in the sensing system, the ultra-light passive wireless signal sensing tag includes: an antenna, a matching network circuit, a first power divider circuit, a second power divider circuit, a rectification and energy management circuit, a multi-dimensional wireless signal feature acquisition circuit, a baseband processing unit, a backscattering circuit, and a solar panel.

[0077] The antenna is connected to the input end of the matching network circuit and the output end of the backscattering circuit; the output end of the matching network circuit is connected to the input end of the first power divider circuit; the output end 1 of the first power divider circuit is connected to the input end of the second power divider circuit; the output end 2 of the first power divider circuit is connected to the input end 1 of the rectification and energy management circuit; the output end of the second power divider circuit is connected to the input end of the multi-dimensional wireless signal feature acquisition circuit; the output end of the multi-dimensional wireless signal feature acquisition circuit is connected to the signal input end of the baseband processing unit; the solar cell panel is connected to the input end 2 of the rectification and energy management circuit; the output end of the rectification and energy management circuit is connected to the power input end of the baseband processing unit; the output end of the baseband processing unit is connected to the input end of the backscattering circuit.

[0078] Among them, the antenna, matching network circuit, first power distributor circuit and second power distributor circuit are used to receive and distribute the radio frequency signal and radio frequency energy sent by the reader; the rectification and energy management circuit is used to convert the radio frequency energy into DC power available to the tag, and store the DC power of the radio frequency rectification and the DC power of the solar panel, and intermittently supply the tag for operation; the solar panel is used to convert solar energy into electrical energy for collection by the energy management circuit; the multi-dimensional wireless signal feature acquisition circuit is used to collect signal features of environmental wireless signals; the baseband processing unit is used to parse the reader command and control the tag to work according to the reader command, and execute the signal recognition and sorting method according to the collected environmental wireless signal features; the backscattering circuit sends a signal to the reader by backscattering under the control of the modulation signal; the sensing tag can be connected to the subsequent circuit through a wired interface.

[0079] like Figure 2 , Figure 3 As shown, in this embodiment, in the ultra-light passive wireless signal sensing tag, the multi-dimensional wireless signal feature acquisition circuit includes: a first envelope detection circuit, a frequency-variable phase-change detection circuit, and a resonant frequency sweeping circuit.

[0080] (1) The first envelope detection circuit adopts Dickson detection structure, such as Figure 3 As shown in (a), the charge pump structure can utilize the energy of the positive and negative half cycles and can generate a DC voltage higher than the input voltage compared to the conventional single diode detection circuit;

[0081] (2) The frequency-variable phase-change detection circuit includes: an adjustable RC differential circuit, a second envelope detection circuit (the specific content is the same as the first envelope detection circuit) and a peak detection circuit; used for wireless signals with frequency and phase changes, including:

[0082] Angle modulation wave, FSK, LFM, frequency hopping, PSK;

[0083] Figure 3The frequency-variable phase-change detection circuit shown in (b) switches between the frequency-variable detection mode and the phase-change detection mode by changing the capacitance value of the adjustable capacitor and controlling the state of the I / O port.

[0084] (3) The resonant frequency sweep circuit is used to extract the approximate spectral characteristics of the wireless signal, such as Figure 3 As shown in (c), the input signal spectrum is swept by utilizing the strong frequency selection performance of the high-Q LC resonant circuit.

[0085] This embodiment also provides an ultra-lightweight passive wireless signal sensing method based on ultra-high frequency RFID technology, and the specific steps are as follows:

[0086] S1. The sensing system starts to work. The reader is controlled by the host computer to send a radio frequency signal to the tag group. The radio frequency signal contains radio frequency energy and commands (the radio frequency signal is divided into two by the antenna, matching network circuit and power divider circuit). The sensing tag is charged and activated in a wireless manner, and the activated tag is controlled to work according to the working mode set by the host computer, so that the tag senses the type and parameters of the wireless signal in the environment.

[0087] The wireless charging and activation of the sensing tag is specifically as follows:

[0088] The ultra-lightweight passive wireless signal sensing tag is placed in the environment to be tested. The antenna receives the RF signal in the environment and the solar panel receives the solar energy and converts it into DC power for its own operation and stores it. When the stored DC power is sufficient for the tag to work normally, the tag is powered. When there is a signal of a certain power in the environment, the tag wakes up and activates. That is, a part of the RF signal enters the rectification and energy management circuit in the form of RF energy (radio frequency energy) to convert the RF energy into DC power available to the tag, and the DC power converted by the RF and the DC power provided by the solar panel are stored. When the stored DC power reaches a certain threshold, the rectification and energy management circuit supplies power to the entire tag circuit. After receiving the DC power provided by the rectification and energy management circuit, the ultra-lightweight passive wireless signal sensing tag starts to operate and can receive command signals from the reader.

[0089] S2. Determine whether the signal in the current environment is a command sent by the reader / writer. If it is a command sent by the reader / writer or other tags, parse the command and execute it. If it is other wireless signals in the environment, use a multi-dimensional wireless signal feature acquisition circuit to extract signal features, and identify and sort the signal through an ultra-lightweight multi-dimensional composite signal recognition and sorting method.

[0090] When the reader sends a RFID general command to the tag, the command signal is demodulated in the form of an RF signal by the first envelope detection circuit that is powered on via the antenna, matching network circuit and power divider circuit and converted into a baseband signal and sent to the baseband processing unit.

[0091] S3. Based on the sorting result obtained in step S2, the data is transmitted back to the reader through the backscattering circuit and the antenna, and then the wireless signal sensing of the environment to be tested is continued until the DC power stored this time is exhausted, the tag is powered off and stops working, until the next time the RF signal and solar energy in the environment make the tag powered on and activated again, and the wireless signal of the environment to be tested is monitored.

[0092] After analysis, the baseband processing unit sends the return signal (baseband signal) to the backscattering circuit, and sends the ASK type general command return signal to the reader by backscattering under the control of the modulation signal. The tag will run in the powered-on state until the DC power stored in the rectification and energy management circuit is exhausted, and then it will be powered off and stop working until the RF signal energy and solar energy in the environment power on and activate the tag again, thus realizing intermittent sensing of the wireless signal in the environment to be tested.

[0093] like Figure 4 As shown, in this embodiment, in step S1, the process of the reader / writer being controlled by the host computer is as follows:

[0094] A1. The reader initializes and takes inventory of the tags around it, records all the tag IDs and updates their timing timestamps, and records duplicate tags only once (this timestamp is used to determine whether the tag has not sent back data for a long time);

[0095] A2. Each tag enters the working state, and determines whether the reader receives the tag signal. If so, the tag signal is analyzed and the tag ID and data are recorded. If not, it goes to step A3;

[0096] A3, determine whether the host computer wants to send a command to the tag, if yes, control the reader to send the command through the antenna, if no, go to step A4;

[0097] A4. Determine whether there is any tag that has timed out and not been updated among the currently counted tags, that is, the countdown timestamp is 0. If so, page it according to its tag ID. Then, if there is a signal returned within T time, update the timestamp and record the tag ID and data. If no return data is received from the tag within T time, determine that the tag is faulty, and then repeat step A4 until there is no tag that has timed out and not been updated among the currently counted tags, then return to step A2.

[0098] The host computer uses the interface to display the data and status returned by the tag and send commands to the tag, such as Figure 5 shown.

[0099] like Figure 6 As shown, in this embodiment, the working process of the ultra-lightweight passive wireless signal sensing tag is as follows:

[0100] B1. After the tag is powered, the baseband processing unit is initialized and enters a dormant state. When there is a wireless signal of a certain strength in the environment, the baseband processing unit is awakened;

[0101] B2. The baseband processing unit determines whether the signal is sent by a known device (reader or other tag) according to the result of envelope detection. If so, the baseband processing unit performs command parsing and implements the command to enter step B4. If not, the baseband processing unit determines that the signal is other wireless signals in the current environment and enters step B3.

[0102] B3, extracting signal features from the environmental signal through a multi-dimensional wireless signal feature acquisition circuit, and executing an ultra-lightweight multi-dimensional composite signal recognition and sorting method according to the signal features to obtain a signal recognition and sorting result, and proceeding to step B4;

[0103] B4. (The baseband processing unit has prepared the signal sorting results or the data that needs to be sent back to the reader / writer) Determine the signal strength in the current environment. If the signal strength in the environment is too low, the signal backscattered back to the reader / writer is weak, and the information that needs to be sent will be stored, and the sleep mode will be ended. If the signal strength in the environment is strong, the amplitude of the current environmental signal will be modulated by backscattering and sent to the reader / writer to complete the signal return.

[0104] Because the backscattered signal has been spread spectrum encoded and has a low transmission rate, it can achieve anti-interference effect. The signal received by the reader can obtain the required data after despreading and decoding. At the same time, the tag can also be connected to the subsequent circuit through a wired interface to transmit the type of current environmental signal, which is beneficial for the subsequent signal processing of the subsequent circuit.

[0105] like Figure 3 As shown, in this embodiment, in the multi-dimensional wireless signal feature acquisition circuit, the frequency-variable phase-change detection circuit and the resonant frequency sweeping circuit have the following specific working processes:

[0106] (1) Frequency-variable phase-change detection circuit;

[0107] The input signal is differentiated by an adjustable RC differential circuit. The RC differential circuit selects appropriate resistance and capacitance values ​​to convert the frequency change in the wireless signal into amplitude change. The expression is as follows:

[0108] V in (t) = Acos(2πf c t+φ(t)) (1)

[0109]

[0110] Z total =R 1 ||Z after (3)

[0111] Among them, V in (t) represents the input time domain signal at time t, A represents the amplitude of the input time domain signal, φ(t) represents the phase-frequency change of the signal, and f c Indicates the carrier frequency, V out (t) represents the output time domain signal amplitude, R 1 Represents the resistance value of the RC differential circuit, Z after Represents the total impedance of the subsequent circuit, Z total It represents the parallel value of the impedance of the RC differential circuit and the impedance of the subsequent circuit, and C represents the capacitance value of the RC differential circuit.

[0112] After differential processing, the signal flows to the second envelope detection circuit or the peak detection circuit. The baseband processing unit switches the frequency-varying detection mode and the phase-varying detection mode by changing the capacitance value of the adjustable capacitor of the frequency-varying phase-varying detection circuit, the peak detection circuit control I / O, and the envelope detection control I / O state, as follows:

[0113] When the peak detection circuit controls the I / O to be set to high voltage, the envelope detection circuit controls the I / O port to be set to high impedance, and the adjustable capacitor value is adjusted so that its RC time constant satisfies equation (4), it works in the frequency variation detection mode, and the signal flows to the envelope detection circuit, and the frequency variation information is extracted through the envelope detection circuit and ADC (analog-to-digital converter).

[0114] When working in frequency-varying detection mode, the time constant of RC is less than the inverse of the modulation signal frequency and greater than the inverse of the carrier signal frequency. The expression is as follows:

[0115]

[0116] Among them, f mod represents the modulation signal frequency, C 1 It represents the capacitance value of C during frequency variation detection.

[0117] The peak value is extracted from the discontinuous point of the phase mutation of the carrier through differential operation, and then the peak value is detected by the passive peak detection circuit. The peak detection part uses the baseband processing unit control pin to initially set the capacitor voltage to 0V, that is, the peak detection circuit controls the I / O to be pre-set to 0V to pull the capacitor voltage to 0V, and then the control pin is set to high impedance, and the envelope detection circuit controls the I / O port to be set to high voltage. The adjustable capacitor value is adjusted so that its RC time constant satisfies formula (5). The unidirectional conduction characteristic of the diode is used to save the peak voltage in the capacitor, and then the voltage amplitude is collected by the ADC, working in the phase change detection mode, and the signal flows to the peak detection circuit.

[0118] When working in the phase change detection mode, because the resistance and capacitance values ​​of the phase change detection circuit satisfy equation (5), the time constant of RC is less than the inverse of the carrier signal frequency, and the expression is as follows:

[0119]

[0120] Among them, C 2 Represents the capacitance value of C during phase change detection.

[0121] In summary, the frequency-variable phase change detection circuit can reflect its phase jump and frequency change information through its peak value and the time domain voltage amplitude detected by the second envelope detection circuit.

[0122] (2) Resonant frequency sweep circuit;

[0123] By using the strong frequency selection performance of the high-Q LC resonant circuit, the input signal spectrum is swept, and the expression is as follows:

[0124]

[0125] Z sum (f) = Z LC (f)||Z after (7)

[0126] V out (f) = V in (f)Z sum (f) (8)

[0127] Among them, the capacitor C var represents the adjustable capacitor, L represents the inductance of the LC resonant network, Z LC (f) represents the impedance value of the LC resonant network at each frequency point, Z L represents the inductive impedance of the LC resonant network, Z C Represents the capacitive impedance of the LC resonant network, Z sum (f) represents the total impedance value of the LC resonant network and the subsequent circuit at each frequency point, f represents the input signal frequency, Z afterRepresents the total impedance of the subsequent circuit, V in (f) represents the amplitude of each frequency point of the input resonant sweep frequency circuit signal, V out (f) represents the amplitude of each frequency point of the output signal of the resonant sweep frequency circuit.

[0128] Among them, the resonant frequency sweeping circuit changes the capacitance value by adjusting the voltage at the control input end of the capacitor, and then changes the resonant frequency point of the LC resonant network by continuously adjusting the capacitance value of the adjustable capacitor, and obtains the average voltage of each frequency point through the subsequent rectifier circuit, and finally obtains the frequency domain information of the wireless signal.

[0129] like Figure 7 As shown, in this embodiment, in step B3, in order to take into account the accuracy of signal recognition and sorting and the size of resource occupation and power consumption, the ultra-lightweight multi-dimensional composite signal recognition and sorting method is proposed, which is specifically as follows:

[0130] C1. Determine whether the tag has deployed a trained signal recognition and sorting model. If not, proceed to step C2. If yes, proceed to step C3.

[0131] Among them, the model is trained by a signal source, an ultra-lightweight passive wireless signal sensing tag and a host computer. The host computer controls the signal source to generate various modulated signals. The ultra-lightweight passive wireless signal sensing tag extracts signal features through a multi-dimensional wireless signal feature acquisition circuit and transmits them back to the host computer through a wired interface. The host computer performs model training based on the returned data.

[0132] C2, the ultra-lightweight multi-dimensional composite signal recognition and sorting method is initialized, and the model training is performed. After the training, the model accuracy is judged. When the model accuracy obtained by training meets the requirements, the model is deployed on the baseband processing unit of the ultra-lightweight passive wireless signal sensing tag. If the model accuracy does not meet the standard, the data is collected again for training until it meets the standard. After the model accuracy meets the standard, it returns to step C1 for re-judgment;

[0133] Among them, the signal features collected by the multi-dimensional wireless signal feature acquisition circuit include: the envelope time domain signal collected by the envelope detection circuit, the envelope after signal differentiation collected by the frequency-variable phase change detection circuit and the amplitude peak of the phase change position, and the spectrum information collected by the resonant sweep frequency circuit.

[0134] C3. The tags deployed with the model will be placed in the environment to be tested to sense the wireless signals in the environment to be tested. The tags use the multi-dimensional wireless signal feature acquisition circuit to collect data, extract feature parameters as input, and send them to the trained signal recognition and sorting model. The model first uses the decision tree method for pre-classification, and classifies the signals into no signal, constant envelope, amplitude modulation, amplitude keying, orthogonal amplitude modulation, and orthogonal frequency division multiplexing through the decision tree;

[0135] Among them, the extracted characteristic parameters include: the mean value, discrete coefficient, difference mean value and difference discrete coefficient of the envelope detection time domain signal; the envelope discrete coefficient, difference discrete coefficient, amplitude peak value and envelope mean ratio of the signal acquired by the frequency-variable phase change detection circuit after differentiation; and the 10 characteristic parameters of the signal acquired by the resonant sweeping circuit, including the number of effective frequency points, effective spectrum width and spectrum flatness within the effective spectrum width.

[0136] The coefficient of dispersion is the ratio of the standard deviation of the observed values ​​to the mean, which is a statistic that describes the degree of data dispersion; the number of effective frequency points refers to the sum of frequency points whose power values ​​are greater than 50% of the maximum power value; the effective spectrum width refers to the width of the minimum spectrum that can include these frequency points; frequency flatness (Wiener entropy) is an indicator used to quantify the shape or uniformity of the signal spectrum, which is the ratio of the geometric mean and the arithmetic mean of the power within the effective spectrum.

[0137] The ultra-lightweight multi-dimensional composite signal identification and sorting method takes the characteristic parameters collected by the multi-dimensional wireless signal characteristic acquisition circuit as input, and first uses the decision tree method for pre-classification. The decision tree method occupies low resources and is suitable for classification scenarios with multiple input characteristic parameters and multiple classification results. The decision tree can be used to classify the signal into six categories: no signal, constant envelope, amplitude modulation, amplitude keying, orthogonal amplitude modulation and orthogonal frequency division multiplexing.

[0138] In the present embodiment, in step C3, the constant envelope class signal is secondary sub-classified by the support vector machine SVM method (this classification method occupies relatively high resources but has high accuracy, and can effectively distinguish the five difficult-to-distinguish signals of angle modulation wave, linear frequency modulation, frequency hopping, frequency shift keying and phase shift keying in the constant envelope class), as follows:

[0139] The ultra-lightweight multi-dimensional composite signal recognition and sorting method is first pre-classified based on a decision tree. The method first combines the constant envelope modulation methods such as angle modulation, linear frequency modulation, frequency hopping, frequency shift keying and phase shift keying into one large category. First, the decision tree method is used to identify and sort the six categories of no signal, constant envelope, amplitude modulation, amplitude keying, orthogonal amplitude modulation and orthogonal frequency division multiplexing through the above ten characteristic parameters. The decision tree architecture diagram is shown in FIG. Figure 8 As shown, this classification method takes up less resources and is suitable for pre-classification of samples with multiple features and multiple classification results. In addition, the five types of signals have relatively large distinctions. Using this method for pre-classification can achieve an accuracy of more than 99.8%.

[0140] For the five types of signals, angle modulation, linear frequency modulation, frequency hopping, frequency shift keying and phase shift keying, they are all signals with constant envelopes and difficult to distinguish. The features of envelope detection time domain signal mean, discrete coefficient, difference mean and difference discrete coefficient cannot be used. For these five types of signals, the SVM method algorithm is used for secondary sub-classification. First, the envelope discrete coefficient, difference discrete coefficient, amplitude peak value of the phase change position and envelope mean ratio after differentiation of the signal collected by the frequency-variable phase change detection circuit are used; the six characteristic parameters of the effective frequency points, effective spectrum width and spectrum flatness within the effective spectrum width of the signal collected by the resonant sweeping circuit are used to sort these five types of signals. First, the input characteristic parameters are upgraded by the Gaussian radial basis function kernel function to increase the discrimination, and the SVM model of five pairs of remainder classes is constructed. The signal that needs to be classified currently extracts the above six characteristic parameters and uses the Gaussian radial basis function kernel function and the SVM model of these five pairs of remainder classes to calculate and obtain the probability values ​​of the five classification results. The signal with the largest comprehensive probability value is the final signal recognition and sorting result.

[0141] The constant envelope signals include: angle modulation wave, linear frequency modulation, frequency hopping, frequency shift keying and phase shift keying.

[0142] In summary, the present invention aims to solve the problems of large volume, high power consumption, high cost, poor layout flexibility, and difficult maintenance of existing wireless signal sensing equipment, and realize low-cost, low-power consumption, high-precision, and highly intelligent monitoring of wireless signals. The present invention integrates ultra-light passive wireless signal sensing circuits and signal identification and sorting methods in board-level electronic tags, which have the beneficial effects of low cost, low power consumption, and convenient maintenance. The system of the present invention has a large sensing range, low cost, low power consumption, and easy deployment. The designed multi-dimensional wireless signal feature acquisition circuit and ultra-light multi-dimensional composite signal identification and sorting method have the beneficial effects of wide recognition range, high recognition accuracy, low resource occupation, and ultra-lightweight, and the software and hardware of the system have the advantage of large secondary development space. The system of the present invention can be applied in many fields, can identify many signal types and modulation modes, and extract wireless signal parameters. It is particularly suitable for scenes such as proctoring scenes and confidential units that require a large number of devices for all-round coverage, and has broad application prospects.

[0143] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. An ultra-lightweight passive wireless signal sensing system based on ultra-high frequency RFID technology, comprising: N ultra-lightweight passive wireless signal sensing tags, 1 reader / writer, and 1 host computer; The ultra-light passive wireless signal sensing tag is used to respond to the command of the reader / writer, monitor the existence of wireless signals in the current environment and their modulation mode, and transmit the collected environmental signal information back to the reader / writer by backscattering or transmit it to the subsequent signal demodulation circuit by wire, and the ultra-light passive wireless signal sensing tag complies with ISO / IEC 18000-6 Type C standard; the reader / writer is controlled by the host computer to send radio frequency signals to the tag group, and the reader / writer runs the anti-collision algorithm of ISO / IEC 18000-6 Type C standard to realize the multi-tag anti-collision function; the host computer is used to control the working state of the reader / writer and display the environmental wireless signal parameter information returned by the tag; In the sensing system, the ultra-light passive wireless signal sensing tag includes: an antenna, a matching network circuit, a first power distributor circuit, a second power distributor circuit, a rectification and energy management circuit, a multi-dimensional wireless signal feature acquisition circuit, a baseband processing unit, a backscattering circuit, and a solar panel; The antenna is connected to the input end of the matching network circuit and the output end of the backscattering circuit; the output end of the matching network circuit is connected to the input end of the first power distributor circuit; the output end (1) of the first power distributor circuit is connected to the input end of the second power distributor circuit; the output end (2) of the first power distributor circuit is connected to the input end (1) of the rectification and energy management circuit; the output end of the second power distributor circuit is connected to the input end of the multi-dimensional wireless signal feature collection circuit; the output end of the multi-dimensional wireless signal feature collection circuit is connected to the signal input end of the baseband processing unit; the solar cell panel is connected to the input end (2) of the rectification and energy management circuit; the output end of the rectification and energy management circuit is connected to the power input end of the baseband processing unit; the output end of the baseband processing unit is connected to the input end of the backscattering circuit; Among them, the antenna, matching network circuit, first power distributor circuit and second power distributor circuit are used to receive and distribute the radio frequency signal and radio frequency energy sent by the reader; the rectification and energy management circuit is used to convert the radio frequency energy into DC power available to the tag, and store the DC power of the radio frequency rectification and the DC power of the solar panel, and intermittently supply the tag for operation; the solar panel is used to convert solar energy into electric energy for collection by the energy management circuit; the multi-dimensional wireless signal feature acquisition circuit is used to collect the signal features of the environmental wireless signal; the baseband processing unit is used to parse the reader command and control the tag to work according to the reader command, and execute the signal recognition and sorting method according to the collected environmental wireless signal features; the backscattering circuit sends a signal to the reader by backscattering under the control of the modulation signal; the sensing tag can be connected to the subsequent circuit through a wired interface; In the ultra-lightweight passive wireless signal sensing tag, the multi-dimensional wireless signal feature acquisition circuit includes: a first envelope detection circuit, a frequency-variable phase-change detection circuit, and a resonant frequency sweeping circuit, which are specifically as follows: (1) The first envelope detection circuit adopts a Dickson detection structure; (2) The frequency-variable phase-change detection circuit includes: an adjustable RC differential circuit, a second envelope detection circuit and a peak detection circuit; used for wireless signals with frequency and phase changes, including: angle-modulated waves, FSK, LFM, frequency hopping, PSK; (3) The resonant frequency sweep circuit is used to extract the approximate spectrum characteristics of the wireless signal; In the multi-dimensional wireless signal feature acquisition circuit, the frequency-variable phase-change detection circuit and the resonant frequency sweeping circuit have the following specific working procedures: (1) Frequency-variable phase-change detection circuit; The input signal is differentiated by an adjustable RC differential circuit. The RC differential circuit selects appropriate resistance and capacitance values ​​to convert the frequency change in the wireless signal into amplitude change. The expression is as follows: V in (t)=Acos(2πf c t+φ(t))(1) With total =R1||Z after (3) Among them, V in (t) represents the input time domain signal at time t, A represents the amplitude of the input time domain signal, φ(t) represents the phase-frequency change of the signal, and f c Indicates the carrier frequency, V out (t) represents the output time domain signal amplitude, R1 represents the resistance value of the RC differential circuit, Z after Represents the total impedance of the subsequent circuit, Z total represents the parallel value of the impedance of the RC differential circuit and the impedance of the subsequent circuit, and C represents the capacitance value of the RC differential circuit; After differential processing, the signal flows to the second envelope detection circuit or the peak detection circuit. The baseband processing unit switches the frequency-varying detection mode and the phase-varying detection mode by changing the capacitance value of the adjustable capacitor of the frequency-varying phase-varying detection circuit, the peak detection circuit control I / O, and the envelope detection control I / O state, as follows: When the peak detection circuit controls the I / O to be set to high voltage, the envelope detection circuit controls the I / O port to be set to high impedance, and the adjustable capacitor value is adjusted so that its RC time constant satisfies equation (4), the system works in the frequency variation detection mode, and the signal flows to the envelope detection circuit, and the frequency variation information is extracted through the envelope detection circuit and the ADC; When working in frequency-varying detection mode, the time constant of RC is less than the inverse of the modulation signal frequency and greater than the inverse of the carrier signal frequency. The expression is as follows: Among them, f mod represents the modulation signal frequency, C1 represents the capacitance value of C during frequency variation detection; The peak value is extracted from the discontinuous point of the phase mutation of the carrier through differential operation, and then the peak value is detected by the passive peak detection circuit; the peak detection part uses the baseband processing unit control pin to initially set the capacitor voltage to 0V, that is, the peak detection circuit controls the I / O to be pre-set to 0V to pull the capacitor voltage to 0V, and then the control pin is set to high impedance, the envelope detection circuit controls the I / O port to be set to high voltage, and the adjustable capacitor value is adjusted so that its RC time constant satisfies formula (5). The unidirectional conduction characteristic of the diode is used to save the peak voltage in the capacitor, and then the voltage amplitude is collected by the ADC, working in the phase change detection mode, and the signal flows to the peak detection circuit; When working in the phase change detection mode, the resistance and capacitance values ​​of the phase change detection circuit satisfy equation (5), and the time constant of RC is less than the inverse of the carrier signal frequency, as shown in the following expression: Wherein, C2 represents the capacitance value of C during phase change detection; (2) Resonant frequency sweep circuit; By using the strong frequency selection performance of the high-Q LC resonant circuit, the input signal spectrum is swept, and the expression is as follows: Z sum (f)=Z LC (f)||Z after (7) V out (f)=V in (f)Z sum (f)(8) Among them, the capacitor C var represents the adjustable capacitor, L represents the inductance of the LC resonant network, Z LC (f) represents the impedance value of the LC resonant network at each frequency point, Z L represents the inductive impedance of the LC resonant network, Z C Represents the capacitive impedance of the LC resonant network, Z sum (f) represents the total impedance value of the LC resonant network and the subsequent circuit at each frequency point, f represents the input signal frequency, Z after Represents the total impedance of the subsequent circuit, V in (f) represents the amplitude of each frequency point of the input resonant sweep frequency circuit signal, V out (f) represents the amplitude of each frequency point of the output signal of the resonant frequency sweep circuit; The resonant frequency sweeping circuit changes the capacitance value by adjusting the voltage at the control input end of the capacitor, and then changes the resonant frequency point of the LC resonant network by continuously adjusting the capacitance value of the adjustable capacitor, and obtains the average voltage of each frequency point through the subsequent rectifier circuit, and finally obtains the frequency domain information of the wireless signal; An ultra-lightweight passive wireless signal sensing method based on ultra-high frequency RFID technology applied to the ultra-lightweight passive wireless signal sensing system, the specific steps are as follows: S1. The sensing system starts to work. The reader / writer is controlled by the host computer to send a radio frequency signal to the tag group. The radio frequency signal contains radio frequency energy and commands, charges the sensing tag wirelessly and activates it, and controls the activated tag to work according to the working mode set by the host computer, so that the tag senses the type and parameters of the wireless signal in the environment. The wireless charging and activation of the sensing tag is specifically as follows: The ultra-lightweight passive wireless signal sensing tag is placed in the environment to be tested. The antenna receives the RF signal in the environment and the solar panel receives the solar energy and converts it into DC power for its own operation and stores it. When the stored DC power is enough for the tag to work normally, the tag is powered. When there is a signal of a certain power in the environment, the tag wakes up and activates. S2. Determine whether the signal in the current environment is a command sent by the reader / writer. If it is a command sent by the reader / writer or other tags, parse the command and execute it. If it is other wireless signals in the environment, use a multi-dimensional wireless signal feature acquisition circuit to extract signal features, and identify and sort the signal through an ultra-lightweight multi-dimensional composite signal recognition and sorting method. S3. Based on the sorting result obtained in step S2, the data is transmitted back to the reader through the backscattering circuit and the antenna, and then the wireless signal sensing of the environment to be tested is continued until the DC power stored this time is exhausted, the tag is powered off and stops working, until the next time the RF signal and solar energy in the environment make the tag powered on and activated again, and the wireless signal of the environment to be tested is monitored.

2. According to claim 1, an ultra-lightweight passive wireless signal sensing system based on ultra-high frequency RFID technology is characterized in that: In step S1, the process of the reader / writer being controlled by the host computer is as follows: A1. The reader initializes and takes inventory of the tags around it, records all the tag IDs and updates their timing timestamps, and records duplicate tags only once; A2. Each tag enters the working state, and determines whether the reader receives the tag signal. If so, the tag signal is analyzed and the tag ID and data are recorded. If not, it goes to step A3; A3, determine whether the host computer wants to send a command to the tag, if yes, control the reader to send the command through the antenna, if no, go to step A4; A4. Determine whether there is any tag that has timed out and not been updated among the currently counted tags, that is, the countdown timestamp is 0. If so, page it according to its tag ID. Then, if there is a signal returned within T time, update the timestamp and record the tag ID and data. If no return data is received from the tag within T time, determine that the tag is faulty, and then repeat step A4 until there is no tag that has timed out and not been updated among the currently counted tags, then return to step A2.

3. The ultra-lightweight passive wireless signal sensing system based on UHF RFID technology according to claim 1 is characterized in that: The working process of the ultra-lightweight passive wireless signal sensing tag is as follows: B1. After the tag is powered, the baseband processing unit is initialized and enters a dormant state. When there is a wireless signal of a certain strength in the environment, the baseband processing unit is awakened; B2. The baseband processing unit determines whether the signal is sent by a known device according to the result of envelope detection. If so, the baseband processing unit performs command parsing and implements the command to enter step B4. If not, the baseband processing unit determines that the signal is other wireless signals in the current environment and enters step B3. B3, extracting signal features from the environmental signal through a multi-dimensional wireless signal feature acquisition circuit, and executing an ultra-lightweight multi-dimensional composite signal recognition and sorting method according to the signal features to obtain a signal recognition and sorting result, and proceeding to step B4; B4. Determine the signal strength in the current environment. If the signal strength in the environment is too low, the signal backscattered back to the reader is weak, and the information that needs to be sent will be stored, and the sleep mode will be ended. If the signal strength in the environment is strong, the amplitude of the current environmental signal will be modulated by backscattering and sent to the reader to complete the signal transmission.

4. The ultra-lightweight passive wireless signal sensing system based on UHF RFID technology according to claim 3 is characterized in that: In step B3, the ultra-lightweight multi-dimensional composite signal identification and sorting method is specifically as follows: C1. Determine whether the tag has deployed a trained signal recognition and sorting model. If not, proceed to step C2. If yes, proceed to step C3. The model is obtained by training a signal source, an ultra-light passive wireless signal sensing tag and a host computer. The host computer controls the signal source to generate various modulation signals. The ultra-light passive wireless signal sensing tag extracts signal features through a multi-dimensional wireless signal feature acquisition circuit and transmits them back to the host computer through a wired interface. The host computer performs model training based on the returned data. C2, the ultra-lightweight multi-dimensional composite signal recognition and sorting method is initialized, and the model training is performed. After the training, the model accuracy is judged. When the model accuracy obtained by training meets the requirements, the model is deployed on the baseband processing unit of the ultra-lightweight passive wireless signal sensing tag. If the model accuracy does not meet the standard, the data is collected again for training until it meets the standard. After the model accuracy meets the standard, it returns to step C1 for re-judgment; Among them, the signal features collected by the multi-dimensional wireless signal feature collection circuit include: the envelope time domain signal collected by the envelope detection circuit, the envelope after signal differentiation collected by the frequency-variable phase change detection circuit and the amplitude peak of the phase change position, and the spectrum information collected by the resonant sweep frequency circuit; C3. The tags deployed with the model will be placed in the environment to be tested to sense the wireless signals in the environment to be tested. The tags use the multi-dimensional wireless signal feature acquisition circuit to collect data, extract feature parameters as input, and send them to the trained signal recognition and sorting model. The model first uses the decision tree method for pre-classification, and classifies the signals into no signal, constant envelope, amplitude modulation, amplitude keying, orthogonal amplitude modulation, and orthogonal frequency division multiplexing through the decision tree; Among them, the extracted characteristic parameters include: the mean value, discrete coefficient, difference mean value and difference discrete coefficient of the envelope detection time domain signal; the envelope discrete coefficient, difference discrete coefficient, amplitude peak value and envelope mean ratio of the signal acquired by the frequency-variable phase change detection circuit after differentiation; and the 10 characteristic parameters of the signal acquired by the resonant sweeping circuit, including the number of effective frequency points, effective spectrum width and spectrum flatness within the effective spectrum width.

5. The ultra-lightweight passive wireless signal sensing system based on UHF RFID technology according to claim 4 is characterized in that: In step C3, the constant envelope signal is subjected to secondary classification using a support vector machine (SVM) method, as follows: The envelope discrete coefficient, difference discrete coefficient, amplitude peak value and envelope mean ratio of the phase change position after differentiation of the signal collected by the frequency-variable phase change detection circuit are used; the six characteristic parameters of the effective frequency points, effective spectrum width and spectrum flatness within the effective spectrum width of the signal collected by the resonant sweeping circuit are used to sort the five types of signals. First, the input characteristic parameters are upgraded by the Gaussian radial basis function kernel function to increase the discrimination, and a five-to-one SVM model is constructed. The six characteristic parameters of the signal that needs to be classified are extracted and calculated using the Gaussian radial basis function kernel function and the five-to-one SVM model to obtain the probability values ​​of the five classification results. The signal with the largest comprehensive probability value is the final signal recognition and sorting result. The constant envelope signals include: angle modulation wave, linear frequency modulation, frequency hopping, frequency shift keying and phase shift keying.

Images