Vehicle identity non-inductive identification method and system based on ultra-high frequency RFID technology

Abstract

The application relates to the fields of Internet of Things and radio frequency identification technology, and discloses a vehicle identity non-inductive identification method and system based on super-high frequency RFID technology, two non-overlapping sampling time windows are established on a side reading and writing terminal of a refueling gun, a first time window adopts power ladder irradiation with unchanged carrier frequency, low power first and high power later, at least one of a carrier frequency, power, polarization or an antenna port is different in a second time window; a phase and intensity sequence of a target label is continuously acquired, trend compatibility, change point correspondence and intensity fluctuation consistency are evaluated, a realness comprehensive score is fused, and the comprehensive score is compared with a threshold value to realize pump starting authorization; when the score is in a transition zone, secondary verification is triggered. The application does not need to change a passive label air interface protocol, has low time delay, is resistant to relay cloning and near metal multipath interference, and is suitable for a vestibule of a gas station and a vehicle fleet oil depot.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of Internet of Things and radio frequency identification technology, in particular to a vehicle identity non-inductive identification method and system based on ultra-high frequency RFID technology. BACKGROUND

[0002] As an important branch of Internet of Things and automatic identification, ultra-high frequency RFID (UHF RFID) can be traced back to the EPC Class-1 Gen2 standard promoted by MIT Auto-ID Center and EPCglobal. The air interface protocol works in the 860 to 960 MHz frequency band, and has become the mainstream since its release in 2004. It was adopted as the ISO / IEC 18000-6C international standard in 2005, laying the foundation for global unified technology and interoperability ecosystem. Gen2 v2 in 2013 further supplemented security and privacy clauses, making its application extend from the supply chain to scenarios with higher security requirements. In the field of transportation, the combination of RFID and vehicle identification has undergone gradual evolution from expressway non-stop toll collection (ETC) to urban parking lot entrance management. As early as 1993, California, USA, launched an RFID-based expressway toll collection pilot, promoting the industrialization of non-cash, unmanned tolling. Subsequently, UHF was gradually used for lane access control, vehicle access management in parks and communities due to its long reading distance and strong concurrent reading and writing capabilities. With the miniaturization and modularization of readers and writers, such as EPC Gen2, dual antenna ports, +30dBm transmission, and other capabilities, the landing and integration of UHF in whole vehicle access and parking fee collection systems have also been accelerated. In summary, UHF RFID vehicle identification has formed a complete chain from standard maturity, industry support, to scenario penetration, and has shown the advantages of automation, low latency, and scalability in parking fee collection, lane control, and park access control.

[0003] However, there are still some key shortcomings in the engineering application of traditional UHF RFID in the forecourt of gas stations: first, many systems still rely on "authorized" as the main system, lack of on-site presence verification for "relay, relay, clone, card borrowing" and other paths, resulting in pump safety depending on black and white lists or high-cost encryption modification, poor compatibility; second, the metal structure is dense in the gas environment, the liquid medium is adjacent, and the near and far field overlap causes multipath and coupling changes, which easily causes phase and RSSI fluctuations, plus the risk of adjacent oil level string reading, resulting in on-site disputes of false authorization or false rejection; third, the vehicle is usually stationary at the oil level, and if only a single time window and a single illumination condition are sampled, there is a lack of dynamic evidence to compare, which is easily confused by transient interference or malicious relay; fourth, the forecourt operation emphasizes the balance between "low latency pump starting" and "anti-misplacement and anti-cheating", but existing systems lack a set of practical, data-driven threshold and hierarchical disposal mechanisms, and parameter drift, seasonal changes and device fine-tuning in long-term operation will cause experience fluctuations. Based on the above shortcomings, the existing technology urgently needs an engineering solution that can complete on-site consistency verification in a short time without changing the passive tag protocol on the vehicle side, has the ability to suppress adjacent oil level interference, and can maintain stable authorization experience through secondary verification and adaptive strategy for a long time, to support the non-inductive pump authorization and risk control of the gas dispenser and the gas gun. Therefore, a vehicle identity non-inductive identification method and system based on ultra-high frequency RFID technology are needed to solve the above problems. SUMMARY

[0004] (I) The technical problem solved: In view of the deficiencies of the prior art, the present application provides a vehicle identity non-inductive identification method and system based on ultra-high frequency RFID technology, which solves the above problems.

[0005] (II) Technical solution: In order to achieve the above purpose, the present application provides the following technical solution: a vehicle identity non-inductive identification method based on ultra-high frequency RFID technology, applied to a read-write terminal provided on a gas gun and a controller in communication connection with the read-write terminal, characterized in that the method comprises: S1: establishing a first sampling time window and a second sampling time window by the read-write terminal, which do not overlap with each other.

[0006] S2: In the first sampling time window, a first illumination condition is applied to at least one currently readable RFID tag, the first illumination condition comprising a first carrier frequency and at least two levels of transmission power switched in a predetermined order, and the first phase sequence and the first intensity sequence of the RFID tag are continuously collected and the time stamp is recorded in the time window.

[0007] S3: applying a second irradiation condition different from the first irradiation condition in at least one parameter including at least one of carrier frequency, transmit power, polarization or antenna port to the RFID tag within the second sampling time window, and continuously collecting a second phase sequence and a second intensity sequence of the RFID tag and recording time stamps within the time window.

[0008] S4: performing consistency evaluation on the phase sequences and intensity sequences collected in the first sampling time window and the second sampling time window; the consistency evaluation at least includes: performing compatibility evaluation on the overall phase change trend of each sampling time window and the expected change range determined according to the field parameters or historical statistics; performing corresponding evaluation on the characteristic changes caused by the difference in irradiation conditions after aligning the time axes of the two sampling time windows; and performing intensity consistency evaluation on the fluctuation patterns of the two intensity sequences over time.

[0009] S5: weighting and fusing the results of the consistency evaluation to obtain a authenticity comprehensive score, and comparing the authenticity comprehensive score with a threshold value, reaching the threshold value to determine that the RFID tag corresponds to the vehicle served by the current fueling gun and output a pump starting authorization instruction, and not reaching the threshold value to enter a secondary verification process or refuse to start the pump.

[0010] Further, the method further comprises, before step S1: obtaining the trigger time of the target vehicle being in place or the gun being lifted by a gun body trigger device, and establishing a unified time reference accordingly, and the start and end times of the first sampling time window and the second sampling time window are aligned with the time reference; wherein the gun body trigger device is at least one of a gun lifting switch, a micro switch, a proximity sensor, an infrared / microwave presence detection, a ground inductor or a millimeter wave radar.

[0011] Further, the method further comprises, between step S1 and step S2: performing a quick inventory by a read-write terminal within a limited time slot, and performing comprehensive evaluation on the currently readable RFID tag according to at least one of the number of responses, signal intensity stability and phase stability to select the target RFID tag.

[0012] Further, between step S3 and step S4, it further comprises: performing phase continuous processing on the first phase sequence and the second phase sequence to eliminate phase jumps; taking the transmit power switching time within the first sampling time window and the difference in irradiation conditions between the two sampling time windows as a reference, identifying identifiable change points in the respective sequences and forming a change point set; performing amplitude normalization on the first intensity sequence and the second intensity sequence and time alignment according to the sampling start and end time.

[0013] Further, after step S5, further comprising: storing the sampling time window information, the phase and intensity sequence summary, the consistency evaluation result, the authenticity comprehensive score and the pump starting conclusion of this time identification; performing statistical analysis on the stored data according to a preset batch or period, and adaptively updating the expected change range and the judgment threshold of the authenticity comprehensive score according to the statistical analysis, so as to adapt to the change of field working conditions.

[0014] Further, the first irradiation condition adopts a low-to-high emission power ladder; the second irradiation condition adopts a second carrier frequency different from the first carrier frequency, and the emission power is set to a lower level in the power ladder, so as to form a stable and distinguishable controlled difference between the first sampling time window and the second sampling time window, for the consistency evaluation.

[0015] Further, in step S4, the compatibility evaluation is specifically: linear fitting of the first phase sequence and the second phase sequence in the respective sampling time windows is performed respectively to obtain respective slope parameters; the slope parameters are compared with a preset expected trend range, and a first trend compatibility result and a second trend compatibility result are respectively output.

[0016] The correspondence evaluation is specifically: time alignment is performed with the start and end time of the two sampling time windows as the reference; a change point is identified in the first phase sequence and the second phase sequence according to a preset threshold, and it is judged whether the appearance time of the corresponding change point falls into a preset tolerance interval, and the feature correspondence result is output according to the judgment.

[0017] The intensity consistency evaluation is specifically: the first intensity sequence and the second intensity sequence are amplitude normalized and time aligned according to the start and end time of the sampling time window; the correlation coefficient of the two sequences is calculated and compared with a preset threshold, and the intensity consistency result is output according to the comparison.

[0018] Further, the two-level verification process specifically includes: switching the pump starting control to a delayed authorization state and keeping the oil level occupied; triggering the license plate recognition module to collect the license plate information of the target vehicle, and comparing the license plate information with the account information bound to the RFID tag to output a license plate comparison result; an additional sampling time window is added by the read-write terminal within a limited time interval, and a third irradiation condition different from the first irradiation condition or the second irradiation condition is used for fast sampling to calculate an additional consistency result; when the license plate comparison result is matched and the additional consistency result reaches a preset additional threshold, a pump starting authorization instruction is output; otherwise, the pump starting is kept to be rejected and recorded as an abnormal event.

[0019] Further, the application also provides a vehicle identity non-inductive identification system based on ultra-high frequency RFID technology, which is applied to a read-write terminal arranged on a fueling gun and a controller in communication connection with the read-write terminal, and characterized in that the controller comprises: a time window establishing module for establishing a first sampling time window and a second sampling time window which are mutually non-overlapping by the read-write terminal.

[0020] A first time window irradiation and collection module is configured to apply a first irradiation condition to at least one currently readable RFID tag in the first sampling time window, the first irradiation condition comprising a first carrier frequency and at least two levels of transmission power switched in a preset order, and continuously collect a first phase sequence and a first intensity sequence of the RFID tag and record a time stamp in the time window.

[0021] A second time window irradiation and collection module is configured to apply a second irradiation condition different from the first irradiation condition in at least one parameter to the RFID tag in the second sampling time window, the parameter comprising at least one of a carrier frequency, a transmission power, a polarization or an antenna port, and continuously collect a second phase sequence and a second intensity sequence of the RFID tag and record a time stamp in the time window.

[0022] A consistency evaluation module is configured to evaluate the phase sequence and the intensity sequence collected in the first sampling time window and the second sampling time window for consistency, the consistency evaluation at least including: compatibility evaluation of the overall phase change trend of each sampling time window with an expected change range determined according to a field parameter or historical statistics; correspondence evaluation of the characteristic change caused by the difference in irradiation conditions after aligning the time axes of the two sampling time windows; and intensity consistency evaluation of the fluctuation pattern of the two intensity sequences over time.

[0023] A fusion determination module is configured to weight and fuse the results of the consistency evaluation to obtain a authenticity comprehensive score, compare the authenticity comprehensive score with a threshold value, and determine that the RFID tag corresponds to a vehicle served by the current fueling gun and output a pump starting authorization instruction when the threshold value is reached, or enter a secondary verification process or refuse to start the pump when the threshold value is not reached.

[0024] (Three) beneficial effects: compared with the prior art, the vehicle identity non-inductive identification method and system based on ultra-high frequency RFID technology have the following beneficial effects: 1, the vehicle identity non-inductive identification method and system based on ultra-high frequency RFID technology, by applying controlled differential illumination conditions in two non-overlapping sampling windows, and evaluating the trend compatibility, feature correspondence and intensity fluctuation pattern consistency of the phase sequence and intensity sequence obtained in the two windows for the same RFID tag, forming a authenticity comprehensive score and taking the threshold as the pump authorization judgment, thereby changing the logic of "reading a legal EPC by default authorization" to "on-site presence is established only when the pump is started"; this method can inhibit relay, cloning and card borrowing for refueling, and solve the defect of "easy to be relayed / cloned fraud"; at the same time, the double-time window comparison and morphological consistency determination weaken the influence of multipath and coupling fluctuation caused by metal and liquid in the refueling environment on the decision, significantly reducing the risk of adjacent fuel level string reading and misauthorization; the total time length of the two windows is controlled in the order of hundreds of milliseconds, which can complete the verification in the moment when the vehicle is in place or the spray gun is close or inserted, without changing the air interface protocol of the vehicle end tag, only realizing it on the read-write terminal and the controller side, avoiding the problems of high encryption modification cost and poor compatibility of traditional methods.

[0025] 2, the vehicle identity non-inductive identification method and system based on ultra-high frequency RFID technology, by weighting and fusing the consistency evaluation results into a single authenticity comprehensive score for decision-making, while introducing a two-level verification process to realize risk review; in peak period, "score meets the standard, authorization" is the main path to ensure the efficiency of oil level turnover; trigger the second verification to bottom out in critical samples, reduce the risk of misstarting the pump and reduce the frequency of manual intervention; combined with log storage and threshold adaptive strategy, the system can continuously fine-tune the expected range and decision threshold with the changes of the on-site environment and seasons, and maintain low false rejection, low false release, and auditable pump starting effect in the long term. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 The vehicle identity non-inductive identification method based on ultra-high frequency RFID technology provided by the present application is shown in the flowchart.

[0027] Figure 2 The vehicle identity non-inductive identification method based on ultra-high frequency RFID technology provided by the present application is shown in the timing diagram.

[0028] Figure 3 The vehicle identity non-inductive identification system based on ultra-high frequency RFID technology provided by the present application is shown in the structural diagram. DETAILED DESCRIPTION

[0029] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of protection of the present application.

[0030] Application scenario: The embodiment is suitable for a forecourt fueling operation scenario. A read-write terminal and a near-field / directional ultra-high frequency RFID antenna are integrated and installed on a handle or an outer side of a gun mouth of a fueling gun. A controller is in communication connection with a pump starting control unit, a license plate recognition camera and a gas station management platform. A gun lifting switch or a micro switch in the gun body is used as a priority trigger source, and can be assisted by a proximity sensor, an infrared / microwave presence detection or a ground inductor, to give a unified time reference of a vehicle being in place or a spray gun being close to a fuel port. In order to suppress adjacent fuel level string reading and multipath coupling caused by metal liquid, the antenna adopts a narrow beam or a near-field coupling type, the working EIRP is controlled and the radiation range is limited by a mechanical shielding plate. If necessary, a second antenna port or different polarization can be configured as an expansion. The system does not change the air interface protocol of a passive UHF tag at the vehicle end, and a read-write terminal applies irradiation conditions with controlled differences in two non-overlapping short time windows and continuously collects phase and intensity sequences of the same target tag. The controller performs consistency evaluation and weighted fusion on the two pieces of data, and completes the pump starting authorization decision in the order of hundreds of milliseconds. When the score is in the transition zone, the system enters a secondary verification, and records the sampling summary and conclusion of this time to a log to support long-term adaptive threshold maintenance. The system aims to realize non-inductive, low-latency and high-reliable authorization of vehicle identity without changing the vehicle and settlement process, improve the resistance to relay, cloning and borrowed card fueling, reduce false pump starting and false rejection caused by adjacent fuel level interference, and maintain stable identification performance and compliance safety under different seasons and passenger flow intensity.

[0031] In order to enable the personnel in the technical field to better understand the present application scheme, the present application will be further described in detail below in combination with the drawings and specific embodiments.

[0032] Please refer to Figures 1-2 , Figure 1 A flowchart of a vehicle identity non-inductive identification method based on ultra-high frequency RFID technology provided by the present application; Figure 1 A flowchart of a vehicle identity non-inductive identification method based on ultra-high frequency RFID technology provided by the present application; Figure 2The vehicle identity non-inductive identification method based on the ultra-high frequency RFID technology provided by the application is a time sequence schematic diagram; the vehicle identity non-inductive identification method based on the ultra-high frequency RFID technology is applied to a read-write terminal arranged on a fuel gun and a controller in communication connection thereof, and characterized in that the method comprises the following steps: S1: a first sampling time window and a second sampling time window that do not overlap with each other are established by the read-write terminal.

[0033] S2: in the first sampling time window, a first irradiation condition is applied to at least one current readable RFID tag, the first irradiation condition comprises a first carrier frequency and at least two levels of transmission power switched in a preset order, and a first phase sequence and a first intensity sequence of the RFID tag are continuously collected and time stamps are recorded in the time window.

[0034] S3: in the second sampling time window, a second irradiation condition different from the first irradiation condition in at least one parameter is applied to the RFID tag, the parameter comprises at least one of the carrier frequency, the transmission power, the polarization or the antenna port, and a second phase sequence and a second intensity sequence of the RFID tag are continuously collected and time stamps are recorded in the time window.

[0035] S4: consistency evaluation is performed on the phase sequence and the intensity sequence collected in the first sampling time window and the second sampling time window; the consistency evaluation at least comprises: compatibility evaluation is performed on the overall change trend of the phase of each sampling time window and the expected change range determined according to the field parameters or historical statistics; after aligning the time axes of the two sampling time windows, corresponding evaluation is performed on the characteristic changes caused by the difference in irradiation conditions; and intensity consistency evaluation is performed on the fluctuation mode of the two intensity sequences with time.

[0036] S5: the results of the consistency evaluation are weighted and fused to obtain a authenticity comprehensive score, and the authenticity comprehensive score is compared with a threshold value; when the threshold value is reached, it is determined that the RFID tag corresponds to the vehicle served by the current fuel gun and an authorized pump starting instruction is output, and when the threshold value is not reached, a secondary verification process is entered or the pump starting is rejected.

[0037] In this embodiment, the read-write terminal and the controller work together in the scenario of the fueling gun. First, the read-write terminal installed on the handle of the fueling gun divides two short sampling time windows that do not overlap each other according to the local time sequence, and performs two controlled and comparable observations on the passive UHF tag carried by the same vehicle. In the first sampling time window, a first irradiation condition containing a fixed carrier frequency and switching multiple transmission power levels in a preset order is applied, and the read-write terminal continuously receives the read-back events of the target tag (each containing at least time stamp, EPC / TID, working carrier frequency, antenna port, RSSI, and physical layer phase metadata) in the time window. The controller only retains the target tag record in ascending order of time stamp, arranges the phase and RSSI fields in the record with time, and forms the first phase sequence and the first intensity sequence, respectively. Then, in the second sampling time window, a second irradiation condition different from the first irradiation condition in at least one of the carrier frequency, transmission power, polarization, or antenna port is applied, and the second phase sequence and the second intensity sequence are formed by the same process (if the hardware outputs IQ samples, the equivalent phase and intensity are calculated on the read-write terminal side first, and then the sequences are entered). Both sequences retain the corresponding time stamp for subsequent alignment and evaluation. The technical principle is that when the operation of the spray gun from "off-site, close to the oil port, and insert into the oil port" occurs, the geometric relationship between the gun-mounted antenna and the vehicle tag changes in the order of hundreds of milliseconds, making the phase change with time showing the overall trend and the change rate. When there is a controlled difference between the two irradiation conditions, the same tag that is truly in place will show corresponding change characteristics and similar strong and weak fluctuation patterns in the two sequences, while relays, clones, or adjacent oil levels are difficult to reproduce this matching relationship under the constraints of short time sequence and controlled difference.

[0038] Next, the controller performs consistency evaluation accordingly: first, compare the overall change trend of each phase sequence with the expected trend range obtained from field parameters or historical statistics; second, perform time alignment based on the start and end times of the two time windows, and verify the corresponding relationship of identifiable changes in the two sequences caused by the controlled difference such as power ladder or frequency switching; third, after amplitude normalization and time alignment of the intensity sequences, evaluate whether the fluctuation patterns over time in the two sequences are consistent.

[0039] The evaluation results are weighted and fused into a comprehensive score of authenticity, which is compared with a threshold value to automatically output the pump start authorization, secondary verification, or pump start rejection instruction, thereby completing the judgment of "whether the target vehicle currently served by the fueling gun is truly in place" in the order of hundreds of milliseconds. This method does not need to change the air interface protocol of the passive tag on the vehicle side, but only relies on the time sequence control and data processing of the read-write terminal, and has the engineering effects of non-intrusive, low latency, anti-relay and clone, and anti-environmental fluctuation, which is suitable for fueling scenarios such as forecourt of gas station, fleet fuel depot, and unattended oil machine.

[0040] For the convenience of further understanding of those skilled in the art, the technical solutions of the method are illustrated below by taking the example of side deployment of a refueling gun: a pair of near-field / directional UHF antennas are integrated outside the handle / mouth of the refueling gun, and a read-write terminal is set to have a sampling frequency of about 500 Hz and an initial threshold value of 0.72; after a trigger is given by a gun lifting or proximity sensor, the read-write terminal establishes two short time windows that do not overlap with each other: a first time window of about 120 ms and a second time window of about 120 ms. A first irradiation condition is applied in the first time window: a carrier frequency of about 920 MHz and a transmission power that is switched in steps from 28 dBm to 30 dBm, and a first phase sequence and a first intensity sequence of the target tag are continuously collected and the time stamp is recorded; then a second irradiation condition is applied in the second time window: the carrier frequency is switched to about 925 MHz and the power is returned to a lower gear (28 dBm), and a second phase sequence and a second intensity sequence are continuously collected and the time stamp is recorded. The controller then performs consistency evaluation: first, check whether the overall trend of the two phase sequences matches the expected trend preset on site (for example, both the two sequences show a stable upward trend and the "speed" level is similar during the process of the gun approaching and inserting); second, after alignment based on the start and end times of the two time windows, observe whether the "step / inflection point" caused by the power step in the first time window appears at the corresponding position in the second time window in relative time; third, compare the fluctuation patterns of the two intensity sequences after amplitude normalization (for example, both the two sequences show a peak in the latter part of the time window). If the authenticity comprehensive score S of the three evaluations is greater than or equal to 0.72, the system immediately outputs a pump starting authorization instruction; if S falls between 0.5 and 0.72, a secondary verification is triggered, and if the matching and additional sampling meet the standards, authorization is granted; if S is less than 0.5, the pump is refused and recorded as an exception. The system also stores the time stamps of the two time windows, the phase / intensity summary, the sub-item evaluation results and S in the log, and adjusts the expected trend range and threshold every about 200 operations to maintain low false rejection and low false start authorization performance under different seasons and passenger flow conditions.

[0041] Further, the method further comprises, before step S1: obtaining a trigger time when a target vehicle is in place or a gun is lifted by a gun body trigger device, and establishing a unified time reference based on the trigger time, and the start and end times of the first sampling time window and the second sampling time window are aligned based on the time reference; wherein the gun body trigger device is at least one of a gun lifting switch, a micro switch, a proximity sensor, an infrared / microwave presence detection, a ground inductor or a millimeter wave radar.

[0042] In this embodiment, the trigger device (any one of the gun lifting switch, micro switch, proximity sensor, infrared / microwave presence detection or oil level sensor) on the gun body side generates a rising edge event when detecting the lifting of the spray gun or the proximity of the spray gun to the oil port; the controller debounces and determines the validity of the event (for example, sets a minimum duration and a minimum interval threshold to filter out false triggers and non-vehicle targets), determines the trigger time as T0, and issues T0 to the read-write terminal as the unified time reference for this identification process; the read-write terminal aligns and locks the start and end times of the first and second sampling time windows in the local time sequence based on T0, so that the two time windows and the spatial movement of the spray gun "off, close, and insert" establish a stable mapping relationship (for example, the first time window corresponds to the proximity to the oil port, and the second time window corresponds to the insertion or close-in stage), thereby reducing the phase and intensity deviation caused by random start, and improving the comparability and reproducibility of the two data. To improve the engineering robustness, the system adopts a main trigger + fallback start strategy: the gun lifting, micro motion or proximity sensor is the main trigger source, and the video or radar is the auxiliary trigger source; when the main trigger is lost or abnormal, the controller falls back to the sliding time window mode within a limited timeout to avoid missing detection. At the same time, the controller and the read-write terminal keep the clock synchronized to ensure that T0 is consistent with the local timestamp; in the continuous operation scene of multiple vehicles or multiple gun lifting, the system implements instance segmentation management for continuous triggers: each trigger binds only one set of S1-S5 process instances, and if the previous instance has not ended, the next instance is generated in queue to prevent the time windows of two identification processes from overlapping.

[0043] Further, the method further comprises, between step S1 and step S2: performing a quick inventory by the read-write terminal within a limited time slot, and comprehensively evaluating the current readable RFID tag according to at least one of the number of responses, signal strength stability, and phase stability to select the target RFID tag.

[0044] In this embodiment, before entering step S2 after establishing the two-sampling time windows in step S1, the read-write terminal first performs EPC C1G2 inventory in a preset fast inventory time slot (e.g. tens of milliseconds, which can be adjusted on site) to collect a set of candidate tags currently readable and form several short samples for each candidate tag, which at least include the response count, the RSSI sequence and the phase sequence. The controller calculates a comprehensive score for each candidate tag: first, the number of responses and the continuous frame appearance rate are measured to measure the coverage and basic readability of the tag; second, the signal strength stability (fluctuation amplitude / dithering rate within the time slot) is selected to preferentially select tags closer to the main lobe of the gun-mounted antenna and less affected by obstructions, and to suppress adjacent oil levels and environmental stray reflections; third, the phase stability is observed to filter tags dominated by too strong multipath or incidental reflections by focusing on the smoothness and short-term consistency over time. The comprehensive score sets a minimum threshold and a ranking rule (e.g. first set a stability standard as a hard condition, then sort by the number of responses, and if necessary, use the latest frame arrival time as a tie-breaking rule), and finally selects the tag with the highest score and meets the threshold as the target RFID tag for this identification, and binds its EPC / TID to the subsequent S2, S3 controlled sampling and S4 consistency evaluation link. If there is no tag that meets the threshold in this time slot, the system triggers a fault-tolerant strategy: retry inventory for a short time; if it still fails, try again with a lower power threshold or a backup antenna port; if it still fails, record the exception and enter the secondary verification process. This method can exclude adjacent oil level vehicles, staff badges, and personal item tags in a very short time, so that the subsequent two-sampling time windows focus on the tag that is most likely to belong to the current refueling vehicle, thereby significantly reducing the influence of false reading and multi-tag interference on the authenticity determination, and improving the reliability and reproducibility of the entire non-inductive pump starting link.

[0045] Further, between step S3 and step S4, there is also: performing phase continuation processing on the first phase sequence and the second phase sequence to eliminate phase jumps; taking the transmission power switching time in the first sampling time window and the difference in illumination conditions between the two sampling time windows as a reference, identifying identifiable change points in the respective sequences and forming a change point set; performing amplitude normalization on the first intensity sequence and the second intensity sequence and time alignment according to the sampling start and end time.

[0046] In this embodiment, after completing the two-stage data collection of step S3 and before entering the consistency evaluation of step S4, the system first performs three pre-processing to improve data availability and evaluation robustness: first, the phase continuous processing is performed on the first phase sequence and the second phase sequence, which expands the jump caused by the measurement phase according to the week folding into a continuous curve in time, so that the curve truly reflects the smooth geometric change in the operation process, avoiding the misjudgment of the folding as a motion feature; second, according to the switching time of the power ladder in the first sampling time window and the difference in illumination conditions between the two sampling time windows (such as different load frequencies, different power levels, different ports / polarizations), automatically identify identifiable change points on the two phase curves and generate a change point set, which can be an inflection point, a step or a local extremum caused by controlled illumination changes, for subsequent correspondence check on the aligned time axis; third, the amplitude normalization and time alignment are performed on the first intensity sequence and the second intensity sequence to eliminate the absolute amplitude offset caused by the load frequency and power difference, so that the comparison focuses on the "fluctuation pattern over time" itself, and the correlation or similarity determination is facilitated on the same reference scale. The above pre-processing does not need to change the passive tag protocol, and only needs to be implemented on the read-write terminal / controller side with a light algorithm, which can be completed in milliseconds; its essential role is to reduce the interference of multipath reflection, instantaneous interference and power switching transition state in the near metal and liquid environment on the consistency evaluation, thereby providing a clearer and more stable input basis for the three types of trend compatibility, change point correspondence and intensity fluctuation consistency, and improving the recognition of the authenticity comprehensive score.

[0047] Further, after step S5, it also includes: storing the identified sampling time window information, the summary of the phase and intensity sequence, the consistency evaluation result, the authenticity comprehensive score and the pump conclusion; performing statistical analysis on the stored data according to the preset batch or period, and adaptively updating the expected change range and the judgment threshold of the authenticity comprehensive score according to the statistical analysis, so as to adapt to the changes of the field working conditions.

[0048] In this embodiment, after the authenticity comprehensive score is completed and the pump starting / rejection decision is made in step S5, the system enters operation and maintenance adjustment: the controller writes the key elements identified this time into the local or platform log in a structured record, including the start and end times of the two sampling time windows, the reading and writing terminals and antenna identifiers, the summaries of the first and second phase and intensity sequences (such as quantile points, extreme values, linear fitting slopes, correlation coefficients, etc. statistical quantities without original privacy data), consistency evaluation sub-item results (trend compatibility, feature correspondence, intensity consistency), final authenticity comprehensive score, and pump starting conclusion and secondary verification state; the log supports indexing retrieval by oil level / gasoline gun number, shift / period, event type, and abnormal samples (such as scores that are too low or secondary verification failures) are automatically labeled for easy review. The system performs statistical analysis according to the preset batch or period (for example, every day, every about 200 operations), calculates the score distribution, sub-item index drift, and false start / false rejection estimation, and accordingly performs small-step adaptive update on the "expected change range" (used as the reference interval for trend compatibility judgment) and "judgment threshold": when the phase "running speed" is found to be overall shifted due to changes in the expected range caused by changes in the expected range, the expected range is automatically shifted or moderately widened; when the false rejection is high or the false start risk is rising in a long period, the threshold is fine-tuned to the target interval, and all adjustments are traceable and can be rolled back. To ensure stability, the update uses amplitude limiting and cold start protection (only effective when the sample size is sufficient, and the single adjustment does not exceed the preset step), and the new and old parameters are compared in gray scale to avoid introducing new bias. Without changing the passive tag and air interface protocol, the system can continuously absorb field data and offset the impact of environmental and equipment changes, maintain a low-latency, low-misreject, and low-misstart pump starting experience for a long time, and provide traceable evidence for safety audit and risk control optimization.

[0049] Further, the first irradiation condition uses a low-to-high emission power ladder; the second irradiation condition uses a second carrier frequency different from the first carrier frequency, and sets the transmission power to a lower level in the power ladder, thereby forming a stable and distinguishable controlled difference between the first sampling time window and the second sampling time window for the consistency evaluation.

[0050] In this embodiment, the first irradiation condition is set to a power step of first low and then high, and the second irradiation condition is selected to have a second carrier frequency different from the first carrier frequency and the power is fixed at the lower level of the step. The two together construct a stable and distinguishable controlled difference between the first and second sampling time windows. The principle is that the power step will form a slight but repeatable energy change in the echo field of the target tag within the first time window, causing the phase and intensity sequence to have a small step or inflection point, providing a time anchor point for the subsequent "change point correspondence". The second time window changes the carrier frequency and maintains low power. On the one hand, it introduces frequency diversity to change the multipath superposition weight and reduce single-frequency contingency. On the other hand, it maintains a link that is not saturated, has a similar link margin as the low-power segment of the first window, and facilitates the comparison of trends and fluctuation patterns across time windows. In engineering implementation, the read-write terminal completes a power switch in the first window according to the pre-designed plan (the switch time should account for a very small proportion of the single-window time to avoid pollution sampling), and completes the carrier frequency switching and locks the low power before the second window starts, ensuring that the two windows do not overlap and the state is stable. This combination does not need to change the passive tag protocol, but only relies on the transmission control of the gun-mounted read-write end and the compliant frequency point switching to implement it; its role is to provide repeatable controlled features and cross-frequency comparison samples for the consistency evaluation of step S4, improve the reliability of trend compatibility determination, enhance the detectability of change points, and form short-time sequence features that are difficult to fake when facing relays, clones, and adjacent oil level string reading. Finally, it supports robust authenticity determination and pump authorization decision-making within hundreds of milliseconds.

[0051] Further, in step S4, the consistency evaluation specifically comprises: performing linear fitting on the first phase sequence and the second phase sequence in the respective sampling time windows to obtain respective slope parameters; and comparing the slope parameters with a preset expected trend range to respectively output a first trend compatibility result and a second trend compatibility result.

[0052] The correspondence evaluation specifically comprises: time alignment based on the start and end times of the two sampling time windows; identifying change points in the first phase sequence and the second phase sequence according to a preset threshold, and determining whether the occurrence times of the corresponding change points fall within a preset tolerance interval, and outputting a feature correspondence result accordingly.

[0053] The intensity consistency evaluation specifically comprises: amplitude normalization of the first intensity sequence and the second intensity sequence, and time alignment according to the start and end times of the sampling time windows; calculating the correlation coefficient of the two sequences and comparing it with a preset threshold, and outputting an intensity consistency result accordingly.

[0054] In this embodiment, firstly, trend compatibility evaluation is performed. The controller performs linear fitting on the first phase sequence and the second phase sequence in their respective sampling time windows, respectively, to obtain a slope parameter reflecting the overall trend and speed of the phase over time; the respective slope is compared with the expected trend range determined based on field calibration or historical statistics, and a first and second trend compatibility result is output, which is used to determine whether the current observation conforms to the geometric constraints of "the lance is out of position, close to, inserted into the oil port" and the typical close-in speed, thereby filtering samples dominated by parking, hand shaking or abnormal reflection. Secondly, correspondence evaluation is performed. Time alignment is completed based on the start and end times of the two sampling time windows, and the change points (inflection points, steps or local extrema) are identified in the aligned two phase sequences according to a preset threshold, and it is checked whether the occurrence times of the corresponding change points in the two sequences fall within a tolerance interval; if they can correspond, it means that the controlled differences such as power steps or frequency conversion have produced reproducible responses in the same "true in-position refueling" process, enhancing the in-situ credibility. Finally, intensity consistency evaluation is performed. The first and second intensity sequences are amplitude-normalized and time-aligned according to their respective time window start and end times, the correlation coefficient of the two sequences is calculated and compared with a preset threshold, and if the consistency requirement is met, it means that the two sequences are similar in energy fluctuation rhythm, which can offset the absolute amplitude difference caused by different carrier frequencies and powers. The above three evaluations each output a clear score / Boolean result, which is input to step S5 according to the weight to improve the discrimination of the authenticity comprehensive score against relay, clone, adjacent oil level string reading and multipath disturbance, while keeping the implementation simple, the calculation amount small and the real-time strong, meeting the online authorization demand of the order of hundreds of milliseconds.

[0055] Further, the secondary verification process specifically includes: switching the pump starting control to a delayed authorization state and maintaining the oil level occupation; triggering the license plate recognition module to collect the license plate information of the target vehicle, and comparing it with the account information bound to the RFID tag, and outputting a license plate comparison result; an additional sampling time window is added by the read-write terminal within a limited time slot, and fast sampling is performed under a third irradiation condition different from the first irradiation condition or the second irradiation condition, and an additional consistency result is calculated; when the license plate comparison result is matched and the additional consistency result reaches a preset additional threshold, an pump starting authorization instruction is output; otherwise, the pump starting is kept rejected and recorded as an abnormal event.

[0056] When the authenticity comprehensive score obtained in step S5 does not reach the threshold value and falls into the preset transition interval in this embodiment, the system enters secondary verification: the controller first switches the pump starting control to a "delayed authorization" state (such as delaying for hundreds of milliseconds to several seconds, during which the oil level is occupied and the spray gun interlock is locked), to gain time for additional verification; then triggers the license plate recognition camera to capture and recognize the license plate number of the current vehicle, and compares it with the account information bound to the RFID tag in the gas station or fleet management platform to obtain the license plate comparison result (match / mismatch); at the same time, the read-write terminal adds an additional sampling time window within a very short time slot, and performs fast sampling under a third irradiation condition different from the first or second irradiation condition (such as changing the frequency again, switching to another power level or standby antenna port), to calculate an additional consistency result as a second piece of evidence independent of the main judgment link. The controller integrates the two results to make a final decision: when the license plate matches and the additional consistency result reaches the preset additional threshold value, the pump starting authorization instruction is immediately output; otherwise, the pump starting is refused, and the key evidence of this secondary verification is written into the log for review and risk control optimization.

[0057] Further, as shown in Figure 2 , Figure 2 The application provides a vehicle identity non-inductive recognition system structure based on ultra-high frequency RFID technology. The application also provides a vehicle identity non-inductive recognition system based on ultra-high frequency RFID technology, which is applied to a read-write terminal arranged on a fuel gun and a controller in communication connection with the read-write terminal, and characterized in that the controller comprises a time window establishing module for establishing a first sampling time window and a second sampling time window which do not overlap with each other by the read-write terminal.

[0058] A first time window irradiation and collection module is configured to apply a first irradiation condition to at least one currently readable RFID tag in the first sampling time window, the first irradiation condition comprising a first carrier frequency and at least two levels of transmission power switched in a preset order, and continuously collect and record time stamps of a first phase sequence and a first intensity sequence of the RFID tag in the time window.

[0059] A second time window irradiation and collection module is configured to apply a second irradiation condition different from the first irradiation condition in at least one parameter to the RFID tag in the second sampling time window, the parameter comprising at least one of a carrier frequency, a transmission power, a polarization or an antenna port, and continuously collect and record time stamps of a second phase sequence and a second intensity sequence of the RFID tag in the time window.

[0060] A consistency evaluation module is configured to evaluate the phase sequence and the intensity sequence collected by the first sampling time window and the second sampling time window. The consistency evaluation at least includes: compatibility evaluation of the overall phase change trend of each sampling time window and the expected change range determined according to the on-site parameters or historical statistics; corresponding evaluation of the characteristic changes caused by the difference in irradiation conditions after aligning the time axes of the two sampling time windows; and intensity consistency evaluation of the fluctuation patterns of the two intensity sequences over time.

[0061] A fusion determination module is configured to weight and fuse the results of the consistency evaluation to obtain a comprehensive authenticity score, compare the comprehensive authenticity score with a threshold value, and determine that the RFID tag corresponds to the vehicle served by the current fueling gun and output a pump starting authorization instruction when the threshold value is reached, or enter a secondary verification process or refuse to start the pump when the threshold value is not reached.

[0062] It should be noted that, in this document, the terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0063] Although embodiments of the present application have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made thereto without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A vehicle identity non-inductive identification method based on ultra-high frequency RFID technology, applied to a read-write terminal arranged on a fueling gun and a controller in communication connection with the read-write terminal, characterized in that, The method comprises: S1: establishing, by a read-write terminal, first and second mutually non-overlapping sampling time windows; S2: in the first sampling time window, applying a first irradiation condition to at least one currently readable RFID tag, the first irradiation condition comprising a first carrier frequency and at least two levels of transmission power switched in a preset order, and continuously collecting a first phase sequence and a first intensity sequence of the RFID tag in the time window and recording a timestamp; S3: in the second sampling time window, applying a second irradiation condition to the RFID tag, the second irradiation condition being different from the first irradiation condition in at least one parameter, the parameter comprising at least one of a carrier frequency, a transmission power, a polarization or an antenna port, and continuously collecting a second phase sequence and a second intensity sequence of the RFID tag in the time window and recording a timestamp; S4: performing consistency evaluation on the phase sequences and intensity sequences collected in the first and second sampling time windows; the consistency evaluation at least comprises: performing compatibility evaluation on the overall phase change trend of each sampling time window and the expected change range determined according to the field parameters or historical statistics; performing corresponding evaluation on the characteristic changes caused by the difference in irradiation conditions after aligning the time axes of the two sampling time windows; and performing intensity consistency evaluation on the fluctuation patterns of the two intensity sequences over time; S5: weighting and fusing the results of the consistency evaluation to obtain a comprehensive authenticity score, and comparing the comprehensive authenticity score with a threshold value, determining that the RFID tag corresponds to the vehicle served by the current fueling gun and outputting a pump starting authorization instruction when the threshold value is reached, and entering a secondary verification process or rejecting the pump starting when the threshold value is not reached; in step S4, the compatibility evaluation is specifically: linear fitting of the first and second phase sequences in their respective sampling time windows to obtain respective slope parameters; comparing the slope parameters with the preset expected trend range to output first and second trend compatibility results, respectively; the corresponding evaluation is specifically: time alignment based on the start and end times of the two sampling time windows; identifying change points in the first and second phase sequences according to a preset threshold, and determining whether the occurrence times of the corresponding change points fall within a preset tolerance interval, and outputting a feature correspondence result accordingly; the intensity consistency evaluation is specifically: amplitude normalization of the first and second intensity sequences and time alignment according to the start and end times of the sampling time windows; calculating the correlation coefficient of the two sequences and comparing it with a preset threshold, and outputting an intensity consistency result accordingly.

2. The vehicle identity non-inductive recognition method based on the ultra-high frequency RFID technology according to claim 1, characterized in that, The method further comprises, before step S1: obtaining, by a gun body trigger device, a trigger time when the target vehicle is in place or the gun is lifted, and establishing a unified time reference based thereon, the start and end times of the first and second sampling time windows being aligned with the time reference; wherein the gun body trigger device is at least one of a gun lifting switch, a microswitch, a proximity sensor, an infrared / microwave presence detection, a ground inductor or a millimeter wave radar. 3.The vehicle identity non-inductive identification method based on ultra-high frequency RFID technology according to claim 1, characterized in that, The method further comprises, between step S1 and step S2, performing a quick inventory by the read-write terminal within a defined time slot, and performing a comprehensive evaluation on the current readable RFID tag according to at least one of the number of responses, signal strength stability and phase stability to select a target RFID tag.

4. The vehicle identity non-contact recognition method based on the ultra-high frequency RFID technology according to claim 1, characterized in that, The method further comprises, between step S3 and step S4, performing phase continuation processing on the first phase sequence and the second phase sequence to eliminate phase jumps; identifying identifiable change points in the respective sequences and forming a change point set, with reference to the transmission power switching time in the first sampling time window and the difference in illumination conditions between the two sampling time windows; and performing amplitude normalization on the first intensity sequence and the second intensity sequence and time alignment according to the sampling start and end time.

5. The method of claim 1, wherein the method is a method of vehicle identity recognition without induction based on ultra-high frequency RFID technology. The method further comprises, after step S5, storing the sampling time window information identified this time, the digest of the phase and intensity sequences, the consistency evaluation result, the authenticity comprehensive score and the pump starting conclusion; and performing statistical analysis on the stored data according to a preset batch or period, and adaptively updating the expected change range and the determination threshold of the authenticity comprehensive score according to the statistical analysis, to adapt to changes in field conditions. 6.The vehicle identity non-inductive identification method based on ultra-high frequency RFID technology according to claim 1, characterized in that: The first illumination condition adopts a low-to-high transmission power ladder; the second illumination condition adopts a second carrier frequency different from the first carrier frequency, and the transmission power is set to a lower level in the power ladder, thereby forming a stable and distinguishable controlled difference between the first sampling time window and the second sampling time window, for the consistency evaluation.

7. The method according to claim 1, wherein the method is characterized by, The two-level verification process specifically comprises: switching the pump starting control to a delayed authorization state and keeping the oil level occupied; triggering the license plate recognition module to collect the license plate information of the target vehicle, and comparing the license plate information with the account information bound to the RFID tag, and outputting a license plate comparison result; performing an additional sampling time window by the read-write terminal within a defined time slot, performing a quick sampling using a third illumination condition different from the first illumination condition or the second illumination condition, and calculating an additional consistency result; when the license plate comparison result is matched and the additional consistency result reaches a preset additional threshold, outputting a pump starting authorization instruction; otherwise, keeping the pump starting rejected and recording as an abnormal event.

8. A vehicle identity non-inductive identification system based on ultra-high frequency RFID technology, applied to a read-write terminal arranged on a fueling gun and a controller in communication connection with the read-write terminal, characterized in that, The method for realizing the vehicle identity non-inductive identification based on the ultra-high frequency RFID technology in any one of claims 1 to 7, the controller comprises: a time window establishing module for establishing a first sampling time window and a second sampling time window which do not overlap each other by a read-write terminal; a first time window irradiation and collection module for applying a first irradiation condition to at least one current readable RFID tag in the first sampling time window, the first irradiation condition comprising a first carrier frequency and at least two levels of transmission power switched in a preset order, and continuously collecting a first phase sequence and a first intensity sequence of the RFID tag in the time window and recording a time stamp; a second time window irradiation and collection module for applying a second irradiation condition different from the first irradiation condition in at least one parameter to the RFID tag in the second sampling time window, the parameter comprising at least one of the carrier frequency, the transmission power, the polarization or the antenna port, and continuously collecting a second phase sequence and a second intensity sequence of the RFID tag in the time window and recording a time stamp; a consistency evaluation module for performing consistency evaluation on the phase sequence and the intensity sequence collected in the first sampling time window and the second sampling time window; the consistency evaluation at least comprises: performing compatibility evaluation on the overall phase change trend of each sampling time window and the expected change range determined according to the field parameters or historical statistics; performing corresponding evaluation on the characteristic changes caused by the difference in irradiation conditions after aligning the time axes of the two sampling time windows; and performing intensity consistency evaluation on the fluctuation patterns of the two intensity sequences over time; a fusion determination module for weighting and fusing the results of the consistency evaluation to obtain a authenticity comprehensive score, comparing the authenticity comprehensive score with a threshold value, determining that the RFID tag corresponds to the vehicle served by the current fueling gun and outputting a pump starting authorization instruction when the threshold value is reached, and entering a secondary verification process or rejecting the pump starting when the threshold value is not reached.

Images