Identity verification system for low-altitude sharing economy platform unmanned aerial vehicle rental service

Abstract

The application discloses a low-altitude sharing economy platform unmanned aerial vehicle leasing service identity verification system, comprising the following modules: a public key disturbance module, which is used for acquiring a user public key and constructing a signature fragment set; an activation signature module, which is used for generating an original identity activation signature; a behavior signature module, which is used for generating a behavior time sequence spectrum and combining the original identity activation signature to generate a relay biological behavior signature; a challenge signature module, which is used for generating a latent challenge area instruction based on the relay biological behavior signature, embedding the latent challenge area instruction in a control channel, and generating a time domain latent signature; a space signature module, which is used for generating a behavior field domain twist signature; a signature integration module, which is used for generating an autophagy closed loop signature structure; and a chain storage module, which is used for performing closed processing on the autophagy closed loop signature structure and writing the autophagy closed loop signature structure into a chain signature data structure. The application combines a multistage behavior signature and a chain encapsulation method, realizes strong binding, high security and full-chain traceability of unmanned aerial vehicle leasing identity authentication.

Description

Technical Field

[0001] This invention relates to the field of identity security authentication technology, and in particular to an identity verification system for drone rental services on a low-altitude shared economy platform. Background Technology

[0002] With the gradual opening of urban low-altitude airspace and the continued rise in demand for drone rentals, low-altitude intelligent equipment dispatch platforms based on the sharing economy model are becoming an important infrastructure for future smart transportation and urban aerial logistics. However, existing drone rental service identity authentication mechanisms generally suffer from the following problems:

[0003] The current methods for verifying the identity of rental users are simplistic, mostly relying on basic verification methods such as static passwords, fingerprints, or facial recognition. These methods are insufficient to effectively defend against high-risk operations such as forged rental requests and unauthorized use of device control privileges, resulting in significant blind spots in identity misuse and flight liability tracing. Existing identity signature methods fail to achieve multi-stage, context-aware authentication throughout the rental process, leading to a disconnect between dynamic control behavior during flight and pre-rental identity authorization, making it difficult to accurately reflect the consistency between the operator's true identity and control behavior. Most platforms lack deep binding mechanisms for the fingerprints of the drone's physical device, relying solely on user-side information for signature construction, ignoring the role of drone sensor behavior and environmental disturbances in enhancing authentication credibility during flight. Traditional chained signature methods do not fully integrate behavioral characteristics and biometric dynamic indicators, lacking dynamic signature mechanisms for control intentions, flight modes, and physical field-related behaviors, failing to achieve continuous identity verification and behavior chain sealing throughout the entire flight process. Furthermore, the authentication results records of most platforms currently remain in centralized databases or single log systems, failing to meet the requirements for verifiability, immutability, and multi-party traceability of identity verification data.

[0004] Therefore, how to provide an identity verification system for drone rental services on a low-altitude shared economy platform is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] One objective of this invention is to propose an identity verification system for drone rental services on a low-altitude shared economy platform. This invention integrates multimodal biometric behavioral feature extraction, hardware fingerprint binding, and chained signature structure encapsulation methods to achieve multi-dimensional dynamic authentication of drone rental users. Through progressive signature nesting of biometric behavioral time-series spectrum, flight behavior features, and equipment information, the system effectively enhances the security, uniqueness, and tamper resistance of authentication. It also supports full-process identity traceability and responsibility confirmation, making it suitable for high-frequency rental and high-risk operation scenarios on low-altitude shared platforms. It boasts advantages such as strong real-time performance, good privacy protection, and high system compatibility.

[0006] The drone rental service identity verification system of the low-altitude sharing economy platform according to an embodiment of the present invention includes the following modules:

[0007] The public key perturbation module is used to obtain the user's public key, decompose the user's public key into multiple elliptic curve point segments, introduce offset perturbations generated by the user's iris image information, and construct a signature fragment set.

[0008] The activation signature module is used to combine the signature fragment set with the drone's unique hardware identifier and timestamp to generate the original identity activation signature;

[0009] The behavior signature module is used to generate a behavior time sequence spectrum and combine it with the original identity activation signature to generate a relay bio behavior signature, wherein the relay bio behavior signature includes a delayed unlocking structure.

[0010] The challenge signature module is used to generate a latent challenge area instruction based on the relay biological behavior signature and embed it in the control channel. The user completes the challenge response at a random time point to generate a time-domain latent signature.

[0011] The spatial signature module is used to generate a spatial behavior map, construct a field knot matrix, and combine the field knot matrix with the temporal latent signature to generate a behavior field knot signature.

[0012] The signature integration module is used to combine the original identity activation signature, relay biological behavior signature, temporal latency signature and behavioral field knot signature to generate an autophagy closed-loop signature structure.

[0013] The chain-based evidence storage module is used to enclose the self-eating closed-loop signature structure and write it into the chain-based signature data structure for identity verification and accountability.

[0014] The identity verification method for drone rental services on a low-altitude sharing economy platform according to an embodiment of the present invention includes the following steps:

[0015] Step 1: Obtain the user's public key, decompose the user's public key into multiple elliptic curve point segments, introduce offset perturbations generated by the user's iris image information into each segment, and construct a signature fragment set;

[0016] Step 2: Combine the set of signature fragments with the drone's unique hardware identifier and timestamp to generate the original identity activation signature;

[0017] Step 3: Collect the user's facial micro-expression movements, fingertip touch rhythm, and thermal imaging breathing frequency to generate a behavioral temporal spectrum, and combine it with the original identity activation signature to generate a relay biological behavioral signature, which contains a delayed unlocking structure;

[0018] Step 4: Generate a latent challenge region instruction based on the relay biological behavior signature, and embed the latent challenge region instruction into the control channel. Generate a time-domain latent signature by having the user complete the challenge response at a random time point.

[0019] Step 5: Collect inertial measurement unit data, global positioning system data, magnetometer data and wind speed sensor data during the UAV flight process, generate a spatial behavior map, construct a field kink matrix, and combine the field kink matrix with the time-domain latent signature to generate a behavioral field kink signature;

[0020] Step Six: After the flight is completed, combine the original identity activation signature, relay biological behavior signature, temporal latency signature, and behavioral field knot signature to generate an autophagy closed-loop signature structure.

[0021] Step 7: Perform structural closure processing on the autophagy closed-loop signature structure and store the autophagy closed-loop signature structure in a chained signature data structure for the purpose of verifying the identity of the drone rental service.

[0022] Optionally, step one specifically includes:

[0023] The system receives user authentication requests through the user terminal and retrieves the elliptic curve cryptography public key bound to the user's identity from the authentication server, which serves as the user's public key.

[0024] The user public key is mapped nonlinearly in an increasing manner on the elliptic curve domain, and the user public key data is divided into variable distances on the elliptic curve domain according to a higher-order ratio function to obtain several elliptic curve point segments of unequal lengths.

[0025] The system collects user iris images, extracts iris texture feature maps using Gabor filtering, and generates iris feature vector sets by combining principal direction gradient histogram with local binary patterns.

[0026] The iris feature vector set is subjected to group hash operation to generate a perturbation mask matrix. The perturbation mask matrix is ​​a set of perturbation control parameters corresponding to the elliptic curve point segment. The perturbation control parameters include perturbation rotation angle, perturbation direction identifier and perturbation weight coefficient.

[0027] According to the perturbation control parameters in the perturbation mask matrix, a point rotation perturbation operation is performed on the corresponding elliptic curve point segments. The point rotation perturbation operation takes the center of the curve parameter of each elliptic curve point segment as the rotation reference, performs an affine rotation transformation in the domain of the elliptic curve according to the perturbation rotation angle, determines the perturbation axis according to the perturbation direction identifier, and adjusts the perturbation amplitude under the action of the perturbation weight coefficient to generate a perturbation point segment sequence.

[0028] The perturbated point segment sequence is recombined according to the index order of the perturbation mask matrix to construct a signature fragment set, and the signature fragment set is used as the root key input.

[0029] Optionally, the unique hardware identifier of the UAV includes the unique fixed serial number of the UAV main control chip, and the initial factory bias parameter set of the gyroscope and accelerometer in the inertial measurement unit;

[0030] The unique, fixed serial number is a hardware fingerprint string burned into the read-only storage area of ​​the UAV main control chip, and it is unique and tamper-proof;

[0031] The initial factory bias parameter set is constructed by collecting multiple sets of raw sensor output data continuously while the UAV is stationary and calculating the steady-state mean vector.

[0032] The unique fixed serial number is concatenated with the initial factory bias parameter set and hashed to obtain the UAV device fingerprint digest, which is then combined with the signature fragment set and timestamp to generate the original identity activation signature.

[0033] Optionally, the process of collecting user facial micro-expression movements, fingertip touch rhythm, and thermal imaging breathing frequency to generate a behavioral temporal spectrum, and combining it with the original identity activation signature to generate a relay biometric behavioral signature, specifically:

[0034] The system collects users' facial micro-expression movements, annotates the instantaneous changes of the eyelids, corners of the mouth, and nasal wing muscles with image sequences, and extracts the parameters of relative displacement, angle change rate, and duration between micro-expression movement frames.

[0035] The system collects the user's fingertip touch rhythm and measures the rhythm frequency, time interval, contact pressure change curve, and multi-finger trajectory direction of multi-point contact on an electrostatic capacitive touch screen.

[0036] The system collects the user's thermal imaging respiratory rate, uses an infrared thermal imaging sensor to capture the periodic temperature change waveforms in the nasal ala and upper lip regions, and calculates the rate of change of heat flux and the stability of respiratory rhythm.

[0037] The user's facial micro-expression parameters, the user's fingertip touch rhythm parameters, and the thermal imaging waveform parameters are initially synchronized according to the sampling timestamp. A reference time axis is constructed using a time window sliding mechanism, and the feature stream of each channel is linearly interpolated and phase normalized to generate a time feature matrix under a unified time scale, where each row corresponds to a type of behavior channel and each column corresponds to a standardized time segment.

[0038] In the time feature matrix, the similarity measurement results between behavioral channel pairs are obtained, and the average cross-correlation coefficient is extracted as a cross-channel temporal correlation index.

[0039] Within each channel, a sliding window is used to calculate the local rate of change curve, and the instantaneous abrupt change segment is identified based on the Z-score anomaly detection algorithm. The density of abnormal peaks is counted as the abrupt change point detection value.

[0040] The overlap of peaks and troughs in each channel within the time window is statistically analyzed, and the proportion of overlapping events in the total period is calculated as a waveform co-change factor.

[0041] A stability scoring function is constructed based on the cross-channel time-series correlation index, abnormal peak density, and waveform co-variation factor to calculate the stability score.

[0042] For each time slice, calculate the corresponding stability score, select the continuous time segment with the stability score greater than the set score threshold as the high stability period, extract the channel feature vector sequence of the high stability period and connect them to form a behavioral time series spectrum;

[0043] The behavioral time series spectrum is feature-compressed and a fusion vector is extracted as a perturbation factor; a hash digest is performed on the original identity activation signature to obtain a signature digest value; the fusion vector and the signature digest value are weighted according to the channel stability score, and a hash mask insertion operation is performed to generate a perturbation mixing result; the perturbation mixing result is signed using the user's private key to obtain the relay biometric behavior signature.

[0044] Optionally, the delayed unlock structure includes a delay timer, an activation threshold condition set, and an unlock signal buffer;

[0045] The delay timer is set to delay verification by at least N seconds from the moment the signature is generated.

[0046] The set of activation threshold conditions includes the similarity threshold between the behavioral time series spectrum and the historical behavioral model, the stability threshold of the change amplitude of the behavioral feature vector, and the lower limit threshold of the consistency score of the cooperative fluctuation between behavioral channels.

[0047] The unlock signal buffer is used to cache signature requests and release relay biological behavior signature results after the delay timer and activation threshold conditions are met.

[0048] Optionally, step four specifically includes:

[0049] The relay biological behavior signature is hashed and mapped, and a challenge seed value is generated by combining it with a preset challenge factor table. The challenge seed value is then input into a pseudo-random number generator to generate a latent challenge area instruction. The latent challenge area instruction includes a time trigger point, a channel disturbance type, and response condition parameters.

[0050] The latent challenge area command is encoded and embedded into the control channel of the flight control system in the form of a disturbance frame, forming a latent behavior challenge that is imperceptible to the user.

[0051] During flight control, a challenge event is initiated at a pseudo-randomly set target time point. The drone's sensors detect whether the user completes the behavioral response according to the preset method. The behavioral response includes sudden attitude control, button rhythm input, and micro-inertial gesture matching of the mobile device.

[0052] If an operation matching the response condition parameters is detected within the challenge window, the challenge response is considered successful, and the challenge response result is recorded. The challenge response result includes the challenge response time, behavior feature code, and challenge number.

[0053] The challenge response result is bound to the relay biological behavior signature, hash mixing is performed, and the user key is called to complete the signature generation, resulting in a time-domain latent signature.

[0054] Optionally, step five specifically includes:

[0055] During the flight of the UAV, angular velocity and acceleration data output by the inertial measurement unit, position information provided by the global positioning system, three-axis geomagnetic data output by the magnetometer, and external disturbance wind field data recorded by the wind speed sensor are collected. Timestamp alignment and synchronous interpolation are performed to construct a continuous temporal feature set. Based on trajectory curvature, attitude rotation rate, magnetic vector offset and wind speed disturbance value, a corresponding spatial behavior map is generated. The spatial behavior map reflects the motion trend and environmental interaction characteristics of the UAV in physical space.

[0056] The trajectory curvature, attitude rotation rate, magnetic vector offset, and wind speed disturbance value in the spatial behavior map are encoded in time sequence to construct a multidimensional time series dataset. Tensor coding method is used to establish the interaction relationship between behavior variables to form a field kink matrix with spinor relationship representation.

[0057] The field knot matrix and the time-domain latent signature are hashed and fused together, and the user's private key is used to perform the signature operation to generate the behavioral field knot signature.

[0058] Optionally, step six specifically includes:

[0059] After the flight mission is completed, the original identity activation signature, relay biological behavior signature, temporal latent signature and behavioral field knot signature are structurally integrated in chronological order. The nested chain encapsulation method is used to recursively bind the signature content of each level as part of the signature input of the next level, and generate an autophagy closed-loop signature structure.

[0060] In the self-eating closed-loop signature structure, signature nodes at each level form a chain data structure through hash digest references, and the final signature node contains a hash aggregate digest of all signature digests to ensure integrity verification.

[0061] The self-devouring closed-loop signature structure is structurally indivisible. Any missing, altered, or broken signature layer will cause the overall signature chain verification to fail. It is configured for non-repudiation verification and on-chain responsibility binding and tracing.

[0062] Optionally, step seven specifically includes:

[0063] Add a chain seal flag to the last-level signature node of the self-eating closed-loop signature structure to lock the reference path; at the same time, calculate the closed hash digest of the entire self-eating closed-loop signature structure and mark it as read-only.

[0064] The closed-loop signature structure after the encapsulation process is stored in a chain signature data structure. The chain signature data structure is constructed using multi-node distributed ledger technology and supports sequential verification and traceability verification of signature nodes.

[0065] The chain-signature data structure is interface-bound with the identity management system of the low-altitude shared economy platform to perform multi-level signature consistency comparison of rental users, actual operation behavior and flight records, so as to realize identity verification and responsibility traceability based on the signature chain.

[0066] The beneficial effects of this invention are:

[0067] This invention addresses the problems of single identity authentication methods, coarse authentication granularity, and unsustainable signature binding in existing low-altitude sharing platforms by constructing a multi-layered nested signature structure based on user biometric perturbation and combining UAV equipment hardware fingerprint and flight behavior sequence. It proposes a composite authentication system based on perturbation weighted signature and chained spatiotemporal behavior encapsulation. During the leasing initiation phase, this invention achieves unique dynamic construction of the identity root key through elliptic curve segment decomposition of the user's public key and iris perturbation reconstruction mechanism. During device activation, the invention introduces the drone's main control chip's fixed serial number and initial inertial bias parameters as a hardware fingerprint to construct the original identity activation signature, ensuring the device binding of the authentication chain. During operation, the invention integrates the user's facial micro-expressions, fingertip rhythm, and thermal imaging breathing frequency to construct a multi-channel behavioral temporal spectrum, and extracts high-confidence periods through a stability scoring mechanism to generate a relay biometric signature. During mid-flight, challenge area commands generated by the relay signature are pseudo-randomly implanted into the control channel, combined with micro-behavioral responses for real-time verification, outputting a time-domain latent signature. During late-flight, spatial behavioral variables during flight are collected to construct a spinor tensor field and generate a coupling kink matrix, which is then combined with the time-domain signature to generate a behavioral field kink signature. Finally, multi-layered signature nodes are encapsulated in chronological order to form an indivisible self-eating closed-loop signature structure, completing the chain-like signature storage. Through the above technical solution, the present invention realizes the dynamic integration and structured binding of identity signature, flight behavior, and device entity, effectively improving the real-time performance, anti-counterfeiting and traceability of identity authentication during the leasing process, and enhancing the platform's closed-loop verification capability for flight behavior responsibility. Attached Figure Description

[0068] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0069] Figure 1 This is a schematic diagram of the identity verification system for drone rental services on the low-altitude shared economy platform proposed in this invention.

[0070] Figure 2 This is an overall flowchart of the identity verification method for drone rental services on the low-altitude shared economy platform proposed in this invention. Detailed Implementation

[0071] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0072] refer to Figure 1 The drone rental service identity verification system of the low-altitude sharing economy platform includes the following modules:

[0073] The public key perturbation module is used to obtain the user's public key, decompose the user's public key into multiple elliptic curve point segments, introduce offset perturbations generated by the user's iris image information, and construct a signature fragment set.

[0074] The activation signature module is used to combine the signature fragment set with the drone's unique hardware identifier and timestamp to generate the original identity activation signature;

[0075] The behavior signature module is used to generate a behavior time sequence spectrum and combine it with the original identity activation signature to generate a relay bio behavior signature, wherein the relay bio behavior signature includes a delayed unlocking structure.

[0076] The challenge signature module is used to generate a latent challenge area instruction based on the relay biological behavior signature and embed it in the control channel. The user completes the challenge response at a random time point to generate a time-domain latent signature.

[0077] The spatial signature module is used to generate a spatial behavior map, construct a field knot matrix, and combine the field knot matrix with the temporal latent signature to generate a behavior field knot signature.

[0078] The signature integration module is used to combine the original identity activation signature, relay biological behavior signature, temporal latency signature and behavioral field knot signature to generate an autophagy closed-loop signature structure.

[0079] The chain-based evidence storage module is used to enclose the self-eating closed-loop signature structure and write it into the chain-based signature data structure for identity verification and accountability.

[0080] refer to Figure 2 The identity verification method for drone rental services on low-altitude shared economy platforms includes the following steps:

[0081] Step 1: Obtain the user's public key, decompose the user's public key into multiple elliptic curve point segments, introduce offset perturbations generated by the user's iris image information into each segment, and construct a signature fragment set;

[0082] Step 2: Combine the set of signature fragments with the drone's unique hardware identifier and timestamp to generate the original identity activation signature;

[0083] Step 3: Collect the user's facial micro-expression movements, fingertip touch rhythm, and thermal imaging breathing frequency to generate a behavioral temporal spectrum, and combine it with the original identity activation signature to generate a relay biological behavioral signature, which contains a delayed unlocking structure;

[0084] Step 4: Generate a latent challenge region instruction based on the relay biological behavior signature, and embed the latent challenge region instruction into the control channel. Generate a time-domain latent signature by having the user complete the challenge response at a random time point.

[0085] Step 5: Collect inertial measurement unit data, global positioning system data, magnetometer data and wind speed sensor data during the UAV flight process, generate a spatial behavior map, construct a field kink matrix, and combine the field kink matrix with the time-domain latent signature to generate a behavioral field kink signature;

[0086] Step Six: After the flight is completed, combine the original identity activation signature, relay biological behavior signature, temporal latency signature, and behavioral field knot signature to generate an autophagy closed-loop signature structure.

[0087] Step 7: Perform structural closure processing on the autophagy closed-loop signature structure and store the autophagy closed-loop signature structure in a chained signature data structure for the purpose of verifying the identity of the drone rental service.

[0088] In this embodiment, step one specifically includes:

[0089] The system receives user authentication requests through the user terminal and retrieves the elliptic curve cryptography public key bound to the user's identity from the authentication server, which serves as the user's public key.

[0090] The user public key is mapped nonlinearly in an increasing manner on the elliptic curve domain, and the user public key data is divided into variable distances on the elliptic curve domain according to a higher-order ratio function to obtain several elliptic curve point segments of unequal lengths.

[0091] The system collects user iris images, extracts iris texture feature maps using Gabor filtering, and generates iris feature vector sets by combining principal direction gradient histogram with local binary patterns.

[0092] The iris feature vector set is subjected to group hash operation to generate a perturbation mask matrix. The perturbation mask matrix is ​​a set of perturbation control parameters corresponding to the elliptic curve point segment. The perturbation control parameters include perturbation rotation angle, perturbation direction identifier and perturbation weight coefficient.

[0093] In this embodiment of the invention, to enhance the individual uniqueness and irreversibility of the elliptic curve segment perturbation process, a technical approach is adopted to generate a perturbation mask matrix by performing grouped hash operations on the iris feature vector set. The specific steps are as follows:

[0094] The iris feature vector set, jointly encoded by Gabor filtering, principal direction gradient histogram, and local binary pattern (LBP), is then divided into fixed lengths. The data is divided into equal groups to obtain several iris sub-vector groups. Each iris sub-vector group is used as a perturbation seed input to the hash function, and a hash digest value is generated using the SHA-3 cryptographic hash algorithm to control the perturbation behavior.

[0095] For each hash digest value, perform structural parsing and extract different bit segments as the source for generating different perturbation parameters:

[0096] Perturbation rotation angle: Truncate 10 bits from the high-order bits of the hash digest value and map them proportionally to a preset angle range. ), used to control the rotation amplitude of elliptic curve point segments around their parameter center;

[0097] Perturbation direction indicator: extracted from the lowest two bits of the median segment, with the following values: The perturbation directions are mapped to the positive and negative quadrants of the X and Y axes, respectively;

[0098] Perturbation weighting coefficient: 10 consecutive bits are truncated from the least significant bit of the hash digest value and normalized to a certain value. The range is used to adjust the perturbation intensity, ensuring that the perturbation amplitude strikes a balance between individual identification accuracy and security.

[0099] Finally, the three sets of perturbation parameters are combined to form a complete perturbation control parameter triplet, and the same operation is performed on all iris sub-vector groups to form a perturbation mask matrix.

[0100] The perturbation mask matrix corresponds one-to-one with the set of elliptic curve point segments. It is called item by item during the elliptic curve point rotation perturbation stage, ensuring that the perturbation operation of each point segment has randomness and non-reconstructability driven by unique biometric features, which greatly enhances the anti-forgery capability and identity binding accuracy of the signature fragment set.

[0101] According to the perturbation control parameters in the perturbation mask matrix, a point rotation perturbation operation is performed on the corresponding elliptic curve point segments. The point rotation perturbation operation takes the center of the curve parameter of each elliptic curve point segment as the rotation reference, performs an affine rotation transformation in the domain of the elliptic curve according to the perturbation rotation angle, determines the perturbation axis according to the perturbation direction identifier, and adjusts the perturbation amplitude under the action of the perturbation weight coefficient to generate a perturbation point segment sequence.

[0102] The perturbated point segment sequence is recombined according to the index order of the perturbation mask matrix to construct a signature fragment set, and the signature fragment set is used as the root key input.

[0103] In this embodiment, the unique hardware identifier of the UAV includes the unique fixed serial number of the UAV main control chip, and the initial factory bias parameter set of the gyroscope and accelerometer in the inertial measurement unit.

[0104] The unique, fixed serial number is a hardware fingerprint string burned into the read-only storage area of ​​the UAV main control chip, and it is unique and tamper-proof;

[0105] The initial factory bias parameter set is constructed by collecting multiple sets of raw sensor output data continuously while the UAV is stationary and calculating the steady-state mean vector.

[0106] The unique fixed serial number is concatenated with the initial factory bias parameter set and hashed to obtain the UAV device fingerprint digest, which is then combined with the signature fragment set and timestamp to generate the original identity activation signature.

[0107] In this embodiment, the process of collecting user facial micro-expression movements, fingertip touch rhythm, and thermal imaging breathing frequency to generate a behavioral temporal spectrum, and combining it with the original identity activation signature to generate a relay biometric behavioral signature, specifically involves:

[0108] The system collects users' facial micro-expression movements, annotates the instantaneous changes of the eyelids, corners of the mouth, and nasal wing muscles with image sequences, and extracts the parameters of relative displacement, angle change rate, and duration between micro-expression movement frames.

[0109] The system collects the user's fingertip touch rhythm and measures the rhythm frequency, time interval, contact pressure change curve, and multi-finger trajectory direction of multi-point contact on an electrostatic capacitive touch screen.

[0110] The system collects the user's thermal imaging respiratory rate, uses an infrared thermal imaging sensor to capture the periodic temperature change waveforms in the nasal ala and upper lip regions, and calculates the rate of change of heat flux and the stability of respiratory rhythm.

[0111] The user's facial micro-expression parameters, the user's fingertip touch rhythm parameters, and the thermal imaging waveform parameters are initially synchronized according to the sampling timestamp. A reference time axis is constructed using a time window sliding mechanism, and the feature stream of each channel is linearly interpolated and phase normalized to generate a time feature matrix under a unified time scale, where each row corresponds to a type of behavior channel and each column corresponds to a standardized time segment.

[0112] In the time feature matrix, the similarity measurement results between behavioral channel pairs are obtained, and the average cross-correlation coefficient is extracted as a cross-channel temporal correlation index.

[0113] In the time feature matrix, the time series of any two behavioral channels are first paired, and the time series of the two behavioral channels are nonlinearly aligned using the dynamic time warping method to obtain a set of shortest path matching sequences. The matching process allows for local compression or expansion of the time axis, so that the behavioral change trends of the two channels can establish a time mapping relationship under the premise of not being completely synchronized.

[0114] Set up a channel and channels The aligned mapped time series is and After alignment, the following normalized cross-correlation coefficients are calculated based on the alignment index. :

[0115] ;

[0116] in, , These represent the mean values ​​of the time series for each channel. , Indicates the corresponding standard deviation. This indicates the number of aligned time segments in the behavioral time series.

[0117] For all channel pairing combinations Repeat the above cross-correlation coefficient calculation. Calculate and average the cross-correlation coefficients of all channel pairs to serve as an indicator of overall behavioral channel consistency, i.e., cross-channel temporal correlation value. :

[0118] ;

[0119] in, Indicates the number of behavior channels.

[0120] Within each channel, a sliding window is used to calculate the local rate of change curve, and the instantaneous abrupt change segment is identified based on the Z-score anomaly detection algorithm. The density of abnormal peaks is counted as the abrupt change point detection value.

[0121] The overlap of peaks and troughs in each channel within the time window is statistically analyzed, and the proportion of overlapping events in the total period is calculated as a waveform co-change factor.

[0122] A stability scoring function is constructed based on the cross-channel time-series correlation index, abnormal peak density, and waveform co-variation factor to calculate the stability score.

[0123] ;

[0124] in, Indicates the stability score. Indicators representing cross-channel time-series correlation. Indicates abnormal peak density, Indicates the waveform covariance factor. and This represents the preset weighting coefficients, which satisfy... ;

[0125] For each time slice, calculate the corresponding stability score, select the continuous time segment with the stability score greater than the set score threshold as the high stability period, extract the channel feature vector sequence of the high stability period and connect them to form a behavioral time series spectrum;

[0126] The behavioral time series spectrum is feature-compressed and a fusion vector is extracted as a perturbation factor; a hash digest is performed on the original identity activation signature to obtain a signature digest value; the fusion vector and the signature digest value are weighted according to the channel stability score, and a hash mask insertion operation is performed to generate a perturbation mixing result; the perturbation mixing result is signed using the user's private key to obtain the relay biometric behavior signature.

[0127] In this embodiment, the delayed unlocking structure includes a delay timer, an activation threshold condition set, and an unlocking signal buffer;

[0128] The delay timer is set to delay verification by at least N seconds from the moment the signature is generated.

[0129] The set of activation threshold conditions includes the similarity threshold between the behavioral time series spectrum and the historical behavioral model, the stability threshold of the change amplitude of the behavioral feature vector, and the lower limit threshold of the consistency score of the cooperative fluctuation between behavioral channels.

[0130] By setting similarity thresholds between the behavioral time-series spectrum and historical behavioral models, stability thresholds for behavioral feature vector change amplitudes, and lower limits for consistency scores of coordinated fluctuations between behavioral channels, the validity of relay biometric signatures is dynamically verified and controlled, serving as conditional triggering criteria in the delayed unlocking structure. When a user generates a relay biometric signature, the system does not immediately allow the signature to proceed to the next verification or chain encapsulation stage, but instead enters a preset delayed judgment phase.

[0131] During this phase, the system compares the current behavior time-series spectrum with the user's historical behavior model to calculate a similarity index; it also assesses whether the change amplitude of the current behavior feature vector is within a preset stable range; and analyzes whether the coordinated fluctuations between multi-channel behavior curves meet the lower limit of the consistency score. An unlocking signal is triggered only when all the above indicators meet the corresponding set thresholds, making the relay biometric signature valid after the delay period and allowing it to proceed to the next authentication chain level.

[0132] The unlock signal buffer is used to cache signature requests and release relay biological behavior signature results after the delay timer and activation threshold conditions are met.

[0133] In this embodiment, step four specifically includes:

[0134] The relay biological behavior signature is hashed and mapped, and a challenge seed value is generated by combining it with a preset challenge factor table. The challenge seed value is then input into a pseudo-random number generator to generate a latent challenge area instruction. The latent challenge area instruction includes a time trigger point, a channel disturbance type, and response condition parameters.

[0135] The latent challenge area command is encoded and embedded into the control channel of the flight control system in the form of a disturbance frame, forming a latent behavior challenge that is imperceptible to the user.

[0136] During flight control, a challenge event is initiated at a pseudo-randomly set target time point. The drone's sensors detect whether the user completes the behavioral response according to the preset method. The behavioral response includes sudden attitude control, button rhythm input, and micro-inertial gesture matching of the mobile device.

[0137] If an operation matching the response condition parameters is detected within the challenge window, the challenge response is considered successful, and the challenge response result is recorded. The challenge response result includes the challenge response time, behavior feature code, and challenge number.

[0138] The challenge response result is bound to the relay biological behavior signature, hash mixing is performed, and the user key is called to complete the signature generation, resulting in a time-domain latent signature.

[0139] In this embodiment, step five specifically includes:

[0140] During the flight of the UAV, angular velocity and acceleration data output by the inertial measurement unit, position information provided by the global positioning system, three-axis geomagnetic data output by the magnetometer, and external disturbance wind field data recorded by the wind speed sensor are collected. Timestamp alignment and synchronous interpolation are performed to construct a continuous temporal feature set. Based on trajectory curvature, attitude rotation rate, magnetic vector offset and wind speed disturbance value, a corresponding spatial behavior map is generated. The spatial behavior map reflects the motion trend and environmental interaction characteristics of the UAV in physical space.

[0141] The trajectory curvature, attitude rotation rate, magnetic vector offset, and wind speed disturbance value in the spatial behavior map are encoded in time sequence to construct a multidimensional time series dataset. Tensor coding method is used to establish the interaction relationship between behavior variables to form a field kink matrix with spinor relationship representation.

[0142] Specifically, the trajectory curvature, attitude rotation rate, magnetic vector offset, and wind speed disturbance value in the spatial behavior map are used as feature variables of four physical dimensions, and time-series encoding is performed in the sampling order under a unified time axis to form a multi-dimensional time series dataset.

[0143] The trajectory curvature is calculated from the rate of change of direction of the flight path per unit time, the attitude rotation rate comes from the differential value of the Euler angle provided by the inertial measurement unit, the magnetic force vector offset is obtained from the difference between the current geomagnetic field and the preset reference geomagnetic field, and the wind speed disturbance value comes from the three-axis wind force change of the wind speed sensor.

[0144] The data from the above dimensions are combined to construct a high-order tensor structure. Tensor coding is used to model the temporal correlation between different behavioral dimensions. The main correlation components are extracted through tensor decomposition, key perturbation regions are further extracted, and spinor coupling intervals are identified to represent the continuous nonlinear change trend of multidimensional behavior in the spatial domain. This forms a field kink matrix containing rotational coupling features, which serves as the core representation of the spatial behavioral structure.

[0145] The field knot matrix and the time-domain latent signature are hashed and fused together, and the user's private key is used to perform the signature operation to generate the behavioral field knot signature.

[0146] In this embodiment, step six specifically includes:

[0147] After the flight mission is completed, the original identity activation signature, relay biological behavior signature, temporal latent signature and behavioral field knot signature are structurally integrated in chronological order. The nested chain encapsulation method is used to recursively bind the signature content of each level as part of the signature input of the next level, and generate an autophagy closed-loop signature structure.

[0148] In the self-eating closed-loop signature structure, signature nodes at each level form a chain data structure through hash digest references, and the final signature node contains a hash aggregate digest of all signature digests to ensure integrity verification.

[0149] The self-devouring closed-loop signature structure is structurally indivisible. Any missing, altered, or broken signature layer will cause the overall signature chain verification to fail. It is configured for non-repudiation verification and on-chain responsibility binding and tracing.

[0150] In this embodiment, step seven specifically includes:

[0151] Add a chain seal flag to the last-level signature node of the self-eating closed-loop signature structure to lock the reference path; at the same time, calculate the closed hash digest of the entire self-eating closed-loop signature structure and mark it as read-only.

[0152] The closed-loop signature structure after the encapsulation process is stored in a chain signature data structure. The chain signature data structure is constructed using multi-node distributed ledger technology and supports sequential verification and traceability verification of signature nodes.

[0153] The chain-signature data structure is interface-bound with the identity management system of the low-altitude shared economy platform to perform multi-level signature consistency comparison of rental users, actual operation behavior and flight records, so as to realize identity verification and responsibility traceability based on the signature chain.

[0154] Example 1:

[0155] To verify the feasibility of this invention in practice, it was applied to a simulated operation environment built by a shared drone rental platform. The scenarios covered typical application modes such as short-distance logistics delivery, emergency visual inspection, and high-frequency round-trip transportation within the park. The system simulated various rental processes, including registration, activation, flight control, challenge interaction, and flight closed-loop confirmation, comprehensively examining the performance of the proposed layered signature structure in user identity verification, behavior consistency verification, and flight responsibility binding.

[0156] In this experimental system, users initiate rental requests via an authentication terminal. The platform uses the encrypted public key bound to the user to perform nonlinear perturbation, generating a set of signature fragments. This, combined with the rental device's unique hardware fingerprint and the current timestamp, generates the original identity activation signature. Before initiating the flight mission, the system guides the user through a behavioral acquisition process based on facial micro-expressions, fingertip rhythm, and thermal imaging breathing frequency, constructing a behavioral temporal spectrum. The system extracts the dynamic coordination relationships between channels, calculates the average cross-correlation and waveform abrupt change, and selects high-confidence time-segment features using a stability scoring function. This is then combined with the original signature to generate a relay biometric signature. This signature contains a delayed unlocking structure, allowing for dynamic determination of behavioral consistency.

[0157] During UAV flight, the system embeds latent challenge commands generated from behavioral signatures into the flight control channel and randomly triggers behavioral response tasks at unpredictable times to detect whether the operator has completed the required attitude control, input rhythm, or micro-inertial device response. Successful challenge results are bound to relay behavioral signatures to generate time-domain latent signatures. During flight, the device synchronously collects trajectory, attitude, magnetic field, and wind disturbance data to generate a spatial behavioral map and construct a field knot matrix, which is further fused with latent signatures to generate behavioral field knot signatures. Finally, the four levels of signatures are uniformly encapsulated into a closed-loop signature structure and written into a multi-node chained ledger.

[0158] The platform evaluates the system by using 100 randomly generated flight missions, comparing the performance of the system with that of a traditional account + facial recognition mechanism across multiple dimensions.

[0159] Table 1 Performance Evaluation Table of Chained Signature Traceability

[0160] Sample number System type Average response latency to challenge (ms) Successful response rate non-response rate Signature chain closure integrity rate User Behavior Consistency Index A01-A25 This invention system 386 95.2% 2.1% 100% 0.916 A26-A50 This invention system 371 96.5% 1.7% 100% 0.928 A51-A75 Control system No challenge mechanism No data No data not applicable 0.701 A76-A100 Control system No challenge mechanism No data No data not applicable 0.682

[0161] The data analysis in Table 1 shows that the system of this invention outperforms the control system in terms of chain signature response capability, integrity guarantee, and user behavior consistency. Firstly, regarding the average latency of challenge response, the system of this invention provides samples A01-A25 and A26-A50 with latency of 386 milliseconds and 371 milliseconds respectively, with the overall latency kept below 400 milliseconds, demonstrating excellent real-time performance. In contrast, the control system, lacking a challenge mechanism, failed to provide latency data, indicating a deficiency in its chain behavior verification capability.

[0162] Regarding response quality, the successful response rates of the system of this invention were 95.2% and 96.5%, respectively, while the non-response rates were only 2.1% and 1.7%, indicating that the user's response stability and system compatibility are good under random challenge conditions. In contrast, the control system had no relevant data available for evaluation without a challenge mechanism, reflecting its lack of ability to verify the operator's identity in real time.

[0163] Regarding the signature chain integrity rate, the system of this invention achieved 100% in both sample groups, indicating that its nested signature structure can effectively complete the signature integration at all stages without data gaps or verification failures. The control system, however, failed to complete the closed-loop structure due to the lack of a challenge process and chain encapsulation mechanism.

[0164] The User Behavior Consistency Index, a core indicator for measuring the stability of user behavior patterns across different stages, was achieved by the system of this invention at 0.916 and 0.928, respectively, which are significantly higher than the control system's 0.701 and 0.682. This indicates that the system has higher accuracy and discriminative power in the coordination of biological behavior modeling and challenge response.

[0165] Table 2 Performance Evaluation Table of Chained Signature Traceability

[0166] Flight number Total number of signature nodes Tampering with simulated location Verification time (ms) Backtracking positioning accuracy Tampering with blocking effect Data integrity score (out of 1) T01 4 Relay behavior node 112 100% success 1.00 T02 4 Temporal Latent Nodes 109 100% success 0.99 T03 4 Last level encapsulation node 121 100% success 1.00 T04 4 Original identity node 116 100% success 0.98

[0167] As can be seen from Table 2, the chain signature system constructed in this invention exhibits extremely high traceability accuracy, response speed and data recovery capability when faced with node-level data tampering, verifying its practicality and robustness in identity verification and accountability.

[0168] Each of the four flight missions, from flight number T01 to T04, was configured with a complete set of four signature nodes. Simulated tampering operations were implemented at the relay behavior node, the time-domain latent node, the final-level encapsulation node, and the original identity node. These nodes cover different key links in the signature chain and are representative. For each tampered node, the system achieved 100% backtracking accuracy without external intervention, demonstrating that the nested signature structure and chain indexing mechanism in this invention can accurately pinpoint the location of anomalies, ensuring the system's ability to immediately detect errors.

[0169] In terms of verification response time, the tamper verification time of each node is between 109ms and 121ms, with an average of only 114.5ms, which is enough to complete the integrity verification of the chain structure and the identification of abnormal nodes. This reflects the real-time characteristics of the system and is suitable for the needs of rapid delivery and instant auditing in dynamic leasing scenarios.

[0170] Meanwhile, judging from the tampering blocking effect, all nodes are marked as successful, indicating that the signature chain has a tampering and chain-breaking mechanism, which can prevent subsequent chains from continuing to take effect when the signatures are inconsistent or the content is replaced, thereby preventing data speculation or subsequent forgery.

[0171] In terms of data integrity scoring, the relay behavior node and the last-level encapsulation node achieved a perfect score of 1.00, indicating that the signature node after chain encapsulation still maintains the consistency of data state even under attack. Due to the large number of dimensions of associated behavior data, fluctuations are more sensitive. The scores of the time-domain latent node and the original identity node decreased slightly (0.99 and 0.98 respectively), but still remained at a high integrity level and did not affect the overall system performance.

[0172] This embodiment constructs a chain-like signature structure that fuses multi-dimensional, multi-source heterogeneous data, enabling dynamic perception of user identity, behavioral consistency judgment, and full-process traceability verification during drone rental. This significantly improves the robustness and security of identity verification. Compared to traditional single authentication mechanisms, this invention effectively prevents risks such as forged control commands, illegal manipulation, and post-event data tampering, ensuring the immutability and traceability of identity authentication results. It provides a highly reliable and trustworthy identity security mechanism for low-altitude shared economy platforms.

[0173] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An identity verification system for drone rental services on a low-altitude shared economy platform, characterized in that: Includes the following modules: The public key perturbation module is used to obtain the user's public key, decompose the user's public key into multiple elliptic curve point segments, introduce offset perturbations generated by the user's iris image information, and construct a signature fragment set. The activation signature module is used to combine the signature fragment set with the drone's unique hardware identifier and timestamp to generate the original identity activation signature; The behavior signature module is used to generate a behavior time sequence spectrum and combine it with the original identity activation signature to generate a relay bio behavior signature, wherein the relay bio behavior signature includes a delayed unlocking structure. The challenge signature module is used to generate a latent challenge area instruction based on the relay biological behavior signature and embed it in the control channel. The user completes the challenge response at a random time point to generate a time-domain latent signature. The spatial signature module is used to generate a spatial behavior map, construct a field knot matrix, and combine the field knot matrix with the temporal latent signature to generate a behavior field knot signature. The signature integration module is used to combine the original identity activation signature, relay biological behavior signature, temporal latency signature and behavioral field knot signature to generate an autophagy closed-loop signature structure. The chain-based evidence storage module is used to enclose the self-eating closed-loop signature structure and write it into the chain-based signature data structure for identity verification and accountability.

2. The drone rental service identity verification system of the low-altitude shared economy platform according to claim 1, characterized in that, The modules are connected in the following way: Step 1: Obtain the user's public key, decompose the user's public key into multiple elliptic curve point segments, introduce offset perturbations generated by the user's iris image information into each segment, and construct a signature fragment set; Step 2: Combine the set of signature fragments with the drone's unique hardware identifier and timestamp to generate the original identity activation signature; Step 3: Collect the user's facial micro-expression movements, fingertip touch rhythm, and thermal imaging breathing frequency to generate a behavioral temporal spectrum, and combine it with the original identity activation signature to generate a relay biological behavioral signature, which contains a delayed unlocking structure; Step 4: Generate a latent challenge region instruction based on the relay biological behavior signature, and embed the latent challenge region instruction into the control channel. The user completes the challenge response at a random time point to generate a time-domain latent signature. Step 5: Collect inertial measurement unit data, global positioning system data, magnetometer data and wind speed sensor data during the UAV flight process, generate a spatial behavior map, construct a field kink matrix, and combine the field kink matrix with the time-domain latent signature to generate a behavioral field kink signature; Step Six: After the flight is completed, combine the original identity activation signature, relay biological behavior signature, temporal latency signature, and behavioral field knot signature to generate an autophagy closed-loop signature structure. Step 7: Perform structural closure processing on the autophagy closed-loop signature structure and store the autophagy closed-loop signature structure in a chained signature data structure for identity verification of drone rental services.

3. The drone rental service identity verification system of the low-altitude shared economy platform according to claim 2, characterized in that, Step one specifically involves: The system receives user authentication requests through the user terminal and retrieves the elliptic curve cryptography public key bound to the user's identity from the authentication server, which serves as the user's public key. The user public key is mapped nonlinearly in an increasing manner on the elliptic curve domain, and the user public key data is divided into variable distances on the elliptic curve domain according to a higher-order ratio function to obtain several elliptic curve point segments of unequal lengths. The system collects user iris images, extracts iris texture feature maps using Gabor filtering, and generates iris feature vector sets by combining principal direction gradient histogram with local binary patterns. The iris feature vector set is subjected to group hash operation to generate a perturbation mask matrix. The perturbation mask matrix is ​​a set of perturbation control parameters corresponding to the elliptic curve point segment. The perturbation control parameters include perturbation rotation angle, perturbation direction identifier and perturbation weight coefficient. According to the perturbation control parameters in the perturbation mask matrix, a point rotation perturbation operation is performed on the corresponding elliptic curve point segments. The point rotation perturbation operation takes the center of the curve parameter of each elliptic curve point segment as the rotation reference, performs an affine rotation transformation in the domain of the elliptic curve according to the perturbation rotation angle, determines the perturbation axis according to the perturbation direction identifier, and adjusts the perturbation amplitude under the action of the perturbation weight coefficient to generate a perturbation point segment sequence. The perturbated point segment sequence is recombined according to the index order of the perturbation mask matrix to construct a signature fragment set, and the signature fragment set is used as the root key input.

4. The drone rental service identity verification system of the low-altitude shared economy platform according to claim 2, characterized in that, The unique hardware identifier of the UAV includes the unique fixed serial number of the UAV main control chip, and the initial factory bias parameter set of the gyroscope and accelerometer in the inertial measurement unit; The unique, fixed serial number is a hardware fingerprint string burned into the read-only storage area of ​​the UAV main control chip, and it is unique and tamper-proof; The initial factory bias parameter set is constructed by collecting multiple sets of raw sensor output data continuously while the UAV is stationary and calculating the steady-state mean vector. The unique fixed serial number is concatenated with the initial factory bias parameter set and hashed to obtain the UAV device fingerprint digest, which is then combined with the signature fragment set and timestamp to generate the original identity activation signature.

5. The drone rental service identity verification system of the low-altitude shared economy platform according to claim 2, characterized in that, The process involves collecting user facial micro-expressions, fingertip touch rhythms, and thermal imaging breathing frequencies to generate a behavioral temporal spectrum. This spectrum is then combined with the original identity activation signature to generate a relay biometric behavioral signature, specifically: The system collects users' facial micro-expression movements, annotates the instantaneous changes of the eyelids, corners of the mouth, and nasal wing muscles with image sequences, and extracts the parameters of relative displacement, angle change rate, and duration between micro-expression movement frames. The system collects the user's fingertip touch rhythm and measures the rhythm frequency, time interval, contact pressure change curve, and multi-finger trajectory direction of multi-point contact on an electrostatic capacitive touch screen. The system collects the user's thermal imaging respiratory rate, uses an infrared thermal imaging sensor to capture the periodic temperature change waveforms in the nasal ala and upper lip regions, and calculates the rate of change of heat flux and the stability of respiratory rhythm. The user's facial micro-expression parameters, the user's fingertip touch rhythm parameters, and the thermal imaging waveform parameters are initially synchronized according to the sampling timestamp. A reference time axis is constructed using a time window sliding mechanism, and the feature stream of each channel is linearly interpolated and phase normalized to generate a time feature matrix under a unified time scale, where each row corresponds to a type of behavior channel and each column corresponds to a standardized time segment. In the time feature matrix, the similarity measurement results between behavioral channel pairs are obtained, and the average cross-correlation coefficient is extracted as a cross-channel temporal correlation index. Within each channel, a sliding window is used to calculate the local rate of change curve, and the instantaneous abrupt change segment is identified based on the Z-score anomaly detection algorithm. The density of abnormal peaks is counted as the abrupt change point detection value. The overlap of peaks and troughs in each channel within the time window is statistically analyzed, and the proportion of overlapping events in the total period is calculated as a waveform co-change factor. A stability scoring function is constructed based on the cross-channel time-series correlation index, abnormal peak density, and waveform co-variation factor to calculate the stability score. For each time slice, calculate the corresponding stability score, select the continuous time segment with the stability score greater than the set score threshold as the high stability period, extract the channel feature vector sequence of the high stability period and connect them to form a behavioral time series spectrum; The behavioral time series spectrum is feature-compressed and a fusion vector is extracted as a perturbation factor; a hash digest is performed on the original identity activation signature to obtain a signature digest value; the fusion vector and the signature digest value are weighted according to the channel stability score, and a hash mask insertion operation is performed to generate a perturbation mixing result; the perturbation mixing result is signed using the user's private key to obtain the relay biometric behavior signature.

6. The drone rental service identity verification system of the low-altitude shared economy platform according to claim 2, characterized in that, The delayed unlocking structure includes a delay timer, an activation threshold condition set, and an unlocking signal buffer; The delay timer is set to delay verification by at least N seconds from the moment the signature is generated. The set of activation threshold conditions includes the similarity threshold between the behavioral time series spectrum and the historical behavioral model, the stability threshold of the change amplitude of the behavioral feature vector, and the lower limit threshold of the consistency score of the cooperative fluctuation between behavioral channels. The unlock signal buffer is used to cache signature requests and release relay biological behavior signature results after the delay timer and activation threshold conditions are met.

7. The drone rental service identity verification system of the low-altitude shared economy platform according to claim 2, characterized in that, Step four specifically involves: The relay biological behavior signature is hashed and mapped, and a challenge seed value is generated by combining it with a preset challenge factor table. The challenge seed value is then input into a pseudo-random number generator to generate a latent challenge area instruction. The latent challenge area instruction includes a time trigger point, a channel disturbance type, and response condition parameters. The latent challenge area command is encoded and embedded into the control channel of the flight control system in the form of a disturbance frame, forming a latent behavior challenge that is imperceptible to the user. During flight control, a challenge event is initiated at a pseudo-randomly set target time point. The drone's sensors detect whether the user completes the behavioral response according to the preset method. The behavioral response includes sudden attitude control, button rhythm input, and micro-inertial gesture matching of the mobile device. If an operation matching the response condition parameters is detected within the challenge window, the challenge response is considered successful, and the challenge response result is recorded. The challenge response result includes the challenge response time, behavior feature code, and challenge number. The challenge response result is bound to the relay biological behavior signature, hash mixing is performed, and the user key is called to complete the signature generation, resulting in a time-domain latent signature.

8. The drone rental service identity verification system of the low-altitude sharing economy platform according to claim 2, characterized in that, Step five specifically involves: During the flight of the UAV, angular velocity and acceleration data output by the inertial measurement unit, position information provided by the global positioning system, three-axis geomagnetic data output by the magnetometer, and external disturbance wind field data recorded by the wind speed sensor are collected. Timestamp alignment and synchronous interpolation are performed to construct a continuous temporal feature set. Based on trajectory curvature, attitude rotation rate, magnetic vector offset and wind speed disturbance value, a corresponding spatial behavior map is generated. The spatial behavior map reflects the motion trend and environmental interaction characteristics of the UAV in physical space. The trajectory curvature, attitude rotation rate, magnetic vector offset, and wind speed disturbance value in the spatial behavior map are encoded in time sequence to construct a multidimensional time series dataset. Tensor coding method is used to establish the interaction relationship between behavior variables to form a field kink matrix with spinor relationship representation. The field knot matrix and the time-domain latent signature are hashed and fused together, and the user's private key is used to perform the signature operation to generate the behavioral field knot signature.

9. The drone rental service identity verification system of the low-altitude sharing economy platform according to claim 2, characterized in that, Step six specifically involves: After the flight mission is completed, the original identity activation signature, relay biological behavior signature, temporal latent signature and behavior field knot signature are structurally integrated in chronological order. The nested chain encapsulation method is used to recursively bind the signature content of each level as part of the signature input of the next level, and generate an autophagy closed loop signature structure. In the self-eating closed-loop signature structure, signature nodes at each level form a chain data structure through hash digest references, and the final signature node contains a hash aggregate digest of all signature digests. The self-devouring closed-loop signature structure is structurally indivisible. Any missing, altered, or broken signature layer will cause the overall signature chain verification to fail. It is configured for non-repudiation verification and on-chain responsibility binding and tracing.

10. The drone rental service identity verification system of the low-altitude shared economy platform according to claim 2, characterized in that, Step seven specifically involves: Add a chain seal flag to the last-level signature node of the self-eating closed-loop signature structure to lock the reference path; at the same time, calculate the closed hash digest of the entire self-eating closed-loop signature structure and mark it as read-only. The closed-loop signature structure after the encapsulation process is stored in a chain signature data structure. The chain signature data structure is constructed using multi-node distributed ledger technology and supports sequential verification and traceability verification of signature nodes. The chain-signature data structure is interface-bound with the identity management system of the low-altitude shared economy platform to perform multi-level signature consistency comparison of rental users, actual operation behavior and flight records, so as to realize identity verification and responsibility traceability based on the signature chain.

Images