Method, system and apparatus for detecting elastic fabric

Abstract

The application provides a kind of elastic fabric detection method, system and equipment, applied to the detection system including elastic fabric sensor network;The sensor is embedded in the elastic fabric;The detection method comprises: establishing the first node and the second node of the elastic fabric sensor network;The first node includes the sensor;First data of the first node is acquired;The first data is encrypted to obtain encrypted first data;The second node is used to perform ciphertext domain fusion processing on the encrypted first data, to obtain the feature matrix of the elastic fabric;The feature matrix represents the correlation between the grid coordinates of the elastic fabric and the data of the sensor;The feature matrix is encrypted and sent to the terminal, so that the terminal decrypts the encrypted feature matrix to generate a detection report of the elastic fabric;The detection report is used to determine the use of the elastic fabric.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent wear and biosensing technology detection, and in particular to a detection method, system and device of elastic fabric. BACKGROUND

[0002] With the rapid development of intelligent wearable technology, accurate detection of temperature and pressure data related to textile touch has become a core requirement for functional clothing design, medical rehabilitation aid development and sports equipment optimization. However, the existing technical system has the following bottlenecks: single data dimension, insufficient touch feature modeling, existing textile touch detection technology is mostly based on independent measurement of a single physical quantity, lacking comprehensive analysis capability of spatio-temporal correlation of multi-modal data such as temperature and pressure, leading to one-sided touch feature extraction, which cannot accurately represent the dynamic comfort change of textiles in complex scenarios. Poor dynamic adaptability, lack of real-time feedback, traditional sensor network uses fixed sampling frequency and static threshold judgment mechanism, which is difficult to adapt to the rapid nonlinear change of touch parameters in the human motion process, and the data processing link delay is high, which cannot realize millisecond-level real-time warning, seriously limiting the application value in dynamic scenarios. Insufficient collaboration between hardware and algorithms, existing solutions focus on sensor hardware optimization, but do not realize deep collaborative design of algorithms and hardware, leading to a prominent contradiction between massive sensor data and limited computing power of edge devices. Lack of privacy protection, traditional solutions do not encrypt sensitive information such as sensor data and system state, in multi-node collaboration or cloud interaction scenarios, there is a risk of privacy leakage such as data interception and sensor collusion attack. There is no effective solution to the above problems. SUMMARY

[0003] The present application provides a detection method, system and device of elastic fabric, which has the advantages of realizing spatio-temporal correlation analysis of multi-modal data, improving dynamic adaptability and ensuring data privacy and security.

[0004] The present application provides a detection method of elastic fabric, which is applied to a detection system including an elastic fabric sensor network; the sensor is embedded in the elastic fabric; the detection method comprises:

[0005] Establishing a first node and a second node of the elastic fabric sensor network; the first node includes the sensor;

[0006] Obtaining first data of the first node;

[0007] Encrypting the first data to obtain encrypted first data;

[0008] The second node is used for performing ciphertext domain fusion processing on the encrypted first data to obtain a feature matrix of the elastic fabric; the feature matrix represents an association between grid coordinates of the elastic fabric and data of the sensor;

[0009] The feature matrix is encrypted and sent to a terminal, so that the terminal decrypts the encrypted feature matrix to generate a detection report of the elastic fabric; the detection report is used for decision-making on the use of the elastic fabric.

[0010] In some embodiments, the sensor includes a temperature sensor and / or a pressure sensor; and the first data of the first node is obtained by:

[0011] The temperature data of the temperature sensor and / or the pressure data of the pressure sensor are obtained when the elastic fabric is subjected to temperature and / or pressure.

[0012] In some embodiments, the sensor further includes a Bluetooth sensor; and the method further includes:

[0013] The Bluetooth sensor is used for network configuration of a non-network configuration node in the elastic fabric sensor network.

[0014] In some embodiments, the first data is encrypted to obtain the encrypted first data by:

[0015] The first data is mapped and encrypted by using a preset homomorphic encryption technology to obtain the encrypted first data.

[0016] In some embodiments, the second node is used for performing ciphertext domain fusion processing on the encrypted first data to obtain a feature matrix of the elastic fabric, including:

[0017] The second node is used for obtaining a state matrix, an observation matrix, a control matrix and a gain matrix of the encrypted first data;

[0018] The state matrix, the observation matrix, the control matrix and the gain matrix are subjected to multi-modal data fusion to obtain the feature matrix of the elastic fabric.

[0019] In some embodiments, the method further includes:

[0020] The dimension of the feature matrix is compressed to obtain a compressed feature matrix; and the compressed feature matrix is used for encryption and sending to the terminal.

[0021] In some embodiments, the method further includes:

[0022] A preset private key is obtained.

[0023] decrypt the encrypted feature matrix by using the private key to obtain a decrypted feature matrix;

[0024] generate a detection report of the elastic fabric based on the decrypted feature matrix.

[0025] In some embodiments, the output end of the sensor is connected in parallel with an RC filter network; the method further comprises:

[0026] obtaining an output signal of the sensor in the no-load state of the elastic fabric;

[0027] performing noise filtering on the output signal to obtain a digital signal corresponding to the output signal;

[0028] performing filtering processing on the digital signal to obtain a filtered digital signal;

[0029] performing noise compensation correction on the filtered digital signal to obtain the first data.

[0030] The embodiment of the application also provides an elastic fabric detection system, which comprises an elastic fabric, a temperature sensor and / or a pressure sensor, and a processing device; the temperature sensor and / or the pressure sensor are embedded in the elastic fabric; the temperature sensor and / or the pressure sensor construct a sensor network; the temperature sensor is connected to a temperature sensor signal line; the pressure sensor is connected to a pressure sensor signal line; the processing device is connected to the temperature sensor signal line and the pressure sensor signal line respectively; wherein,

[0031] the temperature sensor signal line is used for obtaining temperature data of the temperature sensor under the action of temperature of the elastic fabric;

[0032] the pressure sensor signal line is used for obtaining pressure data of the pressure sensor under the action of pressure of the elastic fabric;

[0033] the processing device is used for performing encryption processing on the temperature data and / or the pressure data to obtain encrypted first data; performing ciphertext domain fusion processing on the encrypted first data to obtain a feature matrix of the elastic fabric; the feature matrix represents the correlation between the grid coordinates of the elastic fabric and the data of the sensor; the feature matrix is sent to a terminal after being encrypted, so that the terminal decrypts the encrypted feature matrix to generate a detection report of the elastic fabric; and the detection report is used for deciding the use of the elastic fabric.

[0034] The embodiment of the present application also provides a detection device for the elastic fabric, which comprises a processor and a memory for storing a computer program capable of running on the processor, wherein the processor is used to run the computer program to execute the steps of the detection method.

[0035] The detection method for the elastic fabric provided by the embodiment of the present application is applied to a detection system for the elastic fabric; the sensor is embedded in the elastic fabric; the detection method comprises the following steps: establishing a first node and a second node of a sensor network of the elastic fabric; the first node comprises the sensor; obtaining first data of the first node; performing encryption processing on the first data to obtain encrypted first data; performing ciphertext domain fusion processing on the encrypted first data by using the second node to obtain a feature matrix of the elastic fabric; the feature matrix represents the correlation between the grid coordinates of the elastic fabric and the data of the sensor; the feature matrix is sent to a terminal after being encrypted, so that the terminal generates a detection report of the elastic fabric after decrypting the encrypted feature matrix; and the detection report is used to determine the use of the elastic fabric. By establishing the sensor network node, performing encryption processing on the multi-modal data and performing the ciphertext domain fusion to generate the feature matrix, the technical solution of the present application solves the problems of single data dimension and lack of privacy protection in the traditional technology, and has the advantages of realizing the spatio-temporal correlation analysis of the multi-modal data, improving the dynamic adaptability and guaranteeing the data privacy security. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 A flowchart of the detection method for the elastic fabric provided by the embodiment of the present application is shown in the figure;

[0037] Figure 2 A schematic diagram of the thin film temperature and pressure sensing network in the embodiment of the present application is shown in the figure, wherein the pressure sensor signal line represented by reference numeral 4 is led out from the bottom of the elastic fabric base 6;

[0038] Figure 3 A schematic diagram of the Bluetooth Mesh networking construction in the embodiment of the present application is shown in the figure;

[0039] Figure 4 An application scenario diagram of the detection method for the elastic fabric in the embodiment of the present application is shown in the figure;

[0040] Figure 5 A structural diagram of the detection system for the elastic fabric provided by the embodiment of the present application is shown in the figure;

[0041] Figure 6 A hardware structure diagram of the detection device for the elastic fabric in the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0042] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be described in detail below with reference to the drawings and specific embodiments.

[0043] The various technical features in each of the various embodiments described in the specific embodiments can be combined in various combinations, for example, different combinations of different technical features can form different embodiments, in order to avoid unnecessary repetition, various possible combinations of the various technical features in the present application are not described again.

[0044] It should also be noted here that, in order to avoid obscuring the present application due to unnecessary details, only structures and / or processing steps closely related to the scheme of the present application are shown in the drawings, and other details not closely related to the present application are omitted.

[0045] In addition, it should also be noted that the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or includes elements inherent to such process, method, article or equipment. In the following description, the terms "first, second,... " are only to distinguish different objects, and do not mean that the objects have the same or relationship. It should be understood that the orientation described by the orientation terms "above", "below", "inner", "outer" and the like is the orientation in the normal use state.

[0046] With the rapid development of intelligent wearable technology, accurate detection of temperature and pressure data related to textile touch becomes a core requirement for functional clothing design, medical rehabilitation aid development and sports equipment optimization. However, the existing technical system has the following bottlenecks:

[0047] 1. Single data dimension, insufficient touch feature modeling: existing textile touch detection technology is mostly based on independent measurement of a single physical quantity, lacking comprehensive analysis capability of the spatio-temporal correlation of multi-modal data such as temperature and pressure, leading to one-sided touch feature extraction and inability to accurately represent the dynamic comfort change of textiles in complex scenarios.

[0048] 2. Poor dynamic adaptability, lack of real-time feedback: traditional sensor networks use fixed sampling frequency and static threshold judgment mechanism, which is difficult to adapt to the rapid nonlinear change of touch parameters in the human motion process, and the data processing link delay is high, which cannot realize millisecond-level real-time early warning, seriously limiting the application value in dynamic scenarios.

[0049] 3. Insufficient synergy between hardware and algorithms: existing solutions focus on sensor hardware optimization, but do not achieve deep collaborative design of algorithms and hardware, leading to a prominent contradiction between massive sensor data and limited computing power of edge devices.

[0050] 4. Lack of privacy protection: traditional solutions do not encrypt sensitive information such as sensor data and system state, and in multi-node collaboration or cloud interaction scenarios, there is a risk of data interception, sensor collusion attack and other privacy leakage.

[0051] Based on this, the present application provides a detection method of elastic fabric, as shown in Figure 1 Figure 1 A flowchart of a detection method of elastic fabric provided by an embodiment of the present application is shown, which is applied to a detection system of elastic fabric; the sensor is embedded in the elastic fabric; the detection method comprises:

[0052] Step S101, establishing a first node and a second node of the elastic fabric sensor network; the first node comprises the sensor.

[0053] Step S102, obtaining first data of the first node.

[0054] Step S103, encrypting the first data to obtain encrypted first data.

[0055] Step S104, using the second node to perform ciphertext domain fusion processing on the encrypted first data to obtain a feature matrix of the elastic fabric; the feature matrix represents the correlation between the grid coordinates of the elastic fabric and the data of the sensor.

[0056] Step S105, sending the encrypted feature matrix to a terminal to make the terminal decrypt the encrypted feature matrix to generate a detection report of the elastic fabric; the detection report is used to determine the use of the elastic fabric.

[0057] ​The elastic fabric sensor network refers to a distributed detection system composed of sensors embedded in the fabric, which can be implemented by using an array of temperature sensors and pressure sensors to synchronously collect multi-modal tactile data. The first node refers to a data acquisition unit containing sensors, which can be implemented by using an embedded microcontroller to acquire and preliminarily process raw data. The encryption processing refers to mathematical transformation of the raw data to prevent information leakage, which can be implemented by using a homomorphic encryption algorithm to enable data operations in an encrypted state. The ciphertext domain fusion processing refers to matrix operations performed in the encrypted data space, which can be implemented by using a Kalman filter algorithm to extract features through iterative updates of a state matrix and an observation matrix. The feature matrix refers to a multi-dimensional data structure containing the correlation between grid coordinates and sensor data, which can be implemented by using a sparse matrix encoding method to represent the dynamic tactile distribution of the elastic fabric.

[0058] Specifically, when the elastic fabric is subjected to external force, the temperature or pressure sensor in the first node acquires physical signals, which are converted into raw data after analog-to-digital conversion. The data is converted into ciphertext by a homomorphic encryption algorithm, transmitted to the second node, and subjected to matrix operations without decryption, including state prediction, observation update, and gain adjustment. The fused feature matrix generates a spatial distribution model through a grid coordinate mapping algorithm, which is transmitted to the terminal device after secondary encryption. The terminal generates a detection report containing tactile intensity and distribution characteristics after decryption using a private key.

[0059] As an example, the sensor includes a temperature sensor and / or a pressure sensor; the first node can be referred to as a sub-node; the second node can be referred to as a master node; the master node is responsible for receiving and processing data from the temperature and pressure nodes, and the sub-node contains a temperature sensor and a pressure sensor, which are responsible for acquiring the values of each temperature and pressure sensor and uploading data.

[0060] The present application realizes end-to-end data protection through a double encryption mechanism. The prior art uses a single sensor to independently process data, which cannot establish a spatial correlation model. However, the present application maps discrete data into continuous distribution characteristics through matrix fusion. The existing system relies on fixed threshold values to determine tactile states, while the present application can adapt to nonlinear changes during the deformation of elastic fabric through a dynamic matrix update mechanism.

[0061] Through the above technical solutions, the present application realizes encrypted fusion processing of multi-modal tactile data, solves the problems of single data dimension and insufficient privacy protection in traditional methods. By performing matrix operations in the ciphertext domain, a spatial correlation model of tactile features is established under the premise of ensuring data security, improving the accuracy of detection results in dynamic scenarios. The terminal decryption and report generation mechanism ensures that the end user can obtain complete tactile evaluation data, providing a reliable basis for the optimization design of medical rehabilitation aids.

[0062] The embodiment of the present application provides a detection method of elastic fabric, which solves the problems of single data dimension and lack of privacy protection in traditional technologies by establishing a sensor network node, encrypting multi-modal data and generating a feature matrix through ciphertext domain fusion, and has the advantages of realizing multi-modal data space-time correlation analysis, improving dynamic adaptability and guaranteeing data privacy and security.

[0063] In some embodiments, the sensor includes a temperature sensor and / or a pressure sensor; the first data of the first node is acquired by:

[0064] The temperature data of the temperature sensor and / or the pressure data of the pressure sensor are acquired under the action of temperature and / or pressure on the elastic fabric.

[0065] The temperature sensor refers to a device capable of detecting the temperature change of the surface of the elastic fabric, and can be implemented by a thermistor or a thermocouple, and is used for monitoring the temperature distribution of the fabric in real time when the fabric is heated. The pressure sensor refers to a device capable of sensing the pressure intensity of the elastic fabric, and can be implemented by a piezoelectric film or a capacitive pressure sensor, and is used for capturing the pressure distribution of the fabric when the fabric is pressed. The temperature data and the pressure data refer to the physical quantity signals collected by the temperature sensor and the pressure sensor respectively, and can be converted into digital signals by an analog-to-digital conversion module, and are used to represent the response state of the elastic fabric under different external actions.

[0066] Specifically, when the elastic fabric is subjected to external temperature or pressure, the temperature sensor and the pressure sensor start data collection simultaneously. The temperature sensor senses the temperature change of the surface of the fabric through a thermoelement, and generates temperature data; the pressure sensor detects the stress distribution of the pressure area of the fabric through a pressure-sensitive element, and generates pressure data. The two kinds of data are transmitted to the processing device through signal lines, forming a multi-modal data set. For example, in a medical rehabilitation scene, when a patient wears an elastic fabric to exercise, the temperature sensor can monitor the local skin temperature change, and the pressure sensor can synchronously record the pressure distribution of the joint part, and the two data jointly constitute the first data collected by the first node, providing multi-dimensional input for subsequent feature matrix construction.

[0067] As an example, the temperature sensor: adopts a thin film K-type thermocouple, which has a foldable characteristic, a wide temperature measurement range (-20~250℃), and a high accuracy of 0.4℃, and can accurately measure the temperature of the textile under different environments. The pressure sensor: adopts a flexible thin film pressure-sensitive resistor, which has a foldable characteristic, a wide pressure measurement range (20g~20kg), and can sensitively perceive the deformation of the textile under different pressures. The integrated temperature and pressure sensor constructs a touch feature matrix to realize multi-modal data space-time correlation analysis.

[0068] The application realizes the synchronous acquisition of multiple physical quantities by integrating temperature and pressure sensors, and solves the problem of one-dimensional data leading to the one-sidedness of touch feature modeling. The independent measurement method in the prior art cannot capture the influence of the interaction of temperature and pressure on the fabric performance, while the application can more accurately represent the dynamic response characteristics of the fabric under complex working conditions through multi-modal data fusion.

[0069] Through the above technical solution, the application realizes the synchronous acquisition and joint analysis of temperature and pressure data, and solves the problem of one-dimensional touch feature modeling in the prior art. Through multi-modal data fusion, the dynamic response of the elastic fabric under complex external action can be more comprehensively reflected, providing multi-dimensional data support for subsequent feature matrix construction and improving the accuracy of the detection report for fabric use decision.

[0070] In some embodiments, the sensor further comprises a Bluetooth sensor; and the method further comprises:

[0071] The Bluetooth sensor is used to configure the network for the unconfigured network node in the elastic fabric sensor network.

[0072] The Bluetooth sensor refers to a sensor module integrated with a Bluetooth communication protocol, which can be implemented by a low-power Bluetooth chip, and is used to establish a data transmission channel between devices within a wireless communication range. The module actively sends device identification information in broadcast mode and supports bidirectional data encryption transmission protocol.

[0073] The unconfigured network node refers to an isolated sensor unit that has not yet accessed the sensor network, which can be in a state of not completing network address allocation or not establishing a communication link. Such nodes appear at the beginning of deployment or when the network topology changes, and need to be identified and accessed through a dynamic discovery mechanism.

[0074] Specifically, during the deployment of the elastic fabric sensor network, the Bluetooth sensor periodically sends broadcast signals containing network identification. When the unconfigured network node enters the signal coverage range, its built-in Bluetooth module automatically detects the broadcast packet and triggers a pairing request. A bidirectional authentication mechanism is used in the pairing process, such as a key exchange protocol based on an elliptic curve encryption algorithm, to complete identity verification and allocate a dynamic network address. After the network address allocation is completed, the node is included in the sensor network topology and participates in subsequent data acquisition and transmission tasks.

[0075] As an example, CH573 Bluetooth sensors are used to configure the network for unconfigured network nodes, realizing the rapid formation and flexible configuration of the sensor network.

[0076] The application realizes wireless ad hoc networking through Bluetooth sensors, and the nodes can autonomously complete network discovery and access without physical contact or preset gateway parameters, significantly reducing the deployment complexity.

[0077] By the technical solution, the dynamic expansion capability of the sensor network is achieved, the unconfigured network node can autonomously complete identity authentication and network access, the problem of incomplete network coverage caused by low efficiency of manual configuration in the traditional scheme is effectively solved, meanwhile, wiring interference caused by physical connection is avoided, and the adaptability of the sensor network in the elastic fabric is improved.

[0078] In some embodiments, the encryption processing of the first data to obtain the encrypted first data comprises:

[0079] The first data is mapped and encrypted by using a preset homomorphic encryption technology to obtain the encrypted first data.

[0080] The homomorphic encryption technology refers to an encryption method that allows direct operation on encrypted data and the decryption result is consistent with the plaintext operation result, and specifically, an encryption algorithm based on ring learning error can be used to implement the technology, which enables the encrypted first data to be fused in the ciphertext state. The mapping processing refers to converting the original data into a mathematical structure suitable for homomorphic encryption operation, and specifically, a vector encoding method on a polynomial ring can be used to implement the step, which ensures that the sensor data is standardized in format before encryption.

[0081] Specifically, during the operation of the elastic fabric sensor network, the temperature or pressure data collected by the first node is first converted into a numerical vector of a preset dimension, and then structured data is generated through polynomial ring mapping. The structured data is input into the homomorphic encryption module, and encrypted by using a preconfigured public key to generate an encrypted data packet containing a ciphertext coefficient sequence. When the encrypted data packet is transmitted to the second node through wireless transmission, the ciphertext attribute can prevent sensitive information from being leaked during the transmission process. After receiving the encrypted data, the second node can directly call the ciphertext operation protocol to complete the matrix fusion operation without decryption, and the entire process keeps the data in an encrypted state.

[0082] As an example, the Pailliar partial homomorphic encryption technology is used, the subnode maps and encrypts the original data after fixed-point number mapping, and transmits it to the master node through Bluetooth Mesh. Specifically, the floating-point data (such as temperature and pressure values) collected by the sensor is mapped to an integer through a fixed-point number set Q ( A , B ), Q ( A , B ), A-B-1 , 2 A-B-1 -2 -B , F ( A , B , b)=2 B b mod 2 A Convert to integer field representation; wherein A, B can be understood as parameters in fixed-point number mapping, used for converting floating-point data into integers; b Can be understood as an intermediate variable after the original floating-point data is mapped by the fixed-point number; F ( A , B , b ) can be understood as a fixed-point number mapping function. Use the encryption formula Enc ( d )= G d R N mod N 2 Generate ciphertext, support ciphertext field addition and scalar multiplication operation; wherein, d Can be understood as the original data (such as temperature and pressure value); R Can be understood as a random number in the encryption process, satisfying , Indicates the multiplicative group of module N .

[0083] The present application realizes the whole process of ciphertext processing through homomorphic encryption technology, which maintains the data processing ability while ensuring the data privacy. The symmetric encryption or asymmetric encryption method used in the prior art cannot support the ciphertext field operation, and must rely on data fusion in the plaintext environment, which increases the risk of illegal interception of data.

[0084] Through the above technical scheme, the present application realizes the encryption protection of sensor data in the whole process of collection, transmission and processing, effectively prevents the data leakage problem in the multi-node cooperative scene. The encrypted data can be directly fused and operated, avoiding the calculation power loss caused by repeated encryption and decryption in the traditional scheme, while ensuring the integrity of the feature matrix generation process.

[0085] In some embodiments, the ciphertext domain fusion processing of the encrypted first data by the second node to obtain the feature matrix of the elastic fabric comprises:

[0086] Obtaining the state matrix, observation matrix, control matrix and gain matrix of the encrypted first data by the second node;

[0087] Performing multi-modal data fusion on the state matrix, observation matrix, control matrix and gain matrix to obtain the feature matrix of the elastic fabric.

[0088] In this embodiment, the state matrix refers to a dataset containing the dynamic changes of sensor data. Specifically, it can be implemented using a time series analysis model to dynamically model the encrypted data, characterizing the real-time state evolution of the elastic fabric under external influences. The observation matrix refers to the set of actual sensor measurements, which can be implemented by decoupling multi-source signal components from the encrypted data, reflecting the measured values ​​of physical quantities at different spatial locations. The control matrix refers to the set of quantified parameters representing the influence of external input variables on the system. Specifically, it can be implemented using an adaptive adjustment algorithm to dynamically calibrate the transmission parameters between nodes, balancing network load and data processing accuracy. The gain matrix refers to the set of adjustment parameters for each modality weight during data fusion. Specifically, it can be implemented using a pre-trained multi-objective optimization model to dynamically weight the encrypted data features, optimizing the correlation between multi-source data. Multimodal data fusion refers to the joint computation of heterogeneous data under encrypted conditions. Specifically, it can be implemented using a matrix operation protocol based on homomorphic encryption, establishing a mapping between grid coordinates and sensor data while protecting data privacy.

[0089] Specifically, in the encrypted domain fusion process, the second node first parses the encrypted first data and separates the dynamic feature parameters in the state matrix and the spatial distribution parameters in the observation matrix through homomorphic operations. The control matrix dynamically adjusts the data transmission path according to the real-time network load, and the gain matrix iteratively optimizes the weights of temperature and pressure data. In the multimodal data fusion stage, the four matrices are jointly operated in the encrypted state. For example, spatiotemporal feature coupling is achieved through matrix multiplication, and multi-source data superposition is completed through matrix addition. The final generated feature matrix stores the comprehensive tactile parameters of each grid point of the elastic fabric under different working conditions in encrypted form.

[0090] As an example, after receiving encrypted data, the edge node updates its state using the following formula: Enc ( Z k+1 )=( S - T × C )× Enc ( Z k )⊕ U × Enc ( U k ⊕ V × Enc ( Y k );in, Z k , Z k+1 It can be understood as a state variable, used for state updates in ciphertext processing; SCan be understood as a state matrix, a parameter in ciphertext processing; T Can be understood as an observation matrix, a parameter in ciphertext processing; U Can be understood as a control matrix, a parameter in ciphertext processing; V Can be understood as a gain matrix, a parameter in ciphertext processing; U k 、Y k Can be understood as an input variable, used for ciphertext processing calculation; complete multi-modal data fusion, avoid plaintext exposure.

[0091] The application establishes the non-linear association between data in the ciphertext domain by constructing four types of functional matrices, which not only solves the problem of lack of spatio-temporal association of multi-modal data, but also avoids the risk of privacy leakage in data processing.

[0092] Through the above technical solution, the application can realize dynamic correlation analysis of temperature and pressure data in an encrypted environment, improve the integrity of the modeling of the touch features of elastic fabrics, and at the same time, through matrix operation in the ciphertext domain, ensure the security of data transmission, solving the dual problems of single data dimension and lack of privacy protection in traditional solutions.

[0093] In some embodiments, the method further comprises:

[0094] The dimension of the feature matrix is compressed to obtain a compressed feature matrix; the compressed feature matrix is used to send to the terminal after encryption.

[0095] Wherein, the dimension of the feature matrix is compressed, which means that the dimension number of the matrix is reduced by dimension reduction algorithm, which can be realized by principal component analysis or linear discriminant analysis, through orthogonal transformation to project high-dimensional data to low-dimensional space, remove redundant information and retain key feature dimensions. The compressed feature matrix refers to the low-dimensional data set formed after dimension reduction processing, which can be realized by matrix decomposition or feature selection method, by reducing the data dimension to reduce the amount of transmission data, while maintaining the integrity of the feature association relationship.

[0096] Specifically, after generating the feature matrix representing the correlation between the grid coordinates of the elastic fabric and the sensor data, the dimension of the matrix is compressed by a preset dimension reduction algorithm. For example, principal component analysis is used to extract the principal component with the maximum variance in the matrix, and the original high-dimensional matrix is mapped to a low-dimensional space to eliminate the linear correlation between temperature and pressure data. During the compression process, the eigenvalues and eigenvectors of the covariance matrix are calculated, and the principal components with a contribution rate exceeding a set threshold are selected to construct a low-dimensional feature matrix. The compressed matrix formed in this way not only retains the spatio-temporal correlation of the fabric touch features, but also significantly reduces the data size, facilitating subsequent encrypted transmission.

[0097] As an example, a reduction operator is used. Red ( Q The ciphertext dimension is compressed, and a conservative strategy is used to ensure that the reduced-order region includes the original state estimation range. The formula is: Enc ( )= Red ( Q )× Enc ( Z k+1 );in, Q These parameters can be understood as those of the order reduction operator, which are usually related to the precision of the ciphertext data and the dimension of the state space. The specific values ​​need to be set according to the actual needs of the sensor network. Red(Q) It can be understood as a reduction operator used for ciphertext processing. Its main function is to compress the dimensionality of ciphertext data, reduce computational complexity, and at the same time ensure the accuracy of state estimation. Enc(Z k+1 ) This can be understood as the original ciphertext state variable; Enc ( This can be understood as a ciphertext after order reduction; it reduces computational complexity while ensuring estimation accuracy.

[0098] This invention introduces dimensionality compression processing to proactively reduce the dimension of the feature matrix before data encryption. This reduces the amount of ciphertext data to improve transmission efficiency, avoids the waste of storage resources caused by data redundancy, and ensures the integrity of feature information by retaining principal components.

[0099] Through the above technical solution, this application effectively solves the problem of low transmission efficiency of high-dimensional feature matrices. It actively reduces the data dimension before the encryption process, reduces the communication resource consumption required for ciphertext transmission, and maintains detection accuracy by preserving principal component features, thereby enabling the detection system to operate efficiently under limited bandwidth conditions.

[0100] In some embodiments, the method further includes:

[0101] Obtain the preset private key;

[0102] The encrypted feature matrix is ​​decrypted using the private key to obtain the decrypted feature matrix;

[0103] A test report for the elastic fabric is generated based on the decrypted feature matrix.

[0104] Wherein, the private key refers to a decryption key matched with the public key encryption algorithm, and can be specifically implemented by using an RSA private key or an elliptic curve encryption private key in an asymmetric encryption algorithm, and is used for uniquely authorized decryption of encrypted data. The decrypted feature matrix refers to an original data matrix restored through private key operation, and can be specifically implemented by using a homomorphic decryption algorithm or a hierarchical decryption protocol to ensure that the ciphertext data is restored to parseable plaintext data on the terminal side. The detection report refers to a structured document containing fabric state evaluation results, and can be specifically implemented by generating a visual chart through pattern recognition of the feature matrix by using a machine learning model, and is used for guiding fabric use decision.

[0105] Specifically, the private key matched with the encrypted public key is pre-stored in the terminal device, and when the encrypted feature matrix is received, the private key is called for decryption operation to restore the original data format. The decrypted feature matrix parses the correspondence between the sensor data and the fabric grid through a coordinate mapping algorithm, and automatically generates a detection report containing a temperature distribution map, a pressure thermal map and a comprehensive score index based on a preset fabric performance evaluation model. The entire decryption process is completed in a secure execution environment to ensure that sensitive data is only readable in authorized terminals.

[0106] As an example, the user performs decryption operation through the private key: Dec ( Enc ( d )= L [( Enc ( d ) λ mod N 2 )]× Mmod N ; wherein L(x) is an auxiliary function, for example, L(x)=(x-1) / N; the plaintext result is restored, and data closed loop processing is completed.

[0107] The application binds the private key with the authorized terminal to construct an end-to-end data security link, ensuring multi-node collaborative calculation while realizing controlled access of terminal data.

[0108] Through the above technical solution, the application effectively prevents the feature matrix from being parsed by unauthorized equipment on the terminal side, and avoids the risk of leakage of sensor network data in the final application link. The combination of decryption operation and automatic report generation process improves the generation efficiency and accuracy of detection results on the premise of ensuring data privacy, and provides a reliable basis for fabric use decision.

[0109] In some embodiments, the output end of the sensor is connected in parallel with an RC filter network; the method further comprises:

[0110] In the unloaded state of the elastic fabric, the output signal of the sensor is acquired;

[0111] noise filtering on the output signal to obtain a digital signal corresponding to the output signal;

[0112] filtering processing on the digital signal to obtain a filtered digital signal;

[0113] noise compensation correction on the filtered digital signal to obtain the first data.

[0114] Among them, RC filter network refers to a filter circuit composed of resistance and capacitance, which can be specifically implemented by a low-pass filter with a cutoff frequency of 10Hz-1kHz, for filtering out high-frequency interference signals in the sensor output signal. Noise filtering refers to the preliminary noise reduction processing of the sensor original signal, which can be specifically implemented by using a sliding window average algorithm or a wavelet threshold denoising algorithm, for eliminating the baseline drift caused by environmental electromagnetic interference. Filter processing refers to secondary noise reduction on the digital signal, which can be specifically implemented by using a Butterworth low-pass filter or a Kalman filter, for suppressing random noise in the signal transmission process. Noise compensation correction refers to dynamic compensation of residual noise, which can be specifically implemented by using an adaptive compensation algorithm based on a historical noise database, for eliminating signal distortion caused by temperature drift or circuit noise.

[0115] Specifically, in the unloaded state of the elastic fabric without external force, the sensor output signal only contains environmental noise and circuit background noise. At this time, the RC filter network in parallel is used to preliminarily filter the sensor analog signal, and the high-frequency interference signal is attenuated to below the preset threshold. The digital signal after analog-digital conversion is input into the noise filtering module, and the sliding window average algorithm is used to weight and average the continuous sampling points to eliminate burst pulse interference. Then, the Butterworth low-pass filter is used for frequency domain processing of the signal to suppress out-of-band noise components. Finally, the adaptive compensation algorithm is used to dynamically compensate the residual noise according to the preset unloaded state reference value, so that the sensor output in the unloaded state is stabilized within the error allowable range.

[0116] As an example, the noise suppression two-stage strategy: a. Hardware layer noise filtering algorithm: RC filter network in parallel at the output end of the pressure sensor, parameter configuration: resistance R = 1kΩ, capacitance C = 10μF. The cutoff frequency is calculated as: , effectively attenuating high-frequency noise above 20Hz (such as high-frequency interference generated by motion friction); wherein, f c It can be understood as the cutoff frequency, which is used to calculate the noise attenuation range of hardware filtering. The sensor output signal in the unloaded state is collected, and the noise mean μ n and the standard deviation σ n, for algorithm end noise compensation.b. Algorithm layer dynamic filtering algorithm: adaptive window initialization: integrate three-axis acceleration sensor, real-time monitor human motion state. Static state: acceleration amplitude |a| <0.2g, set filter window N = 50 (corresponding to 100ms). Motion state: |a| ≥ 0.2g, window dynamic shrink to N = 10 (corresponding to 20ms), improve real-time performance. Dynamic moving average filtering: to temperature / pressure digital signal x k Perform sliding window average: . When the window slides, remove outliers exceeding μ n ±3 σ n The mean value is calculated. Noise compensation correction: to the filtered signal y k Trend item elimination, remove low frequency drift noise by median filtering: (2) Energy consumption optimization design: set energy saving mode, automatically switch to microampere level sleep mode in static state; adopt data compression transmission technology, compress 12bit original data to 8bit, reduce energy consumption while ensuring data quality.

[0117] The traditional scheme does not set the no-load state calibration link, and the sensor background noise and environmental interference are directly superimposed on the effective signal, resulting in reduced data acquisition accuracy. The single-stage filtering design in the prior art cannot meet the wideband noise suppression requirement, and lacks a dynamic compensation mechanism, which is easy to produce signal distortion in complex electromagnetic environment.

[0118] Through the above technical solutions, the influence of sensor background noise and environmental interference on data acquisition is effectively reduced, and the signal-to-noise ratio and accuracy of the first data are improved. The multi-stage signal processing process in the no-load state establishes the sensor reference working state, providing a reliable reference for subsequent data acquisition in the load state, avoiding system misjudgment caused by noise accumulation.

[0119] In actual application, the detection method of the elastic fabric can be a detection method of a distributed elastic fabric temperature and pressure detection sensor network for textile touch test, comprising:

[0120] 1. Multi-modal detection: integrate temperature and pressure sensors to build a touch feature matrix to realize multi-modal data space-time correlation analysis.

[0121] 2. Dynamic adaptation: dynamically optimize the sampling frequency combined with human motion state to adapt to rapid changes in touch parameters.

[0122] 3. Cooperative optimization: joint noise reduction of hardware end RC filter and algorithm end dynamic filter to improve signal quality.

[0123] 4. Encryption transmission: Pailliar homomorphic encryption is adopted to encrypt sensor data end-to-end to protect privacy.

[0124] 5. Anti-collusion attack: a fully symmetric polyhedral encryption structure is combined with a reduction operator to prevent collusion attacks and data inference.

[0125] The present application mainly embeds a temperature and pressure sensing array into an elastic fabric, constructs a sensor network through Bluetooth Mesh networking technology, and realizes distributed real-time monitoring of the tactile sensation of textiles. Specifically as follows:

[0126] 1. Intelligent fabric sensor network architecture.

[0127] (1) Sensor layer: temperature sensor: thin film K-type thermocouple is adopted, the thermocouple has a foldable characteristic, the temperature measurement range is wide (-20~250℃), and the accuracy is as high as 0.4℃, which can accurately measure the temperature of textiles in different environments.

[0128] Pressure sensor: flexible thin film pressure-sensitive resistor is adopted, the resistor has a foldable characteristic, the pressure measurement range is wide (20g~20kg), and it can sensitively perceive the deformation of textiles under different pressures.

[0129] Sensor network design: through optimizing sensor layout and wiring method, all measurement data is obtained while reducing the number of wiring, improving the reliability and stability of the sensor network. This content can be understood in combination with Figure 2 , Figure 2 a schematic diagram of the thin film temperature and pressure sensing network in the embodiments of the present application. In Figure 2 , 1 can be a thin film pressure sensor; 2 can be a thin film temperature sensor; 3 can be a sensor network; 4 can be a pressure sensor signal line; 5 can be a temperature sensor signal line; 6 can be an elastic fabric base.

[0130] (2) Communication layer.

[0131] Bluetooth Mesh network construction: including master node and slave node. The master node is responsible for receiving and processing data from the temperature and pressure nodes, and the slave node contains a temperature sensor and a pressure sensor, which is responsible for obtaining the value of each temperature and pressure sensor and uploading data.

[0132] Data transmission optimization: CH573 Bluetooth sensor is used to configure the network for unconfigured nodes, realizing fast formation and flexible configuration of the sensor network.

[0133] Encryption transmission: Pailliar partial homomorphic encryption technology is adopted, the slave node encrypts the original data after mapping to fixed-point numbers, and transmits the data to the master node through Bluetooth Mesh.

[0134] The content can be understood in combination with Figure 3 , Figure 3 is a schematic diagram of the Bluetooth Mesh networking constructed in the embodiments of the present application.

[0135] 2. Data encryption and processing

[0136] (1) Key generation: generate two large prime numbers P and Q , calculate modulus N = PQ and auxiliary parameter λ = lcm ( P -1, Q -1), construct public key pk = ( N , G )( G = N +1) and private key sk =( λ , M )( M is the modular inverse), which are used for subsequent encryption and decryption operations; P and Q can be understood as two large prime numbers used when generating the key P and Q can be understood as two large prime numbers used when generating the key; N can be understood as the modulus, which is calculated from P and Q ; λ can be understood as the auxiliary parameter, λ =lcm( P -1, Q -1), where lcm represents the least common multiple; G can be understood as a component of the public key, G = N + 1; pk can be understood as the public key, pk = ( N , G ); sk can be understood as the private key, sk = ( λ , M ), where M is the modular inverse, satisfying M = ( λ mod N) -1 .

[0137] (2) Data encryption:

[0138] The floating-point data (such as temperature and pressure values) collected by the sensor are collected through fixed-point numbersQ A , B ) mapped to integers, Q A , B ) contains the rational numbers of the interval (-2 A-B-1 , 2 A-B-1 -2 -B ) converted to integer field representation by the mapping function F A , B , b )=2 B b mod 2 A ; wherein A, B can be understood as parameters in the fixed-point number mapping, used to convert floating-point data into integers; b can be understood as an intermediate variable after the original floating-point data is mapped by the fixed-point number; F A , B , b ) can be understood as a fixed-point number mapping function.

[0139] Using the encryption formula Enc d G d R N mod N 2 R is a random number, ) to generate ciphertext, supporting ciphertext field addition and scalar multiplication operations; wherein, d can be understood as the original data (such as temperature and pressure values); R can be understood as a random number in the encryption process, satisfying , denotes the multiplicative group modulo N .

[0140] (3) Ciphertext processing:

[0141] After the edge node receives the encrypted data, it updates the state through the state update formula:

[0142] Enc Z k+1 )=( S - T × C )× Enc Z k )⊕ U × Enc U k ⊕​​​​​​​​​​V × Enc ( Y k );

[0143] wherein, Z k 、 Z k+1 may be understood as a state variable, used for state update in ciphertext processing; S may be understood as a state matrix, a parameter in ciphertext processing; T may be understood as an observation matrix, a parameter in ciphertext processing; U may be understood as a control matrix, a parameter in ciphertext processing; V may be understood as a gain matrix, a parameter in ciphertext processing; U k 、Y k may be understood as an input variable, used for ciphertext processing calculation.

[0144] Complete multi-modal data fusion, avoid plaintext exposure. Wherein S is a state matrix, T is an observation matrix, U is a control matrix, V is a gain matrix.

[0145] (4) Order reduction optimization:

[0146] The order reduction operator Red ( Q ) is used to compress the dimension of ciphertext, and a conservative strategy is used to ensure that the region after order reduction contains the original state estimation range, and the formula is:

[0147] Enc ( )= Red ( Q )× Enc ( Z k+1 );

[0148] wherein, Q may be understood as a parameter of the order reduction operator, which is usually related to the precision of ciphertext data and the dimension of state space, and the specific value needs to be set according to the actual needs of the sensor network; Red(Q) may be understood as an order reduction operator used for ciphertext processing, which mainly functions to compress the dimension of ciphertext data, reduce the calculation complexity, and at the same time ensure the accuracy of state estimation; Enc (Z k+1 ) may be understood as an original ciphertext state variable; Enc ( ) may be understood as a ciphertext after order reduction.

[0149] Reduce the computational complexity while ensuring the estimation accuracy.

[0150] (5) Decryption verification:

[0151] The user performs a decryption operation through a private key:

[0152] Dec ( Enc ( d )= L [( Enc ( d ) λ mod N 2 )]× Mmod N ;

[0153] Where L(x) is an auxiliary function, for example, L(x) = (x-1) / N.

[0154] Restore the plaintext result and complete the data closed-loop processing.

[0155] 3. Hardware-algorithm collaborative optimization

[0156] (1) Noise suppression two-stage strategy:

[0157] a. Hardware layer noise filtering algorithm:

[0158] The output end of the pressure sensor is connected in parallel with an RC filter network, with the parameters configured as: resistance R = 1kΩ, capacitance C = 10μF.

[0159] Cut-off frequency calculation: , effectively attenuating high-frequency noise above 20Hz (such as high-frequency interference generated by motion friction); where, f c It can be understood as the cut-off frequency, which is used to calculate the noise attenuation range of hardware filtering.

[0160] Collect the sensor output signal in the no-load state, and count the noise mean μ n and standard deviation σ n , which is used for algorithm-side noise compensation.

[0161] b. Algorithm layer dynamic filtering algorithm.

[0162] Adaptive window initialization: integrate a three-axis acceleration sensor to monitor the human motion state in real time. Stationary state: acceleration amplitude |a| <0.2g, set the filter window N = 50 (corresponding to 100ms). Motion state: |a| ≥ 0.2g, window dynamically reduced to N = 10 (corresponding to 20ms), improving real-time performance.

[0163] Dynamic moving average filtering: on temperature / pressure digital signal x k Perform sliding window average: When the window slides, remove outliers exceeding μ n ±3 σ n of the mean value calculation.

[0164] Noise compensation correction: on filtered signal y k Trend item elimination, remove low frequency drift noise by median filtering: .

[0165] (2) Energy consumption optimization design: set energy saving mode, automatically switch to microampere level sleep mode in static state; adopt data compression transmission technology, compress 12bit original data to 8bit, reduce energy consumption while ensuring data quality.

[0166] The application scenario diagram of the detection method of the elastic fabric is as shown in Figure 4 , Figure 4 The application scenario diagram of the detection method of the elastic fabric in the embodiment of the application. Its working mode is as follows:

[0167] 1. Initialization and networking: when the system starts, the master node configures the network for the unconfigured network node through the CH573 Bluetooth sensor, and constructs a complete Bluetooth Mesh network. The master node distributes public key pk , and private key sk is stored securely by the user terminal.

[0168] 2. Encrypted data acquisition and transmission: when the textile is stimulated by pressure / temperature, the child node collects data and performs fixed-point number mapping and Pailliar encryption, and transmits it to the master node in real time through Bluetooth Mesh.

[0169] 3. Data processing and analysis: the master node uses the Pailliar encryption algorithm to perform ciphertext domain fusion analysis on the encrypted temperature and pressure data transmitted by the sensor layer. Specifically, the following steps are taken to construct the touch feature matrix:

[0170] (1) Ciphertext state update: based on the historical encrypted data of the sensor, the real-time encrypted measurement value and the control parameters (such as sampling frequency, threshold setting), the state estimation set is updated through homomorphic operation, and the state matrix, observation matrix and gain matrix are combined to complete the spatio-temporal correlation calculation of multi-modal data without decryption.

[0171] (2) Feature extraction and quantification: Extract key feature parameters such as temperature distribution uniformity and pressure change gradient from the updated encrypted data, construct a two-dimensional tactile feature matrix, with rows and columns as fabric surface grid coordinates, and elements as temperature-pressure correlation characteristic values.

[0172] 4. Result output and feedback: The main node transmits the encrypted analysis results to the cloud or the upper computer. Users decrypt the encrypted results through private keys to obtain intuitive plaintext reports for decision-making reference in textile research and development, production, or quality detection.

[0173] 5. Maintenance and expansion: The sensor network is easy to expand and maintain. When adding or reducing sensor nodes, only need to adjust through the configuration function of Bluetooth Mesh network. In addition, the system also supports remote monitoring and maintenance functions, allowing users to monitor the working status of the sensor network at any time and make necessary adjustments.

[0174] The embodiment of the application also provides a detection system 500 for elastic fabric, as shown in Figure 5 , which is a structural diagram of the detection system for elastic fabric provided by the embodiment of the application. The structure and function of the detection system will be described exemplarily in combination with various embodiments. Figure 5

[0175] The detection system 500 includes elastic fabric 501, temperature sensor 502 and / or pressure sensor 503, and processing device 504; the temperature sensor 502 and / or pressure sensor 503 are embedded in the elastic fabric 501; the temperature sensor 502 and / or the pressure sensor 503 construct a sensor network; the temperature sensor 502 is connected to the temperature sensor signal line; the pressure sensor 503 is connected to the pressure sensor signal line; the processing device 504 is connected to the temperature sensor signal line and the pressure sensor signal line respectively; wherein,

[0176] The temperature sensor signal line is used to obtain temperature data of the temperature sensor 502 under the action of temperature on the elastic fabric;

[0177] The pressure sensor signal line is used to obtain pressure data of the pressure sensor 503 under the action of pressure on the elastic fabric;

[0178] ​The processing device 504 is configured to perform encryption processing on the temperature data and / or the pressure data to obtain encrypted first data, perform ciphertext domain fusion processing on the encrypted first data to obtain a feature matrix of the elastic fabric, the feature matrix representing an association between the grid coordinates of the elastic fabric and the data of the sensor, and send the encrypted feature matrix to a terminal to enable the terminal to generate a detection report of the elastic fabric after decrypting the encrypted feature matrix, the detection report being used to determine the use of the elastic fabric.

[0179] The elastic fabric refers to a textile material with stretchable properties, which can be implemented by using polyurethane fiber and cotton fiber blended materials, and the grid coordinates are positioned by nodes formed by interlacing warp and weft yarns. The temperature sensor signal line refers to a conductive line connecting the temperature sensor and the processing device, which can be implemented by using a silver-plated fiber braided wire to maintain signal transmission stability when the fabric is deformed by heat. The ciphertext domain fusion processing refers to data fusion operation in an encrypted state, which can be implemented by using a homomorphic encryption algorithm combined with a Kalman filter equation to complete multi-modal data association analysis while protecting data privacy.

[0180] Specifically, when the elastic fabric is subjected to external temperature or pressure, the temperature sensor and the pressure sensor embedded in the fabric transmit the collected physical quantity data to the processing device through dedicated signal lines. The processing device first performs homomorphic encryption processing on the original data, so that subsequent operations can be performed in an encrypted state. The encrypted temperature data and pressure data are input into the fusion algorithm module, multi-dimensional data association is performed by constructing a state matrix and an observation matrix, and finally a feature matrix reflecting the distribution of sensing data of each grid point of the fabric is generated. The matrix is transmitted to a terminal device after being encrypted twice, and a detection report containing fabric performance parameters is generated by authorized users using a private key.

[0181] In some embodiments, the temperature sensor signal line can be arranged in the warp yarn of the fabric, and the pressure sensor signal line can be arranged in the weft yarn to form a signal transmission network by interlacing warp and weft. The processing device can be integrated on a flexible circuit board at the edge of the fabric, and a low-power Bluetooth module can be used to establish a communication connection with the terminal. The encryption of the feature matrix can use a layered encryption mechanism based on elliptic curve cryptography to implement end-to-end protection during data transmission.

[0182] Compared with the prior art, the traditional detection system independently works with a single type of sensor and cannot synchronously acquire the fabric performance data under the coupling of temperature and pressure. The present application realizes the parallel acquisition of multiple physical quantities by constructing double signal transmission channels, and effectively solves the problem of single data dimension by combining the ciphertext domain fusion algorithm. In the prior art, the sensor network mostly uses plaintext transmission mode, while the present application implements encryption protection in the whole process of data acquisition, processing and transmission, significantly improving the system security.

[0183] Through the above technical solutions, the present application realizes the synchronous encryption acquisition and fusion analysis of temperature and pressure data, solves the technical defects of single data dimension and weak privacy protection of the traditional detection system. The spatiotemporal correlation analysis capability of multi-modal data makes the fabric performance evaluation more comprehensive, and the encrypted transmission mechanism effectively prevents sensitive data from being stolen in the transmission process, providing reliable protection for the safe application of intelligent textiles.

[0184] Among them, the temperature sensor signal line refers to the signal transmission channel connecting the temperature sensor and the processing device, which can be realized by flexible conductive material or metal wire, and is used for transmitting temperature data when the elastic fabric is subjected to temperature action. The pressure sensor signal line refers to the signal transmission channel connecting the pressure sensor and the processing device, which can be realized by pressure-sensitive conductive fabric or strain gauge structure, and is used for transmitting pressure data when the elastic fabric is subjected to pressure action. The processing device refers to a computing module with data encryption and fusion functions, which can be realized by an embedded processor combined with a homomorphic encryption algorithm, and is used for encrypted processing and ciphertext domain fusion operation of multi-modal sensor data.

[0185] Specifically, the temperature sensor signal line and the pressure sensor signal line respectively acquire the physical quantity change of the elastic fabric under the action of temperature or pressure, and transmit the original data to the processing device through the signal line. The processing device homomorphically encrypts the received temperature data and pressure data, so that the data can still be fused in the encrypted state. The encrypted data is fused by the Kalman filtering algorithm to generate a feature matrix reflecting the correlation of the sensing data of each grid point of the elastic fabric. The feature matrix is transmitted to the terminal device after being encrypted twice, and the terminal generates a detection report containing the fabric performance parameters after decryption by the private key, providing data support for subsequent use decision.

[0186] The present application realizes the separation of physical quantity acquisition by independently setting the temperature and pressure sensor signal lines, avoiding signal crosstalk; the processing device directly performs data fusion in the ciphertext domain, which not only guarantees data security but also reduces the decryption operation amount and improves the system real-time performance.

[0187] By the technical scheme, the application solves the problems of poor spatiotemporal correlation of multi-modal data and lack of privacy protection, realizes accurate synchronous collection and safe transmission of temperature and pressure data through signal line isolation and ciphertext domain fusion technology, reduces data processing delay, and provides reliable data basis for fabric performance evaluation in a dynamic scene.

[0188] Figure 6 A hardware structure schematic diagram of the elastic fabric detection equipment is provided for the embodiments of the application. The elastic fabric detection equipment 600 comprises at least one processor 601, a memory 602, and optionally, the elastic fabric detection equipment 600 can further comprise at least one communication interface 603. The various components in the elastic fabric detection equipment 600 are coupled together through a bus system 604. It can be understood that the bus system 604 is used to realize the connection communication between the components. In addition to the data bus, the bus system 604 also comprises a power supply bus, a control bus and a state signal bus. However, in order to clearly illustrate, all kinds of buses are marked as the bus system 604 in the figure. Figure 6

[0189] ​It is to be understood that the memory 602 can be a volatile memory or a nonvolatile memory, and can also include both volatile and nonvolatile memory. Among them, the nonvolatile memory can be a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a Ferromagnetic Random Access Memory (FRAM), a Flash Memory, a magnetic surface memory, an optical disc, or a Compact Disc Read-Only Memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a Random Access Memory (RAM) used as an external cache. By way of example and not limitation, many forms of RAM can be used, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDR SDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Sync Link Dynamic Random Access Memory (SLDRAM), Direct Rambus Random Access Memory (DRRAM).The memory 602 described in the embodiments of the present application is intended to include, but is not limited to, these and any other suitable type of memory.

[0190] The memory 602 in the embodiments of the present application is used to store various types of data to support the operation of the elastic fabric detection device 600. Examples of these data include: any computer program used to operate on the elastic fabric detection device 600, the program implementing the method of the embodiments of the present application can be contained in the memory 602.

[0191] The method disclosed in the above embodiments of the present application can be applied in the processor 601 or implemented by the processor 601. The processor can be an integrated circuit chip having a processing capability of signals. In the implementation process, each step of the above method can be completed by the integrated logic circuit of hardware in the processor or the instruction in the form of software. The above processor can be a general processor, a digital signal processor (DSP), or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. The processor can implement or execute the disclosed methods, steps and logic block diagrams in the embodiments of the present application. The general processor can be a microprocessor or any conventional processor, etc. In combination with the steps of the method disclosed in the embodiments of the present application, the above-mentioned method can be directly embodied as a hardware decoding processor to be executed, or be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium, which is located in the memory, and the processor reads the information in the memory to complete the above-mentioned method steps in combination with the hardware thereof.

[0192] In the exemplary embodiments, the elastic fabric detection device 600 can be implemented by one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers (MCUs), microprocessors (Microprocessors), or other electronic elements, for executing the above-mentioned method.

[0193] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other manners. The embodiments described above are merely exemplary, for example, the division of the units is only a logical function division, and there can be other division manners in actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed coupling, or direct coupling or communication connection between the components can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or in other forms. The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e. can be located in one place or distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments. In addition, the units in the embodiments of the present application can be integrated in a processing module, or each unit can be a separate unit, or two or more units can be integrated in a unit; the integrated unit can be implemented in the form of hardware, or in the form of hardware plus software function unit.

[0194] The above embodiments are merely used to illustrate the technical solutions of the present application, rather than limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A method of detecting an elastic fabric, characterized by, A detection system applied to an elastic fabric sensor network; The sensor is embedded in the elastic fabric; the detection method comprises: establishing a first node and a second node of the elastic fabric sensor network; the first node comprises the sensor; obtaining first data of the first node; encrypting the first data to obtain encrypted first data; performing ciphertext domain fusion processing on the encrypted first data by using the second node to obtain a feature matrix of the elastic fabric; the feature matrix represents the correlation between the grid coordinates of the elastic fabric and the data of the sensor; sending the encrypted feature matrix to a terminal to make the terminal generate a detection report of the elastic fabric after decrypting the encrypted feature matrix; the detection report is used to determine the use of the elastic fabric; wherein, the ciphertext domain fusion processing on the encrypted first data by using the second node to obtain the feature matrix of the elastic fabric comprises: obtaining a state matrix, an observation matrix, a control matrix and a gain matrix of the encrypted first data by using the second node; performing multi-modal data fusion on the state matrix, the observation matrix, the control matrix and the gain matrix to obtain the feature matrix of the elastic fabric.

2. The detection method according to claim 1, characterized in that, The sensor comprises a temperature sensor and / or a pressure sensor; the method for obtaining the first data of the first node comprises: obtaining temperature data of the temperature sensor and / or pressure data of the pressure sensor under the action of temperature and / or pressure on the elastic fabric.

3. The method of claim 1, wherein The sensor further comprises a Bluetooth sensor; the method further comprises: networking the non-networked nodes in the elastic fabric sensor network through the Bluetooth sensor.

4. The detection method according to claim 3, characterized in that, The encryption processing on the first data to obtain the encrypted first data comprises: mapping and encrypting the first data by using a preset homomorphic encryption technology to obtain the encrypted first data.

5. The detection method according to claim 4, characterized in that, The method further comprises: compressing the dimension of the feature matrix to obtain a compressed feature matrix; the compressed feature matrix is used to send to the terminal after encryption.

6. The assay of any one of claims 1-5, wherein, The method further comprises: obtaining a preset private key; decrypting the encrypted feature matrix by using the private key to obtain a decrypted feature matrix; generating a detection report of the elastic fabric based on the decrypted feature matrix.

7. The detection method according to claim 6, characterized in that, The output end of the sensor is connected in parallel with an RC filter network; the method further comprises: obtaining an output signal of the sensor in the no-load state of the elastic fabric; performing noise filtering on the output signal to obtain a digital signal corresponding to the output signal; performing filtering processing on the digital signal to obtain a filtered digital signal; performing noise compensation correction on the filtered digital signal to obtain the first data.

8. A system for detecting elastic fabric, characterized by The detection system comprises an elastic fabric, a temperature sensor and / or a pressure sensor and a processing device; the temperature sensor and / or the pressure sensor is embedded in the elastic fabric; the temperature sensor and / or the pressure sensor constructs a sensor network; The temperature sensor is connected to a temperature sensor signal line; the pressure sensor is connected to a pressure sensor signal line; the processing device is connected to the temperature sensor signal line and the pressure sensor signal line respectively; wherein, The temperature sensor signal line is used to obtain temperature data of the temperature sensor when the elastic fabric is subjected to temperature action; The pressure sensor signal line is used to obtain pressure data of the pressure sensor when the elastic fabric is subjected to pressure action; The processing device is used to perform encryption processing on the temperature data and / or the pressure data to obtain encrypted first data; perform ciphertext domain fusion processing on the encrypted first data to obtain a feature matrix of the elastic fabric; the feature matrix represents the correlation between the grid coordinates of the elastic fabric and the data of the sensor; the feature matrix is sent to a terminal after being encrypted, so that the terminal generates a detection report of the elastic fabric after decrypting the encrypted feature matrix; the detection report is used to determine the use of the elastic fabric; The ciphertext domain fusion processing on the encrypted first data to obtain the feature matrix of the elastic fabric comprises: Obtaining a state matrix, an observation matrix, a control matrix and a gain matrix of the encrypted first data by using the processing device; Performing multi-modal data fusion on the state matrix, the observation matrix, the control matrix and the gain matrix to obtain the feature matrix of the elastic fabric.

9. An apparatus for detecting an elastic fabric, characterized by comprising: The device comprises a processor and a memory for storing a computer program capable of running on the processor, wherein the processor is used to run the computer program, and the steps of the detection method according to any one of claims 1 to 7 are executed.

Images