On-orbit safety identification method

Abstract

The invention discloses an on-orbit security identification method, which comprises the following steps: a system initialization stage: a trusted third party distributes a key for a satellite and initializes a Cuckou filter; in the coarse granularity comparison stage, a satellite divides a remote sensing scene into sub-images, extracts feature vectors, and performs preliminary difference identification through a Cuckou filter and a secure matrix encryption (SME) technology. In the fine granularity identification stage, if potential differences are found, difference sub-images are confirmed through construction of a binary hash tree and MSIP calculation. And the difference is further accurately positioned in the DCTree-based identification stage. And the differential sub-image is encrypted by using a symmetric session key and is safely transmitted to a ground station, so that the confidentiality and integrity of data are ensured. On the premise of protecting the data confidentiality of the remote sensing image, remote sensing satellite on-orbit safety difference data identification is completed, the method is suitable for the situation that satellite resources are limited, and satellite on-orbit data efficient safety processing and data downlink are achieved.

Description

Technical Field

[0001] The invention relates to satellite communication data security technology, and in particular to an on-orbit security identification method. Background Art

[0002] Modern high-resolution satellite constellations can image the global land surface on a daily basis, which helps companies and researchers obtain information about changes, conduct routine monitoring, and understand the physical world. For example, the PlanetScope constellation has about 130 satellites operating at an orbital altitude of 475 kilometers, providing scene products with a pixel resolution of 3 meters and a field of view of 280 to 630 square kilometers. Due to the limitations of ground station construction, downlink transmission cannot meet the high-speed requirements of satellite on-orbit data collection, such as the data volume generated by hyperspectral and ultra-wide field cameras exceeding 20GB per second. Although modern satellites are equipped with space processors, most of them are specially designed for tasks such as satellite altitude control. For those computationally intensive object detection and image classification tasks, simply transplanting ground multiprocessors is not possible due to the harsh radiation environment and limited on-orbit power supply. Under the challenges of limited on-orbit resources and dynamic network topology, heterogeneous methods are proposed to make full use of the collected data. The first solution is through task coordination, that is, energy-efficient distribution of image processing tasks to multiple embedded processors. The second solution is to design lightweight image processing algorithms. The third solution is downlink reduction, which requires on-orbit cooperation between satellites.

[0003] Since seismic activity may still be active years after an earthquake, satellites in orbit generate redundant and duplicate remote sensing images. Considering the scarce downlink bandwidth resources, only insights reflecting changes in seismic activity are worth downloading and further studying. That is, only information reflecting the difference between two adjacent scenes needs to be transmitted to the ground. However, daily-based EAL regional monitoring can only be achieved through the collaboration of remote sensing constellations, and adjacent scenes are usually taken by different satellites. In addition, the on-orbit difference identification of two adjacent scenes also requires collaboration through the interplanetary link between the two satellites. Since the intellectual property rights of each EAL monitoring image belong to the Disaster Relief Center (DRC), each remote sensing image should be kept confidential between the DRC and the satellite. The disclosure of EAL scenes may first lead to misunderstandings, causing negative public opinion and even chaos in the relevant areas. At the same time, each satellite node provides a vulnerable surface that may be hacked or implanted with malware. Passive injection satellites may monitor the message traffic during scene comparison and difference identification and extract useful information as much as possible. Therefore, collaborative comparison and identification should achieve confidentiality, that is, this process can only disclose the difference part but cannot recover the real content.

[0004] Edge computing is proposed to realize data processing and storage at the edge of the network to meet the strict requirements of high dynamics and near real-time response. On-orbit collaborative computing faces the same problems of high dynamics and low propagation delay as mobile edge computing. At the same time, it is also affected by the dynamic topology caused by the orbital trajectory and the on-orbit processing capabilities under the limitations of space electronic equipment and energy supply. Mobile edge computing has extensively studied security and privacy protection protocols, and its ultimate goal is to achieve secure aggregation and integrity verification of data items collected by IoT devices. Current research on mobile edge computing security and privacy protection schemes focuses on confidentiality and integrity protection of data and models.

[0005] Secure inner product functional encryption allows authorized users to use a key to obtain a function associated with a message, enabling computation on encrypted data without interaction. As an example of functional encryption, inner product functional encryption (IPFE) first generates ciphertext for a vector, calculates the decryption key associated with the vector, and decrypts the result to obtain the inner product.

[0006] Like most mobile networks, edge computing security vulnerabilities mainly exist in: 1) the connection between the server and the edge node; 2) the connection between the edge nodes; and 3) the distributed edge nodes themselves. Distributed edge nodes are at risk of physical and remote network attacks, such as malicious code injection, hardware Trojans, etc. At the same time, the different, heterogeneous and personally owned forms of edge nodes bring complex challenges to device management. Therefore, these edge devices cannot be fully trusted, and data items need to be protected during joint data processing. Mobile edge computing has extensively studied security and privacy protection protocols, with the ultimate goal of achieving secure aggregation and integrity verification of data items collected by IoT devices. Another type of security threat is to destroy learning models through data poisoning and avoidance, because IoT devices are the main data source for model training. In addition, there are attacks on edge nodes to infer personal information, because a large number of data items come from personal wearable devices. Current research on mobile edge computing security and privacy protection schemes focuses on confidentiality and integrity protection of data and models. However, satellite networks are subject to highly dynamic network topologies and extremely limited on-orbit resource constraints, and existing mobile edge computing security enhancement schemes cannot meet the challenges unique to the secure change sub-image recognition use case.

[0007] However, existing secure inner product function encryption schemes rely on intensive cryptographic techniques and cannot adapt to the limitations of on-orbit processing capabilities of space electronic equipment.

[0008] It should be noted that the information disclosed in the above background technology section is only used for understanding the background of the present application, and therefore may include information that does not constitute prior art known to ordinary technicians in the field. Summary of the invention

[0009] The main purpose of the present invention is to overcome the defects existing in the above-mentioned background technology and provide an on-orbit safety identification method.

[0010] To achieve the above object, the present invention adopts the following technical solutions: An on-orbit safety identification method for identifying data differences between remote sensing satellites includes the following steps: System initialization: A trusted third party (TA) initializes the satellite formation, selects a prime modulus, initializes the cuckoo filter and selects a hash function; assigns a symmetric reversible matrix as a key to each satellite, generates a random number as a public key, and distributes key pairs securely; shares symmetric session keys between satellites and ground stations (GS) for secure communication; shares symmetric session keys between satellite pairs to ensure secure sharing and management of keys; Coarse-grained comparison: Two satellites perform monitoring tasks sequentially over the target area, dividing the collected remote sensing scene into multiple sub-images, extracting a set of quantitative geographic features from each sub-image to form a multi-dimensional feature vector; each satellite generates a vector containing sub-image features and inserts the vector items into the cuckoo filter; the sub-image feature vectors are encrypted using the secure matrix encryption technology (SME) and ciphertexts are exchanged to identify possible difference sub-images, allowing for false positive results; Fine-grained recognition: Based on the results of coarse-grained comparison, if the coarse-grained comparison determines that the two sets are exactly the same, further verification is performed to determine whether the two sets are completely consistent. If potential differences are found based on the coarse-grained comparison, or if further verification finds that the two sets are not completely consistent, fine-grained recognition is performed on the difference parts by constructing a binary hash tree and using matrix encryption-based secure inner product (MSIP) calculation to confirm the true difference sub-image. DCTree-based recognition: Starting from the root node of DCTree or the internal node that has failed verification, difference recognition is performed layer by layer downward to accurately locate the specific position and content of the difference sub-image until the different dimensions of the sub-image; Data transmission and verification: After difference recognition is completed, if the two images are judged to be the same, no data is sent; if a difference sub-image is identified, the difference sub-image is encrypted using the symmetric session key shared with the ground station and securely transmitted to the ground station to ensure the confidentiality and integrity of the data.

[0011] Further, the system initialization includes: The trusted third party generates a unique symmetric reversible matrix key and public key for each satellite, and generates a derived key shared across satellites through matrix inversion operations; The trusted third party securely distributes the key pair and symmetric session key to the corresponding satellite node to ensure key independence and resistance to network attacks; at the same time, through the prime modulus and hash function, the cuckoo filter is initialized to support the hash mapping and false positive control of subsequent feature vectors.

[0012] Further, the coarse-grained comparison includes: The satellite generates a structured item vector for security matrix encryption calculation by combining the sub-image feature vector with a shared session key through a hash function, and compresses and stores the structured item vector using a cuckoo filter; The satellites of both parties exchange encrypted feature vector ciphertexts, which are generated by a linear combination of random vectors and hash values, and cross-node calculations are implemented based on secure matrix encryption technology; The receiver preliminarily screens the set of sub-images with differences by decrypting and matching with the cuckoo filter, allowing false positive results to enter the fine-grained verification stage.

[0013] Furthermore, the fine-grained identification includes: When there are potential differences in the coarse-grained comparison, the two satellites construct binary hash trees based on the difference sub-image sets and generate randomized hash vectors of the root nodes. The root node ciphertext is cross-encrypted and decrypted through a secure inner product calculation protocol based on matrix encryption, and the consistency of the root node hash value is verified by combining random vector perturbation and symmetric key derivation technology; If the verification fails, DCTree-based recognition is triggered.

[0014] Furthermore, the fine-grained identification includes: For the false positive results of the cuckoo filter in the coarse-grained comparison, a secondary screening is performed through a binary hash tree verification mechanism, wherein the satellites of both parties exchange the randomized hash ciphertexts of the root nodes, and decrypt and compare them with the randomized hash values ​​through secure inner product calculation based on matrix encryption to eliminate false differences caused by misjudgment of the cuckoo filter; if the comparison is consistent, the coarse-grained result is determined to be a false positive, and the difference transmission process is terminated.

[0015] Furthermore, the identification of the DCTree includes: Starting from the root node of the hash tree, child nodes are selected layer by layer and randomized hash ciphertext is generated. The ciphertext is cross-decrypted through the secure inner product protocol, and the consistency of each layer of nodes is verified by combining the XOR operation of the randomized hash value and the symmetric key. If the verification of a node at a certain layer fails, it continues to recurse to the lower layer until the minimum dimension or leaf node of the difference sub-image is located. When the verification of an intermediate node fails, the child node ciphertext is generated layer by layer starting from this node and recursively verified until the difference dimension is located.

[0016] Furthermore, the data transmission and verification includes: The satellite encrypts the identified difference sub-image using a pre-shared symmetric session key and transmits it to the ground station through a secure channel; After receiving the ciphertext, the ground station reconstructs the hash tree based on the decrypted sub-image, and verifies the integrity and authenticity of the difference data by comparing the root node hash value with the original satellite data to prevent tampering or forgery during transmission.

[0017] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the on-orbit safety identification method is implemented.

[0018] A computer program product, which implements the on-orbit safety identification method when run by a processor.

[0019] The present invention has the following beneficial effects: The present invention proposes an on-orbit security identification method, which completes the on-orbit security difference data identification of remote sensing satellites under the premise of protecting the data confidentiality of remote sensing images, and comprehensively utilizes cryptographic technology and multi-party secure computing technology to make it suitable for situations where satellite resources are limited, and realizes efficient and secure processing and data downlink of satellite on-orbit data. The present invention can perform on-orbit data security difference identification for image data of remote sensing satellites in long-term earthquake accelerated geological disasters (EALs) monitoring, and has the following significant advantages: First, the scheme ensures that the true difference data value of the remote sensing satellite is kept confidential between the Disaster Rescue Center (DRC) and the satellite; second, it ensures the security and confidentiality of the data transmitted in the process of collaborative calculation between satellites to find the intersection of difference data; third, it realizes efficient and secure data difference query of satellite collaboration, greatly reducing the calculation overhead and communication overhead.

[0020] Other beneficial effects of the embodiments of the present invention will be further described below. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 The figure is a flow chart of an on-orbit safety identification method according to an embodiment of the present invention.

[0022] Figure 2 Schematic diagram of an on-orbit difference recognition solution according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The following is a detailed description of the embodiments of the present invention. It should be emphasized that the following description is only exemplary and is not intended to limit the scope and application of the present invention.

[0024] Definitions of abbreviations and key terms; Cuckoo Filter The cuckoo filter is a compact variant of the cuckoo hash table, which consists of two or more hash functions. The basic unit of the cuckoo hash table of the cuckoo filter is called an entry, and each entry stores a fingerprint. hour, ,in, Represents the exclusive OR operation; Fingerprint function It takes a certain amount of bits through a hash function, and if any bucket is empty, stores the fingerprint in the free bucket. If both buckets are not empty, it will select a candidate bucket, delete the existing item, and reinsert the deleted item into the other bucket until an empty bucket is found or the maximum number of displacements is exceeded. To check if there is a , first calculate the fingerprint , and then derive two related buckets , if the fingerprint can be recognized, the cuckoo filter returns True; otherwise, it returns False.

[0025] Secure Matrix Encryption

[0026] Secure matrix encryption technology is an encryption system based on the LWE (Learning With Errors) problem, which is used to implement ciphertext matrix operations and privacy-preserving inner product calculations. Suppose there is a vector and one containing N A dataset of vectors: , represents an n-dimensional vector space where each vector component is an integer modulo p, . Security Matrix Encryption ( SME ) technique consists of four steps, formally defined as .

[0027] : This algorithm randomly selects a reversible matrix , calculate its inverse , and form a key . Then it chooses a random number , As a public key .

[0028] First, the algorithm converts each n Dimensional data records , Expand to ,in from Randomly select. Then use Encryption, where is a randomly generated noise vector satisfying .

[0029] : The algorithm first transforms the n-dimensional vector Expand to in from Random selection, Is a positive number.

[0030] Then use Encryption, where is a random integer noise vector satisfying .

[0031] : Given a ciphertext and , , this algorithm performs

[0032] in Representation model The nearest integer of . Representation vector and Closer.

[0033] Matrix encryption-based SecureInnerProduct (MSIP) The MSIP solution is derived from SME technology. SME implements k-means clustering using outsourced cloud, where and The data owner uses the key M to encrypt the data, and the outsourcing cloud uses the public key Γ to recover the calculation results. However, the purpose of this scheme is to realize the inner product calculation between two independent parties, and only one party can recover the inner product. In order to achieve secure inner product calculation, SME is modified: one party holds the secret matrix , the other party holds , θ is a random key. Given a vector and , MSIP can safely calculate the inner product , specifically defined as It should be noted that MSIP uses SME technology, and the security of SME depends on the difficulty of the Learning With Error (LWE) problem.

[0034] : The algorithm randomly selects an asymmetric reversible matrix , calculate its inverse matrix , and generate a delegated key In addition, it will be a random Identified as the system public key .

[0035] : This algorithm encrypts a vector ,

[0036] Randomly select .at the same time, are all random vectors that satisfy .

[0037] Algorithm input vector , generate ciphertext pair

[0038] in are all random vectors that satisfy .

[0039] : Given a pair of ciphertext and key , restore the calculation results:

[0040] The final calculation results are: .

[0041] Difference Check Tree

[0042] The goal is to achieve two vectors in the following two scenarios and In the first scenario, given The goal is to ensure In the second scenario, given , the goal is to find , so that Given a hash function and contains A vector of elements , use the following procedure to construct the vector Corresponding , formally defined as : : Given an input vector , the algorithm first generates the vector ,in Specifically, each node pass To get, among them , , , and .

[0043] The query process consists of two cases. For Case 1, the goal is to ensure . It is used and Build two trees and output them simultaneously For case 2, given two vectors of the same dimension, and . The goal is to output the nodes of the other half of the nodes in each layer of the binary tree. When , the set of candidate nodes for equivalence testing is .

[0044] The present invention proposes an on-orbit secure identification method for data difference identification between remote sensing satellites. The method is a secure on-orbit difference identification scheme that can be used for adjacent scenes between two satellites. The scheme first combines the cuckoo filter with the MSIP to achieve a secure coarse-grained difference comparison between terrain features at the sub-image level, with the existence of false positives. Then, based on the comparison results, the fine-grained identification of different included sub-images is further achieved by combining the MSIP scheme and the DCTree structure. In the anomaly identification scheme of the present invention, the satellite can only learn the difference between the two sets, but cannot know the true value of the two set data, nor the true content of the difference, which can effectively meet the security requirements of confidentiality.

[0045] See also Figure 1 The embodiment of the present invention provides an on-orbit safety identification method for identifying data differences between remote sensing satellites, comprising the following steps: System initialization: A trusted third party (TA) initializes the satellite formation, selects a prime modulus, initializes the cuckoo filter and selects a hash function; assigns a symmetric reversible matrix as a key to each satellite, generates a random number as a public key, and distributes key pairs securely; shares symmetric session keys between satellites and ground stations (GS) for secure communication; shares symmetric session keys between satellite pairs to ensure secure sharing and management of keys; Coarse-grained comparison: Two satellites perform monitoring tasks sequentially over the target area, dividing the collected remote sensing scene into multiple sub-images, extracting a set of quantitative geographic features from each sub-image to form a multi-dimensional feature vector; each satellite generates a vector containing sub-image features and inserts the vector items into the cuckoo filter; the sub-image feature vectors are encrypted using the secure matrix encryption technology (SME) and ciphertexts are exchanged to identify possible difference sub-images, allowing for false positive results; Fine-grained recognition: Based on the results of coarse-grained comparison, if the coarse-grained comparison determines that the two sets are exactly the same, further verification is performed to determine whether the two sets are completely consistent. If potential differences are found based on the coarse-grained comparison, or if further verification finds that the two sets are not completely consistent, fine-grained recognition is performed on the difference parts by constructing a binary hash tree and using matrix encryption-based secure inner product (MSIP) calculation to confirm the true difference sub-image. DCTree-based recognition: Starting from the root node of DCTree or the internal node that has failed verification, difference recognition is performed layer by layer downward to accurately locate the specific position and content of the difference sub-image until the different dimensions of the sub-image; Data transmission and verification: After difference recognition is completed, if the two images are judged to be the same, no data is sent; if a difference sub-image is identified, the difference sub-image is encrypted using the symmetric session key shared with the ground station and securely transmitted to the ground station to ensure the confidentiality and integrity of the data.

[0046] The method of the present invention completes the on-orbit security difference data identification of remote sensing satellites under the premise of protecting the data confidentiality of remote sensing images, and is applicable to the situation where satellite resources are limited, so as to realize efficient and secure processing and data downlink of satellite on-orbit data.

[0047] In a preferred embodiment, the system initialization specifically includes: a trusted third party generates a unique symmetric reversible matrix key and public key for each satellite, and generates a derived key shared across satellites through matrix inverse operations; the trusted third party securely distributes the key pair and the symmetric session key to the corresponding satellite nodes to ensure key independence and resistance to network attacks; at the same time, through the prime modulus and hash function, the cuckoo filter is initialized to support the hash mapping and false positive control of subsequent feature vectors.

[0048] In a preferred embodiment, the coarse-grained comparison specifically includes: the satellite generates a structured term vector for security matrix encryption calculation by combining the sub-image feature vector with a shared session key through a hash function, and compresses and stores the structured term vector using a cuckoo filter; the satellites of both parties exchange encrypted feature vector ciphertexts, which are generated by a linear combination of a random vector and a hash value, and implement cross-node calculations based on security matrix encryption technology; the receiving party preliminarily screens a set of sub-images with differences by decrypting and matching with the cuckoo filter, allowing false positive results to enter the fine-grained verification stage.

[0049] In a preferred embodiment, the fine-grained recognition specifically includes: when there are potential differences in the coarse-grained comparison, the satellites of both parties respectively construct binary hash trees based on the difference sub-image sets to generate a randomized hash vector of the root node; the root node ciphertext is cross-encrypted and decrypted through a secure inner product calculation protocol based on matrix encryption, and the consistency of the root node hash value is verified by combining random vector perturbation and symmetric key derivation technology; if the verification fails, DCTree-based recognition is triggered.

[0050] In a preferred embodiment, the fine-grained identification specifically includes: for the false positive results of the cuckoo filter in the coarse-grained comparison, secondary screening is performed through a binary hash tree verification mechanism, wherein the randomized hash ciphertexts of the root nodes of the two satellites are exchanged, and decrypted and compared with the randomized hash values ​​through secure inner product calculation based on matrix encryption to eliminate false differences caused by misjudgment of the cuckoo filter; if the comparison is consistent, the coarse-grained result is determined to be a false positive, and the difference transmission process is terminated.

[0051] In a preferred embodiment, the identification of the DCTree specifically includes: starting from the root node of the hash tree, selecting child nodes layer by layer and generating randomized hash ciphertext, cross-decrypting the ciphertext through a secure inner product protocol, combining the XOR operation of the randomized hash value with the symmetric key, and verifying the consistency of each layer of nodes; if the verification of a node at a certain layer fails, continue to recurse to the lower layer until the minimum dimension or leaf node of the difference sub-image is located; when the verification of an intermediate node fails, generate child node ciphertext layer by layer starting from the node and recursively verify until the difference dimension is located.

[0052] In a preferred embodiment, the data transmission and verification specifically include: the satellite encrypts the identified difference sub-image using a pre-shared symmetric session key and transmits it to the ground station through a secure channel; after the ground station receives the ciphertext, it reconstructs the hash tree based on the decrypted sub-image, and verifies the integrity and authenticity of the difference data by comparing the root node hash value with the original satellite data to prevent tampering or forgery during transmission.

[0053] The present invention can efficiently and securely complete the on-orbit difference data identification of remote sensing satellites under the premise of protecting the confidentiality of remote sensing image data, meeting the challenges brought by limited satellite resources and dynamic network topology. In the method of the present invention, in order to achieve security protection of on-orbit difference comparison of remote sensing satellites, a satellite collaborative on-orbit difference data security identification scheme based on a matrix encryption-based secure inner product calculation protocol is designed. In this scheme, the satellite can only obtain the result of the image difference data, and cannot know the true value of the data and the true value of the difference data. In order to achieve efficient security identification of on-orbit image difference data of remote sensing satellites, a difference data identification scheme using cuckoo filters and homomorphic matrix encryption technology is also designed to achieve secure difference comparison between the terrain features of the image, and improve the difference identification efficiency of the onboard processor in a resource-constrained environment.

[0054] In the application of the present invention, the collaboration between satellites to reduce the redundancy of on-orbit images can be focused on, which can be achieved through long-term earthquake accelerated geological hazards (EALs) monitoring.

[0055] The specific embodiments of the present invention and algorithm examples thereof are further described below.

[0056] Since geological hazard observation is a long-term process, the scenes acquired by two adjacent satellites may have only slight differences, and only the changes in terrain features may provide valuable information. Due to downlink bandwidth limitations, a pair of adjacent satellites should collaborate in orbit to perform difference comparison, and the second satellite only transmits the changed terrain feature sub-image to the ground station. Figure 2 Two satellites are shown and An example of on-orbit identification. Assume First, it flies over the target observation area and first flies over the ground station. In phase 1, it transmits the entire image to the ground station. In phase 2, and First, the difference comparison is performed collaboratively on orbit. Only the identified different sub-images are transmitted to the ground station.

[0057] The workflow of the main system of the present invention includes four parts: 1) system initialization; 2) coarse-grained comparison; 3) fine-grained recognition; 4) DCTree-based recognition. Its workflow is as follows: 1) System initialization

[0058] During system initialization, we assume that there is a trusted third party (TA) to start the entire satellite formation. The TA first selects two prime modules p and q ,Right now . Then TA initializes a cuckoo filter using the steps defined in Section 3. At the same time, TA selects two hash functions and .

[0059] For each satellite , TA chooses a symmetric reversible matrix As the key. At the same time, TA selects a random number In addition, TA will use the secret public key pair Safely sent to In addition, TA GS and A symmetric session key is shared between , used for remote sensing image transmission.

[0060] For any pair of satellites in the satellite formation and ,TA first shares a symmetric session key between them during initialization Given Secret Key ,TA also performs the following steps: 1) selects a random number and will Send to ;2) Calculate the secret key ;3) Safely sent to In addition, TA also Secret Key Follow the exact same process.

[0061] 2) Coarse-grained comparison

[0062] When the satellite and Sequentially over the target area EAL When monitoring, they use onboard instruments to collect adjacent remote sensing scenes. For ease of comparison, each satellite divides a large scene into a series of d dimensional sub-images (each containing a set of quantitative geographic features), i.e. and . To perform coarse-grained sub-image comparison, First, generate a vector ,in and It's TA and A symmetric session key is then shared between Perform the following steps to Insert the items in the cuckoo filter For each item , , Select One dimensional random vector .at the same time, Compute a structured term vector ,in , , and Insert the items in the cuckoo filter .at last, Formulate and send a message Give .

[0063] receive back, Generate a vector , which contains a series of sub-images , these sub-images are obtained by For each project , , Calculate a vector and by running Generate ciphertext pair , the formula is:

[0064] besides, Constructing the message , and Send to .

[0065] receive back, For each ciphertext run ,in yes A random vector is selected. At the same time, Construct another message and will Send to .receive back, For each ciphertext run Decrypt to get the corresponding structured item , .at last, Each item , With Cuckoo Filter to make a match.

[0066] 3) Fine-grained recognition if Discover All items in ,So Execute the steps of case 1; otherwise, Follow the steps for Case 2.

[0067] Case 1. In this case, The goal is to ensure , because the cuckoo filter may have false positives, it performs the following steps: First run Build a binary hash tree and create the corresponding root node Generate ciphertext ,in .at the same time, Constructing the message , and send it to .

[0068] receive back, First run Build a binary hash tree. At the same time, Pick a random vector , is the root node obtained Generate vector , and calculate the ciphertext .also, calculate .at last, Constructing the message And send to .

[0069] receive back, First of all Decryption; at the same time, calculate , and verify Is it equal to .if , verify and are the same; otherwise, A DCTree-based recognition process is performed to identify processes of different data dimensions.

[0070] Case 2. In this case, The goal is to ensure that, except for the identified changed nodes, the remaining nodes are the same. Items and there are no false positives, then the node Should not belong to .therefore, It should be ensured that all remaining nodes in the binary hash tree are Specifically, The following process is performed for query response generation: First select the necessary nodes in the binary hash tree to be Reconstruct the remaining nodes, expressed as And the depth is .at the same time, for Generate encrypted text ,in .also, use The index of each node is encrypted and expressed as . Crafting Messages , and send it to .

[0071] receive back, First use Index Decrypt to get the corresponding node .at the same time, For each Process, get In addition, it calculates , .in addition, run Get the ciphertext .at last, Crafting Messages Send to .

[0072] receive back, right Decrypt to get the structure value .at the same time, verify Is it equal to .if , verify and remove Otherwise, Execute the DCTree-based recognition process for further recognition.

[0073] After the difference identification process, if and Determine whether two images are the same. Nothing will be sent to GS. Once a different sub-image has been identified, it uses the symmetric session key shared with the GS Encrypt the sub-image and send the ciphertext to GS in a secure way.

[0074] 4) DCTree-based recognition

[0075] To identify the next changing leaf node of the DCTree root node, and The steps defined in case 1 can be repeatedly performed on the hash tree from the root node to the lower layer until the different dimensions of the sub-image. For case 1, the recognition process is performed from the root node to the lower layer. For case 2, the recognition process is performed only from different internal nodes to the lower layer data dimensions.

[0076] 5) Data transmission and verification

[0077] Based on the difference recognition results, if the two images are judged to be the same, no data is sent; if a difference sub-image is identified, the difference sub-image is encrypted using the symmetric session key shared with the ground station and securely transmitted to the ground station to ensure the confidentiality and integrity of the data.

[0078] In summary, the present invention proposes an efficient and secure solution for the on-orbit secure difference identification of remote sensing satellite image data in long-term earthquake accelerated geological disaster (EALs) monitoring. By combining cuckoo filters, matrix encryption-based secure inner product calculation (MSIP) and difference check tree (DCTree) structure, efficient and secure processing and difference identification of satellite image data are achieved under the premise of protecting data confidentiality. This solution not only ensures that the real difference data of remote sensing satellites are strictly confidential between the Disaster Relief Center (DRC) and the satellite to prevent data leakage or unauthorized access, but also ensures the security and confidentiality of data transmission in the collaborative computing process between satellites through encryption technology and security protocols, avoiding the risk of data being tampered with or stolen during transmission and processing. In addition, the present invention makes full use of cryptographic technology and multi-party secure computing technology, optimizes the computing and communication processes, significantly reduces the computing overhead and communication costs, and enables it to adapt to the environment with limited satellite resources, thereby achieving efficient and secure data difference query and processing under the conditions of meeting high dynamic network topology and limited on-orbit resources, and provides an innovative solution for the efficient management and secure transmission of remote sensing satellite on-orbit data.

[0079] An embodiment of the present invention further provides a storage medium for storing a computer program, which at least performs the above method when executed.

[0080] An embodiment of the present invention further provides a control device, comprising a processor and a storage medium for storing a computer program; wherein the processor is configured to execute at least the method described above when executing the computer program.

[0081] An embodiment of the present invention further provides a processor, wherein the processor executes a computer program and at least executes the method described above.

[0082] The storage medium can be implemented by any type of non-volatile storage device, or a combination thereof. Among them, the non-volatile 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 magnetic random access memory (FRAM), a ferromagnetic random access memory (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 storage medium described in the embodiments of the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0083] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0084] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0085] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0086] A person skilled in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, etc. Various media that can store program codes.

[0087] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0088] The methods disclosed in the several method embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method embodiments.

[0089] The features disclosed in several product embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new product embodiments.

[0090] The features disclosed in several method or device embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0091] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art of the present invention, several equivalent substitutions or obvious variations can be made without departing from the concept of the present invention, and the performance or use is the same, which should be regarded as belonging to the protection scope of the present invention.

Claims

1. An on-orbit safety identification method for identifying data differences between remote sensing satellites, characterized in that: The following steps are involved: System initialization: A trusted third party initializes the satellite formation, selects a prime modulus, initializes the cuckoo filter and selects a hash function; assigns a symmetric reversible matrix as a key to each satellite, generates a random number as a public key, and distributes key pairs securely; shares symmetric session keys between satellites and ground stations for secure communication; shares symmetric session keys between satellite pairs to ensure secure sharing and management of keys; Coarse-grained comparison: Two satellites perform monitoring tasks sequentially over the target area, dividing the collected remote sensing scene into multiple sub-images, extracting a set of quantitative geographic features from each sub-image to form a multi-dimensional feature vector; each satellite generates a vector containing sub-image features and inserts the vector items into the cuckoo filter; the sub-image feature vectors are encrypted using secure matrix encryption technology, and ciphertexts are exchanged to identify possible difference sub-images, allowing for false positive results; Fine-grained identification: Based on the results of coarse-grained comparison, if the coarse-grained comparison determines that the two sets are exactly the same, further verification is performed to determine whether the two sets are completely consistent; If potential differences are found based on the coarse-grained comparison, or if further verification shows that the two sets are not completely consistent, the difference parts are fine-grainedly identified by constructing a binary hash tree and using secure inner product calculation based on matrix encryption to confirm the true difference sub-image; DCTree-based recognition: Starting from the root node of DCTree or the internal node that has failed verification, difference recognition is performed layer by layer downward to accurately locate the specific position and content of the difference sub-image until the different dimensions of the sub-image; Data transmission and verification: Based on the difference recognition results, if the two images are judged to be the same, no data is sent; if a difference sub-image is identified, the difference sub-image is encrypted using the symmetric session key shared with the ground station and securely transmitted to the ground station to ensure the confidentiality and integrity of the data.

2. The on-track safety identification method according to claim 1, characterized in that: The system initialization includes: The trusted third party generates a unique symmetric reversible matrix key and public key for each satellite, and generates a derived key shared across satellites through matrix inversion operations; The trusted third party securely distributes the key pair and symmetric session key to the corresponding satellite node to ensure key independence and resistance to network attacks; at the same time, through the prime modulus and hash function, the cuckoo filter is initialized to support the hash mapping and false positive control of subsequent feature vectors.

3. The on-track safety identification method according to claim 1, characterized in that: The coarse-grained comparison includes: The satellite generates a structured item vector for security matrix encryption calculation by combining the sub-image feature vector with a shared session key through a hash function, and compresses and stores the structured item vector using a cuckoo filter; The satellites of both parties exchange encrypted feature vector ciphertexts, which are generated by linear combinations of random vectors and hash values, and cross-node calculations are implemented based on secure matrix encryption technology; The receiver preliminarily screens the set of sub-images with differences by decrypting and matching with the cuckoo filter, allowing false positive results to enter the fine-grained verification stage.

4. The on-track safety identification method according to claim 1, characterized in that: The fine-grained recognition includes: When there are potential differences in the coarse-grained comparison, the two satellites construct binary hash trees based on the difference sub-image sets and generate randomized hash vectors of the root nodes. The root node ciphertext is cross-encrypted and decrypted through a secure inner product calculation protocol based on matrix encryption, and the consistency of the root node hash value is verified by combining random vector perturbation and symmetric key derivation technology; If the verification fails, DCTree-based recognition is triggered.

5. The on-track safety identification method according to claim 1, characterized in that: The fine-grained recognition includes: For the false positive results of the cuckoo filter in the coarse-grained comparison, a secondary screening is performed through a binary hash tree verification mechanism, wherein the satellites of both parties exchange the randomized hash ciphertexts of the root nodes, and decrypt and compare them with the randomized hash values ​​through secure inner product calculation based on matrix encryption to eliminate false differences caused by misjudgment of the cuckoo filter; if the comparison is consistent, the coarse-grained result is determined to be a false positive, and the difference transmission process is terminated.

6. The on-track safety identification method according to claim 1, characterized in that: The identification of the DCTree includes: Starting from the root node of the hash tree, child nodes are selected layer by layer and randomized hash ciphertext is generated. The ciphertext is cross-decrypted through a secure inner product protocol based on matrix encryption, and the consistency of each layer of nodes is verified by combining the XOR operation of the randomized hash value and the symmetric key. If the verification of a node at a certain layer fails, it will continue to recurse to the lower layer until the minimum dimension or leaf node of the difference sub-image is located. When the verification of an intermediate node of the hash tree fails, the child node ciphertext is generated layer by layer starting from this node and recursively verified until the difference dimension is located.

7. The on-track safety identification method according to claim 1, characterized in that: The data transmission and verification include: The satellite encrypts the identified difference sub-image using a pre-shared symmetric session key and transmits it to the ground station through a secure channel; After receiving the ciphertext, the ground station reconstructs the hash tree based on the decrypted sub-image, and verifies the integrity and authenticity of the difference data by comparing the root node hash value with the original satellite data to prevent tampering or forgery during transmission.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the on-orbit safety identification method according to any one of claims 1 to 7 is implemented.

9. A computer program product, characterized in that When the computer program product is executed by a processor, the on-orbit safety identification method according to any one of claims 1 to 7 is implemented.

Images