Reputation-based privacy protection unmanned aerial vehicle task matching method and system applied to mobile crowd sensing

Abstract

The invention discloses a reputation-based privacy protection unmanned aerial vehicle task matching method and system applied to mobile crowd sensing, and the method comprises the steps: encrypting a task through a delegation predicate encryption technology by a task requester, and guaranteeing that only an unmanned aerial vehicle meeting an attribute condition can decrypt the task; the mobile crowd sensing platform is responsible for task distribution and matching without acquiring attribute information of tasks or unmanned aerial vehicles; the ground station provides a safe data transmission channel among the mobile crowd sensing platform, the trusted mechanism and the unmanned aerial vehicle; the unmanned aerial vehicle serves as a task executor and has the capabilities of decrypting the task, collecting data and participating in reputation evaluation. According to the method, a hybrid reputation evaluation mechanism is introduced, direct reputation, indirect reputation, role reputation and historical reputation are integrated, comprehensive evaluation and dynamic updating of the node credibility of the unmanned aerial vehicle are realized, the position, identity and reputation privacy of the unmanned aerial vehicle is protected through an encryption technology, and the reliability of the unmanned aerial vehicle is improved on the premise of protecting the privacy of both parties. The efficient and reliable multi-attribute task matching and reputation management are realized, and the system security and the task execution efficiency are improved.

Description

Technical Field

[0001] This invention relates to encryption technology for protecting the privacy of unmanned aerial vehicles (UAVs), specifically to a reputation-based privacy-preserving UAV task matching method and system applied to mobile swarm intelligence sensing. Background Technology

[0002] Mobile Crowdsensing (MCS) leverages the smart devices of general users to collaboratively complete complex data collection and task execution. Drones, with their flexibility, low cost, and rapid deployment capabilities, have become important task execution tools in MCS environments. However, assessing the reliability of drones during task coordination, and achieving efficient, accurate, and privacy-preserving task matching based on this reliability, remains a critical issue that urgently needs to be addressed.

[0003] Existing research on privacy protection in task matching largely focuses on using techniques such as k-anonymity and differential privacy to hide the location information of drones, but generally neglects the confidentiality of the attributes of the task requester. This allows the platform to still infer sensitive information from the matching results, creating a situation of "one-sided privacy" protection. In addition, most solutions do not fully consider the differences in drones' computing power, payload capacity, reputation, and other multi-dimensional attributes, lacking the ability to support flexible multi-attribute matching and making it difficult to achieve intelligent adaptation between tasks and resources.

[0004] In the area of ​​drone trust management, although some solutions have introduced reputation mechanisms to improve system security, significant limitations remain. For example, some centralized reputation assessment methods rely on a single trusted center for reputation calculation. Liu et al. proposed a non-interactive data trust assessment scheme based on a ground control station. While this scheme achieves a certain degree of privacy protection through hash functions, its architecture is susceptible to single-point-of-failure risks. Furthermore, existing schemes primarily evaluate drones based on static attributes, failing to adequately reflect their dynamic behavior and interaction history, which can easily lead to performance bottlenecks as the node scale expands. On the other hand, while distributed reputation mechanisms avoid central dependence, they often rely on a single evaluation factor, incur high communication overhead, and are vulnerable to collusion by malicious local nodes, resulting in insufficient robustness of the evaluation results. More importantly, most existing reputation schemes do not impose privacy protection on the reputation value itself. Attackers can infer drone behavior trajectories by analyzing changes in reputation values, leading to secondary privacy leaks.

[0005] Therefore, current system design for drone MCS urgently needs an integrated solution that can take into account the privacy of both the tasker and the executor, support accurate matching of multiple attributes, and have a lightweight, scalable, and privacy-protected reputation management mechanism to achieve safe, reliable, and efficient task collaboration in an open and dynamic environment. Summary of the Invention

[0006] Purpose of the invention: The purpose of this invention is to address the shortcomings of existing technologies and provide a reputation-based privacy-preserving drone task matching method and system for mobile swarm intelligence sensing. This method and system can achieve multi-attribute matching while protecting the privacy of task requirements and drone attributes. Furthermore, through a hybrid reputation evaluation mechanism, it enables comprehensive, accurate, and privacy-preserving reputation management of drone nodes, thereby improving the security, reliability, and efficiency of the entire MCS system.

[0007] Technical solution: The present invention provides a reputation-based privacy-preserving drone task matching method for mobile crowd sensing, comprising the following steps:

[0008] Step 1, System Initialization and Key Generation, includes system initialization, task requester registration, and drone registration, where the Trusted Organization (TA) sets a complete set of attributes. and generate a system public key. and system private key Task requester With attribute set And all task attributes The task requester applies for registration with TA, and TA generates a delegated predicate encryption key for them. drones With attribute set And all drone attributes The drone applies for registration with a trusted authority (TA), which generates a predicate decryption key for it. and kana ;

[0009] Step 2: Mobile crowd sensing task matching;

[0010] Step 2.1, Task Encryption and Issuance, i.e., Task Requester Use its delegated predicate to encrypt the key. Task information Encrypt and generate task ciphertext. It is then submitted to the Mobile Crowd Sensing Platform (MCS), which broadcasts the submission.

[0011] Step 2.2, Task Decryption and Confirmation, UAV Receive task ciphertext Then, use its predicate to decrypt the key. Decryption is performed; if the drone meets the mission requirements, decryption is successful, and a confirmation message is returned to the Mobile Swarm Intelligence Platform (MCS); after verifying the confirmation message, the MCS sends a recruitment confirmation message to the drone. ;

[0012] Step 2.3: Upload task data and receive confirmation upon successful completion. The drones are collecting mission data Then, encrypt it as Along with confirmation information The credentials were uploaded together to the Mobile Crowd Sensing Platform (MCS);

[0013] Step 3: Hybrid credit assessment;

[0014] Step 3.1: Multi-reference source reputation assessment. Reputation assessments are conducted between drone nodes based on direct interaction, neighbor recommendations, role information, and historical reputation to generate drone... For drones Individual overall credit Simultaneously, the Mobile Crowd Sensing Platform (MCS) will also generate task feedback reports. And send it to a trusted organization (TA);

[0015] Step 3.2: Global Reputation Update. Trusted organizations (TAs) aggregate the overall reputation of all individuals and task feedback to update the drone's reputation. Global reputation score And mapped to reputation level ;

[0016] Step 3.3: Key and pseudonym update. The trusted authority (TA) generates a new decryption key for the drone based on the updated reputation level. and kana .

[0017] The privacy protection of task matching in this invention is reflected in the following two aspects: During the task publishing and matching process, the task requester encrypts the task using a delegated predicate encryption key authorized by a trusted institution. The published task requirements must be encrypted and meet the system's predefined policy constraints. The mobile swarm intelligence sensing platform completes the ciphertext matching without knowing the specific attribute information. During the task execution and data submission phase, only drones possessing decryption keys matching the task conditions can unlock the task. Simultaneously, their task execution performance is incorporated into a dynamic reputation evaluation system; drones with insufficient reputation will be unable to participate in subsequent high-requirement tasks. By introducing delegated technology and predicate encryption, control over the task requester's publishing authority is achieved; by integrating the evaluation of multi-reference source reputation, the reliability of the task executor is guaranteed, thus achieving privacy protection for both the publisher and the executor throughout the entire task lifecycle.

[0018] Furthermore, the system public key in step 1 and system private key The generation method is as follows:

[0019] Select a security parameter And generate key pairs Define a bilinear mapping ,in , , It is a prime number. The cyclic group, whose generators are respectively Trusted Institution (TA) selects hash function And randomly select parameters Then calculate the system public key. and system private key The formula is as follows:

[0020] ,

[0021] ;

[0022] The detailed method by which a trusted authority (TA) generates a delegated predicate encryption key for a task requester is as follows:

[0023] First, the task requester Choose an asymmetric key pair and and generate a signature. .in, It is the public key used by the task requester for signing. It is its corresponding private key; It is the public key used by the task requester for encryption. It is its corresponding private key; then, the task requester submits its attribute set SA, and the trusted authority generates a key based on the SA. 3D attribute vector Random selection Calculate the delegated predicate encryption key The formula is as follows:

[0024] ;

[0025] The detailed method for generating predicate decryption keys for drones is as follows:

[0026] First, drones Select key pair And generate a signature ;

[0027] Then, the drone submits its attribute set (UA), and the trusted authority generates a predicate vector based on the UA. Random selection Calculate the decryption key ;

[0028] At the same time, trusted institutions randomly select Assigning pseudonyms to drones ;

[0029] in For drones His true identity.

[0030] Furthermore, the specific method for step 2, mobile swarm intelligence sensing task matching, is as follows:

[0031] First, the task requester Random selection Use its delegated predicate to encrypt the key For extended plaintext Encrypt and generate ciphertext. The expression is as follows:

[0032] ;

[0033] in, This is the current timestamp;

[0034] The task requester will send the encrypted message and timestamp. They are sent together to the Mobile Crowd Sensing Platform (MCS), which then broadcasts the task.

[0035] Then, perform the task of decryption and uploading. The detailed method is as follows:

[0036] drones Use its predicate to decrypt the key. calculate:

[0037]

[0038] If and only if In other words, when the drone's attributes meet the mission requirements, the drone can correctly decrypt and obtain the mission message. ;

[0039] Then, the drone returns confirmation information, including pseudonyms, to the Mobile Crowd Sensing Platform (MCS). and task identifier The Mobile Crowd Sensing Platform (MCS) verifies the pseudonym of the drone and checks the timestamp. If the number of recruits for the task is not full, a random selection will be made. Send recruitment confirmation information to drones The expression is as follows: ;

[0040] Successfully received The drones are collecting mission data Then, encrypt it. The expression is: ;

[0041] This refers to hashing the task data;

[0042] Then, along with The credentials are uploaded together to the Mobile Crowd Sensing Platform (MCS).

[0043] Furthermore, the detailed method for multi-reference source reputation assessment in step S3 is as follows:

[0044] drones For another drone Individual overall credit Taking into account the direct reputation of drone nodes Recommended Reputation Role Reputation and historical reputation The formula is as follows:

[0045] ;

[0046] In the above formula, , , and These are the adaptive weights of the corresponding reference sources, and satisfy... ;

[0047] Among them, direct credit For drones With the target drone Reputation score obtained through direct interaction;

[0048] drones Recorded with drones Every interaction And there is a level of satisfaction for each interaction. Place in direct interaction set Among them , express think The information provided is incorrect. express For use The information provided Correct; since new interaction records have higher reference value, a time decay function is introduced as a weight for direct reputation. ;

[0049] in The difference between the current time and the recorded time. As a control factor; ultimately, direct reputation is obtained:

[0050] ;

[0051] Recommended Reputation Value It is a drone node Through common neighbor nodes The obtained information about the target drone node Evaluation;

[0052] To mitigate the impact of biased opinions from neighbors, similarity is introduced. ,in As a weighting factor, the recommendation reputation The calculation is as follows:

[0053] ;

[0054] in For drones For drones direct credibility, For recommending interactive sets, The number of common neighbors;

[0055] For drone nodes Trust levels related to current roles, including police drones, commercial drones, and private drones;

[0056] For target drone node The weighted sum of the last historical global reputation values.

[0057] Furthermore, the detailed method for the global reputation assessment in step S3 is as follows:

[0058] A trusted institution combines the overall reputation of all individuals in this round. and task feedback report , This indicates that the mission was successful. This indicates that the task has failed.

[0059] drones Global reputation The calculation formula is:

[0060] ;

[0061] in, , and For feedback weights, and satisfying ; For the evaluation set of nodes; For individual comprehensive credit and nodes To the node The difference between the average overall credit rating; A collection of feedback reports; The total number of historical credit records; yes The k-th historical global reputation; Let k be the time interval between the current time and the k-th historical reputation record;

[0062] Then, the trusted institution maps the global reputation value to a reputation level. ,calculate:

[0063] ;

[0064] Updated global reputation rating This is further embedded into the drone's decryption vector; subsequently, the trusted authority determines the decryption based on the new reputation level. for Update the corresponding decryption key and kana .

[0065] This invention also discloses a system for a reputation-based privacy-preserving drone task matching method applied to mobile swarm intelligence sensing, comprising a trusted institution, a task requester, a mobile swarm intelligence sensing platform, a ground station, and a drone; the trusted institution is configured to perform system initialization, key generation, reputation assessment and updating, and to allocate and update pseudonyms and decryption keys for the drone; the task requester encrypts the task using a delegated encryption key obtained from the trusted institution and submits the encrypted task to the mobile swarm intelligence sensing platform; the mobile swarm intelligence sensing platform is responsible for receiving, storing, and broadcasting the encrypted task, verifying the drone task confirmation information, aggregating task data, and generating a task feedback report; the ground station provides a secure data transmission channel between the mobile swarm intelligence sensing platform, the trusted institution, and the drone; the drone receives the encrypted task from the platform, uses its decryption key to decrypt and execute the task, collects and uploads sensing data, and participates in distributed reputation assessment based on multiple reference sources.

[0066] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0067] (1) The drone task matching scheme of the present invention enables the MCS platform to complete accurate matching without knowing the task attributes and drone attributes, thereby realizing the privacy protection of both the task party and the drone party.

[0068] (2) The present invention effectively balances the dependence on centralized infrastructure through a hybrid reputation assessment with multiple reference sources, and further improves the accuracy of UAV reputation assessment. Attached Figure Description

[0069] Figure 1 This is a schematic diagram of the system structure of the present invention;

[0070] Figure 2 This is an overall flowchart of the present invention;

[0071] Figure 3 This is a flowchart of the initialization process of the present invention;

[0072] Figure 4 This is a flowchart of the task matching process of the present invention;

[0073] Figure 5 This is a flowchart of the hybrid credit assessment process of the present invention;

[0074] Figure 6 The figures show a comparison of experiments for various technical solutions in the embodiments. Detailed Implementation

[0075] The technical solution of the present invention will be described in detail below, but the scope of protection of the present invention is not limited to the embodiments described.

[0076] like Figure 1 and Figure 2 As shown, the reputation-based privacy-preserving drone task matching method of the present invention, applied to mobile crowd sensing, includes the following steps:

[0077] Step 1, System Initialization and Key Generation, includes system initialization, task requester registration, and drone registration, where the Trusted Organization (TA) sets a complete set of attributes. and generate a system public key. and system private key Task requester With attribute set And all task attributes The task requester applies for registration with TA, and TA generates a delegated predicate encryption key for them. drones With attribute set And all drone attributes The drone applies for registration with a trusted authority (TA), which generates a predicate decryption key for it. and kana ;

[0078] Step 2: Mobile crowd sensing task matching;

[0079] Step 2.1, Task Encryption and Issuance, i.e., Task Requester Use its delegated predicate to encrypt the key. Task information Encrypt and generate task ciphertext. It is then submitted to the Mobile Crowd Sensing Platform (MCS), which broadcasts the submission.

[0080] Step 2.2, Task Decryption and Confirmation, UAV Receive task ciphertext Then, use its predicate to decrypt the key. Decryption is performed; if the drone meets the mission requirements, decryption is successful, and a confirmation message is returned to the Mobile Swarm Intelligence Platform (MCS); after verifying the confirmation message, the MCS sends a recruitment confirmation message to the drone. ;

[0081] Step 2.3: Upload task data and receive confirmation upon successful completion. The drones are collecting mission data Then, encrypt it as Along with confirmation information The credentials were uploaded together to the Mobile Crowd Sensing Platform (MCS);

[0082] Step 3: Hybrid credit assessment;

[0083] Step 3.1: Multi-reference source reputation assessment. Reputation assessments are conducted between drone nodes based on direct interaction, neighbor recommendations, role information, and historical reputation to generate drone... For drones Individual overall credit Simultaneously, the Mobile Crowd Sensing Platform (MCS) will also generate task feedback reports. And send it to a trusted organization (TA);

[0084] Step 3.2: Global Reputation Update. Trusted organizations (TAs) aggregate the overall reputation of all individuals and task feedback to update the drone's reputation. Global reputation score And mapped to reputation level ;

[0085] Step 3.3: Key and pseudonym update. The trusted authority (TA) generates a new decryption key for the drone based on the updated reputation level. and kana .

[0086] like Figure 3 As shown, the system public key in step 1 of this embodiment and system private key The generation method is as follows:

[0087] Select a security parameter And generate key pairs Define a bilinear mapping ,in , , It is a prime number. The cyclic group, whose generators are respectively Trusted Institution (TA) selects hash function And randomly select parameters Then calculate the system public key. and system private key The formula is as follows:

[0088] ,

[0089] ;

[0090] The detailed method by which a trusted authority (TA) generates a delegated predicate encryption key for a task requester is as follows:

[0091] First, the task requester Select key pair and Used for signing and encryption respectively, and for generating signatures. Then, the task requester submits its attribute set SA, and the trusted authority generates an attribute set SA based on the SA. 3D attribute vector Random selection Calculate the delegated predicate encryption key The formula is as follows:

[0092] ;

[0093] The detailed method for generating predicate decryption keys for drones is as follows:

[0094] First, drones Select key pair And generate a signature ;

[0095] Then, the drone submits its attribute set (UA), and the trusted authority generates a predicate vector based on the UA. Random selection Calculate the decryption key ;

[0096] At the same time, trusted institutions randomly select Assigning pseudonyms to drones ;

[0097] in For drones His true identity.

[0098] like Figure 4 As shown, the specific method for mobile crowd sensing task matching in step 2 of this embodiment is as follows:

[0099] First, the task requester Random selection Use its delegated predicate to encrypt the key For extended plaintext Encrypt and generate ciphertext. The expression is as follows:

[0100] ;

[0101] in, This is the current timestamp;

[0102] The task requester will send the encrypted message and timestamp. They are sent together to the Mobile Crowd Sensing Platform (MCS), which then broadcasts the task.

[0103] Then, perform the task of decryption and uploading. The detailed method is as follows:

[0104] drones Use its predicate to decrypt the key. calculate:

[0105]

[0106] If and only if In other words, when the drone's attributes meet the mission requirements, the drone can correctly decrypt and obtain the mission message. ;

[0107] Then, the drone returns confirmation information, including pseudonyms, to the Mobile Crowd Sensing Platform (MCS). and task identifier The Mobile Crowd Sensing Platform (MCS) verifies the pseudonym of the drone and checks the timestamp. If the number of recruits for the task is not full, a random selection will be made. Send recruitment confirmation information to drones The expression is as follows: ;

[0108] Successfully received The drones are collecting mission data Then, encrypt it. The expression is: ;

[0109] This refers to hashing the task data;

[0110] Then, along with The credentials are uploaded together to the Mobile Crowd Sensing Platform (MCS).

[0111] like Figure 5As shown, the detailed method for multi-reference source reputation assessment in step S3 of this embodiment is as follows:

[0112] drones For another drone Individual overall credit Taking into account the direct reputation of drone nodes Recommended Reputation Role Reputation and historical reputation The formula is as follows:

[0113] ;

[0114] In the above formula, , , and These are the adaptive weights of the corresponding reference sources, and satisfy... ;

[0115] Among them, direct credit For drones With the target drone Reputation score obtained through direct interaction;

[0116] drones Recorded with drones Every interaction And there is a level of satisfaction for each interaction. Place in direct interaction set Among them , express think The information provided is incorrect. express For use The information provided Correct; since new interaction records have higher reference value, a time decay function is introduced as a weight for direct reputation. ;

[0117] in The difference between the current time and the recorded time. As a control factor; ultimately, direct reputation is obtained:

[0118] ;

[0119] Recommended Reputation Value It is a drone node Through common neighbor nodes The obtained information about the target drone node Evaluation;

[0120] To mitigate the impact of biased opinions from neighbors, similarity is introduced. ,in As a weighting factor, the recommendation reputation The calculation is as follows:

[0121] ;

[0122] in For drones For drones direct credibility, For recommending interactive sets, The number of common neighbors;

[0123] For drone nodes Trust levels related to current roles, including police drones, commercial drones, and private drones;

[0124] For target drone node The weighted sum of the last historical global reputation values.

[0125] The detailed method for the global reputation assessment in step S3 above is as follows:

[0126] A trusted institution combines the overall reputation of all individuals in this round. and task feedback report , This indicates that the mission was successful. This indicates that the task has failed.

[0127] drones Global reputation The calculation formula is:

[0128] ;

[0129] in, , and For feedback weights, and satisfying ; For the evaluation set of nodes; For individual comprehensive credit and nodes To the node The difference between the average overall credit rating; A collection of feedback reports; The total number of historical credit records; yes The k-th historical global reputation; Let k be the time interval between the current time and the k-th historical reputation record;

[0130] Then, the trusted institution maps the global reputation value to a reputation level. ,calculate:

[0131] ;

[0132] Updated global reputation rating This is further embedded into the drone's decryption vector; subsequently, the trusted authority determines the decryption based on the new reputation level. for Update the corresponding decryption key and kana .

[0133] The aforementioned reputation-based privacy-preserving drone mission matching system applied to mobile swarm intelligence sensing includes a trusted institution, a mission requester, a mobile swarm intelligence sensing platform, a ground station, and drones. The trusted institution is configured to perform system initialization, key generation, reputation assessment and updating, and to allocate and update pseudonyms and decryption keys for drones. The mission requester encrypts the mission using a delegated encryption key obtained from the trusted institution and submits the encrypted mission text to the mobile swarm intelligence sensing platform. The mobile swarm intelligence sensing platform is responsible for receiving, storing, and broadcasting the encrypted mission text, verifying drone mission confirmation information, aggregating mission data, and generating mission feedback reports. The ground station provides a secure data transmission channel between the mobile swarm intelligence sensing platform, the trusted institution, and the drones. The drones receive the encrypted mission text from the platform, use their decryption keys to decrypt and execute the mission, collect and upload sensing data, and participate in distributed reputation assessment based on multiple reference sources.

[0134] The trusted agency is responsible for system initialization and delegated predicate key generation. The generated key is sent to the task requester and drones via a secure channel. The task requester obtains valuable information by issuing a task. Based on the task's requirements, the agency encrypts the task using a delegated encryption key and assigns it to a group of drones for execution via a mobile swarm intelligence platform. The mobile swarm intelligence platform collects and processes the ciphertext, enabling task matching without knowing the task conditions or drone attributes, and sends a reputation feedback report to the trusted agency. The ground station primarily acts as a bridge between the mobile swarm intelligence platform and the drones, as well as between the trusted agency and the drones, handling data transmission. Drones are considered to have varying capabilities and reputations. To accomplish diverse tasks, they can collect data based on various types of sensors and transport supplies according to different payloads and flight capabilities.

[0135] To further verify the beneficial effects and technical feasibility of the present invention, this embodiment uses the Python cryptography library Charm to simulate the operation of the scheme.

[0136] The experiment deployed 100 nodes, selecting 10 attributes for each encryption and decryption operation. For each task, each node generated 20 behavioral instances to evaluate behavior-based reputation. The programming language was Python, and the machine configuration was Ubuntu 18.04, Intel Core i7 (2.9GHz, 16GB RAM).

[0137] like Figure 6 As shown in (a), when the work count is 1, the bar graph of the present invention is significantly shorter than that of other methods, indicating that the execution time is significantly reduced.

[0138] exist Figure 6 In (b), as the number of staff increases from 1 to 10, all three solutions show a linear increase in execution time. However, the execution time of the present invention increases more slowly and is consistently lower than the other two methods.

[0139] Figure 6 (c) demonstrates that the honest behavior of drone nodes leads to a gradual increase in reputation score, which eventually stabilizes in successive evaluation rounds.

[0140] on the contrary, Figure 6 (d) illustrates how persistent malicious behavior leads to a significant decline in reputation score. Furthermore, the mechanism of this invention responds quickly to changes in behavior, allowing for rapid adjustments to the reputation score.

[0141] Figure 6 As shown in (e), the technical solution of the present invention consistently achieves a higher malicious node detection rate across all tested malicious node ratios. Although the detection performance of all solutions shows a downward trend as the proportion of malicious nodes increases, the present invention exhibits superior robustness. It is worth noting that even with 50% malicious nodes, the present invention still maintains a detection rate of over 91%.

[0142] This invention takes into account the information flow direction in cloud-edge data sharing, where data flows from the data sharer to the data receiver. On the data sharer's side, data cannot be directly uploaded to the cloud server, and messages cannot directly reach the receiver. Therefore, it prevents malicious users from sharing illegal files.

[0143] For edge servers, because the messages are transmitted in encrypted form and the edge server does not have the decryption key, it only performs data reprocessing and forwarding operations. Cloud servers, on the other hand, only perform encrypted data storage and retrieval services, thus protecting data privacy.

[0144] For data receivers, this invention employs attribute encryption technology to ensure fine-grained access control at the receiver's end. Only users whose attributes conform to the ciphertext policy can decrypt to obtain the key and generate a download password, thus achieving access control at the receiver's end.

[0145] As described above, although the invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the invention itself. Various changes in form and detail may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A reputation-based privacy-preserving drone task matching method applied to mobile crowd sensing, characterized in that: Includes the following steps: Step 1, System Initialization and Key Generation, includes system initialization, task requester registration, and drone registration, where the Trusted Organization (TA) sets a complete set of attributes. and generate a system public key. and system private key Task requester With attribute set And all task attributes The task requester applies for registration with TA, and TA generates a delegated predicate encryption key for them. ; drones With attribute set And all drone attributes ; The drone applies for registration with a trusted authority (TA), which generates a predicate decryption key for it. and kana ; Step 2: Mobile crowd sensing task matching; Step 2.1, Task Encryption and Issuance, i.e., Task Requester Use its delegated predicate to encrypt the key. Task information Encrypt and generate task ciphertext. It is then submitted to the Mobile Crowd Sensing Platform (MCS), which broadcasts the submission. Step 2.2, Task Decryption and Confirmation, UAV Receive task ciphertext Then, use its predicate to decrypt the key. Decryption is performed; if the drone meets the mission requirements, decryption is successful, and a confirmation message is returned to the Mobile Swarm Intelligence Platform (MCS); after verifying the confirmation message, the MCS sends a recruitment confirmation message to the drone. ; Step 2.3: Upload task data and receive confirmation upon successful completion. The drones are collecting mission data Then, encrypt it as Along with confirmation information The credentials were uploaded together to the Mobile Crowd Sensing Platform (MCS); Step 3: Hybrid credit assessment; Step 3.1: Multi-reference source reputation assessment. Reputation assessments are conducted between drone nodes based on direct interaction, neighbor recommendations, role information, and historical reputation to generate drone... For drones Individual overall credit Simultaneously, the Mobile Crowd Sensing Platform (MCS) will also generate task feedback reports. And send it to a trusted organization (TA); Step 3.2: Global Reputation Update. Trusted organizations (TAs) aggregate the overall reputation of all individuals and task feedback to update the drone's reputation. Global reputation score And mapped to reputation level ; Step 3.3: Key and pseudonym update. The trusted authority (TA) generates a new decryption key for the drone based on the updated reputation level. and kana .

2. The reputation-based privacy-preserving UAV task matching method for mobile swarm intelligence sensing as described in claim 1, characterized in that, The system public key in step 1 and system private key The generation method is as follows: Select a security parameter And generate key pairs Define a bilinear mapping ,in , , It is a prime number. The cyclic group, whose generators are respectively Trusted Institution (TA) selects hash function And randomly select parameters ; Then calculate the system public key. and system private key The formula is as follows: ， ； The detailed method by which a trusted authority (TA) generates a delegated predicate encryption key for a task requester is as follows: First, the task requester Choose an asymmetric key pair and and generate a signature. ; in, It is the public key used by the task requester for signing. It is its corresponding private key; It is the public key used by the task requester for encryption. It is its corresponding private key; Then, the task requester submits its attribute set SA, and the trusted authority generates an attribute set SA based on the SA. 3D attribute vector Random selection Calculate the delegated predicate encryption key The formula is as follows: ； The detailed method for generating predicate decryption keys for drones is as follows: First, drones Select key pair And generate a signature ; Then, the drone submits its attribute set (UA), and the trusted authority generates a predicate vector based on the UA. Random selection Calculate the decryption key Random selection ; At the same time, trusted institutions randomly select Assigning pseudonyms to drones ; in For drones His true identity.

3. The reputation-based privacy-preserving UAV task matching method for mobile swarm intelligence sensing as described in claim 1, characterized in that, The specific method for matching the mobile crowd sensing task in step 2 is as follows: First, the task requester Random selection Use its delegated predicate to encrypt the key For extended plaintext Encrypt and generate ciphertext. The expression is as follows: ； in, This is the current timestamp; The task requester will send the encrypted message and timestamp. They are sent together to the Mobile Crowd Sensing Platform (MCS), which then broadcasts the task. Then, perform the task of decryption and uploading. The detailed method is as follows: drones Use its predicate to decrypt the key. calculate: ； If and only if In other words, when the drone's attributes meet the mission requirements, the drone can correctly decrypt and obtain the mission message. ; Then, the drone returns confirmation information, including pseudonyms, to the Mobile Crowd Sensing Platform (MCS). and task identifier The Mobile Crowd Sensing Platform (MCS) verifies the pseudonym of the drone and checks the timestamp. If the number of recruits for the task is not full, a random selection will be made. Send recruitment confirmation information to drones The expression is as follows: ; Successfully received The drones are collecting mission data Then, encrypt it. The expression is: ; This refers to hashing the task data; Then, along with The credentials are uploaded together to the Mobile Crowd Sensing Platform (MCS).

4. The reputation-based privacy-preserving UAV task matching method for mobile swarm intelligence sensing as described in claim 1, characterized in that, The detailed method for multi-reference source reputation assessment in step S3 is as follows: drones For another drone Individual overall credit Taking into account the direct reputation of drone nodes Recommended Reputation Role Reputation and historical reputation The formula is as follows: ； In the above formula, , , and These are the adaptive weights of the corresponding reference sources, and satisfy... ; Among them, direct credit For drones With the target drone Reputation score obtained through direct interaction; drones Recorded with drones Every interaction And there is a level of satisfaction for each interaction. Place in direct interaction set Among them , express think The information provided is incorrect. express For use The information provided Correct; since new interaction records have higher reference value, a time decay function is introduced as a weight for direct reputation. ; in The difference between the current time and the recorded time. As a control factor; ultimately, direct reputation is obtained: ； Recommended Reputation Value It is a drone node Through common neighbor nodes The obtained information about the target drone node Evaluation; To mitigate the impact of biased opinions from neighbors, similarity is introduced. ,in As a weighting factor, the recommendation reputation The calculation is as follows: ； in For drones For drones direct credibility, For recommending interactive sets, The number of common neighbors; For drone nodes Trust levels related to current roles, including police drones, commercial drones, and private drones; For target drone node The weighted sum of the last historical global reputation values.

5. The reputation-based privacy-preserving UAV task matching method for mobile swarm intelligence sensing as described in claim 1, characterized in that, The detailed method for the global reputation assessment in step S3 is as follows: A trusted institution combines the overall reputation of all individuals in this round. and task feedback report , This indicates that the mission was successful. This indicates that the task has failed. drones Global reputation The calculation formula is: ； in, , and For feedback weights, and satisfying ; For the evaluation set of nodes; For individual comprehensive credit and nodes To the node The difference between average overall credit ratings; A collection of feedback reports; The total number of historical credit records; yes The k-th historical global reputation; Let k be the time interval between the k-th historical reputation record and the current time. Then, the trusted institution maps the global reputation value to a reputation level. ,calculate The formula is as follows: ； Updated global reputation rating This is further embedded into the drone's decryption vector; subsequently, the trusted authority determines the decryption based on the new reputation level. for Update the corresponding decryption key and kana .

6. A system for implementing the reputation-based privacy-preserving drone mission matching method for mobile swarm intelligence sensing as described in any one of claims 1 to 5, characterized in that, The system includes a trusted institution, a task requester, a mobile swarm intelligence platform, a ground station, and a drone; the trusted institution is configured to perform system initialization, key generation, reputation assessment and updating, and to assign and update pseudonyms and decryption keys for the drone; the task requester encrypts the task using a delegated encryption key obtained from the trusted institution and submits the encrypted task to the mobile swarm intelligence platform. The mobile swarm intelligence sensing platform is responsible for receiving, storing, and broadcasting encrypted mission messages, verifying UAV mission confirmation information, aggregating mission data, and generating mission feedback reports; the ground station provides a secure data transmission channel between the mobile swarm intelligence sensing platform, trusted institutions, and UAVs; The drone receives the mission ciphertext from the platform, uses its decryption key to decrypt and execute the mission, collects and uploads perception data, and participates in distributed reputation assessment based on multiple reference sources.

Images