Trusted and privacy-protected low-cost high-quality data acquisition method

Abstract

The invention discloses a credible and privacy-protected low-cost high-quality data acquisition method, and relates to the field of credible computing, privacy protection and low-cost data collection of a crowd-sourcing network. According to the method, the quoted price and the credibility of the worker are sent to the block chain after being subjected to homomorphic encryption, the ratio of the credibility to the quoted price under privacy protection is realized through a homomorphic encryption and re-encryption method, and then the worker with high credibility and low quoted price is selected according to the ratio. And then, the truth value and the trust evaluation are obtained through privacy calculation based on trust truth value discovery between the server and the data requester, so that private, high-quality and low-cost data acquisition is realized.

Description

Technical Field

[0001] The present invention relates to the fields of trusted computing, privacy protection, and low-cost data collection in a crowdsourced network, and more particularly to a method for obtaining low-cost and high-quality data with trust and privacy protection. Background Art

[0002] Mobile Crowdsensing (MCS), as an emerging data collection technology, uses mobile devices equipped with sensors and a large number of user groups to complete sensing tasks, and has shown great application potential in many fields such as military, industry, life and health, and environmental monitoring. However, the wide application of crowdsensing technology faces problems of privacy protection and inaccurate data. The crowdsensing technology mainly collects data by recruiting a large number of people carrying sensing devices, called workers. Due to the wide distribution and large number of workers, the crowdsensing technology can achieve large-scale and timely data acquisition. In the crowdsensing technology, obtaining high-quality data is an important research topic. Only high-quality data can construct high-quality applications, while applications constructed with low-quality data have low quality. However, there are challenges in obtaining high-quality data. The main reason is that in the crowdsensing network, it costs for workers to sense and submit data. The platform needs to give certain rewards to workers to encourage them to participate in data collection. And it is very difficult to verify the quality of the data submitted by workers, that is, after the platform receives the data, it still does not know its quality. In such a situation, malicious and dishonest workers will report false data to defraud rewards.

[0003] On the other hand, the requirements for privacy protection make data collection more challenging. The main goal of data privacy protection is to protect the data content of workers from being known to other parties. Because information such as workers' data and quotes is sensitive or private. If workers are concerned about privacy issues, they are reluctant to participate in data collection, which will hinder the development of crowd sensing. A relatively simple way of privacy protection is that when DR issues tasks, it simultaneously publishes the public key for encrypting data through the platform. Workers encrypt the collected data with the public key and upload it to the platform. Then, the platform transfers the data to DR. DR then decrypts it with the private key to obtain the plaintext data. However, such a privacy protection method has relatively weak privacy protection. Because in this method, DR actually knows the workers' data, and at this time, the workers' data is leaked to DR. Therefore, although the encryption method is used, it only makes the platform unable to know the workers' data, while DR still knows the workers' data. Therefore, in order to protect the data privacy of workers, there are actually many privacy protection studies in subsequent research. However, the current privacy protection studies still have the following deficiencies. First, the current privacy protection rarely takes data quality into account, which will result in the lack of protection for data quality. The reason for this situation is that privacy protection requires hiding workers' information. To obtain high-quality data, it is necessary to evaluate the quality of the data submitted by workers, and then select high-quality workers to collect data to obtain high-quality data. In privacy protection, the platform cannot obtain the workers' data, so it is difficult to conduct quality evaluation, and it is difficult to select workers based on the results of quality evaluation, resulting in challenges in obtaining high quality in privacy protection methods. Moreover, in data acquisition, it is also hoped to reduce costs as much as possible. However, reducing costs requires selecting high-quality and low-cost workers from workers. On the one hand, it is difficult to design a method for selecting high-quality and low-cost workers; more difficultly, in privacy protection, workers' quotes are also private information and cannot be obtained by any other third party. In such a privacy protection situation, it is very difficult to obtain high-quality and low-cost workers. In view of the deficiencies of the prior art in simultaneously meeting high data quality, low quotes, and privacy protection, therefore, there is an urgent need for an inventive method that can comprehensively solve the above problems. Summary of the Invention

[0004] In view of this, the present invention provides a method for obtaining low-cost and high-quality data with trust and privacy protection. Compared with previous strategies, it mainly proposes an effective method to identify trustworthy workers on the premise of ensuring the privacy protection of data content and quotes, so as to select trustworthy workers to collect data, thereby effectively improving the data quality and realizing the privacy protection of data content and quotes. The objectives of the method of the present invention are three. The first is to protect the privacy of the data collected by workers and the privacy of quotes, so that they are not known to any third party other than the workers themselves. The second is to select trustworthy workers to ensure high-quality data collection, and the trustworthiness of the workers should not be known to any irrelevant third party; the third is to select workers with low quotes to reduce costs. The present invention sends the quotes and trustworthiness of workers to the blockchain after homomorphic encryption, and realizes the ratio of trustworthiness and quotes under privacy protection through homomorphic encryption and re-encryption methods, and then selects workers with high trustworthiness and low quotes according to this ratio. Then, through the privacy calculation based on the discovery of trust truth values between the server and the data requester, the truth value and trust evaluation are obtained, thus realizing the acquisition of private, high-quality, and low-cost data.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for obtaining low-cost and high-quality data with trust and privacy protection, comprising the following steps:

[0007] The data requester publishes a task on the data trading center;

[0008] Workers apply to execute the task according to their own will, and use their own public keys to encrypt the quotes for doing the task and upload them to the blockchain of the data trading center;

[0009] The data requester removes the workers with a trustworthiness less than a preset threshold from the set of workers applying for the task, and sends the encrypted ciphertext of the trustworthiness of the remaining workers to the data trading center using the public key;

[0010] The data requester sends the generated re-encryption key to the server using the public key of the server and his own, and the server encrypts the data encrypted with its own public key using the re-encryption key, and then decrypts it with its own public key to obtain the ratio of the workers' trustworthiness and quotes;

[0011] Sort according to the ratio of the workers' trustworthiness and quotes, and give the sorting result to the data requester. The data requester selects workers according to the trust status of the workers, selects K workers from high to low according to the ratio, notifies the workers to execute the task, and the workers execute the task.

[0012] Optionally, for each worker u i , use the generator g i to generate a cyclic group G i , with order qi ; Worker u i Select an x from the set 1, 2, …, q i -1, and then calculate i . The public key generated by each worker is (G , q i , g i , h i ) and is made public, and its private key is x i , which is not made public and is saved privately. The public key of worker u i is denoted by PK i . Respectively, use PK i,u and PK DR to denote the public keys of the data requester and the server respectively, and use x S , x DR to denote the private keys of the data requester and the server respectively; S

[0013] The set of workers is denoted by , the set of trusted workers is denoted by , and the initial trust level of the workers in the set is 1; the trust levels of the remaining other workers are unknown, and the set is denoted by , with a trust level of 0.5; the trust level of the workers is between [0, 1], where 0 represents a low trust level and 1 represents the highest trust level; the set of malicious workers is denoted by , which is initially empty; the trust level of worker u i is

[0014] Optionally, the worker encrypts the quote for the task performed by himself with his own public key and uploads it to the blockchain. The expression is as follows:

[0015]

[0016] PK i,u represents the public key of worker u i , represents the ciphertext after encrypting the quote , represents encrypting i,u with the key PK .

[0017] Optionally, the data requester removes the workers with a trust level less than the preset threshold from the set of workers applying for the task, and encrypts the trust levels of the remaining workers with the public key to generate ciphertext and sends it to the data trading center. The expression is as follows: Remove the workers with a trust level R i less than the threshold from the set of workers applying for the task, and use the public key PK i to encrypt the trust level R​​DR Generate the ciphertext E(R i ) and send it to the data trading center;

[0018] E(R i ) = E(R i , PK DR ) (2).

[0019] Optionally, the server generates a random number z for each worker i , and encrypts it with the worker's public key PKi ,u and sends it to the DTC;

[0020] E(z i ) = E(z i , PK i,u ) (3).

[0021] Optionally, the smart contract on the blockchain uses the performance calculation of adding a constant in homomorphic encryption:

[0022]

[0023] Optionally, after the worker completes the task, the server interacts with the data requester to obtain the weights. The specific steps are as follows:

[0024] The server selects a data as

[0025] Save the true value of the previous round Calculate the distance between each worker and ;

[0026]

[0027] The server sends the difference of each worker to the data requester. The data requester receives the differences of K workers and calculates the true distance from according to the following formula;

[0028]

[0029] Calculate the weight of each worker according to the following formula:

[0030]

[0031] DR then sends the of each worker to the server;

[0032] The server allows to calculate

[0033]

[0034] If the value is greater than the threshold ε, then go back to calculating the distance between each worker and , loop, and continue.

[0035] Optionally, after the worker completes the task, the data requester updates the trust level of the worker. The specific steps are as follows:

[0036] The data requester calculates the average weight of the K workers who are trusted workers belonging to the set for the sensing task. The set of K workers who are trusted workers belonging to the set for the sensing task is The set The number of elements in the set is

[0037]

[0038] Calculate the difference rate between the weight of each worker not in the set and as follows: As shown in the following formula:

[0039]

[0040] If is less than the threshold then increase the trust level of the worker according to Equation (17) Otherwise, decrease the trust level of the worker according to Equation (18)

[0041]

[0042] If the updated trust level of the worker is greater than the upper threshold worker u i is included in the set of trusted workers If the updated trust level of the worker is less than the lower threshold worker u i is included in the set of malicious workers

[0043] Re - assign weights to workers according to their trust levels: If worker u i is included in the set of malicious workers then the weight of worker u i For other workers except this, their weights are their trust levels. Therefore, the weight update calculation method is as follows:

[0044] ​

[0045] Optionally, the server calculates according to the following formula The server will Send it to the DR, and the obtained by the DR is the final true value:

[0046]

[0047] As can be seen from the above technical solutions, compared with the prior art, the present invention provides a method for protecting the privacy of workers' data and quotes from being known by any owner, and capable of selecting high-quality and low-quoted workers on the premise of privacy protection, so as to obtain high-quality data. The experimental results show that the method of the present invention can achieve privacy protection of data content and workers' quotes, and can also select high-quality and low-quoted workers to obtain accurate data. Brief Description of the Drawings

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0049] Figure 1 It is a structural schematic diagram of the present invention;

[0050] Figure 2 The situation of obtaining the average absolute error of data;

[0051] Figure 3 The change of trust degree of different types of workers with data collection. Detailed Embodiments

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0053] Embodiment 1

[0054] The embodiment of the present invention discloses a method for obtaining low-cost and high-quality data with credibility and privacy protection, including the following steps:

[0055] The data requester publishes a task on the data trading center;

[0056] Workers apply to execute tasks according to their own will, and use their own public keys to encrypt and upload their task quotes to the blockchain of the data trading center;

[0057] The data requester removes the workers with a trust level less than the preset threshold from the set of workers applying for the task, and uses the public key to encrypt the trust levels of the remaining workers to generate ciphertext and send it to the data trading center;

[0058] The data requester uses the public key of the server and their own generated re-encryption key to send it to the server. The server encrypts the data encrypted with its own public key using the re-encryption key, and then decrypts it with its own public key to obtain the ratio of the worker's trust level to the quote;

[0059] Sort according to the ratio of the worker's trust level to the quote, and give the sorted result to the data requester. The data requester selects workers according to the trust status of the workers, selects K workers from the highest to the lowest ratio, notifies the workers to execute the task, and the workers execute the task.

[0060] Among them, the ElGamal cryptosystem is a public-key encryption algorithm based on the discrete logarithm problem. The basic definitions and operation processes are as follows:

[0061] (1) Key generation: Use the generator δ in a cyclic group G of order n. Randomly select a Calculate h = δ y . Finally, the public key is represented as (G, n, δ, h), and the private key is x.

[0062] (2) Encryption process: The message m is encrypted using the public key and private key. The encryption process is as follows:

[0063]

[0064] (3) Decryption process: To decrypt c, this study first calculates s = c1 y , according to the key y, the decryption formula is as follows:

[0065] c2·s -1 = mδ x·y ·δ -x·y = m (b)

[0066] (4) Homomorphic property: The ElGamal cryptosystem has a homomorphic property, that is, the encrypted ciphertext can perform homomorphic operations. The specific manifestation is:

[0067]

[0068] E(m1 - m2) = E(m1)·E(m2) -1 #(d)

[0069] E(α·m1) = E(m1)α #(e)

[0070] Initialization: For each worker u i , a cyclic group G i is generated using the generator g i , whose order is q i . Worker u i selects an x i from the set 1, 2, …, q i −1, and then calculates The public key generated by each worker is (G i , q i , g i , h i ) and is made public, and its private key is x i , which is not made public and is saved by itself. In the present invention, the public key of worker u i is denoted by PK i,u ; similarly, DR and the server also generate their own public and private keys, which are denoted by PK DR and PK S to represent the public keys of DR and the server respectively, and x DR , x S to represent the private keys of DR and the server respectively;

[0071] The set of workers is denoted by . DR initially knows that a certain number of workers are trustworthy, and its set is denoted by , and the initial trust level of the workers in its set is 1; the trust levels of the remaining other workers are unknown, and its set is denoted by , and its trust level is 0.5; the trust level of the workers is between [0, 1], where 0 represents a low trust level and 1 represents the highest trust level; the set of malicious workers is denoted by , which is initially empty; the trust level of worker u i is denoted by

[0072] Specifically, the present invention provides a method for obtaining low-cost and high-quality data with trust and privacy protection, including the following steps:

[0073] (1) DR publishes a task on the data trading center, i.e., DTC;

[0074] (2) Workers view the task;

[0075] (3) If a worker is willing to execute the task therein, the following two actions are taken: Worker u i applies to DR for the task; the worker simultaneously encrypts the quotation for doing the task with its own public key and uploads it to the blockchain;

[0076]

[0077] PK i,u represents the public key of worker u i , represents the ciphertext after encrypting the quotation , represents encrypting i,u with the key PK , and the same applies hereinafter;

[0078] (4) DR First, make a selection. Remove the workers with a trust level R i less than the threshold from the set of workers applying for tasks. Then, use the public key PK i to encrypt the remaining workers' trust level R DR to generate the ciphertext E(R i ) and send it to DTC;

[0079] E(R i ) = E(R i , PK DR ) (2)

[0080] (5) The server generates a random number z i for each worker, and encrypts it with the worker's public key PK i,u and sends it to DTC;

[0081] E(z i ) = E(z i , PK i,u ) (3)

[0082] (6) The smart contract on the blockchain uses the performance of adding a constant in homomorphic encryption for calculation:

[0083]

[0084] At this time, DTC already has E(R i ), E(z i ) and E(R i ), E(z i ) and It doesn't mean there is only one worker i. It represents a set, and the i-th worker in it is used to represent;

[0085] (7) is encrypted with PK i,u , while E(R i ) is encrypted with the public key PK DREncrypted. Therefore, these two encrypted numbers cannot be homomorphically computed; thus, this step is converted into data encrypted with the same public key; worker u i First, use the public key PK of DR DR and its own x S to generate a re-encryption key PK DR,c and send it to DTC;

[0086] (8) DTC will re-encrypt and convert it into a ciphertext encrypted with the public key PK of DR DR Although it is now encrypted with the public key of DR, DR does not know the random number z i , so it cannot know

[0087]

[0088] (9) DTC will perform a division operation on E(R i ) and in the encrypted state;

[0089]

[0090] (10) DTC sends to the server. The server only knows z i , and even if it can decrypt, it cannot know and R i ;

[0091] (11) The server calculates

[0092]

[0093] (12) DR uses the public key PK of the server S and its own x DR to generate a re-encryption key PK S,c and sends it to the server. The server encrypts it into data encrypted with its own public key using the re-encryption key PK S,c , and then decrypts it with its own public key as shown in Equation 9 to obtain the ratio of the worker's trustworthiness to the quote. The server still cannot know and R i ;

[0094]

[0095] (13) Sort according to the ratio of the worker's trustworthiness to the quote, and give the sorted result to DR;

[0096] (14) DR selects workers based on the trust status of the workers, and selects K workers from the highest to the lowest ratio:

[0097] (15) Instruct the worker to perform the task, and the worker performs the task; let the data sensed by the \(i\)-th worker be \(d_i\). i There are \(K\) workers who sense \(K\) pieces of data.

[0098] (16) Each worker generates two random numbers. Let the random numbers generated by the \(i\)-th worker be \(\alpha_i\) and \(\gamma_i\). i Each worker uploads the two random numbers it generates to the DR; then, after performing addition and multiplication operations on the data \(\gamma_i\), uploads i \(\alpha_i + \gamma_i\) and \(\alpha_i\times\gamma_i\) \(\alpha_i + \gamma_i\) \(\alpha_i\times\gamma_i\) to the server, and the server obtains \(2K\) pieces of data.

[0099]

[0100] (17) The server interacts with the DR to obtain the weights, as detailed below:

[0101] (a) The server selects a piece of data as

[0102] (b) Save the true value of the previous round, calculate the distance between each worker's \(\alpha_i\) and \(\alpha\).

[0103]

[0104] (c) The server sends the difference of each worker to the DR. The DR receives the differences of \(K\) workers and calculates the distance between the true value and \(\alpha\) according to the following formula:

[0105]

[0106] Calculate the weight of each worker according to the following formula:

[0107]

[0108] The DR then sends the weight of each worker to the server.

[0109] (d) The server asks \(\alpha\) to calculate

[0110]

[0111] (e) If the value of \(\alpha\) is greater than the threshold \(\varepsilon\), then go back to step (b), loop, and continue; otherwise, calculate further;

[0112] (18) The DR updates the trustworthiness of the worker.

[0113] (a) The DR calculates the average weight of the K workers in the set of trustworthy workers for the sensing task. The K workers for the sensing task belong to the set The set of trustworthy workers is set The number of sets is

[0114]

[0115] (b) Calculate the difference rate between the weight of each worker not in the set and as follows: The formula is as follows:

[0116]

[0117] (c) If is less than the threshold then increase the trust level of the worker according to Equation 17 Otherwise, decrease the trust level of the worker according to Equation 18

[0118]

[0119] (d) If the updated worker is greater than the upper threshold worker u i is included in the set of trustworthy workers If the updated worker is less than the lower threshold worker u i is included in the set of malicious workers

[0120] (19) Reassign weights to workers according to their trust levels: If worker u i is included in the set of malicious workers then the weight of worker u i is For other workers, their weights are their trust levels. Therefore, the weight update is summarized as follows:

[0121]

[0122] (20) The DR then sends the of each worker to the server; for

[0123] (21) The server calculates according to the following formula. The server sends to the DR, and the obtained by the DR is the final true value.

[0124]

[0125] Example 2

[0126] The difference between this example and Example 1 is only the following:

[0127] In the crowd-sensing network, when real data of a certain place needs to be obtained and relatively high data quality is required, such as for applications of sensing data like temperature and humidity, air quality, traffic conditions, etc., and the sensed data is uploaded to the platform for processing. In the tasks released by the platform, participants apply to participate in the tasks and provide different data through workers. By applying the method of the present invention to the above data collection application, it is possible to identify the credibility of workers without disclosing any data content and quotes of the workers to any party under the condition of privacy protection, and improve the accuracy of the data.

[0128] The experimental results of the inventive method are given below.

[0129] Figure 1 The network structure diagram of the method of the present invention is given. The method of the present invention has the following components: the network structure consists of a data requester, denoted by DR, a server, and a blockchain. Among them, the blockchain is mainly used to obtain the ratio of the trust degree and the quote of the worker for privacy, so as to guide the selection of workers; the server and the DR cooperate to calculate the true value under the condition of privacy protection; different from the general method of finding the true value in the past, in the process of calculating the true value by the server and the DR, the trust inference is carried out based on the weights of the trusted workers, so as to be able to identify the trust degree of the same worker, and adjust the weights of the workers with the trust degree to obtain a more accurate true value.

[0130] Figure 2 The experimental results of the method of the present invention and other methods in terms of data accuracy are given. There are three methods compared with the method of the present invention: the first one is CRH, which is a widely used method. It recruits K workers to execute a sensing task at the same time, and then takes the weighted average of the K obtained data as the final result; the weight of the worker is related to the distance between the data reported by the worker and the final result. If the worker's data is far from the final result, the weight is small, otherwise, the weight is large; the second method is the average method, that is, it recruits K workers to execute a sensing task at the same time, and then takes the average of the K obtained data as the final result; the last one is the median method, which takes the median data as the final result. It can be seen from the experimental results in Table 1 that the method of the present invention can effectively reduce errors.

[0131] Figure 3The performance of the method of the present invention in identifying the trust level of workers is given. As can be seen from the figure, for trustworthy workers, their trust level continuously rises above the threshold of trustworthy workers; while for malicious workers, their trust level continuously decreases. It can be seen that the method of the present invention has a good ability to identify the trust level of workers, and can guide the selection of trustworthy workers in the selection of workers, thereby improving data quality.

[0132] The experimental results show that the present invention performs excellently in terms of data quality. In the comparison of the method of the present invention with other methods, compared with the CRH method, the average method and the median method, the reduction amplitude of the mean absolute error (MAE) of the data is shown in Table 1 below. It can be seen that the method of the present invention can effectively reduce data errors.

[0133] Table 1: The reduction amplitude of the mean absolute error of the data by the method of the present invention compared with other methods

[0134]

[0135]

[0136] The reason for such an effect is that the method of the present invention can effectively identify the trustworthiness of workers, so as to guide the selection of workers with high trust level and low asking price to participate in data collection and improve data quality. To sum up, the present invention realizes the privacy protection of data and quotes, improves data quality, provides an efficient, reliable and innovative solution for the wide application of crowd intelligence networks, and has important practical application value and promotion prospects.

[0137] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0138] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A low-cost and high-quality data acquisition method for trust and privacy protection, characterized in that, It includes the following steps: The data requester publishes a task on the data trading center; The worker applies to execute the task according to his own will, and encrypts and uploads his quotation for doing the task to the blockchain of the data trading center using his own public key; The data requester removes the workers with a trust level less than the preset threshold from the set of workers applying for the task, and uses the public key to encrypt the trust levels of the remaining workers to generate ciphertext and send it to the data trading center; The data requester sends the re-encryption key generated with his own and the server's public key to the server. The server encrypts the data encrypted with its own public key using the re-encryption key, and then decrypts it with its own public key to obtain the ratio of the worker's trust level to the quotation; Sort according to the ratio of the worker's trust level to the quotation, and give the sorting result to the data requester. The data requester selects workers according to the trust status of the workers, selects K workers from the highest to the lowest ratio, notifies the workers to execute the task, and the workers execute the task.

2. A low-cost and high-quality data acquisition method for trusted and privacy protection according to claim 1, characterized in that For each worker u i , a cyclic group G i is generated using the generator g i , of order q i ; the worker u i selects an x from the set 1, 2, …, q i - 1, and then calculates i The public key generated by each worker is (G , q i , g i , h i ), which is made public, and its private key is x i , which is not made public and is saved privately. The public key of the worker u i is denoted by PK i ; the public keys of the data requester and the server are denoted by PK i,u and PK DR respectively, and the private keys of the data requester and the server are denoted by x S , x DR , x S respectively; The set of workers is represented by and the set of trusted workers is represented by The initial trust level of the workers in the set is 1; the trust levels of the remaining other workers are unknown, and the set is represented by with a trust level of 0.5; the trust level of a worker is between [0, 1], where 0 represents a low trust level and 1 represents the highest trust level; Set of malicious workers is denoted and initially empty; the trust level of worker u i is 3. A low-cost and high-quality data acquisition method for trusted and privacy protection according to claim 1, characterized in that The worker encrypts the quotation for the tasks he has done with his own public key and uploads it to the blockchain. The expression is as follows: PK i,u represents the public key of worker u i , represents the ciphertext after encrypting the quotation , represents encrypting i,u using the key PK .

4. A low-cost and high-quality data acquisition method for trust and privacy protection according to claim 1, characterized in that The data requester removes the workers with a trust level less than the preset threshold from the set of workers for the application task, and uses the public key to encrypt the trust levels of the remaining workers to generate ciphertext and send it to the data trading center. The expression is as follows: Remove the workers with a trust level R i less than the threshold from the set of workers for the application task, and use the public key PK i to encrypt the trust levels R DR of the remaining workers to generate ciphertext E(R i ) and send it to the data trading center; E(R i ) = E(R i , PK DR ) (2).

5. A low-cost and high-quality data acquisition method for trust and privacy protection according to claim 1, characterized in that, The server generates a random number z for each worker i , and encrypts it with the worker's public key PK i,u and sends the encrypted result to the DTC; E(z i ) = E(z i , PK i,u ) (3).

6. A low-cost and high-quality data acquisition method for trust and privacy protection according to claim 1, characterized in that The smart contract on the blockchain uses the performance calculation of adding with a constant in homomorphic encryption:

7. A low-cost and high-quality data acquisition method for trust and privacy protection according to claim 1, characterized in that After the worker completes the task, the server and the data requester interact to obtain the weight. The specific steps are as follows: The server selects a piece of data as Save the true value of the previous round Calculate for each worker and the distance; The server sends the difference of each worker to the data requester. The data requester receives the differences of K workers and calculates the distance to the real one according to the following formula; ; Calculate the weight of each worker according to the following formula: DR will then send each worker's to the server; The server allows to calculate according to the following formula If is greater than the threshold ε, then return to calculating the distance between each worker and , loop, and continue.

8. A low-cost and high-quality data acquisition method for trust and privacy protection according to claim 1, characterized in that After the worker completes the task, the data requester updates the worker's trust level. The specific steps are as follows: The data requester averages the weights of the trusted workers among the K workers of the sensing task belonging to the set The set of trusted workers among the K workers of the sensing task belonging to the set is Set The number of sets is Calculate the difference rate between the weight of each worker not in the set and as follows: The formula is as follows: If is less than the threshold then increase the trust level of the worker according to Equation (17) Otherwise, decrease the trust level of the worker according to Equation (18) If the updated worker is greater than the upper threshold Worker i is included in the set of trusted workers If the updated worker is less than the lower threshold Worker u i is included in the set of malicious workers Re - assign weights to workers according to their trust levels: If worker u i is classified into the set of malicious workers then the weight of worker u i is For other workers, their weights are their trust levels; thus, the weight update calculation method is as follows:

9. A low-cost and high-quality data acquisition method for trust and privacy protection according to claim 7, characterized in that, The server calculates according to the following formula The server will send to the DR, and what the DR gets is the final true value:

Images