CN103957062B - The method of cognitive user credit worthiness is assessed in distributed cognition radio network - Google Patents

Abstract

The invention discloses a method for evaluating the credibility of cognitive users in a distributed cognitive radio network, which mainly solves the problem of trust evaluation in the absence of a central control device in the distributed cognitive radio network. The implementation steps of the present invention are: initializing, selecting the initial reputation evaluation user group, screening the reputation evaluation user group, judging whether the number of remaining reputation evaluation users is less than 2, reselecting the reputation evaluation user group, evaluating the cooperative reputation of the cognitive user, resource allocation, evaluation Cognizant the communication reputation of the user, determine the evaluation reputation value of the reputation evaluation user, and update the reputation value. The present invention can effectively evaluate the network behavior reputation of cognitive users in the distributed cognitive radio network, solve the evaluation, calculation and update of the reputation value of cognitive users more efficiently, determine the reputation status of cognitive users, and ensure trust Assess fairness, improve network efficiency, network security, and network robustness.

Description

A Method for Evaluating the Reputation of Cognitive Users in Distributed Cognitive Radio Networks

technical field

The invention belongs to the technical field of communication, and further relates to a method for evaluating the reputation of cognitive users in a distributed cognitive radio network in the technical field of cognitive radio networks. The invention can solve the problems of reputation value calculation, sensing data fusion and frequency spectrum allocation in the case of no data fusion center in the distributed cognitive radio network, and makes the distributed cognitive radio network run more efficiently, fairly, safely and robustly.

Background technique

There is a lack of a central control device in a distributed cognitive wireless network, so each cognitive user needs to assume the dual responsibilities of a cognitive user and a central base station in a centralized cognitive wireless network. Cognitive users need to complete all the operations of the cognitive cycle. Due to the limitation of the ability of a single cognitive user, it is necessary for cognitive users to cooperate to complete information fusion and decision-making, so that the cognitive cycle can proceed smoothly. The lack of central control equipment has brought problems such as decentralized communication control, collaborative decision-making without a decision maker, fairness and reliability between nodes, lack of a trusted third party for reputation value calculation, and no one in the trust mechanism. Calculations etc. Therefore, in order to solve the trustworthiness of cognitive users in distributed cognitive radio networks, it is necessary to design a reasonable trust management mechanism.

The patent application "Reputation-Based Cognitive Radio Network Hierarchical Cooperative Spectrum Sensing Method" (application number CN201310061675.2, publication number CN103178910A) filed by Zhejiang University discloses a reputation-based cognitive radio network hierarchical cooperative spectrum sensing method. The specific steps are: cluster the sensing nodes, perform local sensing on the nodes, make judgments based on the sensing data of nodes in the cluster, and the fusion center make judgments based on the judgment results and reputation of each cluster head, and update the reputation of each cluster head. The invention combines the reputation mechanism with the network layered cooperative perception using the clustering strategy to reduce the impact of clusters with low reputation on the system perception performance, so that the cognitive radio network can greatly reduce the impact of deep-fading users and attacked users on the system. The adverse effects caused by the system can effectively reduce the communication overhead of the system at the same time. The shortcomings of this method are: although this method considers the credit degree in the cooperative spectrum sensing in the cognitive radio network, the reputation measurement method is not described in detail, and a complete reputation management method has not been formed; the spectrum sensing method applied for by this method It is only applicable to hierarchical networks using clustering strategies, and cannot be applied in distributed cognitive radio networks. In the distributed cognitive radio network, there is a lack of central facilities, and this method cannot efficiently complete spectrum sensing; in addition, this method does not introduce any supervision mechanism, and cannot guarantee the fairness and robustness.

S.Parvinetal. In the article "Towards Trust Establishment for Spectrum selection in Cognitive Radio Networks" published on the 201024th IEEE International Conference on Advanced Information Networking and Applications, a behavior-based trust evaluation method under the central cognitive radio network architecture is proposed. The model considers the relationship between direct trust and indirect trust, which can be easily Detecting bad behaviors of cognitive users in cognitive radio networks. The trust management evaluation steps are as follows: 1. Cognitive user perceives idle spectrum information and sends the sensing information to cognitive user base station; 2. Cognitive user base station obtains direct trust value and indirect trust value and integrates them to obtain a comprehensive trust value; 3. Make spectrum decisions based on trust values. The shortcomings of this method are: firstly, the reputation value calculation, sensing data fusion and spectrum allocation of this method need to be completed based on cognitive user base stations, which is not applicable to distributed cognitive radio networks; secondly, this method does not have Fully combining the characteristics of cognitive wireless networks, the factor of trust measurement is too single, and does not consider the network behavior characteristics of cognitive users in special network situations in cognitive radio networks; again, this method only considers trust generation in trust management However, specific issues such as trust measurement and renewal are not covered. The trust management framework is relatively rough, and the entire trust relationship is not explained in accordance with the idea of cognitive cycle.

Contents of the invention

Aiming at the deficiencies of the above-mentioned prior art, the present invention proposes a method for evaluating the credibility of cognitive users in a distributed cognitive radio network, and completes the case where there is no data fusion center and no decision maker in the distributed cognitive radio network, Accurately and fairly calculating the reputation value for cognitive users, completing the fusion of sensing data, and finally assigning the spectrum fairly with reference to the trust value can ensure the fairness, security and robustness of the cognitive radio network operation.

In order to achieve the above object, the present invention comprises the following steps:

(1) Initialization:

(1a) All the data records in the database of cognitive users are cleared, and for each cognitive user in the radio network, a unique identity is set according to the order of natural numbers, and the identity of the cognitive user is stored in the network parameters of the cognitive user data logging;

(1b) The initial cooperation reputation value and the initial communication reputation value of all cognitive users in the records of the cognitive user database are set to 0.5, the initial evaluation reputation value is set to 0, and the initial total reputation value is set to 0.5;

(1c) record the number of cognitive users in the radio network into the network parameter data record of the cognitive users, and complete the initialization of the radio network;

(1d) Initialize new cognitive users requesting to join the network.

(2) Select the initial reputation evaluation user group:

(2a) judge whether the data record of reputation evaluation user group is empty, if empty, execute step (2b), otherwise, execute step (3);

(2b) Choose a positive real number in the range of [0.2,0.5] as the selection ratio;

(2c) Multiply the number of cognitive users in the radio network with the selection ratio and round to get the number of users in the reputation evaluation user group;

(2d) Calculate the selection factor of the reputation evaluation user group according to the following formula:

Among them, m represents the selection factor of the reputation evaluation user group, N represents the number of cognitive users in the radio network, J represents the number of users in the reputation evaluation user group, Indicates the rounding down operation;

(2e) From the database of cognitive users, select the cognitive users who meet the following identity identification conditions, and record them in the credit evaluation user group data records of cognitive users:

Imodm=1

Among them, I represents the identity of the cognitive user, m represents the selection factor of the reputation evaluation user group, and mod represents the remainder operation;

(2f) Use the cognitive users recorded in the reputation evaluation user group data record of the cognitive users as the reputation evaluation users, and use all the reputation evaluation users as the reputation evaluation user groups.

(3) Screen reputation evaluation user groups:

(3a) Check the reputation value of all reputation evaluation users in the reputation evaluation user group data record in turn, if the total reputation value of the reputation evaluation user is less than 0.5, or the evaluation reputation value of the reputation evaluation user is less than 0, then the reputation evaluation user is removed from Reputation evaluation user group data records are deleted;

(3b) Check successively the number of consecutive credit evaluations of all credit evaluation users in the credit evaluation user group data record, if the reputation evaluation user has continuously participated in reputation evaluation for more than 5 times, delete the reputation evaluation user from the reputation evaluation user group data record.

(4) Determine whether the number of remaining credit evaluation users in the credit evaluation user group data record is less than 2, if less than 2, perform step (5), otherwise, perform step (6).

(5) Re-select the reputation evaluation user group:

(5a) The current reputation evaluation user group randomly publishes a positive integer, which is used as the initial application number of the cognitive user;

(5b) Cognitive users who wish to join the reputation evaluation user group send application information;

(5c) Check whether the reputation value of the cognitive user who sent the application information meets the application conditions. Starting from the initial application number and increasing by 1, set an application number in turn for the qualified cognitive users, and the cognitive user The application number is recorded in the data record of the reputation evaluation user group;

(5d) Calculate the selection factor of the reputation evaluation user group according to the following formula:

Wherein, _m represents the selection factor of the reputation evaluation user group, N represents the number of cognitive users in the data record of the reputation evaluation user group, and J represents the number of users of the reputation evaluation user group described in step (2c), Indicates the rounding down operation;

(5e) From the data records of the reputation evaluation user group, delete the cognitive users who do not meet the following application numbering conditions:

Amodm=1

Among them, A represents the application number of the cognitive user, m represents the selection factor of the reputation evaluation user group, and mod represents the remainder operation;

(5f) Use the cognitive users recorded in the reputation evaluation user group data record of the cognitive users as the reputation evaluation users, and use all the reputation evaluation users as the reputation evaluation user groups.

(6) Evaluate the cooperative reputation of cognitive users:

(6a) Cognitive users perform local spectrum sensing;

(6b) The cognitive user reports the local spectrum sensing information to the reputation evaluation user group, the values of the row vector elements of the spectrum sensing information are 0 and 1, and the number of elements of the row vector is the number of perceived spectrums;

(6c) The reputation evaluation user group performs sensing information fusion on the spectrum sensing information of cognitive users to obtain the final spectrum sensing information;

(6d) The reputation evaluation user group broadcasts the final spectrum sensing information to the cognitive users in the radio network, and the cognitive users record the final spectrum sensing information into the cognitive user's sensing information data records;

(6e) The reputation evaluation user judges whether the final spectrum sensing information is identical to the cognitive user's spectrum sensing information. If they are identical, the cognitive user's spectrum sensing is correct; otherwise, the cognitive user's spectrum sensing is wrong;

(6f) The reputation evaluation user evaluates the cooperation score of the cognitive user. If the cognitive user reports the perception information and the spectrum perception of the cognitive user is correct, the cooperation score of the cognitive user is 1; if the cognitive user reports the perception information, but the cognitive user If the cognitive user's spectrum sensing error is wrong, the cognitive user's cooperation score is 0.5; if the cognitive user does not report the sensing information, the cognitive user's cooperation score is 0.

(7) Resource allocation:

(7a) The cognitive user sends a spectrum request message including the cognitive user's bid for the requested spectrum;

(7b) According to the following formula, the reputation evaluation user group calculates the spectrum competitiveness of the cognitive user who sends the spectrum request information, and records the spectrum competitiveness of the cognitive user into the spectrum request data record of the cognitive user:

CP=T*B

Among them, CP represents the spectrum competitiveness of the cognitive user, T is the total credit value of the cognitive user, and B is the bid of the cognitive user for the requested spectrum;

(7c) The reputation evaluation user group sorts the spectrum competitiveness of cognitive users from large to small;

(7d) The reputation evaluation user group will read the spectrum with the final spectrum sensing information of 0 from the spectrum sensing data record of the cognitive user, record it as a free spectrum list, and allocate the spectrum in the free spectrum list to the competitive Cognitive users from large to small;

(7e) The reputation evaluation user group announces the spectrum allocation results.

(8) Evaluate the communication reputation of cognitive users:

(8a) Cognitive users perform data communication according to spectrum allocation results;

(8b) Reputation Evaluation User perception monitors the communication behavior of each cognitive user, and determines the communication score for the cognitive user according to the communication quality evaluation standard.

(9) Determine the evaluation reputation value of the reputation evaluation user:

(9a) The reputation evaluation user performs a signature operation on the cognitive user cooperation score described in step (6f) and the cognitive user communication score described in step (8b);

(9b) Cooperation score and communication score for cognitive users after credit evaluation users exchange signatures with each other;

(9c) According to the score fusion method, respectively fuse the cooperation score and communication score of the cognitive user to the cognitive user by the reputation evaluation user, and obtain the cooperation overall assessment and communication overall assessment of the cognitive user;

(9d) Judgment is made on the ratings of credit evaluation users. If the difference between the credit evaluation user's cooperation score and communication score for the cognitive user and the cognitive user's overall cooperation rating and communication rating is less than 0.1, the reputation evaluation user's evaluation Fairness, otherwise, the evaluation of reputation evaluation users is unfair;

(9e) Determine the evaluation reputation value of the reputation evaluation user according to the evaluation reputation evaluation method.

(10) Reputation value update:

(10a) The reputation evaluation user group obtains the updated reputation data according to the reputation value update method, and the updated trust data includes the cooperation reputation value, communication reputation value and total reputation value of the cognitive user;

(10b) The reputation evaluation user group broadcasts the updated reputation data to all cognitive users of the cognitive radio;

(10c) The cognitive user records the updated reputation data into the trust data record of the cognitive user.

Compared with the prior art, the present invention has the following advantages:

First, since the present invention uses reputation evaluation user groups to manage trust in distributed cognitive radio networks, it overcomes the need to rely on cognitive user base stations for reputation measurement in the prior art, which does not fully combine the characteristics of distributed cognitive wireless networks. Insufficient, so that the present invention does not need to rely on a trusted third party, it can efficiently solve the evaluation, calculation and update of the reputation value of the cognitive user, and determine the reputation status of the cognitive user.

Second, because the present invention adopts the evaluation cooperation reputation value reputation, the evaluation communication reputation value and the determination evaluation reputation value, refines the method for the real-time reputation change in the trust measurement stage, and overcomes the lack of consideration of trust initialization and real-time trust in the prior art. The lack of evaluation and trust update makes the present invention improve the feasibility of trust evaluation in cognitive radio networks.

Thirdly, since the present invention combines two aspects of trust measurement evaluation and supervision to carry out trust measurement, it overcomes the shortcomings in the prior art that cannot guarantee correct reputation measurement and spectrum sensing when the network is attacked maliciously, making the present invention improve Security and robustness of trust evaluation.

Fourth, since the present invention adopts a plurality of trust evaluation measurement factors of cooperation reputation value reputation, communication reputation value and evaluation reputation value, the measurement factors are diversified, which overcomes the single deficiency of the trust measurement factor in the prior art, and makes the present invention improve trust. Fairness of assessment.

Fifth, because the present invention adopts a complete reputation evaluation user group selection scheme and an evaluation and calculation method of cognitive user reputation value, a complete system flow is formed, which overcomes the shortcomings of the existing methods that do not constitute a complete trust evaluation method, making The present invention more intuitively and accurately reflects the credit status of a cognitive user.

Description of drawings

Fig. 1 is a flowchart of the present invention.

Specific implementation measures

The specific steps of the present invention will be further described below in conjunction with accompanying drawing 1 .

Step 1, initialization.

When the radio network is initialized, all data records in the database of cognitive users are cleared, including network parameter data records of cognitive users, trust data records of cognitive users, perception data records of cognitive users, spectrum requests of cognitive users Data Records and Reputation Evaluation Data Records for User Groups There are five types of data records.

A unique identity is set for each cognitive user in the radio network, and the identity of the cognitive user is stored in the network parameter data record of the cognitive user.

Set the initial cooperation reputation value and initial communication reputation value of all cognitive users in the trust data records of cognitive users to 0.5, the initial evaluation reputation value to 0, and the initial total reputation value to 0.5. The number of cognitive users in the radio network is recorded in the network parameter data record of the cognitive users to complete the initialization of the radio network.

When a new cognitive user requests to join the cognitive radio network, the new cognitive user is initialized: the reputation evaluation user group informs the new cognitive user of the reputation value information of other users in the current radio network. The data in the trust database of the new cognitive user is cleared, and the reputation value information and the number of cognitive users of other cognitive users in the current radio network are stored in the database.

Add 1 to the number of known users in the radio network in the network parameter data record of the known users. Create a reputation record for the new cognitive user in the trust database of all cognitive users.

Set the behavior reputation value, cooperation reputation value and total reputation value of the new cognitive user as the average value of all users in the current radio network, set the evaluation reputation value to 0, and complete the initialization of the new cognitive user.

Step 2, select the initial reputation evaluation user group.

If the data record of the reputation evaluation user group is empty, proceed to the following steps; otherwise, proceed to step 3.

Choose a positive real number in the range of [0.2,0.5] as the selection ratio. The selection ratio can be set dynamically. Multiply the number of cognitive users in the radio network with the selection ratio and round it up to get the number of users in the reputation evaluation user group.

Calculate the selection factor for evaluating user groups according to the following formula:

Among them, m represents the selection factor of the reputation evaluation user group, N represents the number of cognitive users in the radio network, J represents the number of users in the reputation evaluation user group, Indicates a round down operation.

From the database of cognitive users, select cognitive users who meet the following identification conditions to be recorded in the credit evaluation user group data records of cognitive users:

Imodm=1

Among them, I represents the identity of the cognitive user, m represents the selection factor of the reputation evaluation user group, and mod represents the remainder operation.

The cognitive users recorded in the reputation evaluation user group data records of the cognitive users are used as the reputation evaluation users, and all the reputation evaluation users are used as the reputation evaluation user groups.

Step 3, filter reputation evaluation user groups.

Check the reputation value of all reputation evaluation users in the reputation evaluation user group data record in turn, if the total reputation value of the reputation evaluation user is less than 0.5, or the evaluation reputation value of the reputation evaluation user is less than 0, then in order to ensure the security of trust evaluation, the The cognitive user no longer satisfies the conditions of the reputation evaluation user group, and the reputation evaluation user is deleted from the reputation evaluation user group data record.

In turn, check the number of consecutive reputation evaluation users in the reputation evaluation user group data record. If the reputation evaluation user has participated in the reputation evaluation for more than 5 consecutive times, in order to ensure the fairness of the trust evaluation, the user cannot continue to be a reputation evaluation user , delete the reputation assessment user from the reputation assessment user group data record.

Step 4, judge whether the number of remaining reputation evaluation users in the data record of the reputation evaluation user group is less than 2, if the remaining number of reputation evaluation users is less than 2, perform step 5, reselect the reputation evaluation user group, otherwise, perform step 6.

Step 5, reselect the reputation evaluation user group.

The current reputation evaluation user group randomly releases a positive integer, which is used as the initial application number of the cognitive user. Cognitive users who wish to join the reputation evaluation user group send out application information.

The credit evaluation user group checks whether the credit value of the cognitive user who sent the application information meets the application conditions. The application conditions include that the credit value of the cognitive user needs to be greater than 0.5, that the cognitive user has participated in spectrum sensing or data communication at least once, and that the cognitive user The remaining energy of is greater than the energy required to complete at least one trust evaluation. Then, starting from the initial application number and increasing by 1, an application number is set in turn for qualified cognitive users, and the application number of the cognitive user is recorded in the data record of the reputation evaluation user group.

Calculate the selection factor of the reputation evaluation user group according to the following formula:

Wherein, _m represents the selection factor of the reputation evaluation user group, N represents the number of cognitive users in the data record of the reputation evaluation user group, and J represents the number of users of the reputation evaluation user group described in step (2c), Indicates a round down operation.

From the data records of the reputation evaluation user group, select cognitive users who meet the following application numbering conditions, and delete users who do not meet the conditions:

Amodm=1

Among them, A represents the application number of the cognitive user, m represents the selection factor of the reputation evaluation user group, and mod represents the remainder operation.

The cognitive users recorded in the reputation evaluation user group data records of the cognitive users are regarded as the reputation evaluation users, and all the reputation evaluation users are regarded as the reputation evaluation user groups. Using the above method to select reputation evaluation users can ensure the fairness of selection.

Step 6, evaluate the cooperation reputation of cognitive users.

Cognitive users perform local signal detection and report local spectrum sensing information to the reputation evaluation user group. The values of the row vector elements of the spectrum sensing information are 0 and 1, and the number of elements in the row vector is the number of perceived spectrums.

After receiving the spectrum sensing information reported by the cognitive users, the reputation evaluation user group arranges the row vectors of the spectrum sensing information from top to bottom to obtain a matrix with the number of rows equal to the number of cognitive users reporting the spectrum sensing information. The credit evaluation user group compares the number of occurrences of vector elements 0 and 1 in each column vector of the matrix, and arranges the vector elements with the most occurrences in each column vector horizontally to obtain a row vector, which is used as the final spectrum sensing information.

The reputation evaluation user group broadcasts the final spectrum sensing information to the cognitive users in the radio network, and the cognitive users record the final spectrum sensing information into the cognitive user's sensing information data record.

The reputation evaluation user judges the cognitive user's spectrum sensing information: if the final spectrum sensing information is exactly the same as the cognitive user's spectrum sensing information, the cognitive user's spectrum sensing is correct; otherwise, the cognitive user's spectrum sensing is wrong.

The reputation evaluation user determines the cooperation score for the cognitive user: if the cognitive user reports the sensing information and the spectrum sensing is correct, the cognitive user’s cooperation score is 1; if the cognitive user reports the sensing information, but the spectrum sensing is wrong, the recognition The cooperative score of the cognitive user is 0.5; if the cognitive user does not report the perception information, the cooperative score of the cognitive user is 0.

Step 7, resource allocation.

Cognitive users who need to use the spectrum send spectrum request information to the reputation evaluation user group, and the request information includes the cognitive user's bid for the requested spectrum. After the reputation evaluation user group receives the cognitive user's request information, it calculates the spectrum competitiveness of the cognitive user who sent the spectrum request information according to the following formula, and records the spectrum competitiveness of the cognitive user into the cognitive user's spectrum request data record middle:

CP=T*B

Among them, CP represents the spectrum competitiveness of cognitive users, T is the total credit value of cognitive users, and B is the bid of cognitive users for spectrum.

The reputation evaluation user group will read the spectrum whose final spectrum sensing information is 0 from the spectrum sensing data records of cognitive users, and record it as a free spectrum list, sort the spectrum competitiveness of cognitive users from large to small, and list the free spectrum The spectrum in the spectrum list is allocated to the cognitive users in descending order of competitiveness.

After the spectrum allocation is completed, the reputation evaluation user group announces the spectrum allocation results.

Step 8, evaluating the communication reputation of the cognitive user.

Cognitive users allocated to the spectrum perform data communication on the allocated spectrum. During the data communication process of the cognitive users, the reputation evaluation user perceives and monitors the communication behavior of each cognitive user, and evaluates the communication score of the cognitive users: If the cognitive user does not forcibly occupy the spectrum or interfere with the communication of the primary user, the communication score of the cognitive user is 1; if the cognitive user has the behavior of interfering with the primary user, the communication score of the cognitive user is 0.5; If the cognitive user has the behavior of forcibly occupying the spectrum, the communication score of the cognitive user is 0.

Step 9, determining the evaluation reputation value of the reputation evaluation user.

Reputation evaluation users will sign the cognitive user's cooperation score and the cognitive user's communication score, and then exchange the signed cooperation and communication scores of the cognitive user. The signature ensures that the reputation evaluation user's cognitive Non-repudiation of user ratings.

According to the following formula, calculate the weight factor of the rating of the credit evaluation user to the cognitive user, and the weight factor is related to the evaluation reputation value of the reputation evaluation user:

${\mathrm{<span\; class="notranslate">\; \&omega;</span>}}^{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}^{\mathrm{<span\; class="notranslate">\; j</span>}}}{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; \&Element;</span>\mathrm{<span\; class="notranslate">\; J</span>}}{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}^{\mathrm{<span\; class="notranslate">\; j</span>}}}$

Among them, ω ^j represents the weighting factor of the rating of the reputation evaluation user with the identity j to the cognitive user, Indicates the evaluation reputation value of the reputation evaluation user whose identity is j, j represents the identity of the reputation evaluation user, J represents the identity set of reputation evaluation users of the reputation evaluation user group, and Σ represents the sum operation.

According to the following formula, the cooperation scores of cognitive users are integrated to obtain the overall cooperation evaluation of cognitive users:

${\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; \&Element;</span>\mathrm{<span\; class="notranslate">\; J</span>}}{\mathrm{<span\; class="notranslate">\; q</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}^{\mathrm{<span\; class="notranslate">\; j</span>}}$

Among them, Q ₁ represents the cooperative general evaluation of the cognitive user, j represents the identity mark of the reputation evaluation user, and J represents the identity mark set of the evaluation user of the reputation evaluation user group, Represents the cooperation score of the credit evaluation user with the identity ^j to the cognitive user, and ωj represents the weight factor of the credit evaluation user with the identity j of the cognitive user's score.

According to the following formula, the communication scores of cognitive users are integrated to obtain the overall communication evaluation of cognitive users:

${\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; \&Element;</span>\mathrm{<span\; class="notranslate">\; J</span>}}{\mathrm{<span\; class="notranslate">\; q</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}^{\mathrm{<span\; class="notranslate">\; j</span>}}$

Among them, Q ₂ represents the general evaluation of the communication of the cognitive user, j represents the identity of the cognitive user, and J represents the identity of the reputation evaluation user of the reputation evaluation user group, Indicates the communication score of the reputation evaluation user whose identity is ^j to the cognitive user, ω represents the weighting factor of the credit evaluation user’s rating of the cognitive user, and ωj represents the weight factor of the credit evaluation user’s rating of the cognitive user with the identity j weighting factor.

After obtaining the overall evaluation of cooperation and communication of cognitive users, the reputation evaluation user group makes judgments on the ratings of credit evaluation users: The differences between the values are less than 0.1, then the reputation evaluation user's evaluation is fair, otherwise, the reputation evaluation user's evaluation is unfair.

According to the scoring and judging situation of the credit evaluation user, the evaluation reputation value of the reputation evaluation user is determined. If the reputation evaluation user completes the reputation evaluation and the evaluation is fair, the evaluation reputation value of the reputation evaluation user is increased by 0.1; if the reputation evaluation user completes the reputation evaluation but the evaluation is unfair, the evaluation reputation value of the reputation evaluation user is reduced by 0.2; After the reputation evaluation is completed, the evaluation reputation value of the reputation evaluation user is reduced by 0.1.

Since the evaluation reputation value is a real number not less than 0 and not greater than 1, it is necessary to judge and correct the evaluation reputation value of the reputation evaluation user. If the evaluation reputation value of the reputation evaluation user is less than 0, the evaluation reputation value of the reputation evaluation user is corrected to 0; if the evaluation reputation value of the reputation evaluation user is greater than 1, the evaluation reputation value of the reputation evaluation user is corrected to 1.

Step 10, the reputation value is updated.

Calculate the updated cooperation reputation value of the cognitive user according to the following formula:

T ₁ ＝H ₁ ×γ+Q ₁ ×(1-γ)

Among them, T ₁ represents the updated cooperation reputation value of the cognitive user, H ₁ is the cooperation reputation value of the cognitive user before the update, γ represents the time trust correction factor, and its value is a real number between 0 and 1 , Q ₁ represents the overall evaluation of the cognitive user's cooperation.

Calculate the updated communication reputation value of the cognitive user according to the following formula:

T ₂ ＝H ₂ ×γ+Q ₂ ×(1-γ)

Among them, T2 represents the updated communication reputation value of the cognitive user, _H2 is the communication reputation value of the cognitive user before the update, γ represents the time trust correction factor, and its value is a real number between 0 and ₁ , Q ₂ represents the overall communication evaluation of cognitive users.

Calculate the updated total reputation value of the cognitive user according to the following formula:

T＝T ₁ ×α+T ₂ ×β+T ₃ ×(1-α-β)

Among them, T represents the updated total reputation value of the cognitive user, T ₁ represents the updated cooperation reputation value of the cognitive user, T ₂ represents the updated communication reputation value of the cognitive user, and T ₃ represents the cognitive user’s Evaluation reputation value, α and β represent trust weight factors, the values of α and β are real numbers between 0 and 1, and satisfy α+β<1.

The updated trust data includes the cognitive user's cooperation reputation value, communication reputation value and total reputation value. After obtaining the updated reputation data, the reputation evaluation user group broadcasts the updated reputation data to all cognitive users of the cognitive radio, and the cognitive users record the updated reputation data into the trust data records of the cognitive users to complete the reputation The value is updated to complete the evaluation of the reputation value of cognitive users in the distributed cognitive radio network.

Claims (8)

1. assess a method for cognitive user credit worthiness in distributed cognition radio network, comprise the steps:

(1) initialization:

(1a) all empty the data record in the database of cognitive user, to cognitive user each in radio net, set unique identify label according to natural number order, by the identify label of cognitive user stored in the network parametric data record of cognitive user;

(1b) the initial cooperation credit value of all cognitive user in the trust data record of cognitive user and initial communication credit value are set to 0.5, initial assessment credit value is set to 0, and initial total credit value is set to 0.5;

(1c) cognitive user number in radio net is recorded in the network parametric data record of cognitive user, completes radio net initialization;

(1d) to asking the new cognitive user adding network to carry out initialization;

(2) initial credit assessment user group is selected:

(2a) judge whether the data record that credit assessment user organizes is empty, if it is empty, performs step (2b), otherwise, perform step (3);

(2b) in [0.2,0.5] scope an optional arithmetic number as selection percentage;

(2c) cognitive user number in radio net is rounded after being multiplied with selection percentage, obtain the number of users of credit assessment user group;

(2d) selective factor B of credit assessment user group according to the following formula, is calculated:

Wherein, m represents that the selective factor B that credit assessment user organizes, N represent cognitive user number in radio net, and J represents the number of users that credit assessment user organizes, represent downward floor operation;

(2e) from the database of cognitive user, select the cognitive user meeting following formula identify label condition, the credit assessment user charging to cognitive user organizes in data record:

Imodm＝1

Wherein, I represents the identify label of cognitive user, and m represents the selective factor B that credit assessment user organizes, and mod represents modulo operation;

(2f) the credit assessment user of cognitive user is organized the cognitive user recorded in data record, as credit assessment user, using all credit assessment users as credit assessment user group;

(3) credit assessment user group is screened:

(3a) check that credit assessment user organizes the credit value of all credit assessment users in data record successively, if total credit value of credit assessment user is less than 0.5, or the assessment credit value of credit assessment user is less than 0, then this credit assessment user is organized data record from credit assessment user and delete;

(3b) check that all credit assessment users that credit assessment user organizes in data record participate in credit assessment number of times continuously successively, if credit assessment user participates in credit assessment continuously more than 5 times, this credit assessment user is organized data record from credit assessment user and deletes;

(4) judge whether the residue credit assessment number of users that credit assessment user organizes in data record is less than 2, if be less than 2, perform step (5), otherwise, perform step (6);

(5) credit assessment user group is reselected:

(5a) current credit assessment user organizes a random issue positive integer, using this positive integer as the initial application number of cognitive user;

(5b) wish that the cognitive user adding credit assessment user group sends application information;

(5c) check whether the credit value sending the cognitive user of application information meets application condition, with initial application number for starting point, with 1 for increasing progressively, to qualified cognitive user, set an application number successively, the application number of cognitive user is recorded in the data record of credit assessment user group;

(5d) selective factor B of credit assessment user group according to the following formula, is calculated:

Wherein, m represents the selective factor B that credit assessment user organizes, N ₁represent the cognitive user number in the data record of credit assessment user group, J represents the number of users of the credit assessment user group described in step (2c), represent downward floor operation;

(5e) from the data record of credit assessment user group, the cognitive user of discontented foot formula application numbers condition is deleted:

Amodm＝1

Wherein, A represents the application number of cognitive user, and m represents the selective factor B that credit assessment user organizes, and mod represents modulo operation;

(5f) the credit assessment user of cognitive user is organized the cognitive user recorded in data record, as credit assessment user, using all credit assessment users as credit assessment user group;

(6) the cooperation prestige of cognitive user is assessed:

(6a) cognitive user carries out local frequency spectrum perception;

(6b) cognitive user is by local frequency spectrum perception information reporting to credit assessment user group, and the row vector element value of frequency spectrum perception information is 0 and 1, and the number of the element of row vector is the frequency spectrum number of perception;

(6c) the frequency spectrum perception information of credit assessment user group to cognitive user carries out perception information fusion, obtains final frequency spectrum perception information;

(6d) credit assessment user group is by final frequency spectrum perception information broadcasting to the cognitive user in radio net, and final frequency spectrum perception information is recorded in the perception information data record of cognitive user by cognitive user;

(6e) credit assessment user adjudicates with whether the frequency spectrum perception information of cognitive user is identical final frequency spectrum perception information, if identical, then cognitive user frequency spectrum perception is correct, otherwise, cognitive user frequency spectrum perception mistake;

(6f) credit assessment user assesses the cooperation scoring of cognitive user, if cognitive user reports perception information, and cognitive user frequency spectrum perception is correct, then the cooperation scoring of cognitive user is 1; If cognitive user reports perception information, but cognitive user frequency spectrum perception mistake, then the cooperation scoring of cognitive user is 0.5; If cognitive user does not report perception information, then the cooperation scoring of cognitive user is 0;

(7) Resourse Distribute:

(7a) cognitive user sends and comprises the frequency spectrum solicited message of cognitive user to the bid of asked frequency spectrum;

(7b) according to the following formula, credit assessment user organizes the frequency spectrum competitiveness calculating and send the cognitive user of frequency spectrum solicited message, the frequency spectrum competitiveness of cognitive user is recorded in the frequency spectrum request msg record of cognitive user:

CP＝T*B

Wherein, CP represents the frequency spectrum competitiveness of cognitive user, and T is the total credit value of cognitive user, B by cognitive user to the bid of request frequency spectrum;

(7c) credit assessment user group by the frequency spectrum competitiveness of cognitive user according to sorting from big to small;

(7d) credit assessment user group is by from the frequency spectrum perception data record of cognitive user, the final frequency spectrum perception information read is the frequency spectrum of 0, be recorded as idle frequency spectrum list, the frequency spectrum in idle frequency spectrum list distributed to successively the descending cognitive user of competitiveness;

(7e) credit assessment user organizes and announces spectrum allocation may result;

(8) the communication prestige of cognitive user is assessed:

(8a) cognitive user carries out data communication according to spectrum allocation may result;

(8b) credit assessment user awareness monitors the communication behavior of each cognitive user, determines to mark to the communication of cognitive user according to communication quality judgment criteria;

(9) the assessment credit value of credit assessment user is determined:

(9a) two scorings of marking of communicating of the cooperation scoring of the cognitive user described in step (6f) and the cognitive user described in step (8b) are carried out signature operation by credit assessment user;

(9b) scoring of the cooperation to cognitive user after credit assessment user intercourses signature and the scoring that communicates;

(9c) according to scoring fusion method, respectively credit assessment user is merged with the scoring that communicates the cooperation scoring of cognitive user, obtain the cooperation general comment of cognitive user and the general comment that communicates;

(9d) scoring of credit assessment user is adjudicated, if the cooperation scoring of credit assessment user to cognitive user is all less than 0.1 with the cooperation general comment of cognitive user with the difference of the general comment that communicates respectively with the scoring that communicates, then the assessment of credit assessment user is fair, otherwise the assessment of credit assessment user is unfair;

(9e) according to assessment credit value evaluation method, the assessment credit value of credit assessment user is determined;

(10) credit value upgrades:

(10a) credit assessment user group is according to credit value update method, obtains the reputation data after upgrading, and the trust data after renewal comprises the cooperation credit value of cognitive user, communication credit value and total credit value;

(10b) credit assessment user group is by the reputation data after renewal, is broadcast to all cognitive user of cognitive radio;

(10c) cognitive user is by the reputation data after renewal, is recorded in the trust data record of cognitive user.

2. in distributed cognition radio network according to claim 1, assess the method for cognitive user credit worthiness, it is characterized in that, data record in the database of step (1a) described cognitive user, comprises data record five kinds of data record of the network parametric data record of cognitive user, the trust data record of cognitive user, the perception data record of cognitive user, the frequency spectrum request msg record of cognitive user and credit assessment user group.

3. assess the method for cognitive user credit worthiness in distributed cognition radio network according to claim 1, it is characterized in that, the new cognitive user initialization described in step (1d), carry out as follows:

The first step, credit assessment user group, by the new cognitive user of credit value information notification of other users in current radio network;

Second step, empties the data in the trust data storehouse of new cognitive user, by the credit value information of other cognitive user in current radio network and the cognitive user number database stored in cognitive user;

3rd step, in the database of all cognitive user, adds 1 by cognitive user number in the radio net in the network parametric data record of cognitive user;

4th step, in the trust data record of all cognitive user, for new cognitive user creates a prestige record;

5th step, by the communication credit value of new cognitive user, cooperation credit value and total credit value, all be set to the mean value of the credit value of every other cognitive user in radio net, the assessment credit value of new cognitive user is set to 0, complete new cognitive user initialization.

4. assess the method for cognitive user credit worthiness in distributed cognition radio network according to claim 1, it is characterized in that, the perception information described in step (6c) merges, and carries out as follows:

The first step, the row vector of frequency spectrum perception information is arranged in order, obtains the matrix that a line number equals to report the cognitive user number of frequency spectrum perception information after receiving the frequency spectrum perception information that cognitive user reports by credit assessment user group from top to bottom;

Second step, credit assessment user organizes the number of times that in each column vector of comparator matrix, vector element 0 and 1 occurs, records the vector element that in each column vector, occurrence number is many;

3rd step, credit assessment user group is by each column vector, and the vector element that occurrence number is many is transversely arranged successively, obtains a row vector, using this row vector as final frequency spectrum perception information.

5. in distributed cognition radio network according to claim 1, assess the method for cognitive user credit worthiness, it is characterized in that, communication quality judgment criteria described in step (8b) is: if the behavior taking frequency spectrum or interfere with primary users communication does not by force appear in cognitive user, then the communication scoring of cognitive user is 1; If the behavior of interfere with primary users appears in cognitive user, then the communication scoring of cognitive user is 0.5; If the behavior taking frequency spectrum by force appears in cognitive user, then the communication scoring of cognitive user is 0.

6. assess the method for cognitive user credit worthiness in distributed cognition radio network according to claim 1, it is characterized in that, the scoring fusion method described in step (9c), carry out as follows:

The first step, according to the following formula, calculates credit assessment user to the weight factor of the scoring of cognitive user:

${\mathrm{\&omega;}}^j=\frac{T_3^j}{{\mathrm{\&Sigma;}}_{j\&Element;J}{T}_3^j}$

Wherein, ω ^jrepresent identify label be the credit assessment user of j to the weight factor of the scoring of cognitive user, represent that identify label is the assessment credit value of the credit assessment user of j, j represents the identify label of credit assessment user, and J represents the identify label collection of the credit assessment user that credit assessment user organizes, and Σ represents sum operation;

Second step, according to the following formula, merges the cooperation scoring of cognitive user, obtains the cooperation general comment of cognitive user:

$Q_1={\mathrm{\&Sigma;}}_{j\&Element;J}{q}_1^j\&CenterDot;{\mathrm{\&omega;}}^j$

Wherein, Q ₁represent the cooperation general comment of cognitive user, j represents the identify label of credit assessment user, and J represents the identify label collection of the assessment user that credit assessment user organizes, expression identify label is that the credit assessment user of j marks to the cooperation of cognitive user, ω ^jrepresent that identify label is that the credit assessment user of j is to the weight factor of the scoring of cognitive user;

3rd step, according to the following formula, merges the communication scoring of cognitive user, obtains the communication general comment of cognitive user:

$Q_2={\mathrm{\&Sigma;}}_{j\&Element;J}{q}_2^j\&CenterDot;{\mathrm{\&omega;}}^j$

Wherein, Q ₂represent the communication general comment of cognitive user, j represents the identify label of credit assessment user, and J represents the identify label collection of the credit assessment user that credit assessment user organizes, expression identify label is that the credit assessment user of j marks to the communication of cognitive user, and ω represents the weight factor of credit assessment user to the scoring of cognitive user, ω ^jrepresent that identify label is that the credit assessment user of j is to the weight factor of the scoring of cognitive user.

7. assess the method for cognitive user credit worthiness in distributed cognition radio network according to claim 1, it is characterized in that, the assessment credit value evaluation method described in step (9e), carry out according to following rule:

The first step, if credit assessment user completes credit assessment and assessment justice, then the assessment credit value of credit assessment user improves 0.1; If credit assessment user completes credit assessment but assessment is unfair, then the assessment credit value of credit assessment user reduces by 0.2; If credit assessment user does not complete credit assessment, then the assessment credit value of credit assessment user reduces by 0.1;

Second step, adjudicates the assessment credit value of credit assessment user and revises: if the assessment credit value of credit assessment user is less than 0, then the assessment credit value of credit assessment user is modified to 0; If the assessment credit value of credit assessment user is greater than 1, then the assessment credit value of credit assessment user is modified to 1.

8. assess the method for cognitive user credit worthiness in distributed cognition radio network according to claim 1, it is characterized in that, the credit value update method described in step (10a), carry out according to following rule:

The first step, according to following formula, calculates the cooperation credit value after the renewal of cognitive user:

T ₁＝H ₁×γ+Q ₁×(1-γ)

Wherein, T ₁represent the cooperation credit value after the renewal of cognitive user, H ₁for the cooperation credit value before the renewal of cognitive user, γ represents that the time trusts modifying factor, and its value is the real number between 0 and 1, Q ₁represent the cooperation general comment of cognitive user;

Second step, according to the following formula, calculates the communication credit value after the renewal of cognitive user:

T ₂＝H ₂×γ+Q ₂×(1-γ)

Wherein, T ₂represent the communication credit value after the renewal of cognitive user, H ₂for the communication credit value before the renewal of cognitive user, γ represents that the time trusts modifying factor, and its value is the real number between 0 and 1, Q ₂represent the communication general comment of cognitive user;

3rd step, according to the following formula, calculates the total credit value after the renewal of cognitive user:

T＝T ₁×α+T ₂×β+T ₃×(1-α-β)

Wherein, T represents the total credit value after the renewal of cognitive user, T ₁represent the cooperation credit value after the renewal of cognitive user, T ₂represent the communication credit value after the renewal of cognitive user, T ₃represent the assessment credit value of cognitive user, α and β represents the trust weight factor, and the value of α and β is between being the real number between 0 and 1, and meets alpha+beta < 1.