CN104168575A - Multi-user dynamic spectrum sharing method for cognitive radio network - Google Patents

Abstract

The invention discloses a multi-user dynamic spectrum sharing method for a cognitive radio network. The method comprises the steps of initializing an authorization user base station, calculating the direct transmission rate and the cooperation rate, comparing the direct transmission rate with the cooperation rate, establishing a gambling model, establishing a benefit matrix, carrying out Hungary matching, broadcasting matching information and direct transmission by authorized users. According to the method, the characteristics of the cognitive radio network are fully utilized, united gambling of the authorized users and cognitive users is carried out, and therefore the overall efficiency of the cognitive radio network is improved; the Hungary algorithm is utilized for carrying out matching on the authorized users and the cognitive users, and therefore the utilization rate of idle frequency bands is increased and the fairness of spectrum allocation is improved.

Description

In cognition wireless network, multi-user's dynamic spectrum is shared method

Technical field

The invention belongs to communication technical field, a kind of multi-user's dynamic spectrum further relating in cognition wireless network technical field is shared method.The present invention ensureing under the prerequisite of fairness, realizes the optimum spectrum allocation may of multipair authorized user and cognitive user in cognition wireless network, effectively promotes the system energy efficiency of cognition wireless network.

Background technology

Along with the rapid growth of radio communication service demand, the deficient phenomenon of frequency spectrum resource becomes increasingly conspicuous.One of them main cause is under traditional spectrum management policy, and fixed allocation frequency spectrum system causes the availability of frequency spectrum low.The concept of cognitive radio has been proposed for this reason.It can allow undelegated user to enter mandate frequency range, dynamically utilizes mandate frequency range idle on time or space to carry out wireless transmission and improves the availability of frequency spectrum.Cognitive radio technology can allow authorisation network and cognition network coexist in same frequency range, and it makes the use of frequency spectrum more intelligent, effective.

Frequency spectrum share refers to according to the number of cognitive user and its to the demand of channel that by idle mandate spectrum allocation may, to one or more cognitive user, main purpose is effectively to select and utilize idle frequency spectrum by an adaptive strategy.In cognition network, authorized user exchanges certain income by leasing spatial frequency spectrum for to cognitive user.By the cooperation of certain way, the performance of authorisation network is significantly improved, and in return, thereby cognitive user has won the frequency spectrum right to use of a period of time, transmits its data.Utilize dynamic spectrum to share, the flexibility that can effectively improve radio communication, realizes the win-win of authorisation network and cognition network.

The patent application that Nanjing Univ. of Posts and Telecommunications proposes " cognitive radio frequency spectrum based on the equilibrium of supply and demand is shared method " (application number 201110106040.0 application publication number CN 102186176 A) discloses a kind of cognitive radio frequency spectrum based on the equilibrium of supply and demand and has shared method.The concrete steps of the method are: according to equilibrium of supply and demand the Theory Construction authorized user and cognitive user supply and demand model, authorized user provides idle frequency spectrum quantity, cognitive user according to the remaining battery capacity of definite frequency spectrum cost, cognitive user terminal and authorized user to different cognitive users the degree of belief in frequency spectrum access, adopt evolutionary game theory to realize each cognitive user maximum revenue.The deficiency that this patent application exists is: first, the idle frequency spectrum quantity that in the method model, authorized user provides is fixed, and can not dynamically adjust according to the state variation of authorized user.Secondly, only consider demand and the bid of cognitive user, do not considered the efficiency of authorisation network integral body in collaboration frequency spectrum shared procedure, sacrificed the performance of authorized user.

" a kind of based on game theoretic cognitive radio network spectrum allocation method " (application number 201310241296.1 application publication number CN 103281703 A) that Chongqing Mail and Telephones Unvi proposes disclose a kind of based on game theoretic cognitive radio spectrum allocation method.The concrete steps of the method are: by fixing a price to sharing frequency spectrum optimization in cognitive radio networks, set up authorized user associating utility function, adopt cooperative game theoretical optimization cognitive user to share the optimization price of frequency spectrum, realize authorized user associating maximization of utility.The deficiency that this patent application exists is: first, only considered the efficiency of authorized user when setting up utility function, the state of cognitive user in passive selection, network coexisted very flexible.Secondly, do not consider the fairness problem of authorized user and cognitive user in frequency spectrum share process.

Summary of the invention

The present invention is directed to above-mentioned the deficiencies in the prior art, propose multi-user's dynamic spectrum in a kind of cognition wireless network and share method, the method is by adopting Hungary Algorithm, combining the mutual selection of carrying out authorized user and cognitive user on the overall performance basis of considering authorisation network and cognition network, improved the utilance of frequency spectrum, guarantee the fairness between user, promoted the whole efficiency of network simultaneously.

The concrete thought that the present invention realizes above-mentioned purpose is, first cognitive user is sent cooperation request to authorized user base station, and joint game, according to the magnitude relationship of direct transfer speed and the speed that cooperates with cognitive user, is carried out by authorized user and cognitive user in authorized user base station; Then, adopt Hungary Algorithm to carry out the coupling of authorized user and cognitive user, realize the optimum spectrum allocation may of multipair authorized user and cognitive user in cognition wireless network, improve the efficiency of cognition network integral body.

The concrete steps that the present invention realizes above-mentioned purpose are as follows:

(1) initialization authorized user base station:

(1a) cognitive user nodes of each in cognition wireless network, sends initial state information to authorized user base station;

(1b) end of authorized user base station scans initial state information content, obtains the required power of transmission information and cooperation required parameter;

(2) calculate direct transfer speed and cooperation speed:

(2a) utilize the rate equation that direct transfers of authorized user node, calculate the speed that direct transfers of authorized user node;

(2b) utilize authorized user node and cognitive user nodes cooperation transmission rate equation, calculate the speed that cooperates of authorized user node and cognitive user nodes;

(3) relatively direct transfer speed and cooperation speed:

Whether authorized user base station judgement cooperation speed is greater than the speed that direct transfers, if so, perform step (4), otherwise, execution step (8);

(4) set up betting model:

If (4a) communication target of authorized user is for determining the reasonably transmission rate of cooperation parameter maximization system, the utility function of authorized user can be established as:

$U_j=\underset{{\mathrm{\α}}_i}{\max }{R}_{\mathrm{ij}}$

Wherein, U _jthe utility function that represents j authorized user, α _ithe cooperation required parameter that represents i cognitive user nodes, R _ijthe speed that represents j authorized user node and i cognitive user nodes cooperation transmission;

(4b) utilize cognitive user to share the throughput equation that frequency spectrum obtains, calculate cognitive user and share the throughput that frequency spectrum obtains;

If (4c) communication target of cognitive user maximizes the efficiency of system for determining rational power distribution strategies, the utility function of cognitive user can be established as:

$U_i=\underset{P_i}{\max }(\frac{1}{2}{\mathrm{\α}}_i{C}_i-\mathrm{\ω}{\mathrm{\α}}_i{P}_i)$

Wherein, U _ithe utility function that represents i cognitive user, α _ithe cooperation required parameter that represents i cognitive user nodes, C _irepresent the throughput that the shared frequency spectrum of i cognitive user nodes obtains, ω represents to transmit the unit energy that every bit information consumes, P _ithe transmitting power that represents i cognitive user nodes;

(4d) combined authorization user's utility function and the utility function of cognitive user, set up Stackelberg betting model as follows:

$\left\{\begin{array}{c}\underset{{\mathrm{\α}}_i\∈\left(\mathrm{0,1}\right]}{\max }{U}_j\\ \underset{P_i\∈(0,P_i^{\max }]}{\max }{U}_i\end{array}\right.$

Wherein, α _ithe cooperation required parameter that represents i cognitive user nodes, P _ithe transmitting power that represents i cognitive user nodes, the maximum power value that represents i cognitive user nodes, U _jand U _irepresent respectively the utility function of j authorized user and i cognitive user;

(4e) utilize formula with the optimum utility of authorized user and cognitive user, wherein, U _jand U _irepresent respectively the utility function of j authorized user and i cognitive user, P _jand P _irepresent respectively the transmitting power of j authorized user and i cognitive user;

(5) set up benefit matrix:

(5a) when cognitive user number M is greater than authorized user number N, with the number of difference M-N compensation authorized user of the two, form cognitive user and the authorized user matrix of a M * M, element value to the M-N increasing in this matrix authorized user gets 0, all the other element values get corresponding cognitive user and authorized user optimum utility adding and, obtain benefit matrix;

(5b) when cognitive user number M is less than authorized user number N, with the number of difference N-M compensation cognitive user of the two, cognitive user and authorized user form the matrix of a N * N, element value to the N-M increasing in this matrix cognitive user gets 0, all the other element values get corresponding cognitive user and authorized user optimum utility adding and, obtain benefit matrix;

(5c) when cognitive user number M equals authorized user number N, directly form cognitive user and the authorized user matrix of a M * N, to the element value in this matrix get corresponding cognitive user and authorized user optimum utility adding and, obtain benefit matrix;

(6) carry out Hungary's coupling:

(6a) to benefit matrix is capable, approximately subtract, with every row element, deduct the least member value in this row, obtain Matrix C ₁;

(6b) judgment matrix C ₁in each row of each row whether have " 0 ", if so, execution step (6c), otherwise, execution step (6d);

(6c) to Matrix C ₁be listed as approximately and subtract, with every column element, deduct the least member value in these row, obtain Matrix C ₂;

(6d) with minimum lid " 0 " line by Matrix C ₂in comprise " 0 " ranks all cover;

(6e) whether the number of judgement lid " 0 " line equals the dimension of matrix, if so, perform step (6h), otherwise, execution step (6f);

(6f) find out Matrix C ₂in least member value λ that tegmentum " 0 " line does not cover, with each element value that tegmentum " 0 " line covers, deduct λ, with the element value that each lid " 0 " line intersects, add λ;

(6g) repeating step (6e), step (6f), until the number of lid " 0 " line equals the dimension of matrix, obtain Hungary's matrix;

(6h) with " 1 ", substitute whole " 0 " in ranks, with " 0 ", substitute all the other element values in ranks, obtain 0-1 matrix;

(6i) get the 1st of 0-1 matrix and walk to capable and the 1st row of M to the coupling matrix that obtains M * N after N row, get the whole elements " 1 " in coupling matrix, obtain match information;

(7) broadcast match information:

Authorized user base station is broadcast to each authorized user and cognitive user by match information, authorized user and cognitive user cooperation transmission;

(8) the direct transmission information of authorized user.

The present invention compared with prior art has the following advantages:

First, the present invention is fully in conjunction with the feature of cognitive wireless, authorized user and cognitive user are carried out to joint game, overcome an one-sided deficiency of considering authorized user efficiency or cognitive user efficiency in prior art, make the present invention increase the income of cognition wireless network, improved the whole efficiency of cognition wireless network.

Second, the present invention adopts Hungary Algorithm to carry out the coupling of authorized user and cognitive user, overcome the deficiency of not considering different authorized users and cognitive user fairness in spectrum allocation may process in prior art, make the present invention improve the fairness of spectrum allocation may and the utilance of idle frequency spectrum, realized the mutual reciprocity and mutual benefit between multi-user.

Accompanying drawing explanation

Fig. 1 is flow chart of the present invention;

Fig. 2 is the flow chart of cooperation transmission process of the present invention;

Fig. 3 is analogous diagram of the present invention.

Embodiment:

Below in conjunction with accompanying drawing, the present invention is described in further detail.

Application scenarios of the present invention is cognition wireless network, in network, authorized user and cognitive user are in same geographical position, adopt centralized networking mode, by center base station, control communications, cognitive user is used idle frequency range by demand application separately, be assigned to after idle frequency range, data communication is carried out in authorized user and cognitive user cooperation.

With reference to accompanying drawing 1, concrete steps of the present invention are further described.

Step 1, initialization authorized user base station.

Each cognitive user nodes sends initial state information to authorized user base station.

The end of authorized user base station scans initial state information content, obtains the required power of transmission and cooperation required parameter.

Step 2, calculates direct transfer speed and cooperation speed.

Utilize the rate equation that direct transfers of authorized user node, calculate the speed that direct transfers of authorized user node:

$R_j={\log }_2(1+\frac{P_j{\left|g_j\right|}^2}{N_j})$

Wherein, R _jthe speed that direct transfers that represents j authorized user node, P _jthe transmitting power that represents j authorized user node, g _jrepresent that authorized user base station is to the channel gain of j authorized user node, N _jthe noise variance that represents j authorized user node.

With reference to accompanying drawing 2, three time slots of cooperation transmission process have been described.PU-BS in Fig. 2 represents authorized user base station, PU _j(j=1,2) represent authorized user, ST _i(i=1,2,3) represent cognitive user transmitter, SR _i(i=1,2,3) represent cognitive user receiver, with black arrow, represent the direction that information is transmitted.During the first time slot, authorized user and authorized user base station send information to cognitive user.During the second time slot, cognitive user sends to authorized user and authorized user base station the information that the first time slot receives simultaneously.During the 3rd time slot, cognitive user transmitting terminal is to self receiving terminal transmission information.

Utilize authorized user node and cognitive user nodes cooperation transmission rate equation, calculate the speed that cooperates of authorized user node and cognitive user nodes:

$R_{\mathrm{ij}}=\min \left\{\min \right[2(1-{\mathrm{\α}}_i){\log }_2(1+\frac{P_{\mathrm{ji}}{\left|h_{\mathrm{ij}}\right|}^2}{N_i}),{\mathrm{\α}}_i{\log }_2(1+\frac{P_i{\left|h_{\mathrm{ij}}\right|}^2}{N_j})],(1-{\mathrm{\α}}_j)R_j\}$

Wherein, R _ijthe speed that represents j authorized user node and i cognitive user nodes cooperation transmission, α _ithe cooperation required parameter that represents i cognitive user nodes, P _jirepresent that j authorized user node of the first time slot sends the power of information, P to i cognitive user nodes _irepresent that i cognitive user nodes of the second time slot sends the power of information, h to authorized user base station _ijexpression cognition is the channel gain to j authorized user node with i family node, N _iand N _jrepresent respectively the noise variance of i cognitive user nodes and j authorized user node, R _jthe speed that direct transfers that represents j authorized user node.

Step 3, the speed that relatively direct transfers and cooperation speed.

Whether authorized user base station judgement cooperation speed is greater than the speed that direct transfers, and if so, performs step 4, otherwise, execution step 8.

Step 4, sets up betting model.

In the shared process of dynamic spectrum, the communication target of authorized user and cognitive user is intended to determine that rational cooperation parameter or power distribution strategies maximize the performance of cognitive wireless.Therefore,, according to the target of system transmission rate and system energy efficiency, the utility function of setting up respectively authorized user and cognitive user is as follows:

If the communication target of authorized user is for determining the reasonably transmission rate of cooperation parameter maximization system, the utility function of authorized user can be established as:

$U_j=\underset{{\mathrm{\α}}_i}{\max }{R}_{\mathrm{ij}}$

Wherein, U _jthe utility function that represents j authorized user, α _ithe cooperation required parameter that represents i cognitive user nodes, R _ijthe speed that represents j authorized user node and i cognitive user nodes cooperation transmission.

Utilize cognitive user to share the throughput equation that frequency spectrum obtains, calculate cognitive user and share the throughput that frequency spectrum obtains:

$C_i=\frac{1}{2}{\mathrm{\α}}_i{\log }_2(1+\frac{P_i{\left|h_{\mathrm{ii}}\right|}^2}{N_i+{\mathrm{\Σ}}_{j=1,j\≠i}^M{P}_j{\left|h_{\mathrm{ij}}\right|}^2})$

Wherein, C _irepresent the throughput that the shared frequency spectrum of i cognitive user nodes of the 3rd time slot obtains, α _ithe cooperation required parameter that represents i cognitive user nodes, P _iand P _jrepresent respectively the transmitting power of i cognitive user nodes and j cognitive user nodes, h _iirepresent that i cognitive user nodes transmitting terminal is to the channel gain of i cognitive user nodes receiving terminal, h _ijrepresent that i cognitive user nodes is to the channel gain of j cognitive user nodes, N _ithe noise variance that represents i cognitive user nodes, M represents the number of cognitive user nodes.

If the communication target of cognitive user maximizes the efficiency of system for determining rational power distribution strategies, the utility function of cognitive user can be established as:

$U_i=\underset{P_i}{\max }(\frac{1}{2}{\mathrm{\α}}_i{C}_i-\mathrm{\ω}{\mathrm{\α}}_i{P}_i)$

Wherein, U _ithe utility function that represents i cognitive user, α _ithe cooperation required parameter that represents i cognitive user nodes, C _irepresent the throughput that the shared frequency spectrum of i cognitive user nodes obtains, ω represents to transmit the unit energy that every bit information consumes, P _ithe transmitting power that represents i cognitive user nodes.

Combined authorization user's utility function and the utility function of cognitive user, set up Stackelberg betting model as follows:

$\left\{\begin{array}{c}\underset{{\mathrm{\α}}_i\∈\left(\mathrm{0,1}\right]}{\max }{U}_j\\ \underset{P_i\∈(0,P_i^{\max }]}{\max }{U}_i\end{array}\right.$

Wherein, α _ithe cooperation required parameter that represents i cognitive user nodes, P _ithe transmitting power that represents i cognitive user nodes, the maximum power value that represents i cognitive user nodes, U _jand U _irepresent respectively the utility function of j authorized user and i cognitive user.

Utilize formula with the optimum utility of authorized user and cognitive user, wherein, U _jand U _irepresent respectively the utility function of j authorized user and i cognitive user, P _jand P _irepresent respectively the transmitting power of j authorized user and i cognitive user.

Step 5, sets up benefit matrix.

In the shared process of dynamic spectrum, authorized user and cognitive user need realize coupling one by one.Therefore,, according to the different situations of cognitive user number M and authorized user number N, set up benefit matrix as follows:

When cognitive user number M is greater than authorized user number N, with the number of difference M-N compensation authorized user of the two, form cognitive user and the authorized user matrix of a M * M, element value to the M-N increasing in this matrix authorized user gets 0, all the other element values get corresponding cognitive user and authorized user optimum utility adding and, obtain benefit matrix.

When cognitive user number M is less than authorized user number N, with the number of difference N-M compensation cognitive user of the two, cognitive user and authorized user form the matrix of a N * N, element value to the N-M increasing in this matrix cognitive user gets 0, all the other element values get corresponding cognitive user and authorized user optimum utility adding and, obtain benefit matrix.

When cognitive user number M equals authorized user number N, cognitive user and authorized user directly form the matrix of a M * N, to the element value in this matrix get corresponding cognitive user and authorized user optimum utility adding and, obtain benefit matrix.

Step 6, carries out Hungary's coupling.

(6a) to benefit matrix is capable, approximately subtract, with every row element, deduct the least member value in this row, obtain Matrix C ₁.

(6b) judgment matrix C ₁in each row of each row whether have " 0 ", if so, execution step (6c), otherwise, execution step (6d).

(6c) to Matrix C ₁be listed as approximately and subtract, with every column element, deduct the least member value in these row, obtain Matrix C ₂.

(6d) with minimum lid " 0 " line by Matrix C ₂in comprise " 0 " ranks all cover.

(6e) whether the number of judgement lid " 0 " line equals the dimension of matrix, if so, perform step (6h), otherwise, execution step (6f).

(6f) find out Matrix C ₂in least member value λ that tegmentum " 0 " line does not cover, with each element value that tegmentum " 0 " line covers, deduct λ, with the element value that each lid " 0 " line intersects, add λ.

(6g) repeating step (6e), step (6f), until the number of lid " 0 " line equals the dimension of matrix, obtain Hungary's matrix.

(6h) with " 1 ", substitute whole " 0 " in ranks, with " 0 ", substitute all the other element values in ranks, obtain 0-1 matrix.

(6i) get the 1st of 0-1 matrix and walk to capable and the 1st row of M to the coupling matrix that obtains M * N after N row, get the whole elements " 1 " in coupling matrix, obtain match information.

Step 7, broadcast match information.

Authorized user base station is broadcast to each authorized user and cognitive user by match information, authorized user and cognitive user cooperation transmission.

Step 8, the direct transmission information of authorized user.

Below in conjunction with 3 pairs of effects of the present invention of accompanying drawing, be further described.

Fig. 3 shows that by relatively the present invention program, traditional scheme 1, traditional scheme 2 authorized user and cognitive user performance are along with the design sketch of authorized user and cognitive user number of variations.Traditional scheme 1 expression authorized user is initiatively selected cognitive user, and traditional scheme 2 expression cognitive user are initiatively selected authorized user.

With reference to accompanying drawing 3 (a), abscissa represents the number of authorized user and cognitive user, and ordinate represents the effectiveness of authorisation network, triangle solid line represents the present invention program, square solid line represents traditional scheme 1, and continuous circular line represents traditional scheme 2, and rhombus dotted line represents direct transmission.From Fig. 3 (a), can find out, increase along with authorized user and cognitive user number, the effectiveness of four kinds of scheme authorisation network rises gradually, wherein, frequency spectrum share scheme of the present invention has on average promoted about 1.6 times with respect to the effectiveness of traditional frequency spectrum share scheme 2 authorisation network, obviously promoted the performance of authorized user, and far above the performance of direct transmission.

With reference to accompanying drawing 3 (b), abscissa represents the number of authorized user and cognitive user, and ordinate represents the effectiveness of cognition network, and triangle solid line represents the present invention program, and five-pointed star solid line represents traditional scheme 1, and continuous circular line represents traditional scheme 2.From Fig. 3 (b), can find out, increase along with authorized user and cognitive user number, the effectiveness of three kinds of scheme cognition networks rises gradually, wherein, though frequency spectrum share scheme of the present invention has certain loss with respect to the effectiveness of traditional frequency spectrum share scheme 1 cognition network, loss is less than 20% of traditional frequency spectrum share scheme 2 cognition network effectiveness.

Claims (4)

1. in cognition wireless network, multi-user's dynamic spectrum is shared a method, comprises the steps:

(1) initialization authorized user base station:

(1a) cognitive user nodes of each in cognition wireless network, sends initial state information to authorized user base station;

(1b) end of authorized user base station scans initial state information content, obtains the required power of transmission information and cooperation required parameter;

(2) calculate direct transfer speed and cooperation speed:

(2a) utilize the rate equation that direct transfers of authorized user node, calculate the speed that direct transfers of authorized user node;

(2b) utilize authorized user node and cognitive user nodes cooperation transmission rate equation, calculate the speed that cooperates of authorized user node and cognitive user nodes;

(3) relatively direct transfer speed and cooperation speed:

Whether authorized user base station judgement cooperation speed is greater than the speed that direct transfers, if so, perform step (4), otherwise, execution step (8);

(4) set up betting model:

If (4a) communication target of authorized user is for determining the reasonably transmission rate of cooperation parameter maximization system, the utility function of authorized user can be established as:

$U_j=\underset{{\mathrm{\α}}_i}{\max }{R}_{\mathrm{ij}}$

Wherein, U _jthe utility function that represents j authorized user, α _ithe cooperation required parameter that represents i cognitive user nodes, R _ijthe speed that represents j authorized user node and i cognitive user nodes cooperation transmission;

(4b) utilize cognitive user to share the throughput equation that frequency spectrum obtains, calculate cognitive user and share the throughput that frequency spectrum obtains;

If (4c) communication target of cognitive user maximizes the efficiency of system for determining rational power distribution strategies, the utility function of cognitive user can be established as:

$U_i=\underset{P_i}{\max }(\frac{1}{2}{\mathrm{\α}}_i{C}_i-\mathrm{\ω}{\mathrm{\α}}_i{P}_i)$

Wherein, U _ithe utility function that represents i cognitive user, α _ithe cooperation required parameter that represents i cognitive user nodes, C _irepresent the throughput that the shared frequency spectrum of i cognitive user nodes obtains, ω represents to transmit the unit energy that every bit information consumes, P _ithe transmitting power that represents i cognitive user nodes;

(4d) combined authorization user's utility function and the utility function of cognitive user, set up Stackelberg betting model as follows:

$\left\{\begin{array}{c}\underset{{\mathrm{\α}}_i\∈\left(\mathrm{0,1}\right]}{\max }{U}_j\\ \underset{P_i\∈(0,P_i^{\max }]}{\max }{U}_i\end{array}\right.$

Wherein, α _ithe cooperation required parameter that represents i cognitive user nodes, P _ithe transmitting power that represents i cognitive user nodes, the maximum power value that represents i cognitive user nodes, U _jand U _irepresent respectively the utility function of j authorized user and i cognitive user;

(4e) utilize formula with the optimum utility of authorized user and cognitive user, wherein, U _jand U _irepresent respectively the utility function of j authorized user and i cognitive user, P _jand P _irepresent respectively the transmitting power of j authorized user and i cognitive user;

(5) set up benefit matrix:

(5a) when cognitive user number M is greater than authorized user number N, with the number of difference M-N compensation authorized user of the two, form cognitive user and the authorized user matrix of a M * M, element value to the M-N increasing in this matrix authorized user gets 0, all the other element values get corresponding cognitive user and authorized user optimum utility adding and, obtain benefit matrix;

(5b) when cognitive user number M is less than authorized user number N, with the number of difference N-M compensation cognitive user of the two, cognitive user and authorized user form the matrix of a N * N, element value to the N-M increasing in this matrix cognitive user gets 0, all the other element values get corresponding cognitive user and authorized user optimum utility adding and, obtain benefit matrix;

(5c), when cognitive user number M equals authorized user number N, cognitive user and authorized user directly form the matrix of a M * N, to the element value in this matrix get corresponding cognitive user and authorized user optimum utility adding and, obtain benefit matrix;

(6) carry out Hungary's coupling:

(6a) to benefit matrix is capable, approximately subtract, with every row element, deduct the least member value in this row, obtain Matrix C ₁;

(6b) judgment matrix C ₁in each row of each row whether have " 0 ", if so, execution step (6c), otherwise, execution step (6d);

(6c) to Matrix C ₁be listed as approximately and subtract, with every column element, deduct the least member value in these row, obtain Matrix C ₂;

(6d) with minimum lid " 0 " line by Matrix C ₂in comprise " 0 " ranks all cover;

(6e) whether the number of judgement lid " 0 " line equals the dimension of matrix, if so, perform step (6h), otherwise, execution step (6f);

(6f) find out Matrix C ₂in least member value λ that tegmentum " 0 " line does not cover, with each element value that tegmentum " 0 " line covers, deduct λ, with the element value that each lid " 0 " line intersects, add λ;

(6g) repeating step (6e), step (6f), until the number of lid " 0 " line equals the dimension of matrix, obtain Hungary's matrix;

(6h) with " 1 ", substitute whole " 0 " in ranks, with " 0 ", substitute all the other element values in ranks, obtain 0-1 matrix;

(6i) get the 1st of 0-1 matrix and walk to capable and the 1st row of M to the coupling matrix that obtains M * N after N row, get the whole elements " 1 " in coupling matrix, obtain match information;

(7) broadcast match information:

Authorized user base station is broadcast to each authorized user and cognitive user by match information, authorized user and cognitive user cooperation transmission;

(8) the direct transmission information of authorized user.

2. in cognition wireless network according to claim 1, multi-user's dynamic spectrum is shared method, it is characterized in that, the rate equation that direct transfers of the authorized user node described in step (2a) is as follows:

$R_j={\log }_2(1+\frac{P_j{\left|g_j\right|}^2}{N_j})$

Wherein, R _jthe speed that direct transfers that represents j authorized user node, P _jthe transmitting power that represents j authorized user node, g _jrepresent that authorized user base station is to the channel gain of j authorized user node, N _jthe noise variance that represents j authorized user node.

3. in cognition wireless network according to claim 1, multi-user's dynamic spectrum is shared method, it is characterized in that, the authorized user node described in step (2b) is as follows with the rate equation that cooperates of cognitive user nodes:

$R_{\mathrm{ij}}=\min \left\{\min \right[2(1-{\mathrm{\α}}_i){\log }_2(1+\frac{P_{\mathrm{ji}}{\left|h_{\mathrm{ij}}\right|}^2}{N_i}),{\mathrm{\α}}_i{\log }_2(1+\frac{P_i{\left|h_{\mathrm{ij}}\right|}^2}{N_j})],(1-{\mathrm{\α}}_j)R_j\}$

Wherein, R _ijthe speed that represents j authorized user node of the first time slot and i cognitive user nodes cooperation transmission, α _ithe cooperation required parameter that represents i cognitive user nodes, P _jirepresent that j authorized user node of the second time slot sends the power of information, P to i cognitive user nodes _irepresent that i cognitive user nodes sends the power of information, h to authorized user base station _ijexpression cognition is the channel gain to j authorized user node with i family node, N _iand N _jrepresent respectively the noise variance of i cognitive user nodes and j authorized user node, R _jthe speed that direct transfers that represents j authorized user node.

4. in cognition wireless network according to claim 1, multi-user's dynamic spectrum is shared method, it is characterized in that, the throughput equation that the cognitive user described in step (4b) is shared frequency spectrum acquisition is as follows:

$C_i=\frac{1}{2}{\mathrm{\α}}_i{\log }_2(1+\frac{P_i{\left|h_{\mathrm{ii}}\right|}^2}{N_i+{\mathrm{\Σ}}_{j=1,j\≠i}^M{P}_j{\left|h_{\mathrm{ij}}\right|}^2})$

Wherein, C _irepresent the throughput that the shared frequency spectrum of i cognitive user nodes of the 3rd time slot obtains, α _ithe cooperation required parameter that represents i cognitive user nodes, P _iand P _jrepresent respectively the transmitting power of i cognitive user nodes and j cognitive user nodes, h _iirepresent that i cognitive user nodes transmitting terminal is to the channel gain of i cognitive user nodes receiving terminal, h _ijrepresent that i cognitive user nodes is to the channel gain of j cognitive user nodes, N _ithe noise variance that represents i cognitive user nodes, M represents the number of cognitive user nodes.