CR-NOMA resource allocation scheme combining Hungary algorithm and genetic algorithm

Abstract

The invention discloses a Hungary algorithm and genetic algorithm combined CR-NOMA resource allocation scheme, which relates to a wireless communication technology, and comprises the following steps: decomposing a constructed optimization problem P0 into a secondary user channel matching problem P1 and a secondary user power allocation problem P2; solving a secondary user channel matching problem P1 by using a Hungary algorithm, designing a secondary user pairing clustering strategy to obtain an optimal channel matching matrix, and solving a secondary user power distribution problem P2 by using an adaptive genetic algorithm to obtain an optimal power distribution scheme; and iteratively executing the steps until the adaptive genetic algorithm reaches the maximum number of iterations, and outputting the optimal channel matching scheme and the optimal power allocation scheme of the secondary users. According to the method, an optimization problem P0 is decomposed into a secondary user channel matching problem P1 and a secondary user power distribution problem P2, so that the problem solving complexity is reduced; and solving a secondary user power distribution problem P2 by using a self-adaptive genetic algorithm, and improving the global search capability by dynamically adjusting the crossover and mutation probability.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a CR-NOMA resource allocation scheme combining a Hungarian algorithm and a genetic algorithm. Background Art

[0002] With the advent of the "Internet of Everything" era, the number of wireless terminals has exploded, and the demand for spectrum resources in mobile networks is increasing. How to solve the problem of efficient multiple access (MA) of large-scale terminals with limited spectrum resources has become a key issue facing current wireless networks.

[0003] The fifth generation of mobile communications has proposed a new power domain non-orthogonal multiple access (NOMA) technology, which supports the simultaneous transmission of multiple users on the same frequency by multiplexing power resources, and adopts a more complex serial interference cancellation (SIC) technology to demodulate multiple access. For the problem of insufficient spectrum resources, cognitive radio (CR) technology is seen as an effective solution. It allows secondary users (SUs) to share the authorized spectrum resources of primary users (PUs) without excessively affecting the transmission performance of primary users (PUs), so as to support flexible and dynamic data transmission of terminals, and also facilitate the random transmission of short packets of large-scale IoT terminals. Considering that CR technology can provide flexible wireless spectrum resources, and NOMA technology supports flexible and efficient multiple access, the combination of the two is expected to greatly improve the utilization of wireless spectrum, thereby solving the contradiction between insufficient spectrum resources and the sharp increase in random access requirements of large-scale terminals.

[0004] At present, CR-NOMA research is being carried out one after another. The research work mainly focuses on how to design more efficient resource allocation schemes to improve access efficiency and system throughput. Most of the attention is paid to small-scale CR terminal scenarios. There are fewer CR-NOMA research works for random access of large-scale IoT terminals. In addition, existing research often uses high-complexity exhaustive methods or general sequential matching methods when dealing with secondary user pairing. Therefore, it is necessary to study efficient NOMA transmission optimization schemes for large-scale CR terminal scenarios to improve CR-NOMA transmission performance. Summary of the invention

[0005] In view of the problems existing in the existing CR-NOMA resource allocation scheme, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is how to design an efficient CR-NOMA resource allocation scheme that can simultaneously optimize user-channel matching and power allocation, maximize system throughput while satisfying various system constraints, and has low computational complexity and is suitable for practical systems.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a CR-NOMA resource allocation scheme combining a Hungarian algorithm and a genetic algorithm, which includes: constructing an optimization problem P in a CR-NOMA downlink system; 0 , and the optimization problem P 0 Decomposed into the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 ; Use the Hungarian algorithm to solve the secondary user channel matching problem P 1 , design a secondary user pairing clustering strategy to obtain an optimal channel matching matrix; based on the optimal channel matching matrix, an adaptive genetic algorithm is used to solve the secondary user power allocation problem P 2 , get the optimal power allocation solution; iterative execution solves the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 steps until the adaptive genetic algorithm reaches the maximum number of iterations and outputs the optimal channel matching scheme and optimal power allocation scheme for the secondary user.

[0009] As a preferred solution of the CR-NOMA resource allocation scheme combining the Hungarian algorithm and the genetic algorithm described in the present invention, wherein: the CR-NOMA downlink system includes a primary base station PBS and a cognitive base station SBS; the primary base station PBS includes a number of primary users PU, wherein the number of channels is M and the channel bandwidth is B 0 = B / M, the channel set is represented by CH = {c 1 ,c 2 ,...,c M The cognitive base station SBS adopts the Underlay mode to share the channel of the primary user PU and transmits downlink signals to N secondary users SU. The secondary user set is U m = {u 1,m ,u 2,m ...,u n,m ,...,u N,m}, the total downlink transmission power is Among them, u n represents the nth secondary user SU, p mis the transmission power allocated by the cognitive base station SBS to channel m; the construction process of the optimization problem includes the following steps: using NOMA technology to share the same channel m for signal transmission, and dividing N secondary users SU into m clusters of K, where the optional power level set of the mth cluster is PT M,K = {Pt 1,1 ,Pt 2,1 ,...,Pt m,k}; Calculate the cognitive base station SBS through the channel m with power level pt m,k To the secondary user u n The downlink achievable rate of the sending signal is as follows:

[0010] R n,m,k =B 0 log 2 (1+SINR n,m,k );

[0011] In the formula, R n,m,k is the downlink achievable rate of the nth secondary user SU in the kth group on channel m, SINR n,m,k is the received signal-to-interference-noise ratio of the nth secondary user SU in the kth group on the channel m. The specific formula is as follows:

[0012]

[0013] In the formula, h n,m =g n,m PL -ε (d n ) is the gain from the cognitive base station SBS to the nth secondary user SU in the mth channel, g n,m is the Rayleigh fading factor of the nth secondary user SU on channel m, ε is the path loss, PL -ε (d n,m ) is the path loss function, d n,m is the distance from the cognitive base station SBS to the nth secondary user SU in the mth channel, η n,m is the additive Gaussian white noise power from the cognitive base station SBS to the nth secondary user SU in the mth channel, pt m,k =α m,k p m is the kth power level in the mth cluster, α m,k ∈[0,1] is the power level pt m,k The total transmit power p on channel m m The proportion of for And k>l, pt m,k >pt m,l .

[0014] As a preferred solution of the CR-NOMA resource allocation solution combining the Hungarian algorithm and the genetic algorithm described in the present invention, it also includes: based on the downlink achievable rate, calculating the total downlink throughput T of the cognitive base station SBS to the secondary user sum , the specific formula is as follows:

[0015]

[0016] In the formula, θ n,m is the channel matching relationship variable of the nth secondary user SU on channel m, m∈{1,2,...,M}, n∈{1,..,N}, is the kth power level pt in the mth cluster m,k The downlink power level selection variable of the nth secondary user SU, if θ n,m = 0, then the cognitive base station SBS does not transmit the signal of the nth secondary user SU through channel m. If θ n,m =1, then the cognitive base station SBS transmits the signal of the nth secondary user SU through channel m, if Then the cognitive base station SBS does not select the kth power level pt in the mth cluster m,k The transmitted signal of the nth secondary user SU, if θ n,m =1, the cognitive base station SBS does not select the kth power level pt in the mth cluster m,k The transmitted signal of the nth secondary user SU.

[0017] According to the total downlink throughput, taking the secondary user throughput maximization as the objective function, C 1 , C 2 , C 3 , C 4 , C 5 , C 6 and C 7 As constraints, establish the optimization problem P 0 , and the optimization problem P 0 Decomposed into the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 .

[0018] Among them, C 1 The constraint condition that each secondary user SU can only occupy one subchannel, C 2 As the constraint that each secondary user SU can only select one power level on each channel, C 3 is the proportion constraint of different power levels on each channel, C 4 is the constraint condition for the ordering relationship of the K power levels on each channel, C 5 is the constraint condition of the total transmission power of the cognitive base station SBS, C6 is the interference constraint condition for the primary user, C 7 is the constraint condition of the minimum transmission rate of the secondary user SU.

[0019] As a preferred solution of the CR-NOMA resource allocation solution combining the Hungarian algorithm and the genetic algorithm described in the present invention, wherein: the optimization problem P 0 The relevant formula is as follows:

[0020]

[0021] In the formula, p max is the upper limit of the transmission power, p mask,m is the maximum interference tolerance of the primary user PU on channel m, is the lower limit of the rate of the nth secondary user SU on channel m.

[0022] The secondary user channel matching problem P 1 The relevant formula is as follows:

[0023]

[0024] The secondary user power allocation problem P 2 The relevant formula is as follows:

[0025]

[0026] As a preferred solution of the CR-NOMA resource allocation solution combining the Hungarian algorithm and the genetic algorithm described in the present invention, wherein: the secondary user channel matching problem P 1 The solution is as follows: Set the gain h of any channel m to N secondary SUs n,m =g n,m PL -ε (d n ), sorted in descending order, and |h 1,m 2 >|h 2,m | 2 >...>|h n,m | 2 >...>|h N,m | 2 ; Use the Hungarian algorithm to select the secondary user set U m = {u 1,m ,u 2,m ...,u n,m ,...,u N,m} selects two users to pair into user cluster m and performs power domain NOMA transmission on channel m, performs secondary user pairing clustering and intra-cluster user power level selection strategy; from the secondary user set U of channel m m = {u1,m ,u 2,m ...,u n,m ,...,u N,m} select the current unpaired minimum gain as the weak user u of channel m m2 , and calculate the achievable rate R of weak user pairing m2,m ; Traverse all channels and obtain the matching weak user u on each channel m2 , and will be added to the set of users that have matched the channel In; from the secondary user set U of channel m m = {u 1,m ,u 2,m ...,u n,m ,...,u N,m} removes users with matching channels and obtains the pruned strong user set According to the strong user set Select strong user u m1 , calculate the achievable rate of the optimal strong user and weak user pairing; traverse the strong user set Get the total rate of user clusters after pairing on the channel and construct the initial channel benefit matrix C M×M =[c n,m ]; for the initial channel benefit matrix C M×M =[c n,m ] to optimize and obtain the optimal channel matching matrix; according to the optimal channel matching matrix, an adaptive genetic algorithm is used to solve the secondary user power allocation problem P 2 , and get the initial power allocation scheme.

[0027] As a preferred solution of the CR-NOMA resource allocation solution combining the Hungarian algorithm and the genetic algorithm described in the present invention, wherein: the secondary user power allocation problem P 2 The solution is as follows: The cognitive base station SBS randomly generates S groups of M-dimensional power allocation coefficient vectors α s =[α s,1 ,α s,2 ,...,α s,m ,...,α s,M] as the initial population, each vector represents a power allocation scheme; based on the power allocation scheme and the optimal channel matching matrix, the total system throughput is calculated as the fitness function f(s) of the population; according to the fitness function f(s), a new generation population is selected using a roulette wheel method, and the individuals with the highest fitness are retained as the new generation population; a two-point crossover operation is performed on the new generation population, and the crossover probability is dynamically adjusted according to the state of the current new generation population; a mutation operation is performed on the new generation population after the two-point crossover operation, and the mutation probability is dynamically adjusted according to the state of the current new generation population; the above steps are repeated until the number of iterations g of the adaptive genetic algorithm reaches the maximum number of generations g max , select the individual with the highest fitness from the final population as the optimal NOMA power allocation coefficient vector on the channel, and output the optimal power allocation scheme.

[0028] As a preferred solution of the CR-NOMA resource allocation solution combining the Hungarian algorithm and the genetic algorithm described in the present invention, the specific formula of the fitness function f(s) is as follows:

[0029]

[0030] in, is the power distribution coefficient in the population individual s when α s , the pairing of the nth secondary user SU and channel m, if Then the nth secondary user SU is multiplexed on channel m, if Then the nth secondary user SU is not multiplexed on channel m, is the total rate of the user cluster when the secondary user SU is multiplexed on channel m.

[0031] In a second aspect, an embodiment of the present invention provides a CR-NOMA resource allocation method combining a Hungarian algorithm and a genetic algorithm, which includes: an optimization problem construction and decomposition module for constructing an optimization problem P in a CR-NOMA downlink system 0 , and the optimization problem P 0 Decomposed into the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 ; A secondary user channel matching module, used to solve the secondary user channel matching problem P using the Hungarian algorithm 1 , design a secondary user pairing clustering strategy to obtain an optimal channel matching matrix; a secondary user power allocation module is used to solve the secondary user power allocation problem P based on the optimal channel matching matrix using an adaptive genetic algorithm 2 , get the optimal power allocation solution; output module, used to iteratively execute to solve the secondary user channel matching problem P 1 and the secondary user power allocation problem P2 steps until the adaptive genetic algorithm reaches the maximum number of iterations and outputs the optimal channel matching scheme and optimal power allocation scheme for the secondary user.

[0032] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the CR-NOMA resource allocation scheme combining the Hungarian algorithm and the genetic algorithm as described in the first aspect of the present invention are implemented.

[0033] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the CR-NOMA resource allocation scheme combining the Hungarian algorithm and the genetic algorithm as described in the first aspect of the present invention are implemented.

[0034] The beneficial effects of the present invention are as follows: by optimizing the problem P 0 Decomposed into the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 , reducing the complexity of problem solving; using the Hungarian algorithm to solve the secondary user channel matching problem P 1 Compared with the exhaustive method, it significantly reduces the amount of calculation and ensures the optimality of the matching result; an adaptive genetic algorithm is used to solve the secondary user power allocation problem P 2 By dynamically adjusting the crossover and mutation probabilities, the risk of falling into the local optimal solution can be effectively avoided and the global search capability can be improved; by iteratively executing the Hungarian algorithm and the adaptive genetic algorithm, the joint optimization of channel matching and power allocation can be achieved, while ensuring the convergence of the algorithm and maximizing the system throughput. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0036] Figure 1 This is a CR-NOMA system diagram of the CR-NOMA resource allocation solution combining the Hungarian algorithm and the genetic algorithm in Example 1.

[0037] Figure 2 This is a downlink CR-NOMA system resource allocation flow chart of the CR-NOMA resource allocation solution combining the Hungarian algorithm and the genetic algorithm in Example 1.

[0038] Figure 3This is a user matching flowchart of the CR-NOMA resource allocation solution combining the Hungarian algorithm and the genetic algorithm in Example 1.

[0039] Figure 4 This is a Hungarian-adaptive genetic algorithm flow chart of the CR-NOMA resource allocation solution that combines the Hungarian algorithm and the genetic algorithm in Example 1.

[0040] Figure 5 Example 2 is a simulation result diagram of the Hungarian algorithm solving the channel-sub-user matching scheme in the CR-NOMA resource allocation scheme that combines the Hungarian algorithm and the genetic algorithm.

[0041] Figure 6 Example 2 is a simulation result diagram of using an adaptive genetic algorithm to solve a power allocation scheme in a CR-NOMA resource allocation scheme that combines the Hungarian algorithm and the genetic algorithm. DETAILED DESCRIPTION

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0043] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0044] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0045] Example 1

[0046] Reference Figure 1 to Figure 4 , which is the first embodiment of the present invention, and provides a CR-NOMA resource allocation scheme combining the Hungarian algorithm and the genetic algorithm, including:

[0047] S1: Optimization Problem P Built on CR-NOMA Downlink System 0 , and the optimization problem P 0 Decomposed into the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 .

[0048] Specifically, Figure 1As shown in FIG. 1 , the CR-NOMA downlink system includes a primary base station PBS and a cognitive base station SBS; the primary base station PBS includes several primary users PU, where the number of channels is M and the bandwidth of each channel is B. 0 = B / M, the channel set is represented by CH = {c 1 ,c 2 ,...,c M}.

[0049] Furthermore, the cognitive base station SBS adopts an Underlay mode to share the primary user channel and transmits downlink signals to N secondary users SU. The secondary user set is U = {u 1 ,u 2 ,...,u n ,....,u N},u n represents the nth user; the total downlink transmission power of the secondary base station SBS is where p m is the transmit power allocated by the secondary base station to channel m.

[0050] Furthermore, NOMA technology is used to share the same channel m for signal transmission, and N secondary user SUs are divided into Q clusters with K SUs in each group, where the optional power level set of the mth cluster is PT M,K = {Pt 1,1 ,Pt 2,1 ,...,Pt m,k}.

[0051] Specifically, the optional power level set of the mth cluster is defined as PT m ={pt m,1 ,pt m,2 ,...,pt m,k ,...,pt m,K}, where pt m,k =α m,k p m is the kth power level in the mth cluster, α m,k ∈[0,1] is the power level pt m,k The total transmit power p on channel m m The proportion of for And k>l, pt m,k >pt m,l .

[0052] Furthermore, the construction process of the optimization problem includes the following steps: calculating the cognitive base station SBS through the channel m at the power level pt m,k To the secondary user u n The downlink achievable rate of the sending signal is as follows:

[0053] R n,m,k =B 0 log 2 (1+SINR n,m,k );

[0054] In the formula, R n,m,k is the downlink achievable rate of the nth secondary user SU in the kth group on channel m, SINR n,m,k is the received signal-to-interference-noise ratio of the nth secondary user SU in the kth group on the channel m. The specific formula is as follows:

[0055]

[0056] In the formula, h n,m =g n,m PL -ε (d n ) is the gain from the cognitive base station SBS to the nth secondary user SU in the mth channel, g n,m is the Rayleigh fading factor of the nth secondary user SU on channel m, ε is the path loss, PL -ε (d n,m ) is the path loss function, d n,m is the distance from the cognitive base station SBS to the nth secondary user SU in the mth channel, η n,m is the additive Gaussian white noise power from the cognitive base station SBS to the nth secondary user SU in the mth channel, pt m,k =α m,k p m is the kth power level in the mth cluster.

[0057] Furthermore, based on the downlink achievable rate, the total downlink throughput T from the cognitive base station SBS to the secondary user is calculated. sum , the specific formula is as follows:

[0058]

[0059] In the formula, θ n,m is the channel matching relationship variable of the nth secondary user SU on channel m, m∈{1,2,...,M}, n∈{1,..,N}, is the kth power level pt in the mth cluster m,k The downlink power level selection variable of the nth secondary user SU, if θ n,m = 0, then the cognitive base station SBS does not transmit the signal of the nth secondary user SU through channel m. If θ n,m =1, then the cognitive base station SBS transmits the signal of the nth secondary user SU through channel m, if Then the cognitive base station SBS does not select the kth power level pt in the mth cluster m,kThe transmitted signal of the nth secondary user SU, if θ n,m =1, the cognitive base station SBS does not select the kth power level pt in the mth cluster m,k The transmitted signal of the nth secondary user SU.

[0060] Specifically, according to the total downlink throughput, the secondary user throughput maximization is taken as the objective function, C 1 , C 2 , C 3 , C 4 , C 5 , C 6 and C 7 As constraints, establish the optimization problem P 0 , and the optimization problem P 0 Decomposed into the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 .

[0061] It should be noted that C 1 The constraint condition that each secondary user SU can only occupy one subchannel, C 2 As the constraint that each secondary user SU can only select one power level on each channel, C 3 is the proportion constraint of different power levels on each channel, C 4 is the constraint condition for the ordering relationship of the K power levels on each channel, C 5 is the constraint condition of the total transmission power of the cognitive base station SBS, C 6 is the interference constraint condition for the primary user, C 7 is the constraint condition of the minimum transmission rate of the secondary user SU.

[0062] Furthermore, the optimization problem P 0 The relevant formula is as follows:

[0063]

[0064] In the formula, p max is the upper limit of the transmission power, p mask,m is the maximum interference tolerance of the primary user PU on channel m, is the lower limit of the rate of the nth secondary user SU on channel m.

[0065] Furthermore, the secondary user channel matching problem P 1 The relevant formula is as follows:

[0066]

[0067] Specifically, the secondary user power allocation problem P 2 The relevant formula is as follows:

[0068]

[0069] S2: Solve the secondary user channel matching problem P using the Hungarian algorithm 1 , design the secondary user pairing clustering strategy and obtain the optimal channel matching matrix.

[0070] Specifically, it is assumed that the cognitive base station SBS knows the status information of all downlink channels. 1 The solution is as follows: Set the gain h of any channel m to N secondary SUs n,m =g n,m PL -ε (d n ), sorted in descending order, and |h 1,m | 2 >|h 2,m | 2 >...>|h n,m | 2 >...>|h N,m | 2 .

[0071] Furthermore, the Hungarian algorithm is used to select the secondary user set U m = {u 1,m ,u 2,m ...,u n,m ,...,u N,m In the algorithm, two users are selected to pair into user cluster m and power domain NOMA transmission is performed on channel m, and the secondary user pairing clustering and intra-cluster user power level selection strategy are performed.

[0072] It should be noted that the number of secondary users and the number of power levels in each cluster of users, that is, on each channel, are both K=2, and the total number of available channels in the system is M=N / 2. m1 ,u m2} represents the user pair of cluster m carried on channel m, where the cognitive base station SBS to user u m1 ,u m2 The channel attenuation coefficient h m1 ,h m2 Satisfy |h m1 | 2 >|h m2 | 2 Corresponding to the user cluster, the set of optional power levels on channel m is PT m ={pt m1 ,pt m2}, to ensure that the user can successfully detect the downlink signal through SIC, for the weak user u m2 Assign a larger transmit power level ptm2 =α m p m ; For strong user u m1 Allocate a smaller transmit power level pt m1 =(1-α m ) m , α m ∈(0.5,1) represents user u in cluster m m2 The power allocation factor.

[0073] Furthermore, from the secondary user set U of channel m m = {u 1,m ,u 2,m ...,u n,m ,...,u N,m} select the current unpaired minimum gain as the weak user u of channel m m2 , and calculate the achievable rate R of weak user pairing m2,m ; Traverse all channels and obtain the matching weak user u on each channel m2 , and will be added to the set of users that have matched the channel In; from the secondary user set U of channel m m = {u 1,m ,u 2,m ...,u n,m ,...,u N,m} removes users with matching channels and obtains the pruned strong user set

[0074] It should be noted that since the secondary user uses SIC technology to receive the downlink NOMA signal, the weak user u m2 The signal will be demodulated first from the received signal, and the strong user signal paired with it will be regarded as noise.

[0075] Specifically, the achievable rate R m2,m The specific formula is as follows:

[0076]

[0077] In the formula, h m2,m is the channel attenuation coefficient from the cognitive base station SBS to the user, η m2,m For SBS to user u m2 Additive white Gaussian noise power on downlink channel m.

[0078] Further, according to the strong user set Select strong user u m1 , calculate the achievable rate of the optimal strong user and weak user pairing; traverse the strong user set Get the total rate of the user cluster after pairing on the channel n∈{1,2,...,N / 2},m∈{1,2,...,M}, construct the initial channel benefit matrix C M×M =[c n,m ].

[0079] It should be noted that c n,m As the benefit matrix C M×M The element in the mth row and nth column, thus obtaining the initial benefit matrix C of the Hungarian algorithm M×M =[c n,m ].

[0080] Furthermore, the specific formula for the achievable rate of the optimal strong user and weak user pairing is as follows:

[0081]

[0082] In the formula, h n,m is the channel attenuation coefficient from the cognitive base station SBS to the user, η n,m For SBS to user u m2 Additive white Gaussian noise power on downlink channel m.

[0083] Specifically, calculate the strong user u n In any channel c m ∈CH={c 1 ,c 2 ,...,c M} and the weak user u on the channel m2,m Paired into user cluster m, the achievable rate of user cluster m N = 2M is:

[0084]

[0085] In the formula, the first item in the brackets is user u n In channel c m The achievable rate on the channel c is m The weak user u that has been matched m2,m The achievable rate; h n,m , η n,m For SBS Xeon users n The channel attenuation coefficient and the additive white Gaussian noise power on the downlink channel m; h u2,m , η u2,m is the weak user u on the SBS to channel m u2,m The channel attenuation coefficient and the additive white Gaussian noise power on the downlink channel m.

[0086] Furthermore, for the initial channel benefit matrix CM×M =[c n,m ] to optimize and obtain the optimal channel matching matrix; the benefit matrix C = [c n,m ] Perform row simplification and column simplification operations, that is, subtract the minimum number of each row from the minimum number of the row, and subtract the minimum number of each column from the minimum number of the column, to obtain a new benefit matrix Each row and column of the new benefit matrix has at least one 0 element; cover the new benefit matrix with the least horizontal and vertical lines All 0 elements in; Find the new benefit matrix The minimum number not covered by horizontal and vertical lines in New Benefit Matrix All rows not covered by the horizontal and vertical lines in the Operation, add the overwritten column The optimal channel matching matrix is ​​obtained by Check the new benefit matrix Whether the total number of horizontal and vertical lines covering 0 elements in the matrix reaches half of the matrix dimension M, that is, M / 2.

[0087] Furthermore, if the new benefit matrix If the total number of horizontal and vertical lines covering 0 elements in does not reach half of the matrix dimension M, repeat the above steps until the above conditions are met; if the new benefit matrix If the total number of horizontal and vertical lines covering 0 elements in reaches half of the matrix dimension M, the Hungarian algorithm converges and outputs the optimal channel matching matrix That is, the current optimal user-channel matching matrix Ω = [ω n,m ], where ω n,m =1 for strong user u n Multiplex channel c m ,ω n,m =0 for strong user u n Unmultiplexed channel c m If ω n,m =1, user u n That is, strong user u on channel m m1 , and the paired cluster is the weak user u m2 .

[0088] Specifically, according to the optimal channel matching matrix, an adaptive genetic algorithm is used to solve the secondary user power allocation problem P 2 , and get the initial power allocation scheme.

[0089] S3: Based on the optimal channel matching matrix, an adaptive genetic algorithm is used to solve the secondary user power allocation problem P 2 , and obtain the optimal power allocation solution.

[0090] Specifically, the secondary user power allocation problem P 2 The solution is as follows: The cognitive base station SBS randomly generates S groups of power allocation coefficient vectors α with a length of M s =[α s,1 ,α s,2 ,...,α s,m ,...,α s,M ], s∈{1,2,...,S} is used as the initial population, and each vector represents a power allocation scheme; where α s is the chromosome of individual s, M is the number of channels, S is the population size, α s,m is the mth gene of individual s, that is, the secondary user power allocation factor in the mth cluster occupying channel m. To satisfy the optimization problem P 0 The constraints C1, α s,m The initial value is a uniformly distributed random number in the interval (0.5, 1).

[0091] It should be noted that the initial value of the population matrix is The initial value of the number of iterations is g, and the maximum number of iterations is g max .

[0092] Furthermore, the chromosome α of individual s in the population s =[α s,1 ,α s,2 ,...,α s,m ,...,α s,M ], s∈{1,2,...,S}, as the secondary user power allocation coefficient vector of all current M channels, based on the power allocation scheme {u m1 ,u m2} s and the optimal channel matching matrix Calculate the total system throughput as the fitness function f(s) of each species in the population. The specific formula is as follows:

[0093]

[0094] In the formula, In the population individual s, when the power allocation coefficient is α s , the pairing of the nth secondary user SU and channel m, if Then the nth secondary user SU is multiplexed on channel m, if Then the nth secondary user SU is not multiplexed on channel m, is the total rate of the user cluster when the secondary user SU is multiplexed on channel m.

[0095] Furthermore, according to the fitness function f(s), a roulette wheel method is used to select S new generation populations to participate in subsequent mutations, and the individuals with the highest fitness are retained as the new generation population. The specific method for selecting population chromosomes is: the continuous [0,1] interval is regarded as a roulette wheel, and the position interval of individual s∈{1,2,...,S} on the roulette wheel is Repeat the roulette wheel selection S times, each time generating a [0,1] uniformly distributed random number. If the random number falls in the interval In the above example, individual s is selected as an individual in the new population; before the start of this population selection, the chromosome with the largest fitness value in the population is selected as the optimal chromosome and directly used as the last individual in the new population.

[0096] It should be noted that the probability of any individual in the population being selected is equal to its fitness divided by the sum of the fitness of all individuals, that is,

[0097]

[0098] Specifically, the two-point crossover method is performed on the new generation population, and the crossover probability is dynamically adjusted according to the current state of the new generation population; the probability of crossover of each population individual is P cro (s) is calculated as:

[0099]

[0100] In the formula, f max and f avg are the maximum fitness and average fitness of the individuals in the current population, i.e., all the individuals in the population before the crossover occurs. f′(s) is the maximum fitness of the individuals in the two populations that have mutated. k 1 and k 2 is a constant in the interval [0,1].

[0101] It should be noted that the two-point crossover method operates as follows: traverse the parent population, select two parent individuals s and s+1 each time, generate a [0,1] uniformly distributed random number for them, and if the random number is greater than the current crossover probability P cro (s), the two do not perform gene crossover; if the random number is less than the current crossover probability P cro (s), randomly select two genes on their chromosomes for exchange to form two new parent individuals.

[0102] Furthermore, the new generation population after the two-point crossover operation is mutated, and the mutation probability is dynamically adjusted according to the state of the current new generation population; the probability P of all population individuals undergoing gene mutation after gene crossover is calculated. mut (s) is:

[0103]

[0104] In the formula, f max and f avg are the maximum fitness and average fitness of the individuals in the current population, i.e., before the mutation occurs; f(s) is the fitness of the individual s that has mutated; k 3 and k 4 is a constant in the interval [0,1].

[0105] It should be noted that the mutation operation method is as follows: a [0,1] uniformly distributed random number is generated for each parent individual after crossover. If the random number is greater than the gene mutation probability P mut (s), the individual does not perform gene mutation; if the random number is less than the gene mutation probability P mut (s), randomly select a gene point on the chromosome of the individual as the mutation point, randomly generate a normally distributed random number with a mean of 0 and a variance of 0.1 as the mutation amount for the gene here, add the mutation amount to the current value of the mutation point gene, and generate a new offspring individual.

[0106] Furthermore, the above steps are repeated until the number of iterations g of the adaptive genetic algorithm reaches the maximum number of generations g max , select the individual with the highest fitness from the final population as the optimal NOMA power allocation coefficient vector on the channel, and output the optimal power allocation scheme.

[0107] S4: Iterative execution to solve the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 steps until the adaptive genetic algorithm reaches the maximum number of iterations and outputs the optimal channel matching scheme and optimal power allocation scheme for the secondary user.

[0108] Furthermore, this embodiment also provides a CR-NOMA resource allocation system combining the Hungarian algorithm and the genetic algorithm, including: an optimization problem construction and decomposition module for constructing the optimization problem P in the CR-NOMA downlink system 0 , and the optimization problem P 0 Decomposed into the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 ; A secondary user channel matching module, used to solve the secondary user channel matching problem P using the Hungarian algorithm 1 , design a secondary user pairing clustering strategy to obtain an optimal channel matching matrix; a secondary user power allocation module is used to solve the secondary user power allocation problem P based on the optimal channel matching matrix using an adaptive genetic algorithm 2 , get the optimal power allocation solution; output module, used to iteratively execute to solve the secondary user channel matching problem P 1and the secondary user power allocation problem P 2 steps until the adaptive genetic algorithm reaches the maximum number of iterations and outputs the optimal channel matching scheme and optimal power allocation scheme for the secondary user.

[0109] This embodiment also provides a computer device, which is suitable for the CR-NOMA resource allocation scheme that combines the Hungarian algorithm and the genetic algorithm, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the CR-NOMA resource allocation scheme that combines the Hungarian algorithm and the genetic algorithm proposed in the above embodiment.

[0110] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0111] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the following steps are implemented: constructing an optimization problem P in the CR-NOMA downlink system 0 , and the optimization problem P 0 Decomposed into the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 ; Use the Hungarian algorithm to solve the secondary user channel matching problem P 1 , design a secondary user pairing clustering strategy to obtain an optimal channel matching matrix; based on the optimal channel matching matrix, an adaptive genetic algorithm is used to solve the secondary user power allocation problem P 2 , get the optimal power allocation solution; iterative execution solves the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 steps until the adaptive genetic algorithm reaches the maximum number of iterations and outputs the optimal channel matching scheme and optimal power allocation scheme for the secondary user.

[0112] In summary, the present invention solves the optimization problem P 0 Decomposed into the secondary user channel matching problem P 1 and the secondary user power allocation problem P 2 , reducing the complexity of problem solving; using the Hungarian algorithm to solve the secondary user channel matching problem P 1 Compared with the exhaustive method, it significantly reduces the amount of calculation and ensures the optimality of the matching result; an adaptive genetic algorithm is used to solve the secondary user power allocation problem P 2 By dynamically adjusting the crossover and mutation probabilities, the risk of falling into the local optimal solution can be effectively avoided and the global search capability can be improved; by iteratively executing the Hungarian algorithm and the adaptive genetic algorithm, the joint optimization of channel matching and power allocation can be achieved, while ensuring the convergence of the algorithm and maximizing the system throughput.

[0113] Example 2

[0114] Reference Figure 5-6 , which is the second embodiment of the present invention, and this embodiment provides a CR-NOMA resource allocation scheme combining the Hungarian algorithm and the genetic algorithm. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0115] Specifically, Figure 5 As shown in the figure, the parameter settings are: the number of primary users M = 5, the number of secondary users N = 10, and each cluster has 2 users. The available bandwidth B = 50MHz, the noise variance σ 2 =-107dBM / Hz. The power allocation coefficient in each user cluster is α=0.75, and the power of the cognitive base station SBS is [1:1:10]w.

[0116] Further, such as Figure 6 As shown in the figure, the parameter settings are: the number of primary users M = 5, the number of secondary users N = 10, and each cluster has 2 users. The available bandwidth B = 50MHz, the noise variance σ 2 =-107dBM / Hz. Crossover probability P cro (s) is 0.6, and the mutation probability is P mut (s) 0.1; crossover probability dynamic adjustment parameter k 1 =0.4; k 2 =0.6, the mutation probability dynamically adjusts the parameter k 3 =0.9; k 4 =0.1, the power value of cognitive base station SBS is [1:1:10]w.

[0117] Furthermore, observe Figure 5 and Figure 6It can be seen that under the dimension of cognitive base station SBS power, the CR-NOMA system can achieve a higher throughput by adopting the Hungarian channel allocation scheme in the present invention compared with sequential matching and random matching. The Hungarian-Adaptive Genetic Algorithm (H-AGA) proposed in this scheme also improves the system throughput and optimizes the allocation scheme of channel resources in the system compared with the Hungarian-Genetic Algorithm (H-GA).

[0118] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A CR-NOMA resource allocation scheme combining Hungarian algorithm and genetic algorithm, characterized by: include, An optimization problem P0 is constructed in the CR-NOMA downlink system, and the optimization problem P0 is decomposed into a secondary user channel matching problem P1 and a secondary user power allocation problem P2; The Hungarian algorithm is used to solve the secondary user channel matching problem P1, a secondary user pairing clustering strategy is designed, and an optimal channel matching matrix is ​​obtained; Based on the optimal channel matching matrix, an adaptive genetic algorithm is used to solve the secondary user power allocation problem P2 to obtain an optimal power allocation solution; The steps of solving the secondary user channel matching problem P1 and the secondary user power allocation problem P2 are iteratively executed until the adaptive genetic algorithm reaches a maximum number of iterations, and an optimal channel matching solution and an optimal power allocation solution for the secondary user are output.

2. The CR-NOMA resource allocation scheme combining the Hungarian algorithm and the genetic algorithm as claimed in claim 1, characterized in that: The CR-NOMA downlink system includes a primary base station PBS and a cognitive base station SBS; the primary base station PBS includes a number of primary users PU, where the number of channels is M, the channel bandwidth is B0 = B / M, and the channel set is represented by CH = {c1, c2, ..., c M The cognitive base station SBS adopts the Underlay mode to share the channel of the primary user PU and transmits downlink signals to N secondary users SU. The secondary user set is U = {u1,u2...,u n ,...,u N }, the total downlink transmission power is m∈{1,2,...,M}, where u n represents the nth secondary user SU, p m is the transmission power allocated by the cognitive base station SBS to the channel m; the construction process of the optimization problem includes the following steps: NOMA technology is used to share the same channel m for signal transmission, and N secondary user SUs are divided into m clusters with K SUs in each group, where the optional power level set of the mth cluster is PT M,K = {Pt 1,1 ,Pt 2,1 ,...,Pt m,k }; Calculate the cognitive base station SBS through the channel m with power level pt m,k To the secondary user u n The downlink achievable rate of the sending signal is as follows: R n,m,k =B0 log2(1+SINR n,m,k ); In the formula, R n,m,k is the downlink achievable rate of the nth secondary user SU in the kth group on channel m, SINR n,m,k is the received signal-to-interference-noise ratio of the nth secondary user SU in the kth group on the channel m. The specific formula is as follows: In the formula, h n,m =g n,m PL -ε (d n ) is the gain from the cognitive base station SBS to the nth secondary user SU in the mth channel, g n,m is the Rayleigh fading factor of the nth secondary user SU on channel m, ε is the path loss, PL -ε (d n,m ) is the path loss function, d n,m is the distance from the cognitive base station SBS to the nth secondary user SU in the mth channel, η n,m is the additive white Gaussian noise power from the cognitive base station SBS to the nth secondary user SU in the mth channel, pt m,k =α m,k p m is the kth power level in the mth cluster, α m,k ∈[0,1] is the power level pt m,k The total transmit power p on channel m m The proportion of k∈Lab={1,2,...,K}, for l∈Lab and k>l, pt m,k >pt m,l .

3. The CR-NOMA resource allocation scheme combining the Hungarian algorithm and the genetic algorithm as claimed in claim 2, characterized in that: Also includes, Based on the downlink achievable rate, calculate the total downlink throughput T from the cognitive base station SBS to the secondary user sum , the specific formula is as follows: In the formula, θ n,m is the channel matching relationship variable of the nth secondary user SU on channel m, m∈{1,2,...,M}, n∈{1,..,N}, is the kth power level pt in the mth cluster m,k The downlink power level selection variable of the nth secondary user SU, if θ n,m = 0, then the cognitive base station SBS does not transmit the signal of the nth secondary user SU through channel m. If θ n,m =1, then the cognitive base station SBS transmits the signal of the nth secondary user SU through channel m, if Then the cognitive base station SBS does not select the kth power level pt in the mth cluster m,k The transmitted signal of the nth secondary user SU, if θ n,m =1, the cognitive base station SBS does not select the kth power level pt in the mth cluster m,k The transmission signal of the nth secondary user SU; According to the total downlink throughput, taking the maximization of the secondary user throughput as the objective function and C1, C2, C3, C4, C5, C6 and C7 as constraints, an optimization problem P0 is established, and the optimization problem P0 is decomposed into a secondary user channel matching problem P1 and a secondary user power allocation problem P2; Among them, C1 is the constraint condition that each secondary user SU can only occupy one subchannel, C2 is the constraint condition that each secondary user SU can only choose one power level on each channel, C3 is the constraint condition for the proportion of different power levels on each channel, C4 is the constraint condition for the size order relationship of K power levels on each channel, C5 is the constraint condition for the total transmission power of the cognitive base station SBS, C6 is the interference constraint condition for the primary user, and C7 is the constraint condition for the minimum transmission rate of the secondary user SU.

4. The CR-NOMA resource allocation scheme combining the Hungarian algorithm and the genetic algorithm as claimed in claim 3, characterized in that: The relevant formula for the optimization problem P0 is as follows: In the formula, p max is the upper limit of the transmission power, p mask,m is the maximum interference tolerance of the primary user PU on channel m, is the lower limit of the rate of the nth secondary user SU on channel m; The relevant formula for the secondary user channel matching problem P1 is as follows: The relevant formula for the secondary user power allocation problem P2 is as follows:

5. The CR-NOMA resource allocation scheme combining the Hungarian algorithm and the genetic algorithm as claimed in claim 4, characterized in that: The solution to the secondary user channel matching problem P1 is as follows: The gain h of any channel m to N secondary users SU n,m =g n,m PL -ε (d n ), sorted in descending order, and |h 1,m 2 >|h 2,m 2 >...>|h n,m 2 >...>|h N,m 2 ; The Hungarian algorithm is used to select the secondary user set U m = {u 1,m ,u 2,m ...,u n,m ,...,u N,m } selects two users to pair into user cluster m and performs power domain NOMA transmission on channel m, performs secondary user pairing clustering and intra-cluster user power level selection strategy; From the secondary user set U of channel m m = {u 1,m ,u 2,m ...,u n,m ,...,u N,m } select the current unpaired minimum gain as the weak user u of channel m m2 , and calculate the achievable rate R of weak user pairing m2,m ; Traverse all channels and obtain the matching weak user u on each channel m2 , and will be added to the set of users that have matched the channel middle; From the secondary user set U of channel m m = {u 1,m ,u 2,m ...,u n,m ,...,u N,m } removes users with matching channels and obtains the pruned strong user set According to the strong user set Select strong user u m1 , calculate the achievable rate of the optimal strong user and weak user pairing; Traverse the strong user collection Get the total rate of user clusters after pairing on the channel and construct the initial channel benefit matrix C M×M =[c n,m ]; For the initial channel benefit matrix C M×M =[c n,m ] to optimize and obtain the optimal channel matching matrix; According to the optimal channel matching matrix, an adaptive genetic algorithm is used to solve the secondary user power allocation problem P2 to obtain an initial power allocation solution.

6. The CR-NOMA resource allocation scheme combining the Hungarian algorithm and the genetic algorithm as claimed in claim 5, characterized in that: The solution to the secondary user power allocation problem P2 is as follows: The cognitive base station SBS randomly generates a power allocation coefficient vector α of S groups with a length of M s =[α s,1 ,α s,2 ,...,α s,m ,...,α s,M ] as the initial population, where element α s,m represents the power allocation factor in the mth channel, and each vector represents a power allocation scheme in M ​​channels; Based on the power allocation scheme and the optimal channel matching matrix, the total system throughput is calculated as the fitness function f(s) of each species in the population; According to the fitness function f(s), a new generation of population is selected using a roulette wheel method, and the individuals with the highest fitness are retained as the new generation of population; Performing a two-point crossover operation on the new generation population, wherein the crossover probability is dynamically adjusted according to the state of the current new generation population; The new generation population after the two-point crossover operation is mutated, and the mutation probability is dynamically adjusted according to the state of the current new generation population; Repeat the above steps until the number of iterations g of the adaptive genetic algorithm reaches the maximum number of generations g max , select the individual with the highest fitness from the final population as the optimal NOMA power allocation coefficient vector on the channel, and output the optimal power allocation scheme.

7. The CR-NOMA resource allocation scheme combining the Hungarian algorithm and the genetic algorithm as claimed in claim 6, characterized in that: The specific formula of the fitness function f(s) is as follows: in, In the population individual s, when the power allocation coefficient is α s , the pairing of the nth secondary user SU and channel m, if Then the nth secondary user SU is multiplexed on channel m, if Then the nth secondary user SU is not multiplexed on channel m, is the total rate of the user cluster when the secondary user SU is multiplexed on channel m.

8. A CR-NOMA resource allocation scheme combining a Hungarian algorithm and a genetic algorithm, based on the CR-NOMA resource allocation scheme combining a Hungarian algorithm and a genetic algorithm according to any one of claims 1 to 7, characterized in that: Also includes, An optimization problem construction and decomposition module is used to construct an optimization problem P0 in a CR-NOMA downlink system and decompose the optimization problem P0 into a secondary user channel matching problem P1 and a secondary user power allocation problem P2; A secondary user channel matching module is used to solve the secondary user channel matching problem P1 by using the Hungarian algorithm, design a secondary user pairing clustering strategy, and obtain an optimal channel matching matrix; A secondary user power allocation module, configured to solve the secondary user power allocation problem P2 by using an adaptive genetic algorithm based on the optimal channel matching matrix to obtain an optimal power allocation solution; The output module is used to iteratively execute the steps of solving the secondary user channel matching problem P1 and the secondary user power allocation problem P2 until the adaptive genetic algorithm reaches the maximum number of iterations, and output the optimal channel matching solution and the optimal power allocation solution for the secondary user.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the CR-NOMA resource allocation solution combining the Hungarian algorithm and the genetic algorithm as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the CR-NOMA resource allocation solution combining the Hungarian algorithm and the genetic algorithm as described in any one of claims 1 to 7 are implemented.

Images