A Heterogeneous Network Multi-Objective Resource Joint Allocation Method

Abstract

The present invention discloses a method for joint allocation of multi-objective resources in a heterogeneous network. In the environment of a Macro-Femtocell two-tier heterogeneous cellular network, considering two system evaluation indexes of spectral efficiency and relative energy consumption, a multi-objective optimization function is constructed, and the joint allocation result of multi-objective resources is quickly obtained through a quantum political optimization mechanism, so as to maximize the relative energy consumption while ensuring the spectral efficiency. The present invention can solve the problem of joint allocation of multi-objective resources in real life. The designed method has stable performance and can find the optimal multi-objective resource joint allocation scheme in a short time, maximizing the relative energy consumption of the network while ensuring the spectral efficiency, realizing green communication, saving hardware resources and avoiding resource waste.

Description

Technical Field

[0001] The present invention belongs to the field of wireless communication, and relates to a method for jointly allocating multi-objective resources in a heterogeneous network, in particular to a method for jointly allocating multi-objective resources in a heterogeneous network based on a quantum political optimization mechanism. Background Art

[0002] With the rapid development of intelligent devices and wireless mobile applications, the number of users has increased sharply, and there are higher requirements for data transmission rate, energy consumption and quality of service. Network resources such as spectrum and energy have become an important factor affecting system performance. By deploying low-power and low-cost femtocells to jointly form a two-tier heterogeneous network with macrocells, the indoor signal coverage can be effectively improved, thereby improving resource utilization. Due to the shortage of spectrum resources, the Macro-Femtocell two-tier heterogeneous network generally adopts the spectrum sharing method. However, this method will bring serious co-tier interference and cross-tier interference, thus weakening the performance of the network. On the other hand, due to the dense deployment of femtocells, energy consumption has also become an issue that cannot be ignored. How to reduce energy consumption while improving spectrum efficiency is an urgent problem to be solved at present.

[0003] Reasonable resource allocation can effectively reduce the interference problem in the network, and is of great significance for improving spectrum utilization and reducing the energy consumption of data transmission. The joint allocation of spectrum resources and power resources is a commonly used resource allocation method in femtocells. The power allocation problem and the spectrum allocation problem are a continuous optimization problem and an integer programming problem respectively. However, both power and spectrum are limited resources, and both will affect the size of the signal-to-noise ratio and jointly act on the quality of system performance. Therefore, the heterogeneous network resource joint allocation technology has received extensive attention from scholars.

[0004] Through the retrieval of existing literature, Zhang Haibo et al. published "Cluster-based Resource Allocation in Femtocell Two-tier Networks" in Systems Engineering and Electronics (2017, 39(4): 899-904), which proposed a cluster-based resource management scheme. The genetic simulated annealing algorithm was used to cluster femtocells, and the heuristic algorithm and KKT conditions were used to allocate subchannels and power to home base station users respectively, meeting the data rate requirements of users and effectively improving the spectral efficiency. However, this method has a slow convergence speed and a long execution time, and there is great room for improvement. Mohammad Goudarzi et al. published "A fog-driven dynamic resource allocation technique in ultra dense femtocell networks" in Journal of Network and Computer Applications (2019, 145: 102407), which proposed a hierarchical technique of dynamic distributed clustering and fog-driven resource allocation, optimizing the overall network throughput while reducing interference in the network. However, this method has a high computational complexity and does not consider the system energy consumption problem.

[0005] The retrieval results of existing literature show that existing resource joint allocation methods are all single-objective optimization algorithms, with a narrow scope of application and high computational complexity. It is difficult to meet the requirement of reducing network energy consumption while improving spectral efficiency, and cannot well meet the multi-objective requirements in different scenarios. Summary of the Invention

[0006] Aiming at the above-mentioned existing technologies, the technical problem to be solved by the present invention is to provide a multi-objective resource joint allocation method for two-tier heterogeneous networks based on the quantum political optimization mechanism. In the Macro-Femtocell two-tier heterogeneous cellular network environment, considering the two system evaluation indicators of spectral efficiency and relative energy consumption, a multi-objective optimization function is constructed, and the multi-objective resource joint allocation result is quickly obtained through the quantum political optimization mechanism, ensuring spectral efficiency while maximizing relative energy consumption.

[0007] To solve the above technical problems, a multi-objective resource joint allocation method for heterogeneous networks of the present invention includes:

[0008] Step 1, establish a multi-objective resource joint allocation model for the Macro-Femtocell two-tier heterogeneous network;

[0009] Step 2, initialize the quantum members of the quantum political group and set parameters;

[0010] Step 3: Calculate the fitness function values of all the quantum members of the political parties and establish the initial elite solution set;

[0011] Step 4: Select the global optimal quantum member from the elite solution set. For each political party and each constituency, establish the elite member set and the elite constituency set, and select the political party quantum leader and the constituency quantum winner from them;

[0012] Step 5: Calculate the trust values of each political party quantum member according to the fitness function values;

[0013] Step 6: Use the simulated quantum rotation gate to evolve the political party quantum members;

[0014] Step 7: Map the updated quantum members and calculate the fitness function values of the mapped states, and update the elite solution set;

[0015] Step 8: Determine whether the maximum number of iterations is reached. If so, output all the elite solutions; otherwise, let t = t + 1 and return to Step 4 to continue the iteration; according to all the elite solutions in the elite solution set, obtain the multi-objective resource joint allocation scheme.

[0016] Furthermore, the establishment of the multi-objective resource joint allocation model for the Macro-Femtocell two-tier heterogeneous network in Step 1 includes:

[0017] Assume that the Macro-Femtocell two-tier heterogeneous network system consists of a macro base station with a radius of R m and F home base stations with a radius of R f . There are a total of N users in the system competing for the right to use Q sub-channels, where N = N m + F × N f , N m is the number of users of the macro base station, N f is the number of users of each home base station, and the set of home base station labels is The set of sub-channel labels is

[0018] On sub-channel s, the signal-to-interference-plus-noise ratio of user u connected to home base station f is The signal-to-interference-plus-noise ratio obtained by macro base station user m on sub-channel s is where, and p s respectively represent the actual transmission powers allocated when home base station f, neighboring home base station f′, and nearby macro base station occupy sub-channel s, is the channel gain between home base station f and its user u, and g u,s are the channel gains between neighboring home base station f′ and nearby macro base station and user u of home base station f, respectively. is the channel gain between the femtocell f and the macrocell user m. and d u,s represent the path losses from the femtocell f, the neighboring femtocell f′, and the nearby macrocell to the user u under the femtocell f, respectively. d m,s and are the path losses from the macrocell and the femtocell f to the macrocell user m, respectively. σ 2 is the power spectral density of additive white Gaussian noise. Then, according to the Shannon formula, the total system data rate can be expressed as where B represents the total system bandwidth;

[0019] The spectral efficiency of the system is defined as The relative energy consumption of the system is defined as where is a feasible spectrum allocation scheme for N users, is a feasible power allocation scheme for N users. P r and P ∑ represent the rated power and the actual transmission power of the whole system, respectively. They are and P M and P F are the total transmit powers of the macrocell and each femtocell, respectively. p i and represent the th macrocell user and the th user of the femtocell f, respectively;

[0020] The actual transmission powers of all macrocell users in the system shall not exceed the total transmit power of the macrocell, and the actual transmission powers of all users in any femtocell f shall not exceed its total transmit power. The powers of macrocell users and femtocell users shall satisfy their respective minimum and maximum power settings; the multi-objective resource joint allocation problem considering spectral efficiency and relative energy consumption is the maximum optimization equation:

[0021] Furthermore, in step two, initialize the quantum members of the quantum political community and set the parameters including:

[0022] Set the maximum number of iterations T, the iteration number label t, t ∈ [1, T]; assume that there are L1 political parties in the quantum political community, and each political party has L2 quantum members, who are running for election in L3 constituencies. The population size of the quantum political community is L = L1 × L2; during the initialization process, the quantum members in the quantum political community are generated randomly. The jth quantum member of political party i in constituency a in the tth generation can be expressed as Let Among them, is the k-th dimensional component of, and the number of them is 2N. is the j-th quantum member of political party i in constituency a of the (t - 1)-th generation, where i = 1, 2,..., L1, j = 1, 2,..., L2, a = 1, 2,..., L3, k = 1, 2,..., 2N; for the quantum members and mapping can obtain the corresponding mapped states and The first N decision variables of the mapped state represent the spectrum allocation of all users, the (N + 1)-th to the (N + N) m -th decision variables represent the power allocation of macro base station users, and the last segment represents the power allocation of all home base station users. The mapping rules are as follows: Where is the ceiling symbol.

[0023] Furthermore, in step three, calculate the fitness function values of all political party quantum members and establish the initial elite solution set, including:

[0024] For the entire quantum political group, substitute the mapped states of all quantum members into the fitness function for calculation to obtain the fitness function values of all quantum members. During the initialization process, let Where represents the fitness function value of the objective z of the mapped state of the (t - 1)-th generation, represents the fitness function value of the objective z of the mapped state of the t-th generation, where i = 1, 2,..., L1, j = 1, 2,..., L2, a = 1, 2,…, L3, z = 1 or 2; according to the multi-objective function values, obtain the initial elite solution set, where all non-dominated solutions of the initial quantum political group form the elite solution set

[0025] Furthermore, in step four, select the global optimal quantum member from the elite solution set. For each political party and each constituency, establish the elite member set and the elite constituency set, and select the political party quantum leader and the constituency quantum winner, including:

[0026] Select the quantum member with the largest crowding degree from the elite solution set , and call it the global optimal quantum member of the quantum political group. Then the global optimal quantum member of the t-th generation quantum political group is h g (t) = [h g,1 (t), h g,2 (t), …, h g,2N (t)]; all non-dominated solutions of political party i form the elite political party set H i, select the quantum member with the largest crowding degree in the elite political party set H i as the political party quantum leader. The quantum leader of political party i is denoted as i = 1, 2, ..., L1. The set of all political party quantum leaders is All non-dominated solutions in each constituency a form the elite constituency set C a , select the quantum member with the largest crowding degree in the elite constituency set C a as the constituency quantum winner. The quantum winner of constituency a is denoted as a = 1, 2, ..., L3. The set of all constituency quantum winners is

[0027] Furthermore, the calculation method of the crowding degree is as follows: Arrange the fitness function values U z in ascending order. When z = 1 or 2, define the crowding degrees of the quantum members corresponding to the minimum and maximum values of the fitness function of objective z as ∞; the calculation formula for the crowding degrees of other quantum members is: r = 2, 3, ..., o - 1, where and are respectively the fitness function values of objective z of the next mapping state and the previous mapping state of the current mapping state , and are the maximum and minimum values of the fitness function of objective z, and o represents the number of quantum members in the elite solution set elite political party set H i or elite constituency set C a ; perform the above calculations on the crowding degrees corresponding to each objective function of each quantum member, and the sum of each crowding degree component is the final crowding degree value of the quantum member.

[0028] Furthermore, calculating the trust degree value of each political party quantum member in step five includes:

[0029] The trust degree value of the j-th quantum member of political party i in constituency a in the t-th generation is calculated as where and are the fitness function values of the system spectral efficiency and the system relative energy consumption corresponding to the t-th generation mapping state , and are the fitness function values of the system spectral efficiency and the system relative energy consumption corresponding to the (t - 1)-th generation mapping state , and represent the maximum and minimum values of the fitness function of the system spectral efficiency in the t-th generation respectively, and represent the maximum and minimum values of the fitness function of the relative energy consumption of the t-th generation system, respectively.

[0030] Furthermore, the evolution of the political party quantum members using the simulated quantum rotation gate in step six includes:

[0031] Update the k-th dimensional component of the j-th quantum member of political party i in constituency a. The quantum rotation angle and the update equation of the quantum member are respectively:

[0032]

[0033]

[0034] where c1 and c2 are learning factors, is the k-th dimensional quantum rotation angle of the j-th quantum member of political party i in constituency a after update, and h g,k (t) is the k-th dimensional component of the global optimal quantum member, is the k-th dimensional component of the j-th quantum member of political party i in constituency a after update, represents the distance, and its value is as follows:

[0035]

[0036] where, is the trust value of the j-th quantum member of political party i in constituency a in the t-th generation, is the k-th dimensional component of the j-th quantum member of political party i in constituency a in the t-th generation, is the k-th dimensional component of the j-th quantum member of political party i in constituency a in the (t - 1)-th generation, m * (t) first takes the k-th dimensional component of the quantum leader of political party i and then takes the k-th dimensional component of the quantum winner in constituency a

[0037] Furthermore, in step seven, map the updated quantum members, calculate the fitness function value of the mapped state, and update the elite solution set, including:

[0038] Measure all the updated quantum members according to the mapping rule to obtain the corresponding mapped state, calculate the fitness function value of each quantum member, and update the elite solution set The specific judgment method for whether a new solution is eligible to join the elite solution set is as follows: If a quantum member is in a non-dominated state when compared with all the elite solutions in the elite solution set, then it is eligible to join the elite solution set, and the dominated elite solutions in the elite solution set are removed from the elite solution set.

[0039] Advantages of the present invention: The present invention provides a multi-objective resource joint allocation method with low energy consumption and high spectral efficiency for a Macro-Femtocell dual-tier heterogeneous cellular network. Compared with the prior art, the advantages of the present invention are as follows:

[0040] (1) The present invention integrates the continuously optimized power allocation problem and the integer programming-based spectrum allocation problem, solves the joint allocation problem of power resources and spectrum resources. In view of the situation that the existing resource joint allocation methods cannot solve multi-objective problems, a multi-objective resource joint allocation method with a novel quantum political optimization mechanism is designed, which can solve the multi-objective resource joint allocation problem in real life. The designed method has stable performance and can find the optimal multi-objective resource joint allocation scheme in a short time.

[0041] (2) For the problem that existing resource joint allocation methods often increase energy consumption while ensuring spectral efficiency, the multi-objective resource joint allocation method with a quantum political optimization mechanism proposed by the present invention can solve the multi-objective optimization problem, ensure spectral efficiency while maximizing the relative energy consumption of the network, achieve green communication, save hardware resources, and avoid resource waste.

[0042] (3) Compared with the traditional political optimization algorithm, the quantum political optimization method designed by the present invention simplifies the steps of the original political optimization algorithm. During the campaign process, the quantum state evolution rule is adopted, which improves the convergence, reduces the time complexity of the algorithm, and has strong optimization ability. In addition, the evolution rules can be combined and utilized according to different problems to obtain better performance, and it can also be transplanted to other complex engineering problems, with good popularization. Description of the Drawings

[0043] Figure 1 It is a schematic diagram of a multi-objective resource joint allocation method based on a quantum political optimization mechanism.

[0044] Figure 2 It is the non-dominated solutions searched by the quantum political optimization mechanism and the particle swarm optimization mechanism considering both spectral efficiency and relative energy consumption (the number of femto base stations is 5).

[0045] Figure 3 It is the non-dominated solutions searched by the quantum political optimization mechanism and the particle swarm optimization mechanism considering both spectral efficiency and relative energy consumption (the number of femto base stations is 10). Detailed Embodiment

[0046] The present invention will be further described below in conjunction with the specification drawings and embodiments.

[0047] In conjunction with Figure 1 , the present invention includes the following steps:

[0048] Step 1: Establish a multi-objective resource joint allocation model for the Macro-Femtocell two-tier heterogeneous network, specifically including:

[0049] Assume that the Macro-Femtocell two-tier heterogeneous network system consists of a macro base station with a radius of R m and F home base stations with a radius of R f . There are N users in the system competing for the right to use Q sub-channels, where N = N m + F×N f , N m is the number of macro base station users, N f is the number of users of each home base station, and the set of home base station labels is The set of sub-channel labels is

[0050] On sub-channel s, the signal-to-interference-plus-noise ratio of user u connected to home base station f is The signal-to-interference-plus-noise ratio obtained by macro base station user m on sub-channel s is Among them, and p s respectively represent the actual transmission powers allocated when home base station f, neighboring home base station f', and nearby macro base station occupy sub-channel s, is the channel gain between home base station f and its user u, and g u,s are the channel gains between neighboring home base station f' and nearby macro base station and user u of home base station f respectively, is the channel gain between home base station f and macro base station user m, and d u,s respectively represent the path losses from home base station f, neighboring home base station f', and nearby macro base station to user u under home base station f, d m,s and are the path losses from the macro base station and home base station f to macro base station user m respectively, and σ 2 is the power spectral density of additive white Gaussian noise. Then, according to the Shannon formula, the total system data rate can be expressed as Among them, B represents the total system bandwidth.

[0051] The spectral efficiency of the system is defined as The relative energy consumption of the system is defined as Among them, is a feasible spectrum allocation scheme for N users, is a feasible power allocation scheme for N users, P r and P ∑ respectively represent the rated power and actual transmission power of the entire system, respectively and P M and P F are the total transmission powers of the macro base station and each home base station respectively, and respectively represent the actual transmission powers of the i-th macro base station user and the -th user of home base station f.

[0052] The actual transmission powers of all macro base station users in the system shall not exceed the total transmission power of the macro base station, and the actual transmission powers of all users in any home base station f shall not exceed its total transmission power. In addition, the power of macro base station users and the power of home base station users shall respectively satisfy between their set minimum and maximum powers to ensure normal services for all users. In summary, the multi-objective resource joint allocation problem considering both spectral efficiency and relative energy consumption is the following maximum optimization equation:

[0053] Step 2: Initialize the quantum members of the quantum political group and set parameters;

[0054] Set the maximum number of iterations T, the iteration number label t, t ∈ [1, T]. Assume that there are L1 political parties in the quantum political group, and there are L2 quantum members in each political party, competing in L3 constituencies. The population size of the quantum political group is L = L1 × L2. During the initialization process, the quantum members in the quantum political group are generated randomly. The j-th quantum member of political party i in constituency a of the t-th generation can be expressed as Let where, is the k-th component of, and the number of them is 2N, is the j-th quantum member of political party i in constituency a of the (t - 1)-th generation, i = 1, 2,..., L1, j = 1, 2,..., L2, a = 1, 2,..., L3, k = 1, 2,..., 2N. For the quantum members and mapping can be performed to obtain the corresponding mapped states and The first N decision variables of the mapped state represent the spectral allocation of all users, the (N + 1)-th to the (N + N m -th decision variables represent the power allocation of macro base station users, and the last segment represents the power allocation of all home base station users. The mapping rules are as follows: where is the ceiling symbol.

[0055] Step 3: Calculate the fitness function values of all political party quantum members and establish an initial elite solution set;

[0056] For the entire quantum political group, substitute the mapped states of all quantum members into the fitness function for calculation to obtain the fitness function values of all quantum members. During the initialization process, let where represents the fitness function value of the target z of the mapped state in the (t - 1)-th generation , and represents the fitness function value of the target z of the mapped state in the t-th generation . Here, i = 1, 2, ..., L1, j = 1, 2, ..., L2, a = 1, 2, ..., L3, and z = 1 or 2. According to the multi-objective function values, an initial elite solution set is obtained, where the elite solution set consists of all non-dominated solutions of the initial quantum political group

[0057] Step 4: Select the globally optimal quantum member from the elite solution set. For each political party and each constituency, establish an elite member set and an elite constituency set, and select the political party quantum leader and the constituency quantum winner from them;

[0058] Select the quantum member with the largest crowding degree from the elite solution set . This quantum member is called the globally optimal quantum member of the quantum political group. Then, the globally optimal quantum member of the t-th generation quantum political group is h g (t) = [h g,1 (t), h g,2 (t), ..., h g,2N (t)]. All non-dominated solutions of political party i form the elite political party set H i . Select the quantum member with the largest crowding degree in the elite political party set H i as the political party quantum leader. The quantum leader of political party i is denoted as for i = 1, 2, ..., L1. The set of all political party quantum leaders is All non-dominated solutions of each constituency a form the elite constituency set C a . Select the quantum member with the largest crowding degree in the elite constituency set C a as the constituency quantum winner. The quantum winner of constituency a is denoted as for a = 1, 2, …, L3. The set of all constituency quantum winners is

[0059] The calculation method of the crowding degree is as follows: Arrange the fitness function values U z in ascending order, where z = 1 or 2. Define the crowding degrees of the quantum members corresponding to the minimum and maximum fitness function values of the target z as ∞. The calculation formula for the crowding degree of other quantum members is: where and are respectively the current mapped states The fitness function value of the target z of the subsequent mapping state and the fitness function value of the target z of the previous mapping state and are the maximum and minimum values of the fitness function of the target z, and o represents the elite solution set Elite political party set H i or the number of quantum members in the elite constituency set C a For each quantum member, the crowding degree corresponding to each objective function is calculated as described above, and the sum of the crowding degree components is the final crowding degree value of the quantum member.

[0060] Step Five, calculate the trust degree value of each political party quantum member according to the fitness function value.

[0061] The trust degree value of the j-th quantum member of political party i in constituency a in the t-th generation is calculated by the formula where and are the fitness function values of the system spectral efficiency and the system relative energy consumption corresponding to the mapping state in the t-th generation , and are the fitness function values of the system spectral efficiency and the system relative energy consumption corresponding to the mapping state in the (t - 1)-th generation , and represent the maximum and minimum values of the fitness function of the system spectral efficiency in the t-th generation respectively and represent the maximum and minimum values of the fitness function of the system relative energy consumption in the t-th generation respectively.

[0062] Step Six, use the simulated quantum rotation gate to evolve the political party quantum members;

[0063] Update the k-th dimension component of the j-th quantum member of political party i in constituency a. The quantum rotation angle and the update equation of the quantum member are as follows respectively:

[0064]

[0065]

[0066] where c1 and c2 are learning factors is the k-th dimension quantum rotation angle of the updated j-th quantum member of political party i in constituency a, and h g,k (t) is the k-th dimension component of the global optimal quantum member is the k-th dimension component of the updated j-th quantum member of political party i in constituency a represents the distance, and its value is as follows:

[0067]

[0068] Among them, is the trust value of the j-th quantum member of political party i in constituency a in the t-th generation, is the k-th dimensional component of the j-th quantum member of political party i in constituency a in the t-th generation, is the k-th dimensional component of the j-th quantum member of political party i in constituency a in the (t - 1)-th generation, m * (t) First, take the k-th dimensional component of the quantum leader of political party i Then, take the k-th dimensional component of the quantum winner in constituency a

[0069] Step 7: Map the updated quantum members and calculate the fitness function value of the mapped state, and update the elite solution set;

[0070] Measure all the updated quantum members according to the mapping rules to obtain the corresponding mapped states, calculate the fitness function value of each quantum member, and update the elite solution set The specific judgment method for whether a new solution is eligible to join the elite solution set is as follows: If a quantum member is in an undominated state when compared with all the elite solutions in the elite solution set, then it is eligible to join the elite solution set, and the dominated elite solutions in the elite solution set will be removed from the elite solution set.

[0071] Step 8: Determine whether the maximum number of iterations has been reached. If so, output all the elite solutions; otherwise, let t = t + 1 and return to Step 4 to continue the iteration. According to all the elite solutions in the elite solution set, obtain the multi-objective resource joint allocation scheme, and in actual communication, users can select one of the allocation schemes as the final result according to the target requirements.

[0072] In Figure 2 and Figure 3 , the multi-objective resource joint allocation method designed by the present invention based on the quantum political optimization mechanism is denoted as MOQPO, and the resource joint allocation method based on the multi-objective particle swarm optimization mechanism is denoted as MOPSO.

[0073] For the Macro-Femtocell two-tier heterogeneous network model, R m = 500m, B = 10MHz, Q = 40, N m = 45, σ 2 = -174dBm / Hz, P M = 40dBm, P F = 23dBm, The minimum transmit power of both macro base station users and femto base station users is 15 dBm. The maximum transmit power of macro base station users is 23 dBm, and the maximum transmit power of femto base station users is 20 dBm, where 1 dBm = 10 0.1 mW. The settings of channel gain and path loss refer to the reference document "Guidelines for evaluation of radio transmission technologies for IMT-2000" (International Telecommunication Union, 1997: 44 - 46). The parameter settings for the quantum political optimization mechanism are as follows: the population size of the quantum political group is 100, T = 500, L1 = L2 = L3 = 10, c1 = 0.8, c2 = 0.3. To facilitate the comparison of the performance of the proposed multi-objective resource joint allocation method based on the quantum political optimization mechanism with the existing multi-objective particle swarm method, the multi-objective particle swarm method is applied to the multi-objective resource joint allocation problem, Figure 2 and Figure 3 the two methods are compared. The population size and the number of termination iterations of the multi-objective particle swarm method are the same as those of the quantum political optimization method. The settings of other parameters of the multi-objective particle swarm optimization mechanism refer to the reference document "Multi-objective particle swarm algorithm based on tree-structured unbounded archive" (Control and Decision, 2020, 35(11): 2675 - 2686).

[0074] Figure 2 The non-dominated solutions searched by the quantum political optimization mechanism and the multi-objective particle swarm optimization mechanism considering both spectral efficiency and relative energy consumption at the same time. There are 5 femto base stations with a radius of 50 m deployed in the network. The number of macro base station users is 45, and the number of users per femto base station is 5.

[0075] Figure 3 The non-dominated solutions searched by the quantum political optimization mechanism and the multi-objective particle swarm optimization mechanism considering both spectral efficiency and relative energy consumption at the same time. There are 20 femto base stations with a radius of 30 m deployed in the network. The number of macro base station users is 25, and the number of users per femto base station is 2.

[0076] From Figure 2 and Figure 3 it can be seen that in the sparse and dense deployment of the Macro-Femtocell two-tier heterogeneous network, due to the complexity of the problem, the number of non-inferior solutions obtained by either the quantum political optimization mechanism or the multi-objective particle swarm optimization mechanism is small. However, the quantum optimization mechanism designed in the present invention can obtain better non-dominated solutions, which shows the effectiveness of the proposed multi-objective resource joint allocation method based on the quantum political optimization mechanism, which can reduce energy consumption while ensuring spectral efficiency, realizing green communication, energy conservation and environmental protection.

Claims

1. A method for joint allocation of multi-objective resources in a heterogeneous network, characterized in that, Including: Step 1: Establish a multi-objective resource joint allocation model for the Macro-Femtocell two-tier heterogeneous network; The multi-objective resource joint allocation problem considering both spectral efficiency and relative energy consumption is a maximum optimization equation: where, U1 is the spectral efficiency of the system; U2 is the relative energy consumption of the system; Step 2: Initialize the quantum members of the quantum political group and set parameters; Step 3: Calculate the fitness function values of all political party quantum members and establish an initial elite solution set; Step 4: Select the global optimal quantum member from the elite solution set. For each political party and each constituency, establish an elite member set and an elite constituency set, and select the political party quantum leader and the constituency quantum winner from them; Step 5: Calculate the trust degree values of each political party quantum member according to the fitness function values; Step 6: Use the simulated quantum rotation gate to evolve the political party quantum members; Update the k-th dimensional component of the j-th quantum member of political party i in constituency a. Its quantum rotation angle and the update equation of the quantum member are respectively: where, c1 and c2 are learning factors, is the k-th dimensional quantum rotation angle of the j-th quantum member of Party i in the updated constituency a, h g,k (t) is the k-th dimensional component of the global optimal quantum member, is the k-th dimensional component of the j-th quantum member of Party i in the updated constituency a, represents the distance, and its value is as follows: Among them, is the trust value of the j-th quantum member of political party i in constituency a in the t-th generation, is the k-th dimensional component of the j-th quantum member of political party i in constituency a in the t-th generation, is the k-th dimensional component of the j-th quantum member of political party i in constituency a in the (t - 1)-th generation, m * (t) First, take the k-th dimensional component of the quantum leader of political party i Then, take the k-th dimensional component of the quantum winner in constituency a Step 7: Map the updated quantum members and calculate the fitness function values of the mapped states, and update the elite solution set; Measure the corresponding mapped states for all updated quantum members according to the mapping rules, calculate the fitness function values of each quantum member, and update the elite solution set The specific judgment method for whether a new solution is eligible to join the elite solution set is as follows: If a quantum member is in a non-dominated state when compared with all elite solutions in the elite solution set, it is eligible to join the elite solution set, and the dominated elite solutions in the elite solution set are removed from the elite solution set; Step 8: Determine whether the maximum number of iterations is reached. If so, output all the elite solutions; otherwise, let t = t + 1 and return to Step 4 to continue the iteration; according to all the elite solutions in the elite solution set, obtain the multi-objective resource joint allocation scheme.

2. The heterogeneous network multi-objective resource joint allocation method according to claim 1, wherein: The establishment of the multi-objective resource joint allocation model for the Macro-Femtocell two-tier heterogeneous network described in Step 1 includes: Suppose the Macro-Femtocell two-tier heterogeneous network system consists of a macro base station with a radius of R m and F femto base stations with a radius of R f . There are N users in the system competing for the right to use Q sub-channels, where N = N m + F × N f , N m is the number of users of the macro base station, N f is the number of users of each femto base station, and the set of femto base station labels is The set of sub-channel labels is On sub-channel s, the signal-to-interference-plus-noise ratio (SINR) of user u connected to home base station f is The SINR obtained by macro base station user m on sub-channel s is Where and p s respectively represent the actual transmission powers allocated when home base station f, neighboring home base station f′, and nearby macro base station occupy sub-channel s. is the channel gain between home base station f and its user u. and g u,s are the channel gains between neighboring home base station f′ and nearby macro base station and user u of home base station f respectively. is the channel gain between home base station f and macro base station user m. and d u,s respectively represent the path losses from home base station f, neighboring home base station f′, and nearby macro base station to user u under home base station f. d m,s and are the path losses from the macro base station and home base station f to macro base station user m respectively. σ 2 is the power spectral density of additive white Gaussian noise. Then, according to the Shannon formula, the total system data rate can be expressed as Where B represents the total system bandwidth; The spectral efficiency of the system is defined as The relative energy consumption of the system is defined as where is a feasible spectrum allocation scheme for N users, is a feasible power allocation scheme for N users, P r and P Σ represent the rated power and the actual transmission power of the whole system respectively, which are and P M and P F are the total transmit powers of the macro base station and each home base station respectively, and represent the actual transmission powers of the th macro base station user and the th user of home base station f respectively; The actual transmission power of all macro base station users in the system shall not exceed the total transmission power of the macro base station. The actual transmission power of all users in any home base station f shall not exceed its total transmission power. The power of macro base station users and the power of home base station users shall respectively satisfy between their set minimum and maximum powers.

3. A heterogeneous network multi-objective resource joint allocation method according to claim 1, characterized in that: The initialization of the quantum members of the quantum political group and the setting of parameters described in Step 2 include: Set the maximum number of iterations \(T\), the iteration number label \(t\), where \(t\in[1,T]\); assume that there are \(L_1\) political parties in the quantum political group, and each political party has \(L_2\) quantum members, who are running for office in \(L_3\) constituencies respectively. The group size of the quantum political group is \(L = L_1\times L_2\); during the initialization process, the quantum members in the quantum political group are generated randomly. The \(j\)-th quantum member of political party \(i\) in constituency \(a\) of the \(t\)-th generation can be expressed as Let where is the \(k\)-th dimensional component of, and the number of them is \(2N\). is the \(j\)-th quantum member of political party \(i\) in constituency \(a\) of the \((t - 1)\)-th generation, \(i = 1,2,\cdots,L_1\), \(j = 1,2,\cdots,L_2\), \(a = 1,2,\cdots,L_3\), \(k = 1,2,\cdots,2N\); map the quantum members and to obtain the corresponding mapped states and The first \(N\) decision variables of the mapped state represent the spectrum allocation of all users, the \((N + 1)\)-th to the \((N+N)\) m th decision variables represent the power allocation of macro base station users, and the last segment represents the power allocation of all femto base station users. The mapping rules are as follows: where is the ceiling symbol.

4. A heterogeneous network multi-objective resource joint allocation method according to claim 1, characterized in that: The calculation of the fitness function values of all political party quantum members and the establishment of the initial elite solution set described in Step 3 include: For the entire quantum political group, substitute the mapping states of all quantum members into the fitness function for calculation to obtain the fitness function values ​​of all quantum members. During the initialization process, let in, Represents the (t-1)th generation mapping state The fitness function value of the target z, Represents the tth generation mapping state The fitness function value of the target z, i = 1, 2, ..., L1, j = 1, 2, ..., L2, a = 1, 2, ..., L3, z = 1 or 2; according to the multi-objective function value, the initial elite solution set is obtained, where all non-dominated solutions of the initial quantum political group constitute the elite solution set 5. A heterogeneous network multi-objective resource joint allocation method according to claim 1, characterized in that: The selection of the global optimal quantum member from the elite solution set, the establishment of the elite member set and the elite constituency set for each political party and each constituency, and the selection of the political party quantum leader and the constituency quantum winner from them described in Step 4 include: Select from the elite solution set the quantum member with the largest crowding degree, and call it the global optimal quantum member of the quantum political group. Then the global optimal quantum member of the t-th generation quantum political group is h g (t) = [h g,1 (t), h g,2 (t),..., h g,2N (t)]; All non-dominated solutions of political party i form the elite political party set H i , and select the quantum member with the largest crowding degree in the elite political party set H i as the quantum leader of the political party. The quantum leader of political party i is denoted as i = 1, 2,..., L1, and the set of all quantum leaders of political parties is All non-dominated solutions of each constituency a form the elite constituency set C a , and select the quantum member with the largest crowding degree in the elite constituency set C a as the quantum winner of the constituency. The quantum winner of constituency a is denoted as a = 1, 2,..., L3, and the set of all quantum winners of constituencies is 6. A heterogeneous network multi-objective resource joint allocation method according to claim 5, characterized in that: The calculation method of the crowding degree is as follows: the fitness function value U z is sorted in ascending order, z = 1 or 2, and the crowding degrees of the quantum members of the minimum and maximum values of the fitness function of the target z are defined as ∞; The calculation formula for the crowding degree of other quantum members is as follows: r = 2, 3,..., o - 1, where and are the fitness function values of the target z of the next mapped state and the fitness function value of the target z of the previous mapped state of the current mapped state respectively, and are the maximum and minimum values of the fitness function of the target z, and o represents the elite solution set the elite political party set H i or the elite constituency set C a the number of quantum members in; the above calculations are performed on the crowding degrees corresponding to the respective objective functions of each quantum member, and the sum of the respective crowding degree components is the final crowding degree value of the quantum member.

7. A heterogeneous network multi-objective resource joint allocation method according to claim 1, characterized in that: The calculation of the trust degree values of each political party quantum member described in Step 5 includes: The trust value of the j-th quantum member of political party i in constituency a of the t-th generation is calculated by the formula where and are the fitness function values of the system spectral efficiency and the system relative energy consumption corresponding to the mapping state of the t-th generation , and are the fitness function values of the system spectral efficiency and the system relative energy consumption corresponding to the mapping state of the (t - 1)-th generation , and represent the maximum and minimum values of the fitness function of the system spectral efficiency of the t-th generation respectively and represent the maximum and minimum values of the fitness function of the system relative energy consumption of the t-th generation respectively

Images