A method for spectrum sharing in integrated satellite-terrestrial networks based on joint power and frequency allocation

Abstract

The application discloses a kind of based on joint power and frequency allocation integrated satellite-ground network spectrum sharing method, it belongs to integrated satellite-ground network spectrum sharing field.The present application solves the problem of serious co-channel interference when integrated satellite-ground spectrum sharing, and the high computational complexity of existing spectrum sharing method.The method of the present application is specifically as follows: step one, constructing integrated satellite-ground network spectrum sharing model;Step two, based on the integrated satellite-ground network spectrum sharing model constructed, integrated satellite-ground network resource optimization allocation problem is constructed;Step three, the problem constructed in step two is decoupled into power allocation subproblem and frequency resource scheduling subproblem;On the interference coordination unit, according to the power allocation subproblem, the power allocation result is calculated and returned to the base station;The base station calculates the number of resource blocks allocated to each ground user according to the returned power allocation result and frequency resource scheduling subproblem, and obtains the resource block allocation result.The method of the present application can be applied to integrated satellite-ground network spectrum sharing.

Description

Technical Field

[0001] This invention belongs to the field of spectrum sharing in integrated space-ground networks, and specifically relates to a method for spectrum sharing in integrated space-ground networks. Background Technology

[0002] The L, S, and C bands, which can be used independently to achieve global coverage, are almost entirely allocated. Due to limitations imposed by terminal antennas, link budgets, and ITU regulations, bands such as Ku, Ka, and Q / V are difficult to use for direct satellite-to-mobile phone connections, while bands at 2 GHz and below can support this. However, only a small amount of spectrum in the L and S bands is available for satellite systems, with the majority allocated to terrestrial mobile communications. Currently, satellite and terrestrial mobile communications allocate frequency bands exclusively. To achieve direct satellite-to-mobile broadband communication, existing rules need to be broken, allowing terrestrial networks to be deployed in satellite bands and satellite systems to be deployed in terrestrial mobile communication bands. 6G requires more mid-to-low frequency bands to increase deployment capacity and coverage at low cost. Considering the wide coverage and low spectrum utilization of satellites, sharing spectrum with terrestrial networks can alleviate the resource shortage problem.

[0003] Satellite networks can supplement the performance of terrestrial networks and provide access anytime, anywhere. However, communication resources are limited, and the explosive growth of data traffic from user devices puts pressure on service quality. With the deepening of satellite-terrestrial integration, co-channel interference caused by spectrum sharing between satellite and terrestrial networks has become a serious problem. Although increasing the protected area and limiting the number of terminals and transmission power can reduce interference, increasing the protected area reduces the available frequency resources of the network, and limiting the number of terminals or transmission power also affects its capacity. Integrated satellite-terrestrial networks require joint management of frequency, time, and spatial resources to meet the service quality requirements of various users and traffic. Furthermore, terrestrial networks typically implement resource allocation algorithms for one or a few base stations within a small area. However, satellite coverage radii range from tens to thousands of kilometers, making the joint resource management of a large number of base stations and satellites extremely complex. In integrated satellite-terrestrial networks, distributed resource management strategies are more efficient than centralized resource management across the entire network.

[0004] Integrated space-ground networks can more fully utilize power and spectrum resources through joint resource allocation, achieving higher capacity and more efficient bandwidth utilization. Research on related spectrum sharing solutions has been conducted both domestically and internationally. This involves constructing objective optimization models to maximize user throughput while using user service quality as a constraint for joint optimization. Time-frequency resource allocation algorithms employed in integrated space-ground networks can handle dynamically changing user numbers and timeliness requirements. However, existing spectrum sharing methods have high computational complexity. Integrated space-ground networks require more flexible spectrum sharing, necessitating research into resource allocation strategies with lower computational complexity. Summary of the Invention

[0005] The purpose of this invention is to address the severe co-channel interference and high computational complexity of existing spectrum sharing methods in satellite-ground integrated network spectrum sharing, and to propose a spectrum sharing method based on joint power and frequency allocation.

[0006] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows:

[0007] A spectrum sharing method for a space-ground integrated network based on joint power and frequency allocation, the method specifically includes the following steps:

[0008] Step 1: Construct a spectrum sharing model for an integrated satellite-ground network. The model includes satellites, satellite terminals, base stations, ground users, and interference coordination units, with the uplink of ground users sharing frequencies with the downlink of satellites.

[0009] Step 2: Based on the spectrum sharing model of the space-ground integrated network constructed in Step 1, construct the resource optimization and allocation problem of the space-ground integrated network.

[0010] Step 3: Decouple the problem constructed in Step 2 into a power allocation sub-problem and a frequency resource scheduling sub-problem;

[0011] In the interference coordination unit, the power allocation result is calculated based on the power allocation subproblem and returned to the base station;

[0012] The base station calculates the number of resource blocks allocated to each ground user based on the returned power allocation results and the frequency resource scheduling sub-problem, thus obtaining the resource block allocation result.

[0013] The beneficial effects of this invention are:

[0014] This invention addresses the channel uncertainty and low frequency utilization issues of interference links in integrated satellite-ground spectrum-sharing networks. It designs a joint power and frequency resource allocation algorithm considering interference protection constraints. To meet the interference protection requirements of satellite terminals, the base station controls the transmit power and frequency resources of ground users. The optimization problem aims to maximize the throughput of the ground network, constrained by interference protection for satellite terminals and quality of service for ground users. The optimization problem is decomposed into power allocation and frequency resource scheduling subproblems. In the power allocation subproblem, a power allocation algorithm based on interference probability constraints is used to consider channel uncertainty, and a suboptimal solution for power allocation is obtained through iterative optimization. In the frequency resource scheduling subproblem, an improved heuristic algorithm is used to handle the non-convexity of the problem while reducing the algorithm's time complexity. Attached Figure Description

[0015] Figure 1This is a schematic diagram of the space-ground integrated network spectrum sharing model in an embodiment of the present invention;

[0016] Figure 2 This is a schematic diagram of the resource management architecture of the space-ground integrated network in an embodiment of the present invention;

[0017] Figure 3 This is a graph showing the relationship between the maximum transmit power of a ground user and the channel gain ratio at different base station and satellite terminal distances in this embodiment of the invention.

[0018] Figure 4 This is a graph showing the relationship between the interference power received by the satellite terminal and the channel gain ratio at different distances between the base station and the satellite terminal in this embodiment of the invention.

[0019] Figure 5 This is a graph showing the relationship between the maximum transmit power of a ground user and the channel gain ratio under different power allocation algorithms in this embodiment of the invention.

[0020] Figure 6 This is a graph showing the relationship between the interference power received by the satellite terminal and the channel gain ratio under different power allocation algorithms in this embodiment of the invention.

[0021] Figure 7 This is a graph showing the relationship between terrestrial network throughput and the number of users in a cellular cell in an embodiment of the present invention.

[0022] In the figure, for each number of ground users, the four corresponding bar charts from left to right are, respectively, the interior point method, the power allocation algorithm based on interference probability constraints (i.e., the method of this invention), the power allocation algorithm based on compact constraint constraints, and the frequency exclusive access algorithm.

[0023] Figure 8 This is a graph showing the relationship between terrestrial network throughput and the distance between the base station and the satellite terminal in an embodiment of the present invention.

[0024] In the figure, for each distance, the four corresponding bar charts from left to right are, respectively, the interior point method, the power allocation algorithm based on interference probability constraints (i.e., the method of this invention), the power allocation algorithm based on compact constraint constraints, and the frequency exclusive access algorithm. Detailed Implementation

[0025] The present application will now be described in further detail with reference to specific embodiments and accompanying drawings. Obviously, the described embodiments are merely a part of the embodiments of the present invention, and not all of them. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are all within the scope of protection of the present invention.

[0026] Specific Implementation Method 1: The spectrum sharing method for integrated space-ground networks based on joint power and frequency allocation described in this implementation method specifically includes the following steps:

[0027] Step 1: Construct a spectrum sharing model for an integrated satellite-ground network. The model includes satellites, satellite terminals, base stations, ground users, and interference coordination units, with the uplink of ground users sharing frequencies with the downlink of satellites.

[0028] Step 2: Based on the spectrum sharing model of the space-ground integrated network constructed in Step 1, construct the resource optimization and allocation problem of the space-ground integrated network.

[0029] Step 3: Decouple the problem constructed in Step 2 into a power allocation sub-problem and a frequency resource scheduling sub-problem;

[0030] In the interference coordination unit, the power allocation result is calculated based on the power allocation subproblem and returned to the base station;

[0031] The base station calculates the number of resource blocks allocated to each ground user based on the returned power allocation results and the frequency resource scheduling sub-problem, thus obtaining the resource block allocation result.

[0032] Considering the interference protection constraints of satellite terminals and the uncertainty of channel state information in ground interference links, this study aims to maximize the throughput of the ground network while ensuring the quality of service for ground users. Unlike existing methods that only consider resource allocation with perfect channel state information, this study introduces a probabilistic model for aggregated interference analysis to ensure the stable operation of satellite terminals.

[0033] Specific Implementation Method Two: This implementation method differs from Specific Implementation Method One in that n are deployed around the satellite terminal. bs There are several base stations sharing a frequency, and the number of ground users in the j-th base station is... j = 1, 2, ..., n bs Base stations are used to allocate time and frequency resource blocks to ground users.

[0034] The other steps and parameters are the same as in Specific Implementation Method 1.

[0035] Specific Implementation Method 3: This implementation method differs from Specific Implementation Method 1 or 2 in that the satellite downlink adopts orthogonal frequency division multiple access, and the ground user uplink adopts single-carrier frequency division multiple access.

[0036] Other steps and parameters are the same as in specific implementation method one or two.

[0037] Specific Implementation Method Four: This implementation method differs from Specific Implementation Methods One to Three in that the specific process of step two is as follows:

[0038] Step 21: The transmission power of the ground user is:

[0039]

[0040] Among them, P i m P represents the maximum transmit power allowed for the i-th terrestrial user in the shared frequency band. i max Pi represents the maximum transmit power of the i-th ground user, P0 represents the nominal receive power of the base station, and Mi represents the maximum transmit power of the i-th ground user. i This represents the number of resource blocks allocated to the i-th ground user, where α represents the path loss compensation factor, 0 < α ≤ 1. This represents the channel gain from the i-th ground user to the base station (here, the base station is the one associated with user i). This represents the antenna gain from the i-th ground user to the base station. Let Δ represent the path loss from the i-th ground user to the base station. i This represents the modulation compensation factor for the i-th ground user;

[0041] Step 22: The aggregated interference power constraint of the ground user uplink to the satellite terminal is as follows:

[0042]

[0043] Where, n r n represents the set of resource blocks that overlap with the protected bandwidth. ue P represents the total number of ground users. i This represents the transmit power of the i-th ground user. This represents the channel gain from the i-th ground user to the satellite terminal on resource block k. This represents the receiving antenna gain from the i-th ground user to the satellite terminal. Z represents the path loss from the i-th ground user to the satellite terminal. ik Let Z represent the channel fading variable for the i-th ground user on resource block k, and Z ik Follows the mean m z The standard deviation is σ z Log-normal distribution (mean m) z Equal to 0dB, standard deviation σ z (equal to 10dB), δ ik δ represents the resource block indicator variable. ik When δ equals 1, it means that resource block k is allocated to the i-th ground user. ik When ε equals 0, it means that resource block k is not allocated to the i-th ground user. s This indicates the interference protection threshold for the satellite terminal;

[0044] Steps two and three: Since the random variable Z cannot be directly measured ik Therefore, using a statistical model, the constraints in step two can be expressed as probabilistic constraints that the aggregation interference needs to satisfy:

[0045]

[0046] Where, ρ s The probability threshold representing aggregation interference. Expressing the request The probability of;

[0047] Step 24: With the goal of maximizing the uplink throughput of ground users, the resource optimization allocation problem is constructed as follows:

[0048]

[0049] stδ ijk ∈{0,1};

[0050]

[0051]

[0052]

[0053]

[0054] Among them, U ij The system utility function (similar in form to formula (6)) represents n. bs U represents the total number of base stations. ij The value depends on the power allocation matrix. and resource block allocation matrix P ijk δ is the power of the k-th resource block allocated by the j-th base station to the i-th ground user. ijk When δ equals 1, it means that resource block k is allocated to the i-th ground user of the j-th base station, δ ijk When n equals 0, it means that resource block k will not be allocated to the i-th ground user of the j-th base station. ue This represents the total number of ground users in the terrestrial network, i.e., satisfying... M is the total number of ground users in the j-th base station. ij This represents the number of resource blocks allocated by the j-th base station to the i-th ground user. Z represents the channel gain from the i-th ground user at the j-th base station to the satellite terminal on resource block k. ijkThis represents the channel fading variable of the i-th ground user at the j-th base station on resource block k.

[0055] In Equation (4), constraints 1-3 indicate that resource blocks are allocated to each user in a continuous manner, constraint 2 indicates that a resource block can only be allocated to one user, and constraint 3 indicates that resource blocks allocated to a specific user must be adjacent. Constraint 4 indicates that the user's transmit power is limited to a feasible range. The optimization problem couples the frequency scheduling problems of multiple base stations through interference protection constraints, greatly increasing the computational complexity.

[0056] The other steps and parameters are the same as those in one of the specific implementation methods one to three.

[0057] Specific Implementation Method Five: This implementation method differs from Specific Implementation Methods One to Four in that the power allocation sub-problem is:

[0058]

[0059] Where U(P) represents the objective function of the power allocation subproblem. Z represents the channel gain from the i-th ground user to the satellite terminal. i Represents the channel fading variable for the i-th ground user;

[0060] The objective function U(P) is:

[0061]

[0062] Where N represents the noise power at the base station receiver.

[0063] The other steps and parameters are the same as those in one of the specific implementation methods one to four.

[0064] To clarify the relationship between power allocation and path loss, we correlate the user's transmit power with path loss. Assuming that we do not consider the uncertainty of the channel state, power allocation can be represented as a deterministic problem:

[0065]

[0066] Using the Karush-Kuhn-Tucker conditions, the transmit power of user i satisfies the following:

[0067]

[0068] Where μ represents the multiplier, and the superscript "+" indicates: if If it is a positive value, then like If it is a negative value, then

[0069] The first term of minimizing equation (8) is considered as the solution to the water-filling algorithm, and the second term limits the user's transmit power. The channel gain ratio of user i is defined as follows: The base station is deployed at a location far from the satellite terminal, and the path loss meets the requirements. This indicates that user i has a larger channel gain ratio. According to formula (8), users with larger channel gain ratios tend to be assigned equal power levels.

[0070] Based on the relationship between transmit power and channel gain ratio, a power allocation algorithm based on interference probability constraints is proposed in Implementation Method Six. It is assumed that the interference power caused by users with higher transmit power tends to a fixed value, and a step function is used to approximate the equal interference power.

[0071] Specific Implementation Method Six: This implementation method differs from Specific Implementation Methods One to Five in that, in the interference coordination unit, the power allocation result is calculated based on the power allocation sub-problem and returned to the base station; specifically:

[0072] Step 1: Initialize the number of ground users with non-zero transmit power n=1, initialize the throughput of ground users u1=0, and initialize the maximum throughput that has been saved u2=0;

[0073] Step 2: Denote the channel gain ratio of the i-th ground user as... according to Sort users in each region in descending order;

[0074] If u2≥u1 and n<n ue Then calculate the interference power level ξ(n) of the top n ground users and continue to step 3;

[0075] Otherwise, the transmit power of each ground user obtained in the last iteration will be used as the final power allocation result;

[0076] Step 3: Calculate the transmit power of each ground user based on the interference power level ξ(n):

[0077]

[0078] P 1,i =0, i=n+1,n+2,…,n ue

[0079] Step 4: Update the ground user throughput based on the transmit power of each ground user. The updated ground user throughput is as follows:

[0080]

[0081] Step 5: Determine whether the updated ground user throughput is greater than u2;

[0082] If the updated ground user throughput is greater than u2, then set u2 equal to the updated ground user throughput, and set n = n + 1. Use the updated u1 and u2 to return to step 2.

[0083] If the updated ground user throughput is less than or equal to u2, then let n = n + 1, and use the updated u1 to return to step 2 (at this time, do not update u2, and use the most recently updated u2 to execute step 2).

[0084] The other steps and parameters are the same as those in one of the specific implementation methods one to five.

[0085] Specific Implementation Method Seven: This implementation method differs from Specific Implementation Methods One through Six in that the channel gain ratio... for:

[0086]

[0087] The other steps and parameters are the same as those in one of the specific implementation methods one to six.

[0088] Specific Implementation Method Eight: This implementation method differs from Specific Implementation Methods One through Seven in that the calculation of the interference power level ξ(n) of the top n ground users is specifically as follows:

[0089] The constraints in formula (5) are transformed into:

[0090]

[0091] According to the Fenton-Wilkinson (FW) approximation, we obtain e is the base of the natural logarithm, and the variable x i Follow the mean variance is The normal distribution, i.e. The variable y follows a mean of m y variance is The normal distribution, i.e. m y β2lnw1-lnw2 / 2, r ij Let m represent the correlation coefficient, then the mean is m. y and variance They are respectively:

[0092] The probability constraint of formula (9) is then transformed into:

[0093]

[0094] Where Q is the right-tail function of the standard normal distribution; the base of log is 10.

[0095] The expression for obtaining the interference power level ξ(n) is:

[0096]

[0097] The superscript "-1" indicates taking the reciprocal.

[0098] The other steps and parameters are the same as those in any of the specific implementation methods one to seven.

[0099] Specific Implementation Method Nine: This implementation method differs from Specific Implementation Methods One through Eight in that the frequency resource scheduling sub-problem is:

[0100] For each base station, establish the following frequency resource scheduling sub-problem, taking the j-th base station as an example:

[0101]

[0102] Among them, U i (P ik ,δ ik ) is the objective function of the frequency resource scheduling subproblem (similar in function form to formula (6)). P is the number of users within base station j. ik P is the power of the k-th resource block used by the i-th ground user. i lim It is the power allocation result for the i-th ground user returned by the base station.

[0103] The other steps and parameters are the same as those in one of the specific implementation methods one to eight.

[0104] Specific Implementation Method Ten: This implementation method differs from Specific Implementation Methods One to Nine in that the base station calculates the number of resource blocks allocated to each ground user based on the returned power allocation results and the frequency resource scheduling sub-problem, specifically as follows:

[0105] Step (1), Definition This is a measure of the amount of information transmitted by user i on resource block k. B represents the bandwidth of the resource block, i = 1, 2, ..., n ue k = 1, 2, ..., n r ;

[0106] Step (2), The k-th column of the metric matrix Λ, where k = 1, 2, ..., n r Then we obtain the metric matrix Λ;

[0107] Initialize c = 1;

[0108] Step (3): Determine whether the set R of unallocated resource blocks is empty (the set R is an empty set containing all resource blocks during initialization);

[0109] If the set of unallocated resource blocks R is not empty, then proceed to step (4);

[0110] If the set of unallocated resource blocks R is empty, then the resource block allocation is completed;

[0111] Step (4): Based on the metric matrix Λ, select the resource block k with the c-th largest metric value from the set of unallocated resource blocks R, and represent the c-th largest metric value as...

[0112] Step (5): Let Γ be the set of resource blocks that have been allocated to user i;

[0113] If k is adjacent to a resource block in Γ or Then, the resource block k with the c-th largest metric value is assigned to user i, and the metric value related to resource block k is removed from the metric matrix to obtain the updated metric matrix. Resource block k is deleted from the unallocated resource block set R to obtain the updated unallocated resource block set. Let c = 1, and then return to step (3).

[0114] Otherwise, let c = c + 1 and return to step (3).

[0115] The other steps and parameters are the same as those in any of the specific implementation methods one to nine.

[0116] Example

[0117] This invention proposes a spectrum sharing method for integrated space-ground networks based on joint power and frequency allocation, specifically including the following steps:

[0118] Step 1: Based on satellites, satellite terminals, base stations, ground users, and interference coordination units, a spectrum sharing model for an integrated satellite-ground network is constructed, with ground user uplinks and satellite downlinks sharing frequencies.

[0119] In step 1, the satellite downlink uses Orthogonal Frequency Division Multiple Access (OFDMA), and the ground uplink uses Single-Carrier Frequency Division Multiple Access (SCDMA). Base stations sharing the same frequency are deployed around the satellite terminal to reduce interference from ground users to the satellite terminal, such as... Figure 1 As shown in the diagram, assuming the number and location of base stations and ground users are known, the base station allocates time-frequency resource blocks to the relevant users during scheduling.

[0120] Step 2: Based on the spectrum sharing model of the integrated space-ground network constructed in Step 1, the base station adjusts the transmit power level of the ground users according to the channel conditions of the ground users, and ensures that the aggregated interference power of the ground users' uplink to the satellite terminal probabilistically meets the interference protection constraints, thus constructing the resource optimization allocation problem of the integrated space-ground network.

[0121] Specifically, the transmission power of ground users and the potential interference power level to satellite terminals depend on the distance between the user and the base station. The transmission power of ground users is:

[0122]

[0123] Among them, P i max P0 represents the user's maximum transmit power, and M represents the base station's nominal receive power. i This represents the number of resource blocks allocated to user i, and α represents the path loss compensation factor (0 < α ≤ 1). This represents the channel gain from the user to the base station. Indicates the user antenna gain. Δ represents path loss. i This represents the modulation compensation factor.

[0124] In this embodiment of the invention, the interference link is modeled using a dual-ray ground reflection model, and spatially uncorrelated log-normal fading is introduced to simulate channel uncertainty. Therefore, it is necessary to ensure that the aggregated interference power of the user uplink to the satellite terminal is lower than the interference protection threshold.

[0125]

[0126] Where, n r Represents the set of resource blocks that overlap with the protected bandwidth, n ue P represents the number of ground users. i This represents the transmit power of user i. This represents the channel gain from the user to the satellite terminal on resource block k. Indicates the gain of the satellite terminal receiving antenna. Z represents path loss. ik Let m represent the channel fading variable that follows a log-normal distribution with mean m. z The standard deviation is σ z δ ik δ represents the resource block indicator variable. ik 1 and 0 respectively represent resource block k being allocated and not allocated to user i, ε s This represents the interference protection threshold for the satellite terminal. Based on the location information of the ground user and the satellite terminal, the channel gain can be estimated using a path loss model; however, the random variable Z cannot be directly measured.ik .

[0127] Therefore, to approximate the aggregation of disturbances and ensure that the disturbance constraints are satisfied with probability, the statistical model used is as follows:

[0128]

[0129] Where, ρ s This represents the probability threshold for aggregation interference, i.e., during a portion of the time ρ s Within this scope, aggregated interference is allowed to exceed the protection threshold. In this embodiment of the invention, the integrated space-ground network considers optimizing resource allocation for ground users to improve spectrum utilization while ensuring the communication performance of satellite terminals. The resource management architecture of the integrated space-ground network is as follows: Figure 2 As shown, the interference coordination unit has an interface to communicate with all base stations near the satellite terminal. Users return channel state information of communication and interference links to the interference coordination unit through the base stations. Furthermore, the interference coordination unit has an interface to communicate with the satellite; the satellite terminal feeds back the channel state information of the satellite downlink to the interference coordination unit and updates the interference protection threshold. Finally, the base station provides services to users within its coverage area through power control and resource block scheduling to maximize the uplink throughput of users, thus constructing the resource allocation problem as follows:

[0130]

[0131] stδ ijk ∈{0,1}; (4a)

[0132]

[0133]

[0134]

[0135]

[0136] The system utility function depends on the power allocation matrix. and resource block allocation matrix The number of users on a terrestrial network is related to the number of users served by each base station, i.e., it satisfies... in, n is the number of users in base station j. bs This refers to the number of base stations. Constraints (4a)-(4c) indicate that resource blocks are allocated to each user in a contiguous manner, constraint (4b) indicates that a resource block can only be allocated to one user, and constraint (4c) indicates that resource blocks allocated to a specific user must be adjacent. Constraint (4d) indicates that the user's transmit power is limited to a feasible range, where M... ijThis represents the number of resource blocks allocated by base station j to user i. The optimization problem couples the frequency scheduling problems of multiple base stations through interference protection constraints, greatly increasing computational complexity.

[0137] Step 3: The resource allocation problem constructed in Step 2 is non-convex and highly complex. Therefore, we consider decoupling the optimization problem into power allocation and frequency resource scheduling sub-problems to find a suboptimal solution. The power allocation algorithm is centrally computed by the interference coordination unit, and the power allocation results are returned to the corresponding base stations. The base stations modify the metric of available resource blocks for each user based on the returned maximum transmit power and use the frequency resource scheduling algorithm to calculate the number of resource blocks allocated to each user.

[0138] In this embodiment of the invention, the power control subproblem only considers users sharing the spectrum and determines power allocation to maximize user throughput. Therefore, it is not necessary to consider the association between users and base stations or resource blocks. The corresponding superscripts and subscripts in the optimization problem can be removed, and the power allocation problem can be expressed as:

[0139]

[0140] Where U(P) represents the objective function that depends on power allocation, and P i Let represent the total transmit power of user i in the protected frequency band. If the optimization problem is a convex instance, a concave objective function is required, and the constraints describe a convex set. Therefore, a concave sum law function is used as the objective function:

[0141]

[0142] in, Let N represent the channel gain from user i to the associated base station, and let N represent the noise power at the base station receiver.

[0143] Step 4: Based on the power allocation subproblem derived in Step 3, considering only ground users sharing the spectrum, determine the power allocation to maximize the throughput of the ground network. Based on the relationship between user transmit power and channel gain ratio, a power allocation algorithm based on interference probability constraints is proposed, and the maximum allowable transmit power for each user is obtained through iterative solution.

[0144] Specifically, to clarify the relationship between power allocation and path loss, the user's transmit power is correlated with path loss. Assuming that the uncertainty of the channel state is not considered, power allocation can be represented as a deterministic problem:

[0145]

[0146] Using the Karush-Kuhn-Tucker conditions, the transmit power of user i satisfies the following:

[0147]

[0148] Where μ represents the multiplier. The first term of the minimization problem is considered a solution to the water-filling algorithm, and the second term limits the user's transmit power. The channel gain ratio for user i is defined as...

[0149] In this embodiment of the invention, the base station is deployed at a location far from the satellite terminal, and the path loss satisfies... This indicates that user i has a larger channel gain ratio. According to equation (8), users with larger channel gain ratios tend to be assigned equal power levels. Since equation (5) has no analytical solution, the interior point method is used to solve it. In the simulation, the terrestrial network uses 3 base stations, which are 8km, 12km and 16km away from the satellite terminal, respectively. Each base station has 30 evenly distributed users in its coverage area. The simulation parameter values ​​are shown in Table 1.

[0150] Table 1

[0151]

[0152] like Figure 3 As shown, users with low channel gain ratios are assigned lower transmit power, resulting in lower interference levels to satellite terminals. By adjusting the transmit power, the throughput of users with high channel gain ratios can be increased. Figure 4 The relationship between the channel gain ratio and the interference level is consistent with the conclusion of equation (8). Users with high channel gain ratios have an average effect on interference to satellite terminals, which can reduce the probability of interference deviating from the average value. The scheduling period of resource blocks in the terrestrial network is 1ms, and the time complexity of the interior-point method is... ε represents the target accuracy. Therefore, the interior-point method has high time complexity and is difficult to apply to power allocation problems.

[0153] Based on the relationship between transmit power and channel gain ratio, a power allocation algorithm based on interference probability constraints is proposed. It is assumed that the interference power caused by users with higher transmit power tends to a fixed value, and a step function is used to approximate the equal interference power. The algorithm flow is as follows:

[0154] First, users are arranged in descending order based on their channel gain ratios. Then, using an approximation method based on interference probability constraints and a step function, the number of users allocated non-zero transmit power and the interference power level are determined. Finally, the user's transmit power is determined based on the interference power level. Therefore, the interference protection constraint in equation (5) is transformed into:

[0155]

[0156] According to the Fenton-Wilkinson (FW) approximation, we obtain variable xi y and y follow a normal distribution. Using the FW approximation, the mean and variance of the variable y are respectively m y =2lnu1-lnu2 / 2 and in r ij Representing the correlation coefficient, the mean and variance of y are obtained as follows: and

[0157]

[0158] Therefore, the probabilistic constraint on aggregation interference is transformed into:

[0159]

[0160] Using the upper bound of the probability constraint and the right-tail function of the standard normal distribution, the closed-form expression for the interference power level ξ(n) is obtained as follows:

[0161]

[0162] Algorithm 1 in Table 2 is the pseudocode for a power allocation algorithm based on interference probability constraints, where u1 represents the throughput implemented by the user, and P1 represents n. ue An array of power allocation values ​​for each user. Algorithm 1 aims to allocate transmit power to users with similar interference power levels, finding the number of users with maximum throughput in each loop operation. Calculating u1 and ξ(n) requires O(n^2) time. ue ) operations, iterative loop requires O(n) times. ue After ) operations, the time complexity of the algorithm is obtained as .

[0163] Table 2

[0164]

[0165] Step 5: Based on the frequency resource scheduling subproblem obtained in Step 3, a single-carrier frequency division multiple access (FDMA) constraint is adopted, limiting resource blocks to a single user, with all resource blocks allocated to the same user being adjacent. Using the user transmit power obtained in Step 4 as a constraint on the satellite terminal, a recursive maximum expansion algorithm is used to obtain the resource allocation scheme.

[0166] In this embodiment of the invention, the base station modifies the metric of the corresponding resource block based on the user's maximum transmit power, and then uses a frequency resource scheduling algorithm to calculate the number of resource blocks allocated to the user. The resource block allocation is updated in each time slot to adapt to dynamic communication needs. However, the power allocation algorithm does not need to run in every time slot, but should be sufficient to address the time dependency of channel state information. The returned power is taken as the user's maximum transmit power P. ilim, to meet the protection threshold of the satellite terminal, i.e. Therefore, equation (4) can be decoupled into a frequency resource scheduling problem for a single base station:

[0167]

[0168] The frequency domain resource scheduling problem for a single base station with continuous resource block constraints is a mixed-integer nonlinear programming problem, typically solved using heuristic approximation algorithms. By appropriately modifying the user transmit power constraints, the frequency resource scheduling problem in equation (12) is solved using the recursive maximum expansion algorithm, the pseudocode of which is Algorithm 2 in Table 3. Definition For user i on resource block c, The metric of resource blocks within the shared frequency band is modified based on the power allocation results. If the transmit power allocated to a user is zero, the metric of the resource blocks within the shared frequency band is set to zero to ensure that the algorithm will not allocate resource blocks within the shared frequency band to that user. However, the algorithm can allocate resource blocks not in the shared frequency band to any user.

[0169] Table 3

[0170]

[0171]

[0172] Step 6: Adjust the resource block allocation strategy according to the resource allocation scheme obtained in Step 5, so as to maximize spectrum utilization while meeting the satellite terminal interference protection requirements.

[0173] In step 6, a simulation analysis is performed on the proposed spectrum sharing method for integrated space-ground networks based on joint power and frequency allocation. The simulation parameter values ​​are shown in Table 1. Figure 5 and Figure 6 This paper analyzes the performance of power allocation algorithms in terms of user transmit power and interference levels. When the channel gain ratio is high, the transmit power allocated by the power allocation algorithm based on interference probability constraints is close to that of the interior-point method. When the channel gain ratio is low, the transmit power allocated by the interior-point method is non-zero, while the transmit power allocated by the power allocation algorithms based on interference probability constraints and compact constraint constraints is zero. Users allocated non-zero transmit power have a larger power allocation space. The power allocation algorithm based on compact constraint constraints reduces the feasible region of power through compact constraints, resulting in lower user transmit power and interference levels. Therefore, the performance of the power allocation algorithm based on interference probability constraints is closer to that of the interior-point method, with lower time complexity.

[0174] exist Figure 7This paper analyzes the relationship between throughput and the number of users achieved by a joint power and frequency allocation algorithm, with the base station 10km away from the satellite terminal. Compared with a frequency-exclusive access algorithm, this algorithm uses an interior-point method for power allocation and a recursive maximum expansion algorithm for frequency scheduling, with the satellite and terrestrial networks not sharing spectrum. When the number of users in a cell is small, not all resource blocks can be fully utilized, resulting in lower network throughput. The interior-point method and the power allocation algorithm based on interference probability constraints achieve similar throughput. Figure 8 This paper analyzes the impact of the distance between the base station and the satellite terminal on throughput, with a terrestrial cell having 16 users. The greater the distance between the base station and the satellite terminal, the lower the interference level for users, and the higher the throughput achieved. With a fixed number of users and resource blocks in the network, the throughput achieved by the frequency-exclusive access algorithm does not change with distance. Therefore, the integrated space-ground network can achieve low time complexity and full utilization of spectrum resources by using an algorithm based on joint power and frequency allocation.

[0175] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by hardware associated with program instructions. The program can be stored in a computer-readable storage medium, which may include ROM, RAM, disk, or optical disk, etc.

[0176] The above examples of the present invention are merely illustrative of the computational model and process of the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is impossible to exhaustively list all possible implementations here. Any obvious variations or modifications derived from the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A spectrum sharing method for a space-ground integrated network based on joint power and frequency allocation, characterized in that, The method specifically includes the following steps: Step 1: Construct a spectrum sharing model for an integrated satellite-ground network. The model includes satellites, satellite terminals, base stations, ground users, and interference coordination units, with the uplink of ground users sharing frequencies with the downlink of satellites. The satellite terminal is surrounded by... The first base station sharing a frequency, the first The number of ground users in each base station is , Base stations are used to allocate time and frequency resource blocks to ground users; Step 2: Based on the spectrum sharing model of the space-ground integrated network constructed in Step 1, construct the resource optimization and allocation problem of the space-ground integrated network. The specific process of step two is as follows: Step Two:

1. The transmission power of the ground user is: （1） in, Indicates the first Transmission power of each ground user Indicates the first Maximum transmit power for each ground user This indicates the base station's nominal received power. Indicates assignment to the first The number of resource blocks per ground user This represents the path loss compensation factor. Indicates the first Channel gain from ground user to base station, , Indicates the first Antenna gain from ground user to base station, Indicates the first Path loss from a ground user to a base station, Indicates the first Modulation compensation factor for each ground user; Step 22: The aggregated interference power constraint of the ground user uplink to the satellite terminal is as follows: （2） in, This represents the set of resource blocks that overlap with the protected bandwidth. This represents the total number of ground users. Indicates the first Transmission power of each ground user Indicates the first From ground users to satellite terminals in resource blocks Channel gain on , Indicates the first The receiving antenna gain from a ground user to a satellite terminal. Indicates the first The path loss from a ground user to a satellite terminal. Indicates the first A ground user in the resource block The channel fading variable on, and Follow the mean Standard deviation is The log-normal distribution Indicates a resource block indicator variable. equal When, it indicates a resource block. Assigned to the One ground user, equal When, it indicates a resource block. Not assigned to the first One ground user, This indicates the interference protection threshold for the satellite terminal; Step Two Three: Express the constraints of Step Two Two as probabilistic constraints that the aggregation interference needs to satisfy: （3） in, The probability threshold representing aggregation interference. Expressing the request The probability of; Step 24: With the goal of maximizing the uplink throughput of ground users, the resource optimization allocation problem is constructed as follows: （4） in, Represents the system utility function. Indicates the total number of base stations. The value depends on the power allocation matrix. and resource block allocation matrix , It is the first The base station is assigned to the first The first ground user used The power size of each resource block, equal When, it indicates that the resource block will be... Assigned to the The first base station One ground user, equal When, it indicates that the resource block will be... Not assigned to the first The first base station One ground user, This represents the total number of ground users in the terrestrial network, i.e., satisfying... , It is the first The total number of ground users in each base station Indicates the first The base station is assigned to the first The number of resource blocks per ground user Indicates the first The first base station From ground users to satellite terminals in resource blocks Channel gain on Indicates the first The first base station A ground user in the resource block Channel fading variables; Step 3: Decouple the problem constructed in Step 2 into a power allocation sub-problem and a frequency resource scheduling sub-problem; In the interference coordination unit, the power allocation result is calculated based on the power allocation subproblem and returned to the base station; The base station calculates the number of resource blocks allocated to each ground user based on the returned power allocation results and the frequency resource scheduling sub-problem, thus obtaining the resource block allocation result.

2. The method for spectrum sharing in a space-ground integrated network based on joint power and frequency allocation according to claim 1, characterized in that, The satellite downlink uses orthogonal frequency division multiple access, while the ground user uplink uses single-carrier frequency division multiple access.

3. The method for spectrum sharing in a space-ground integrated network based on joint power and frequency allocation according to claim 2, characterized in that, The power allocation sub-problem is: （5） in, Let the objective function of the power allocation subproblem be denoted as . Indicates the first Channel gain from a ground user to a satellite terminal, Indicates the first Channel fading variables for each ground user; The objective function for: （6） in, This indicates the noise power at the base station receiver.

4. The method for spectrum sharing in a space-ground integrated network based on joint power and frequency allocation according to claim 3, characterized in that, The interference coordination unit calculates the power allocation result based on the power allocation sub-problem and returns it to the base station; specifically: Step 1: Initialize the number of ground users allocated non-zero transmit power. Initialize the throughput of ground users ,initialization ; Step 2, place the first The channel gain ratio for each ground user is denoted as . ,according to Sort users in each region in descending order; like and Then the calculation is ranked first. Interference power level of ground users , And continue with step 3; Otherwise, the transmit power of each ground user obtained in the last iteration will be used as the final power allocation result; Step 3: Based on the interference power level Calculate the transmit power of each ground user: Step 4: Update the ground user throughput based on the transmit power of each ground user. The updated ground user throughput is as follows: Step 5: Determine if the updated ground user throughput is greater than... ; If the updated ground user throughput is greater than Then let Equal to the updated ground user throughput, then let Utilize the updated and Return to step 2; If the updated ground user throughput is less than or equal to Then let and utilize the updated Return to step 2.

5. The method for spectrum sharing in a space-ground integrated network based on joint power and frequency allocation according to claim 4, characterized in that, The channel gain ratio for: 。 6. The method for spectrum sharing in a space-ground integrated network based on joint power and frequency allocation according to claim 5, characterized in that, The calculations are prioritized. Interference power level of ground users Specifically: The constraints in formula (5) are transformed into: （9） , It is the base of the natural logarithm, the variable. Follow the mean variance is The normal distribution, i.e. , , ; The probability constraint of formula (9) is then transformed into: （10） in, It is the right-tail function of the standard normal distribution; Then obtain the interference power level The expression is: （11） The superscript "-1" indicates taking the reciprocal.

7. A spectrum sharing method for a space-ground integrated network based on joint power and frequency allocation according to claim 6, characterized in that, The frequency resource scheduling sub-problem is: （12） in, It is the objective function of the frequency resource scheduling subproblem. It is a base station Number of users within, It is the first The first ground user used The power size of each resource block, It is the base station returning the first Power allocation results for each ground user.

8. A spectrum sharing method for a space-ground integrated network based on joint power and frequency allocation according to claim 7, characterized in that, The base station calculates the number of resource blocks allocated to each ground user based on the returned power allocation results and the frequency resource scheduling sub-problem, specifically as follows: Step (1), Definition For users In resource block The measure of the amount of information transmitted. , For the bandwidth of the resource block, , ; Step (2), As a metric matrix The List, Then we obtain the metric matrix. ; initialization ; Step (3): Determine the set of unallocated resource blocks Is it an empty set? If no resource block set is allocated If it is not an empty set, then proceed to step (4). If no resource block set is allocated If it is an empty set, then the resource block allocation is completed; Step (4): Based on the metric matrix Unallocated resource block set Select the one with the first Large-scale resource blocks , will the Large metric values ​​are represented as ; Step (5), let It has already been assigned to the user A collection of resource blocks; like and adjacent or Then the first Large-scale resource blocks Assigned to user and remove resource blocks from the metric matrix. The relevant metrics are used to obtain the updated metric matrix, which is based on the set of unallocated resource blocks. Delete resource block This yields the updated set of unallocated resource blocks; let Then return to step (3); Otherwise, let Return to step (3) for execution.

Images