A resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions

Abstract

A resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions, comprising: a base station performing channel estimation on the instantaneous channel state information (CSI) of an uplink channel and performing fitting estimation on the distribution obeyed by the channel estimation error; modeling a band weight and rate optimization problem; the base station alternately iteratively solving the established band weight and rate optimization problem, which is divided into a subcarrier power allocation problem and a multi-user subcarrier allocation problem; after completing the iteration, judging the result of the multi-user subcarrier allocation problem to obtain a final value of a subcarrier allocation factor with a value range of a discrete variable set {0,1}; determining the subcarrier power allocation after the iteration to obtain a resource allocation plan; sending the resource allocation plan to multiple users, and the multiple users performing carrier aggregation according to the resource allocation plan; the present invention can flexibly allocate subcarriers and channel power under different user service priorities, and has strong robustness under non-ideal channel state information conditions.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communications, and in particular relates to a resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions. Background Art

[0002] Carrier Aggregation (CA) is a communication technology that aggregates multiple component carriers (CC) for use. As a key technology in LTE-Advanced and 5G networks, carrier aggregation aims to increase data transmission speed by increasing the bandwidth of wireless communication systems. The classic usage scenarios of carrier aggregation are divided into intra-band continuous carrier aggregation, intra-band non-continuous carrier aggregation, and inter-band carrier aggregation. With the rapid development of wireless communication services, especially the reality of long-term coexistence of 2G, 3G, 4G, and 5G multi-standard and multi-band, the fragmentation of operators' communication spectrum has become increasingly serious, and the advantage of CA technology in aggregating carriers of different frequency bands has been reflected. CA allows operators to effectively integrate fragmented spectrum resources to provide larger transmission bandwidth, thereby meeting the growing demand for data transmission.

[0003] Subcarrier selection and power allocation within a CA system are key to ensuring the performance of carrier aggregation systems. Traditional resource allocation algorithms in carrier aggregation systems include random carrier selection and round-robin. These traditional algorithms assume perfect channel state information for the base station (and user end). However, due to channel estimation errors, feedback, and quantization errors, perfect channel state information is often difficult to obtain in actual deployments. Furthermore, these traditional algorithms assume that all users have the same service priority, making them unsuitable for optimizing carrier aggregation for multiple users with varying service priorities.

[0004] Patent application publication number CN101765123A discloses a multi-user MIMO-OFDM uplink resource allocation method. However, this scheme groups users based on spatial correlation, preventing the sum rate performance from reaching optimal limits. Furthermore, this scheme is unsuitable for scenarios where users have varying service priorities, and the base station's use of continuous interference cancellation is too complex.

[0005] Patent application publication number CN113056015B discloses a power allocation method for a NOMA downlink system under non-ideal channel state information. This method measures the QoS requirements of weak users by using the probability of interruption, significantly improving user fairness while ensuring overall system throughput. However, because this solution focuses on fairness and is based on a downlink transmission system, it cannot be directly applied to uplink multi-carrier aggregation systems. Summary of the Invention

[0006] In order to overcome the deficiencies of the above-mentioned prior art, the object of the present invention is to provide a resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions, which can be applied to scenarios where the uplink multi-user service priorities are different and channel estimation errors exist. At the same time, in the solution process, the resource optimization problem of this scenario is decoupled into 1) power allocation under a given subcarrier allocation and 2) subcarrier selection under a given power allocation. The two subproblems are solved alternately, thereby achieving performance close to the theoretical optimal performance of resource allocation while maintaining a low algorithm complexity. In addition, the non-ideal CSIT conditions considered in the present invention are more consistent with actual communication scenarios, and the resource allocation algorithm under non-ideal CSIT conditions is more robust to channel estimation errors.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0008] A resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions comprises the following steps:

[0009] Step 1: The base station performs channel estimation on the instantaneous channel state information (CSI) of the uplink channel and performs fitting estimation on the distribution obeyed by the channel estimation error.

[0010] Step 2: The base station models a weighted sum rate optimization problem based on the number of multi-user service priority carriers, aggregation capability, number of users, and user transmit power preset by the uplink carrier aggregation system, combined with the channel estimation results of step 1.

[0011] In step 3, the base station solves the weighted and rate optimization problem modeled in step 2. The solution is divided into two steps: 1) solving the subcarrier power allocation problem under a given multi-user subcarrier allocation scheme; 2) solving the multi-user subcarrier allocation problem under a given subcarrier power allocation scheme, and alternately iterating the subcarrier power allocation problem and the multi-user subcarrier allocation problem.

[0012] Step 4: The base station iteratively solves the subcarrier power allocation problem and the multi-user subcarrier allocation problem according to step 3. After the iterations are completed, the base station determines the result of the multi-user subcarrier allocation problem and obtains the final value of the subcarrier allocation factor whose value range is the discrete variable set {0, 1}.

[0013] In step 5, the base station determines the subcarrier power allocation after iteration based on the final value of the subcarrier allocation factor in step 4, and obtains a resource allocation plan; the obtained resource allocation plan is sent to multiple users, and the multiple users perform carrier aggregation according to the resource allocation plan.

[0014] The specific method of step 1 is:

[0015] The multi-antenna base station completes channel estimation based on the uplink transmission pilot of the single-antenna user, using Denotes the channel estimate of user u on subcarrier n;

[0016] The actual value of the channel is expressed as is the channel estimation error, and the error satisfies the mean and variance of Normal distribution;

[0017] For the system band weight and rate under non-ideal CSIT channel estimation, the following expressions are obtained:

[0018]

[0019] in, represents variance; using w u As the preset user weight in the system; b u,n represents the subcarrier allocation factor, which describes whether user u uses subcarrier n for data transmission; p u,n The power allocation represents the transmit power of user u on subcarrier n; e u,n 2 represents the channel error.

[0020] The specific method of step 2 is:

[0021] According to the channel estimation result obtained in step 1, the subcarrier power allocation p in step 1 is used u,n and subcarrier allocation factor b u,n Construct two factor matrices P U×N and B U×N , where P U×N The internal value of the matrix is ​​the subcarrier power allocation p u,n Matrix B U×N The internal value is the subcarrier allocation factor b u,n , its value is 0 or 1; when b u,n =1, indicating that user u uses subcarrier n for aggregation; otherwise, user u does not use subcarrier n;

[0022]

[0023] The uplink carrier aggregation system weight and rate optimization problem is modeled as follows:

[0024]

[0025] p u,n ≥0(c)

[0026]

[0027] b u,n∈{0,1}(f)

[0028] In the objective function (a) The data transmission rate represents the use of subcarrier n by user u; {P, B} is the power and subcarrier optimization variable set; w u Represents the user's weight preset by the system. The size of the weight value is related to the user's service priority, that is, the higher the priority, the greater the weight factor; p u,n 、e u,n 2 、 The definition is the same as step 1; constraint (b) the total transmission power of user u using multi-carrier aggregation should be less than its maximum transmission power; constraint (c) is the power allocation p of the user on a certain subcarrier u,n Should be greater than or equal to 0; Constraint (d) is that each subcarrier can only serve one user in the uplink carrier aggregation system; Constraint (e) is that the maximum number of subcarriers allowed for user u does not exceed z u ; Constraint (f) indicates that the subcarrier selection state can only take values ​​0 and 1.

[0029] The specific method of step 3 is:

[0030] The subcarrier allocation factor b in the optimization problem constraint (f) established in step 2 u,n Relax the value range from the discrete integer set {0,1} to the continuous interval [0,1], and continuously adjust the optimized variable subcarrier power allocation p of the optimization problem established in step 2 through iterative optimization. u,n and subcarrier allocation factor b u,n , until the objective function of the optimization problem established in step 2, that is, the weight and rate converge; before starting the iteration, initialize the subcarrier allocation factor b u,n All To perform power allocation in the initial iteration, the specific iteration part includes the following two steps:

[0031] Step 3.1, solving the subcarrier power allocation subproblem under a given multi-user subcarrier allocation scheme;

[0032] Step 3.1.1, calculate the subcarrier activation mark ρ u (k);

[0033]

[0034] Arrange the channel gains in descending order, ρ u (k) is used as a carrier activation flag. Its characteristics are: 1) The corresponding channel is activated only when it is positive; 2) Its value decreases as the value of k increases; when all ρ is calculated, u (k) value, all ρ uThe maximum k corresponding to the non-zero value in (k) * The value is used as the number of active subcarriers under the optimal power allocation strategy; n0 represents the noise power spectrum density.

[0035] Step 3.1.2: Use the number of active subcarriers k calculated in step 3.1.1 * Calculate the power allocation corresponding to the active subcarriers;

[0036]

[0037] Among them, n corresponds to the subcarrier number, u corresponds to the user number; p u,n As a result of power distribution;

[0038] Step 3.2: Based on the power allocation result obtained in step 3.1, solve the multi-user subcarrier allocation problem under the given subcarrier power allocation. The problem is converted into:

[0039]

[0040] 0≤b u,n ≤1(1d)

[0041] Compared with the problem constructed in step 2, the value range of the subcarrier allocation factor in constraint (1d) becomes the interval [0,1], and the definitions of the remaining variables are the same as those in step 2.

[0042] The specific method of step 4 is:

[0043] The decision rule is introduced to constrain the integer characteristics of the subcarrier allocation factor after iteration; each subcarrier can only serve one user, and the continuous subcarrier allocation factor b in the [0,1] interval is u,n Make a decision and map it back to the discrete set {0,1}. The specific decision rules are:

[0044]

[0045] Then we get the matrix B U×N The final result is then the matrix B U×N Bring it into step 3.1, perform the one-time subcarrier activation status judgment of step 3.1.1 and power allocation of step 3.1.2, and calculate the subcarrier power allocation p u,n To determine the matrix P U×N , and calculate the uplink carrier aggregation system band weight and rate to complete the uplink carrier aggregation system resource allocation.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] From the perspective of algorithm applicability, this invention takes into account non-ideal channel state information, making it more suitable for actual communication scenarios with channel estimation errors. Furthermore, this solution can address scenarios where users have different service priorities, allocating resources to users of different priorities by adjusting weights, thus offering the advantage of strong applicability.

[0048] 2. From the perspective of algorithmic complexity, the present invention decomposes the resource allocation problem of the carrier aggregation system into a power allocation problem and a subcarrier allocation problem, solving them in an alternating iterative manner. This allows for near-optimal algorithm performance while maintaining low computational complexity. This makes it easy to implement.

[0049] In summary, the present invention proposes a resource allocation scheme for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions. Compared with existing schemes, the resource allocation scheme proposed in the present invention can be applied to scenarios where the uplink multi-user service priorities are different and there are channel estimation errors. At the same time, in the solution process, the present invention decouples the resource optimization problem of the scenario into 1) power allocation under a given subcarrier allocation, and 2) subcarrier selection under a given power allocation, and solves the two sub-problems alternately. Thereby, while obtaining performance close to the optimal performance of resource allocation theory, the algorithm complexity is maintained at a low level. In addition, the non-ideal CSIT conditions considered in the present invention are more in line with actual communication scenarios, and the resource allocation algorithm under the proposed non-ideal CSIT conditions is more robust to channel estimation errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a flow chart of the method of the present invention.

[0051] Figure 2 This is a system and rate comparison diagram between the allocation scheme of the present invention when users do not distinguish priorities, the random selection scheme, and the optimal traversal method.

[0052] Figure 3 This is a system and rate comparison diagram between the user priority differentiation scheme of the present invention, the random selection allocation scheme, and the optimal traversal method.

[0053] Figure 4 This is a comparison diagram of the resource allocation algorithm of the present invention using the proposed robust algorithm and a non-robust algorithm (using imperfect channel estimation as perfect channel estimation). DETAILED DESCRIPTION

[0054] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0055] like Figure 1 As shown, a resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions includes the following steps:

[0056] Step 1: The base station performs channel estimation on the instantaneous channel state information (CSI) of the uplink channel and performs fitting estimation on the distribution obeyed by the channel estimation error.

[0057] The specific method of step 1 is:

[0058] The multi-antenna base station completes channel estimation based on the uplink transmission pilot of the single-antenna user, using g u,n Denotes the channel estimate of user u on subcarrier n;

[0059] In practical applications, it is difficult to obtain ideal channel state information. In non-ideal CSIT situations, multiple factors must be considered, such as delay, feedback error, and estimation error. In this context, the actual value of the channel is expressed as is the channel estimation error, and the error satisfies the mean and variance of Normal distribution;

[0060] For the system band weight and rate under non-ideal CSIT channel estimation, they can be obtained by the following expressions:

[0061]

[0062] in, and The definition of has been described above; represents variance; using w u As the preset user weight in the system; b u,n represents the subcarrier allocation factor, which describes whether user u uses subcarrier n for data transmission; p u,n The power allocation represents the transmit power of user u on subcarrier n; e u,n 2 represents the channel error.

[0063] Step 2: The base station models a weighted sum rate optimization problem based on the number of multi-user service priority carriers, aggregation capability, number of users, and user transmit power preset by the uplink carrier aggregation system, combined with the channel estimation results of step 1.

[0064] The specific method of step 2 is:

[0065] According to the channel estimation result obtained in step 1, the subcarrier power allocation p in step 1 is used u,n and subcarrier allocation factor b u,n Construct two factor matrices P U×N and B U×N , where P U×N The internal value of the matrix is ​​the subcarrier power allocation p u,n Matrix B U×NThe internal value is the subcarrier allocation factor b u,n , its value is 0 or 1; when b u,n =1, indicating that user u uses subcarrier n for aggregation; otherwise, user u does not use subcarrier n;

[0066]

[0067] The uplink carrier aggregation system band weight and rate optimization problem can be modeled as follows:

[0068]

[0069] p u,n ≥0(c)

[0070]

[0071] b u,n ∈{0,1}(f)

[0072] In the objective function (a) The data transmission rate represents the use of subcarrier n by user u; {P, B} is the power and subcarrier optimization variable set; w u Represents the user's weight preset by the system. The size of the weight value is related to the user's service priority, that is, the higher the priority, the greater the weight factor; p u,n 、e u,n 2 、 The definition is the same as step 1; constraint (b) the total transmission power of user u using multi-carrier aggregation should be less than its maximum transmission power; constraint (c) is the power allocation p of the user on a certain subcarrier u,n Should be greater than or equal to 0; Constraint (d) is that each subcarrier can only serve one user in the uplink carrier aggregation system; Constraint (e) is that the maximum number of subcarriers allowed for user u does not exceed z u ; Constraint (f) indicates that the subcarrier selection state can only take values ​​0 and 1.

[0073] In step 3, the base station solves the weighted and rate optimization problem modeled in step 2. The solution is divided into two steps: 1) solving the subcarrier power allocation problem under a given multi-user subcarrier allocation scheme; 2) solving the multi-user subcarrier allocation problem under a given subcarrier power allocation scheme, and alternately iterating the subcarrier power allocation problem and the multi-user subcarrier allocation problem.

[0074] The specific method of step 3 is:

[0075] The subcarrier allocation factor b in the optimization problem constraint (f) established in step 2 u,nRelax the value range from the discrete integer set {0,1} to the continuous interval [0,1], and continuously adjust the optimized variable subcarrier power allocation p of the optimization problem established in step 2 through iterative optimization. u,n and subcarrier allocation factor b u,n , until the objective function of the optimization problem established in step 2, that is, the weight and rate converge; before starting the iteration, initialize the subcarrier allocation factor b u,n All To perform power allocation in the initial iteration, the specific iteration part includes the following two steps:

[0076] Step 3.1, solving the subcarrier power allocation subproblem under a given multi-user subcarrier allocation scheme;

[0077] Step 3.1.1, calculate the subcarrier activation flag ρ u (k);

[0078]

[0079] Arrange the channel gains in descending order, ρ u (k) is used as a carrier activation flag. Its characteristics are: 1) The corresponding channel is activated only when it is positive; 2) Its value decreases as the value of k increases; when all ρ is calculated, u (k) value, all ρ u The maximum k corresponding to the non-zero value in (k) * The value is used as the number of active subcarriers under the optimal power allocation strategy; n0 represents the noise power spectrum density.

[0080] Step 3.1.2: Use the number of active subcarriers k calculated in step 3.1.1 * Calculate the power allocation corresponding to the active subcarriers;

[0081]

[0082] Among them, n corresponds to the subcarrier number, u corresponds to the user number; p u,n As a result of power distribution;

[0083] Step 3.2: Based on the power allocation result obtained in step 3.1, solve the multi-user subcarrier allocation problem under the given subcarrier power allocation. The problem is converted into:

[0084]

[0085] 0≤b u,n ≤1(1d)

[0086] The meaning of the constraint in the above formula is the same as that in step 2, except that the value range of the subcarrier allocation factor in constraint (1d) is changed from {0,1} to the interval [0,1].

[0087] Step 4: The base station iteratively solves the subcarrier power allocation problem and the multi-user subcarrier allocation problem according to step 3. After the iterations are completed, the base station determines the result of the multi-user subcarrier allocation problem and obtains the final value of the subcarrier allocation factor whose value range is the discrete variable set {0, 1}.

[0088] The specific method of step 4 is:

[0089] The decision rule is introduced to constrain the integer characteristics of the subcarrier allocation factor after iteration; specifically, in principle, each subcarrier can only serve one user, so it is necessary to allocate the factor b to the continuous subcarriers in the interval [0,1]. u,n Make a decision and map it back to the discrete set {0,1}. The specific decision rules are:

[0090]

[0091] Then we get the matrix B U×N The final result is then the matrix B U×N Bring it into step 3.1, perform the one-time subcarrier activation status judgment of step 3.1.1 and power allocation of step 3.1.2, and calculate the subcarrier power allocation p u,n Then determine the matrix P U×N , and calculate the uplink carrier aggregation system band weight and rate to complete the uplink carrier aggregation system resource allocation.

[0092] In step 5, the base station determines the subcarrier power allocation after iteration based on the final value of the subcarrier allocation factor in step 4. At this point, a resource allocation plan is obtained; the obtained resource allocation plan is sent to multiple users, and the multiple users perform carrier aggregation according to the resource allocation plan.

[0093] Simulation parameter description

[0094] Consider a multi-user uplink carrier aggregation system consisting of a base station, U users, and N subcarriers. The noise power spectral density is n0, the uplink power is controlled by the current system signal-to-noise ratio, the number of carriers that can be aggregated by a single user is M, and there is no interference between the subcarriers of the base station.

[0095] Simulation 1: When different user priority settings are used, the system and rate comparison between the scheme proposed in this invention, the random selection allocation scheme, and the optimal traversal method is shown in Figure 2. Figure 2 、 Figure 3 .

[0096] Table 1 Simulation parameter settings

[0097]

[0098] like Figure 2 、 Figure 3 The figures show a comparison between the alternating iterative algorithm, the random subcarrier allocation method, and the optimal ergodic method, respectively, when user weight parameters are the same and when user weight parameters are different. A larger user weight parameter indicates a higher user priority. It can be concluded that the alternating iterative method for subcarrier allocation in the present invention can achieve a more optimal subcarrier allocation solution than random subcarrier allocation, and brings the system band weight and rate closer to those of the optimal ergodic method.

[0099] Simulation 2: Comparison between the proposed robust algorithm and the non-robust algorithm (i.e., using the imperfect channel estimation as the perfect channel estimation) in the resource allocation algorithm. Figure 4 shown.

[0100]

[0101] Figure 4 Comparing the proposed robust algorithm with a non-robust algorithm for resource allocation shows that the robust algorithm achieves higher weights and rates than the non-robust algorithm. This is because the robust algorithm takes channel errors into account when allocating resources, providing more stable and reliable results than the non-robust algorithm.

[0102] The present invention discloses a resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions. This scheme considers non-ideal channel state information and allocates subcarriers and power resources through an alternating iterative approach. Furthermore, this scheme can meet the requirements of different user service priorities and adjust resource allocation based on weights. Compared with existing methods, this method is less complex, can flexibly allocate subcarriers and channel power under different user service priorities, and is more robust under non-ideal channel state information conditions.

Claims

1. A resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions, characterized in that: The following steps are involved: Step 1: The base station performs channel estimation on the instantaneous channel state information (CSI) of the uplink channel and performs fitting estimation on the distribution obeyed by the channel estimation error. Step 2: The base station models a weighted sum rate optimization problem based on the number of multi-user service priority carriers, aggregation capability, number of users, and user transmit power preset by the uplink carrier aggregation system, combined with the channel estimation results of step 1. Step 3: The base station solves the weight and rate optimization problem modeled in step 2. The solution consists of two steps: 1) solving the subcarrier power allocation problem given a multi-user subcarrier allocation scheme; 2) solving the multi-user subcarrier allocation problem given a subcarrier power allocation scheme. The subcarrier power allocation problem and the multi-user subcarrier allocation problem are solved alternately and iteratively. Step 4: The base station iteratively solves the subcarrier power allocation problem and the multi-user subcarrier allocation problem according to step 3. After the iterations are completed, the base station determines the result of the multi-user subcarrier allocation problem and obtains the final value of the subcarrier allocation factor whose value range is the discrete variable set {0, 1}. Step 5: The base station determines the subcarrier power allocation after iteration based on the final value of the subcarrier allocation factor in step 4, and obtains a resource allocation solution; The obtained resource allocation scheme is sent to multiple users, and the multiple users perform carrier aggregation according to the resource allocation scheme.

2. The resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions according to claim 1, characterized in that: The specific method of step 1 is: The multi-antenna base station completes channel estimation based on the uplink transmission pilot of the single-antenna user, using g u,n Denotes the channel estimate of user u on subcarrier n; The actual value of the channel is expressed as is the channel estimation error, and the error satisfies the mean and variance of Normal distribution; For the system band weight and rate under non-ideal CSIT channel estimation, the following expressions are obtained: in, represents variance; using w u As the preset user weight in the system; b u,n represents the subcarrier allocation factor, which describes whether user u uses subcarrier n for data transmission; p u,n The power allocation represents the transmit power of user u on subcarrier n; e u,n 2 represents the channel error.

3. The resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions according to claim 2, characterized in that: The specific method of step 2 is: According to the channel estimation result obtained in step 1, the subcarrier power allocation p in step 1 is used u,n and subcarrier allocation factor b u,n Construct two factor matrices P U×N and B U×N , where P U×N The internal value of the matrix is ​​the subcarrier power allocation p u,n Matrix B U×N The internal value is the subcarrier allocation factor b u,n , its value is 0 or 1; when b u,n =1, indicating that user u uses subcarrier n for aggregation; Otherwise, user u does not use subcarrier n; The uplink carrier aggregation system weight and rate optimization problem is modeled as follows: In the objective function (a), The data transmission rate represents the use of subcarrier n by user u; {P, B} is the power and subcarrier optimization variable set; w u Represents the user's weight preset by the system. The size of the weight value is related to the user's service priority, that is, the higher the priority, the greater the weight factor; p u,n 、e u,n 2 、 The definition is the same as step 1; constraint (b) the total transmission power of user u using multi-carrier aggregation should be less than its maximum transmission power; constraint (c) is the power allocation p of the user on a certain subcarrier u,n Should be greater than or equal to 0; Constraint (d) is that each subcarrier can only serve one user in the uplink carrier aggregation system; Constraint (e) is that the maximum number of subcarriers allowed for user u does not exceed z u ; Constraint (f) indicates that the subcarrier selection state can only take values ​​0 and 1.

4. The resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions according to claim 3, characterized in that: The specific method of step 3 is: The subcarrier allocation factor b in the optimization problem constraint (f) established in step 2 u,n Relax the value range from the discrete integer set {0,1} to the continuous interval [0,1], and continuously adjust the optimized variable subcarrier power allocation p of the optimization problem established in step 2 through iterative optimization. u,n and subcarrier allocation factor b u,n , until the objective function of the optimization problem established in step 2, that is, the weight and rate converge; before starting the iteration, initialize the subcarrier allocation factor b u,n All To perform power allocation in the initial iteration, the specific iteration part includes the following two steps: Step 3.1, solving the subcarrier power allocation subproblem under a given multi-user subcarrier allocation scheme; Step 3.1.1, calculate the subcarrier activation mark ρ u (k); Arrange the channel gains in descending order, ρ u (k) As a carrier activation flag, its characteristics are: 1) The corresponding channel is activated only when it is positive; 2). Its value decreases as the k value increases; when all ρ is calculated u (k) value, all ρ u The maximum k corresponding to the non-zero value in (k) * The value is used as the number of active subcarriers under the optimal power allocation strategy; n0 represents the noise power spectrum density; Step 3.1.2: Use the number of active subcarriers k calculated in step 3.1.1 * Calculate the power allocation corresponding to the active subcarriers; Among them, n corresponds to the subcarrier number, u corresponds to the user number; p u,n As a result of power distribution; Step 3.2: Based on the power allocation result obtained in step 3.1, solve the multi-user subcarrier allocation problem under the given subcarrier power allocation. The problem is converted into: Compared with the problem constructed in step 2, the value range of the subcarrier allocation factor in constraint (1d) becomes the interval [0,1], and the definitions of the remaining variables are the same as those in step 2.

5. The resource allocation method for a multi-user uplink carrier aggregation system under non-ideal channel estimation conditions according to claim 4, characterized in that: The specific method of step 4 is: The decision rule is introduced to constrain the integer characteristics of the subcarrier allocation factor after iteration; each subcarrier can only serve one user, and the continuous subcarrier allocation factor b in the [0,1] interval is u,n Make a decision and map it back to the discrete set {0,1}. The specific decision rules are: Then we get the matrix B U×N The final result is then the matrix B U×N Bring it into step 3.1, perform the one-time subcarrier activation status judgment of step 3.1.1 and power allocation of step 3.1.2, and calculate the subcarrier power allocation p u,n To determine the matrix P U×N , and calculate the uplink carrier aggregation system band weight and rate to complete the uplink carrier aggregation system resource allocation.

Images