Low-complexity MIMO precoding design method for guaranteeing service quality

Abstract

The invention discloses a low-complexity MIMO (Multiple Input Multiple Output) precoding design method for guaranteeing service quality, which comprises the following steps of: firstly, determining a performance index of a communication system, and constructing a precoding design problem model which takes maximization of the index as a target function and takes the service quality, base station transmitting power and the like as constraint conditions; secondly, constructing a feasible signal to interference plus noise ratio region, and converting the proposed precoding design problem into a signal to interference plus noise ratio distribution problem of maximizing a performance index in the feasible signal to interference plus noise ratio region; and finally, solving a signal to interference plus noise ratio distribution problem and mapping the obtained optimal user signal to interference plus noise ratio into a corresponding precoding matrix. According to the method, a high-dimensional precoding matrix design problem is converted into a low-dimensional signal to interference plus noise ratio distribution problem, so that the optimal system performance is approached with relatively low calculation complexity, and the method is widely applicable to various communication system performance indexes.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technologies, and in particular, to a low-complexity MIMO precoding design method for ensuring quality of service. Background Art

[0002] Multiple-Input Multiple-Output (MIMO) systems are the cornerstone of modern wireless communication systems and are extremely important technologies in mobile communications. They can effectively combat channel fading and achieve high-speed data transmission, and precoding technologies can provide rich spatial diversity and multiplexing gains for MIMO communication systems.

[0003] Existing precoding designs usually have high computational complexity. Generally speaking, existing methods usually focus on directly optimizing the precoding matrix or optimizing the user power under a given precoding vector. However, the high variable dimension leads to a large amount of computational overhead in directly optimizing the precoding matrix, while the method of optimizing power under a given precoding vector cannot fully utilize the spatial gain of MIMO and brings non-negligible performance losses. Therefore, low-complexity MIMO system precoding design methods play a crucial role in improving system performance and reducing implementation complexity.

[0004] In addition, existing low-complexity precoding design methods often only target specific system performance metrics and usually cannot ensure quality of service, with great limitations. Summary of the Invention

[0005] Object of the Invention: The present invention provides a low-complexity MIMO precoding design method for ensuring quality of service, which transforms the high-dimensional precoding matrix design problem into a low-dimensional signal-to-interference-plus-noise ratio (SINR) allocation problem, approximates the optimal system performance with low computational complexity, and is widely applicable to various communication system performance metrics.

[0006] Technical Solution: A low-complexity MIMO precoding design method for ensuring quality of service according to the present invention includes the following steps:

[0007] Step 1: Determine the communication system performance metrics, and construct a precoding design problem model with maximizing this metric as the objective function and quality of service and base station transmit power as the constraint conditions;

[0008] Step 2: Construct a feasible SINR region, and transform the precoding design problem in Step 1 into a SINR allocation problem of maximizing the performance metric within the feasible SINR region;

[0009] Step 3: Solve the user SINR allocation problem proposed in Step 2, and map the obtained optimal user SINRs to the corresponding precoding matrices.

[0010] Furthermore, in step 1, the objective function of the communication system performance metric is a function of the user signal-to-interference-plus-noise ratio, denoted as f(γ 1 ,γ 2 ,…,γ K ), where γ k is the signal-to-interference-plus-noise ratio of user k, defined as the ratio of the average power of the effective signal to the average power of the interference signal plus noise.

[0011] Furthermore, in step 1, the quality-of-service constraint means that the signal-to-interference-plus-noise ratio of any receiving user is not lower than a given threshold; the transmit power constraint of the base station means that the transmit power of the base station does not exceed a given power.

[0012] Furthermore, in step 1, the precoding design problem model additionally includes any constraint g j (γ 1 ,γ 2 ,…,γ K )≤0, where g j (γ 1 ,γ 2 ,…,γ K ) is a function of the user signal-to-interference-plus-noise ratio.

[0013] Furthermore, in step 2, the constructed feasible signal-to-interference-plus-noise ratio region is the set of signal-to-interference-plus-noise ratios that simultaneously satisfy the quality-of-service constraint and the transmit power constraint of the base station, specifically:

[0014]

[0015] where is the channel vector corresponding to user k, is the precoding vector corresponding to user k, γ th =(γ 1,th ,…,γ K,th ) T is the signal-to-interference-plus-noise ratio threshold of the user, is the variance of the noise at user k, Psum is the given power of the base station, and N and K represent the number of base station antennas and the number of users, respectively.

[0016] Furthermore, in step 2, the obtained signal-to-interference-plus-noise ratio allocation problem is an optimization problem with the objective of maximizing the system performance metric in step 1 within the feasible signal-to-interference-plus-noise ratio region, specifically:

[0017]

[0018] Furthermore, in step 3, the obtained optimal user signal-to-interference-plus-noise ratio is denoted as Optionally, mapping the obtained optimal user signal-to-interference-plus-noise ratio (SINR) to the corresponding precoding matrix includes direct mapping and mapping based on the optimal structure.

[0019] Furthermore, in the direct mapping, let be the optimal solution to the following problem:

[0020]

[0021] Then the mapped precoding matrix is

[0022] Furthermore, in the mapping based on the optimal structure, the optimal precoding matrix must satisfy the following structure:

[0023]

[0024] where Λ = Diag(λ 1 , λ 2 , …, λ K ), H = [h 1 , …, h K ,

[0025] Then the mapped precoding matrix is

[0026] Beneficial effects: The present invention transforms the high-dimensional precoding matrix design problem into a low-dimensional SINR allocation problem, and replaces matrix inversion and other operations with scalar operations, thereby greatly reducing the computational complexity of precoding matrix design; the present invention can effectively protect the quality of service of all users and is widely applicable to the performance indicators of various communication systems, enhancing generality and flexibility, and can effectively approximate the optimal system performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a schematic flow chart of the method of the present invention.

[0028] Figure 2 is a schematic diagram of the system rate performance of the precoding design method for guaranteeing quality of service proposed by the present invention.

[0029] Figure 3 is a schematic diagram of the worst SINR of the precoding design method for guaranteeing quality of service proposed by the present invention.

[0030] Figure 4 is a schematic diagram of the complexity performance of the precoding design method for guaranteeing quality of service proposed by the present invention.

[0031] Figure 5 is a schematic diagram of the symbol error rate performance of the precoding design method for guaranteeing quality of service proposed by the present invention. Detailed implementation manners

[0032] As shown Figure 1 in the figure, a low-complexity MIMO precoding design method for guaranteeing service quality includes the following steps:

[0033] Step 1: Determine the performance indexes of the communication system, and construct a precoding design problem model with maximizing the indexes as the objective function and with service quality and base station transmission power as the constraint conditions;

[0034] Step 2: Construct a feasible signal-to-interference-plus-noise ratio (SINR) region, and transform the precoding design problem in Step 1 into a SINR allocation problem of maximizing the performance indexes within the feasible SINR region;

[0035] Step 3: Solve the user SINR allocation problem proposed in Step 2, and map the obtained optimal user SINR to the corresponding precoding matrix.

[0036] In Step 1, an optional expression of the optimization problem is:

[0037]

[0038] where the objective function f(γ 1 , γ 2 , …, γ K ) is a function about the user SINR, and the function g j (γ 1 , γ 2 , …, γ K ) is a convex function about the user SINR.

[0039] In Step 2, all the feasible SINR regions corresponding to the precoding matrix are the following set

[0040]

[0041] For several optional approximate sets of the above set, including but not limited to the following sets:

[0042]

[0043] where α k , β kj are respectively defined as

[0044]

[0045] Thus, the transformed SINR allocation problem is:

[0046]

[0047] In step 3, denote the optimal solution of the problem after the above transformation as The mapping scheme from the optimal signal-to-interference-plus-noise ratio (SINR) to the corresponding precoding matrix includes, but is not limited to, direct mapping and mapping based on the optimal structure:

[0048] Direct mapping: Denote as the optimal solution of the following problem:

[0049]

[0050] Mapping based on the optimal structure: The optimal precoding matrix must satisfy the following structure:

[0051]

[0052] where Λ = Diag(λ 1 , λ 2 , …, λ K ),

[0053]

[0054] Then the mapped precoding matrix is

[0055] Combined with specific examples, the specific application process and performance advantages of the proposed design framework are respectively demonstrated under the conditions of maximizing the system sum rate performance, maximizing the system sum rate based on the zero-forcing criterion, and minimizing the symbol error rate performance, fully demonstrating the flexibility of the proposed precoding design framework.

[0056] (1) Maximizing the system sum rate

[0057] In this example, in the simulation example provided by the present invention, set the number of base station antennas N = 128, the number of users K = N / 2 = 64, and the user signal-to-interference-plus-noise ratio threshold is set to The maximum transmit power P = 10 dBW.

[0058] The system performance function f(γ 1 , γ 2 , …, γ K ) in step 1 is selected as The corresponding optimization problem is:

[0059]

[0060] The feasible signal-to-interference-plus-noise ratio region is selected as

[0061]

[0062] where α k , β kj are respectively defined as

[0063]

[0064] Thus, the signal-to-interference-plus-noise ratio (SINR) allocation problem is as follows:

[0065]

[0066] This problem can be effectively solved by methods such as the gradient method and the interior point method. Here, the quadratic transformation and the water-filling algorithm are used to solve this problem. First, introduce the quadratic transformation auxiliary variable The approximate problem is equivalent to the following problem

[0067]

[0068] where A k (Y), B k (Y), C k (Y) are defined as:

[0069]

[0070] Next, based on the alternating optimization method, the iterative process of the above approximate problem is as follows:

[0071]

[0072] where when updating γ (n+1) , the following water-filling algorithm can be used to solve the sub-problem composed of the original variables. Specifically, we first construct the auxiliary variable μ and define the function where the relevant parameters are defined as

[0073] and After analysis, the optimal solution of the sub-problem satisfies where μ ＊ satisfies

[0074]

[0075] The above problem can be effectively solved by the bisection method, and then the sub-problem is solved. After calculating the optimal signal-to-interference-plus-noise ratio , the corresponding precoding matrix is obtained according to the optimal structure mapping.

[0076] Figure 2 The achievable rates of multiple precoding schemes versus the maximum power are given. Among them, baseline schemes 1 and 2 are based on the weighted minimum mean square error algorithm, and baseline scheme 3 is a zero-forcing precoding scheme based on maximizing the sum rate. It can be seen that the proposed scheme of the present invention can achieve the same performance as baseline schemes 1 and 2, and has a 10% performance gain compared to baseline scheme 3.

[0077] Figure 3 The curves of the minimum signal-to-interference-plus-noise ratio (SINR) versus the maximum transmit power for multiple precoding schemes are given. It can be seen that the baseline scheme 1 cannot meet the quality-of-service (QoS) constraints of users at low transmit powers, while the proposed scheme in the present invention can guarantee the QoS constraints of the worst user under any circumstances.

[0078] Figure 4 The curves of the average central processor running time versus the number of base station antennas for multiple precoding schemes are given. It can be seen that the proposed scheme in the present invention has the lowest average running time, with time gains of more than 10 dB and 20 dB compared to the baseline schemes 1 and 2, respectively.

[0079] (2) Maximizing the system sum rate based on the zero-forcing criterion

[0080] The precoding design based on the zero-forcing criterion, i.e., zero-forcing precoding, is one of the commonly used linear precodings. By zero-forcing the interference between users, the precoding design is simplified. In the simulation example provided by the present invention, the number of base station antennas N = 128, the number of users K = N / 2 = 64, and the user SINR threshold is set to the maximum transmit power P sum = 10 dBW.

[0081] In this example, the communication system performance metric f(γ 1 , γ 2 , …, γ K ) in step 1 is selected as The corresponding optimization problem is:

[0082]

[0083] The feasible SINR region in step 2 is

[0084]

[0085] where The corresponding SINR allocation problem is:

[0086]

[0087] This problem can be effectively solved by the water-filling algorithm. Denote its optimal solution as Then the corresponding precoding matrix is:

[0088]

[0089] (3) Minimizing the system symbol error rate

[0090] In the simulation example provided by the present invention, the number of base station antennas N = 24, the number of users K = 18, and the user SINR threshold is set to The maximum transmission power P = 10 dBW, and the modulation mode is selected as QPSK. At this time, the symbol error rate of the user can be expressed as

[0091]

[0092] where When minimizing the system and symbol error rate, the performance function f(γ 1 ,γ 2 ,…,γ K ) of the communication system can be selected as:

[0093]

[0094] In this example, the optimization problem corresponding to step 1 is:

[0095]

[0096] The feasible signal-to-interference-plus-noise ratio region in step 2 is selected as

[0097]

[0098] where α k ,β kj are respectively defined as

[0099]

[0100] Thus, the signal-to-interference-plus-noise ratio allocation problem is:

[0101]

[0102] This problem can be effectively solved by using methods such as the gradient method and block coordinate descent. After calculating the optimal signal-to-interference-plus-noise ratio, the corresponding precoding matrix can be obtained.

[0103] In the simulation example provided by the present invention, the number of base station antennas N = 24, the number of users K = 18, and the user signal-to-interference-plus-noise ratio threshold is set as Baseline schemes 4 and 5 are schemes designed based on maximizing the sum rate, and baseline scheme 6 is a zero-forcing precoding scheme based on minimizing the bit error rate.

[0104] Figure 5 By giving the curve of the average symbol error rate of various precoding schemes with respect to the maximum power, it can be seen that the scheme of the present invention has the optimal symbol error rate performance.

Claims

1. A low-complexity MIMO precoding design method for ensuring service quality, characterized in that: The steps include: Step 1: Determine the performance index of the communication system, and construct a precoding design problem model with maximizing the index as the objective function and taking the quality of service and base station transmission power as constraints; Step 2: Construct a feasible signal-to-interference-noise ratio region, and transform the precoding design problem in step 1 into a signal-to-interference-noise ratio allocation problem of maximizing the performance index within the feasible signal-to-interference-noise ratio region; Step 3: Solve the user signal to interference plus noise ratio allocation problem mentioned in step 2, and map the obtained optimal user signal to interference plus noise ratio into a corresponding precoding matrix.

2. The low-complexity MIMO precoding design method for ensuring service quality as claimed in claim 1, characterized in that: In step 1, the objective function of the communication system performance indicator is a function of the user signal to interference and noise ratio, where the user signal to interference and noise ratio is defined as the ratio of the average power of the effective signal to the average power of the interference signal plus the noise.

3. The low-complexity MIMO precoding design method for ensuring service quality as claimed in claim 1, characterized in that: In step 1, the service quality constraint means that the signal to interference and noise ratio of any receiving user is not lower than a given threshold; the transmission power constraint of the base station means that the transmission power of the base station is not higher than a given power.

4. The low-complexity MIMO precoding design method for ensuring service quality as claimed in claim 1, characterized in that: In step 1, the precoding design problem model may additionally include any constraint function on the user signal-to-interference-and-noise ratio.

5. The low-complexity MIMO precoding design method for ensuring service quality as claimed in claim 1, characterized in that: In step 2, the feasible signal to interference plus noise ratio region constructed is a set of signal to interference plus noise ratios that simultaneously satisfy the quality of service constraint and the transmission power constraint of the base station.

6. The low-complexity MIMO precoding design method for ensuring service quality as claimed in claim 1, characterized in that: In step 2, the signal to interference and noise ratio allocation problem obtained is an optimization problem in a feasible signal to interference and noise ratio region with maximizing the system performance index in step 1 as the objective function.

7. The low-complexity MIMO precoding design method for ensuring service quality as claimed in claim 1, characterized in that: In step 3, optionally mapping the optimal user signal to interference plus noise ratio to a corresponding precoding matrix includes direct mapping and mapping based on an optimal structure.

8. The low-complexity MIMO precoding design method for ensuring service quality as claimed in claim 7, characterized in that: The precoding matrix after mapping in direct mapping is a precoding matrix with minimum transmission power that satisfies the user signal to interference plus noise ratio equal to the optimal signal to interference plus noise ratio.

9. The low-complexity MIMO precoding design method for ensuring service quality as claimed in claim 7, characterized in that: In the mapping based on the optimal structure, according to the conditions satisfied by the optimal precoding matrix, the parameters corresponding to the optimal signal to interference and noise ratio are calculated, and then the optimal precoding matrix is ​​obtained.

Images