Angle domain hybrid beamforming method suitable for a cell-free millimeter wave MIMO system

Abstract

The application provides an angle domain hybrid beamforming method suitable for a cell-free millimeter wave MIMO system. Firstly, the method uses angle domain hybrid beamforming instead of full digital beamforming to reduce hardware power consumption; secondly, a user-centered service scheme is adopted in the system, and the transmission power and the power consumption of the backhaul line are reduced through reasonable user association. The method respectively uses variables to represent whether the user is associated with the access point, the beam selection of different access points and the digital beamforming vector, then models the system user association and beamforming design problem as an energy efficiency maximization problem under the user service quality constraint and the power consumption constraint; through sparse approximation, Dinkelbach method, equivalent transformation and first-order Taylor expansion, the initial non-convex problem is converted into a convex problem which can be solved iteratively to obtain the optimal energy efficiency of the system. Compared with the early centralized MIMO system and the full digital beamforming method, the application improves the system energy efficiency while ensuring the user service quality.

Description

Technical Field

[0001] This invention relates to a beamforming method for cellular-free millimeter-wave MIMO systems, and more specifically to an angle-domain hybrid beamforming method suitable for cellular-free millimeter-wave MIMO systems. Background Technology

[0002] In recent years, with the exponential growth in service quality requirements and connection density, technologies such as millimeter-wave (mmWave) and massive MIMO (Multiple-Input and Multiple-Output) have been proposed to meet the needs of next-generation mobile communications. By using the shorter wavelength millimeter-wave band, the arrangement of massive antenna arrays becomes possible, allowing more energy to be concentrated in a narrower beam, compensating for its high path loss. However, deploying a large number of antennas in a millimeter-wave MIMO system leads to high energy consumption due to the RF chain, severely reducing the system's energy efficiency.

[0003] Cellular-free millimeter-wave MIMO systems distribute access points and connect them to a central unit for centralized processing. While providing macro diversity, centralized processing reduces interference between access points, thereby lowering the required transmission power and improving system energy efficiency.

[0004] Furthermore, to address the high hardware power consumption issue caused by massive MIMO, hybrid beamforming technology has been proposed and widely applied. Hybrid beamforming combines analog beamforming based on low-power phase shifters with digital beamforming based on small-sized RFs, allowing the number of RF chains to be much smaller than the number of antennas, effectively reducing system power consumption. Among them, angle-domain hybrid beamforming based on beam selection has been proven to be a hybrid beamforming method that can approach the performance of fully digital beamforming with a smaller number of RF chains. Currently, most common beam selection methods are based on relatively simple heuristic algorithms or traversal search. However, when considering applying angle-domain hybrid beamforming to non-cellular millimeter-wave MIMO system architectures to further improve system energy efficiency, heuristic algorithms suitable for single base stations are often difficult to extend to distributed scenarios, and traversal search methods also introduce significant complexity as the number of access points increases. Summary of the Invention

[0005] Technical issues

[0006] To further improve the energy efficiency of cellular-free millimeter-wave MIMO systems by applying angle-domain hybrid beamforming, this invention provides an angle-domain hybrid beamforming method suitable for cellular-free millimeter-wave MIMO systems.

[0007] Technical solution

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] An angle-domain hybrid beamforming method for non-cellular millimeter-wave MIMO systems includes the following steps:

[0010] S1: Establish a communication model for a cellular-free millimeter-wave MIMO system based on angle-domain hybrid beamforming, specifically:

[0011] Configure M service units and K single-antenna users, equipped with N t Antenna ULA, N RF A cellular millimeter-wave MIMO system with RF access points, where the set of access points is represented as... The user set is represented as Channel state information has been obtained based on channel estimation and shared among different access points through central unit control.

[0012] Based on the Saleh-Valenzuela millimeter-wave channel model, the uplink channel between access point m and user k is modeled as follows:

[0013] h k,m =A k,m β k,m (1)

[0014] in, L is the number of independent and identically distributed paths. For N t The antenna ULA steering vector is represented as:

[0015]

[0016] λ is the carrier wavelength, and D is the antenna spacing. Let be the random arrival angle of the l-th path. Furthermore, Let m be the complex gain of the l-th path between access point m and user k;

[0017] This invention considers angle-domain hybrid beamforming to reduce the number of RF chains, thereby reducing hardware power consumption and improving energy efficiency. Due to the weak scattering characteristics of millimeter-wave channels, It is assumed to be distributed within a small interval, which leads to h k,m It exhibits strong correlation. When the discrete Fourier transform matrix F is used as the analog beamforming matrix, the channel is transformed into the angular domain, i.e. Large non-zero values ​​exist only at a few consecutive locations (called spatial points), and the number of spatial points must be the same as the number of RF chains available for digital beamforming. Therefore, the selection of spatial points becomes particularly important when the number of RF chains is limited. Since each spatial point represents a discrete angle 2π / N in the angular domain... t Therefore, selecting a spatial point for transmission can be visualized as sending a beam in the corresponding direction; hence, the selection of a spatial point is also called beam selection. The set of available beams for each access point is defined as follows: This indicates the beam selection at access point m. This indicates that beam b is selected, and vice versa.

[0018] In this non-cellular millimeter-wave MIMO system architecture, a user-centric service scheme is adopted, which selects some rather than all access points to serve a particular user in order to reduce transmit power and load-based backhaul power consumption, thereby improving system energy efficiency. Definition This indicates the association between the user and the base station. This indicates that access point m provides services to user k, and vice versa. In summary, the signal-to-interference-plus-noise ratio (SIR) at user k is expressed as:

[0019]

[0020] in, For the corresponding digital beamformer, Let be the noise variance at user k. The achievable rate of user k is...

[0021] S2: Establish a power consumption model for a cellular-free millimeter-wave MIMO system based on angle-domain hybrid beamforming, specifically:

[0022] The total power consumption of each access point is expressed as:

[0023]

[0024] in, Static hardware power consumption, expressed as

[0025]

[0026] Among them, P lo P ps P DAC and P RF These represent the power consumption of the local oscillator, phase shifter, DAC, and RF chain, respectively. The power consumption of the return line is expressed as

[0027]

[0028] in, Power dissipation, This refers to the backhaul line capacity. When only data exchange between the access point and the central unit is considered, the backhaul rate is... It is assumed to be the sum of the reachable rates of all users served. Let ∈ be the corresponding transmitted power, and ∈ be the efficiency of the power amplifier.

[0029] Because of P ps Much smaller than P DAC and P RF Therefore, compared to the static hardware power consumption of P, lo +N t (2P DAC +P RF The all-digital beamforming method, hybrid beamforming, can significantly reduce power consumption and improve system energy efficiency;

[0030] S3: The user association and beamforming design problem of a non-cellular millimeter-wave MIMO system is modeled as an energy efficiency maximization problem under user service quality constraints and power consumption constraints, specifically:

[0031]

[0032]

[0033]

[0034]

[0035]

[0036]

[0037]

[0038] in, For system energy efficiency, it is expressed as Constraint C1 is the power consumption constraint P of the access point. m Constraint C2 is the user service quality constraint r k Constraints C3 and C4 are caused by the limited number of RF chains.

[0039] When user k or beam b is not associated with or selected by access point m, the corresponding digital beamformer w will be set to 0. ρ can be represented by the zero norm of the corresponding w, and then obtained by approximating the zero norm with a one-norm.

[0040]

[0041]

[0042] Where, α k,m , It can be done

[0043]

[0044]

[0045] Iterative updates are performed, with τ1>0 and τ2>0 as fixed variables to provide stability.

[0046] Furthermore, since energy efficiency optimization introduces a fractional structure into the objective function, the Dinkelbach method is used to approximate the objective function as a linear structure.

[0047]

[0048] Among them, ν can be accessed through

[0049]

[0050] Iterative updates, when When the same optimal solution as the original problem is obtained, ν is the desired optimal energy efficiency.

[0051] Furthermore, we introduce the auxiliary variable γ. k , Furthermore, (P1) can be transformed into

[0052]

[0053]

[0054]

[0055]

[0056]

[0057]

[0058]

[0059] Furthermore, by utilizing phase ambiguity and in The first-order Taylor expansion transforms (P2) constraint C5 into a convex set.

[0060]

[0061] This transforms the original problem into a convex problem that can be solved iteratively;

[0062] Due to sparse approximation, Dinkelbach method, and Taylor expansion (α)k,m , ν、 and All of them have iterative structures, so we can let α k,m , ν、 and Common iterations are used to reduce complexity and the number of iterations. First, the MRT algorithm is used to calculate... The initial values ​​are determined, and α is initialized according to (10), (11), (13), and (3) respectively. k,m , ν and With a fixed ν, iteratively solve the original problem until convergence. After convergence, update ν based on the obtained results and repeat the above steps until convergence is achieved, thus obtaining the optimal energy efficiency.

[0063] Beneficial effects

[0064] This invention discloses an angle-domain hybrid beamforming method suitable for non-cellular millimeter-wave MIMO systems. It employs a user-centric service scheme, maximizing system energy efficiency through joint optimization of user association, beam selection, and digital beamforming. This invention considers the transmit power of the access point, static hardware power consumption, and load-based backhaul power consumption. It transforms the initial non-convex problem into an iteratively solvable convex problem through sparse approximation, the Dinkelbach method, equivalent transformation, and first-order Taylor expansion. Compared to earlier centralized system architectures and all-digital beamforming methods, this invention achieves significant improvements in system energy efficiency. Attached Figure Description

[0065] Figure 1 The present invention provides a flowchart of an angle-domain hybrid beamforming method suitable for non-cellular millimeter-wave MIMO systems.

[0066] Figure 2 This is a comparison diagram of the energy efficiency simulation of the embodiments of the present invention with a centralized system architecture and a fully digital beamforming system. Detailed Implementation

[0067] The technical solutions provided by the present invention will be described in detail below with reference to specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0068] To achieve the above objectives, the present invention provides the following technical solution:

[0069] An angle-domain hybrid beamforming method suitable for cellular-free millimeter-wave MIMO systems includes:

[0070] S1: Establish a communication model for a cellular-free millimeter-wave MIMO system based on angle-domain hybrid beamforming, specifically:

[0071] Configure M service units and K = 4 single-antenna users, equipped with N t =128 / M antenna ULA, N RF =8 / MRF access point non-cellular millimeter-wave MIMO system, where the set of access points is represented as The user set is represented as Channel state information has been obtained based on channel estimation and shared among different access points through central unit control.

[0072] Based on the Saleh-Valenzuela millimeter-wave channel model, the uplink channel between access point m and user k is modeled as follows:

[0073] h k,m =A k,m β k,m (1)

[0074] in, L=20 represents the number of independent and identically distributed paths, including one line-of-sight path and L-1 non-line-of-sight paths. For N t =128 antenna ULA steering vector, expressed as:

[0075]

[0076] λ is the carrier wavelength, D is the antenna spacing, and λ / D = 0.5. Let be the random angle of arrival for the l-th path, assumed to be within the interval of 1°. Furthermore, Let PL(dB) be the complex gain of the l-th path between access point m and user k. The path loss is calculated as: PL(dB) = α + 10βlog 10 (d)+ξ, where The specific parameters are α = 61.4 / 72.0 dB, β = 2 / 2.92 dB, and δ = 5.8 / 8.7 dB, corresponding to line-of-sight and non-line-of-sight paths, respectively. The selectable beam set for each access point is defined as follows: This indicates the beam selection at access point m. This indicates that beam b is selected, and vice versa.

[0077] In this non-cellular millimeter-wave MIMO system architecture, a user-centric service scheme is adopted, defining... This indicates the association between the user and the base station. This indicates that access point m provides services to user k, and vice versa. In summary, the signal-to-interference-plus-noise ratio (SIR) at user k is expressed as:

[0078]

[0079] in, For the corresponding digital beamformer, Let be the noise variance at user k. The achievable rate of user k is...

[0080] S2: Establish a power consumption model for a cellular-free millimeter-wave MIMO system based on angle-domain hybrid beamforming, specifically:

[0081] The total power consumption of each access point is expressed as:

[0082]

[0083] in, Static hardware power consumption, expressed as

[0084]

[0085] Among them, P lo =22.5mW, P ps =21.6mW, P DAC =200mW and P RF =31.6mW represents the power consumption of the local oscillator, phase shifter, DAC, and RF chain, respectively. The power consumption of the return line is expressed as

[0086]

[0087] in, Power dissipation, This refers to the backhaul line capacity. When only data exchange between the access point and the central unit is considered, the backhaul rate is... It is assumed to be the sum of the reachable rates of all users served. The corresponding transmission power is given by ∈ = 0.25, which represents the efficiency of the power amplifier.

[0088] S3: The user association and beamforming design problem of a non-cellular millimeter-wave MIMO system is modeled as an energy efficiency maximization problem under user service quality constraints and power consumption constraints, specifically:

[0089]

[0090]

[0091]

[0092]

[0093]

[0094]

[0095]

[0096] in, For system energy efficiency, it is expressed as Constraint C1 is the power consumption constraint P of the access point. m Constraint C2 is the user service quality constraint r k Constraints C3 and C4 are caused by the limited number of RF chains.

[0097] When user k or beam b is not associated with or selected by access point m, the corresponding digital beamformer w will be set to 0. ρ can be represented by the zero norm of the corresponding w, and then obtained by approximating the zero norm with a one-norm.

[0098]

[0099]

[0100] Where, α k,m , It can be done

[0101]

[0102]

[0103] Iterative updates are performed, with τ1>0 and τ2>0 as fixed variables to provide stability.

[0104] Furthermore, since energy efficiency optimization introduces a fractional structure into the objective function, the Dinkelbach method is used to approximate the objective function as a linear structure.

[0105]

[0106] Among them, ν can be accessed through

[0107]

[0108] Iterative updates, when When the same optimal solution as the original problem is obtained, ν is the desired optimal energy efficiency.

[0109] Furthermore, we introduce the auxiliary variable γ. k , Furthermore, (P1) can be transformed into

[0110]

[0111]

[0112]

[0113]

[0114]

[0115]

[0116]

[0117] Furthermore, by utilizing phase ambiguity and in The first-order Taylor expansion transforms (P2) constraint C5 into a convex set.

[0118]

[0119] This transforms the original problem into a convex problem that can be solved iteratively;

[0120] Let α k,m , ν、 and Common iterations are used to reduce complexity and the number of iterations. First, the MRT algorithm is used to calculate... The initial values ​​are determined, and α is initialized according to (10), (11), (13), and (3) respectively. k,m , ν and With a fixed ν, iteratively solve the original problem until convergence. After convergence, update ν based on the obtained results and repeat the above steps until convergence is achieved, thus obtaining the optimal energy efficiency.

[0121] Figure 2 The figure shows a comparison between the simulation results of this specific implementation method and the results of common system architectures and beamforming algorithms. The figure considers a centralized system architecture with M=1 and a non-cellular millimeter-wave MIMO system architecture with M=2. Each architecture employs fully digital beamforming and angle-domain hybrid beamforming, respectively. The beamforming vectors are optimized using algorithms. Considering a circular region of R=80m, users and access points are randomly and uniformly distributed, and the user service quality constraint is r. k = 1bps / Hz.

[0122] Simulation results demonstrate that the proposed algorithm significantly improves system energy efficiency. Firstly, due to the reduction in RF chains, the energy efficiency achieved by the angle-domain hybrid beamforming algorithm is significantly better than that of the all-digital beamforming algorithm. Secondly, comparing the curves for all-digital beamforming and angle-domain hybrid beamforming reveals that the non-cellular millimeter-wave MIMO system architecture, by deploying more access points, provides macro diversity, reduces communication distance, and enhances system energy efficiency. Furthermore, for angle-domain hybrid beamforming, the increased number of access points leads to a reduction in the number of phase shifters. Therefore, using angle-domain hybrid multiple access in the non-cellular millimeter-wave MIMO system architecture not only reduces transmission power consumption and RF chain-based hardware power consumption but also further reduces phase shifter-based hardware power consumption by increasing the number of access points. Thus, the energy efficiency improvement from angle-domain hybrid beamforming is more pronounced in non-cellular millimeter-wave MIMO systems.

Claims

1. An angle-domain hybrid beamforming method suitable for non-cellular millimeter-wave MIMO systems, characterized in that, Includes the following steps: S1: Establish a communication model for a cellular-free millimeter-wave MIMO system based on angle-domain hybrid beamforming, specifically: Configure M service units and K single-antenna users, equipped with N t Antenna ULA, N RF A cellular millimeter-wave MIMO system with RF access points, where the set of access points is represented as... The user set is represented as Channel state information has been obtained based on channel estimation and shared among different access points through central unit control. Based on the Saleh-Valenzuela millimeter-wave channel model, the uplink channel between access point m and user k is modeled as follows: h k,m =A k,m b k,m , (1) in, L is the number of independent and identically distributed paths. For N t The antenna ULA steering vector is represented as: Where λ is the carrier wavelength and D is the antenna spacing. Let be the random arrival angle of the l-th path; Let m be the complex gain of the l-th path between access point m and user k; The access point selects a beam in the discrete Fourier transform matrix F to simulate beamforming, thereby converting the channel to the angular domain. The selectable beam set for each access point is defined as follows: This indicates the beam selection at access point m. This indicates that beam b is selected, and vice versa; definition This indicates the association between the user and the base station. This indicates that access point m provides services to user k, and vice versa; in summary, the signal-to-interference-plus-noise ratio (SIR) at user k is expressed as... in, For the corresponding digital beamformer, Let be the noise variance at user k; let be the achievable rate of user k. S2: Establish a power consumption model for a cellular-free millimeter-wave MIMO system based on angle-domain hybrid beamforming, specifically: The total power consumption of each access point is expressed as: in, Static hardware power consumption, expressed as Among them, P lo P ps P DAC and P RF These represent the power consumption of the local oscillator, phase shifter, DAC, and RF chain, respectively. The power consumption of the return line is expressed as in, For power dissipation, The backhaul line capacity; when only considering data exchange between the access point and the central unit, the backhaul rate... Assume it is the sum of the reachable rates of all users served; Let ∈ be the corresponding transmission power, and ∈ be the efficiency of the power amplifier; S3: The user association and beamforming design problem of a non-cellular millimeter-wave MIMO system is modeled as an energy efficiency maximization problem under user service quality constraints and power consumption constraints, specifically: in, For system energy efficiency, it is expressed as Constraint C1 is the power consumption constraint P of the access point. m Constraint C2 is the user service quality constraint r k Constraints C3 and C4 are caused by the limited number of RF chains.

2. The angle-domain hybrid beamforming method for a non-cellular millimeter-wave MIMO system according to claim 1, characterized in that, When user k or beam b is not associated with or selected by access point m, the corresponding digital beamformer w will be set to 0. ρ can be represented by the zero norm of the corresponding w, and then obtained by approximating the zero norm with a one-norm. in, It can be done Iterative updates are performed, with τ1>0 and τ2>0 being fixed variables used to provide stability.

3. The angle-domain hybrid beamforming method for a non-cellular millimeter-wave MIMO system according to claim 2, characterized in that, Energy efficiency optimization introduces a fractional structure into the objective function, and uses the Dinkelbach method to approximate the objective function as a linear structure. Among them, ν can be accessed through Iterative updates, when When the same optimal solution as the original problem is obtained, ν is the optimal energy efficiency sought.

4. The angle-domain hybrid beamforming method for a non-cellular millimeter-wave MIMO system according to claim 3, characterized in that, Introducing auxiliary variables Then (P1) is transformed 5. The angle-domain hybrid beamforming method for a non-cellular millimeter-wave MIMO system according to claim 4, characterized in that, Utilizing phase ambiguity and in The first-order Taylor expansion transforms (P2) constraint C5 into a convex set. C5: This transforms the original problem into a convex problem that can be solved iteratively.

6. The angle-domain hybrid beamforming method for a non-cellular millimeter-wave MIMO system according to claim 5, characterized in that, α k,m , ν、 and To reduce complexity and the number of iterations, the two methods are combined and iterated together. Specifically, the MRT algorithm is first used to calculate... The initial values ​​are determined, and α is initialized according to (10), (11), (13), and (3) respectively. k,m , ν and With a fixed ν, iteratively solve the original problem until convergence. After convergence, update ν based on the obtained results and repeat the above steps until convergence is achieved, thus obtaining the optimal energy efficiency.

Images