Precoding method and system for wi-fi based multi-user multiple-input multiple-output system

Abstract

The application provides a precoding method and system of a Wi-Fi based multi-user multi-in multi-out system, and relates to the technical field of communication. The method comprises the following steps: receiving a plurality of weighting matrices respectively fed back by a plurality of receiving end devices; constructing an initial precoding matrix based on the weighting matrix corresponding to each of the receiving end devices; and performing signal-to-noise ratio equalization processing on the initial precoding matrix to obtain a target precoding matrix. That is, the initial precoding matrix is constructed based on the weighting matrices respectively fed back by all the receiving end devices, and the initial precoding matrix is a precoding matrix satisfying a basic multi-user interference suppression constraint. Then, the initial precoding matrix is subjected to signal-to-noise ratio equalization processing to obtain the target precoding matrix. When the data stream is subjected to weighting processing based on the target precoding matrix, the signal-to-noise ratio of the data stream can be balanced while eliminating the interference between users, and the receiving performance is improved.

Description

Technical Field

[0001] This application relates to the field of communication technology, and more specifically, to a precoding method and system for a Wi-Fi-based multi-user multiple-input multiple-output system. Background Technology

[0002] Multi-User Multiple-Input Multiple-Output (MU-MIMO) refers to a system where multiple user terminals (i.e., receivers) simultaneously utilize the same frequency domain resources. The transmitter generates a weighting matrix through precoding to cancel out interference from other users, fully utilizing the transmitter's antenna resources to significantly increase system throughput. The most crucial aspect of MU-MIMO systems is eliminating interference from other users' data streams to the current user.

[0003] Currently, in multi-user multiple-input multiple-output (MIMO) systems, the transmitting end typically uses the weighted matrix fed back by each user terminal for precoding and transmits beamforming frames to eliminate interference from other users. Among these, the mainstream precoding schemes mainly include two types: zero-forcing (ZF) and block diagonalization (BD). ZF precoding constructs a precoding matrix such that its column space is orthogonal to the channel subspace of the non-target user, thereby mathematically forcing the elimination of interference between users. BD precoding further extends to multi-stream scenarios, achieving cross-user and cross-datastream interference collaborative suppression by jointly projecting the user channels into the null space.

[0004] However, both of the above precoding methods prioritize eliminating interference between users as their primary optimization goal, without considering how to balance the signal-to-noise ratio of the data stream, resulting in poor reception performance of user terminals. Summary of the Invention

[0005] The purpose of this application is to provide a precoding method and system for a Wi-Fi-based multi-user multiple-input multiple-output system, in order to address the shortcomings of the prior art and solve the technical problems existing in the prior art.

[0006] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows: In a first aspect, embodiments of this application provide a precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system, applied to a transmitting device, the method comprising: Receive the weighted matrix fed back from multiple receiving devices; An initial precoding matrix is ​​constructed based on the weighting matrix corresponding to each of the receiving devices. The initial precoding matrix is ​​subjected to signal-to-noise ratio equalization to obtain the target precoding matrix.

[0007] Optionally, the step of performing signal-to-noise ratio equalization on the initial precoding matrix to obtain the target precoding matrix includes: Determine the transformation matrix, and perform signal-to-noise ratio equalization on the initial precoding matrix based on the transformation matrix to obtain the target precoding matrix.

[0008] Optionally, determining the transformation matrix and performing signal-to-noise ratio equalization on the initial precoding matrix based on the transformation matrix to obtain the target precoding matrix includes: Determine the Hada code transformation matrix corresponding to each of the receiving devices, and determine the initial precoding submatrix corresponding to each receiving device in the initial precoding matrix; The target precoding matrix is ​​obtained based on the Hada code transformation matrix corresponding to each of the receiving devices and the initial precoding submatrix corresponding to the initial precoding moment.

[0009] Optionally, determining the Hada code transformation matrix corresponding to each of the receiving devices, and determining the initial precoding submatrix corresponding to the receiving device in the initial precoding matrix, includes: Determine the data stream dimension and the number of data stream columns corresponding to the receiving device; Based on the data stream dimension corresponding to the receiving device, determine the Hada code transformation matrix corresponding to the receiving device; Based on the number of data stream columns corresponding to the receiving device, the initial precoding submatrix corresponding to the receiving device in the initial precoding matrix is ​​determined.

[0010] Optionally, obtaining the target precoding matrix based on the Hada code transformation matrix corresponding to each of the receiving devices and the corresponding initial precoding submatrix in the initial precoding moments includes: Iterate through all receiving devices. For the current receiving device, multiply the initial precoding submatrix corresponding to the current receiving device in the initial precoding moment with the Hada code matrix corresponding to the current receiving device, and use the product as the precoding submatrix corresponding to the current receiving device. After the traversal is completed, the target precoding matrix is ​​obtained based on the precoding submatrices corresponding to all receiving devices.

[0011] Optionally, determining the transformation matrix and performing signal-to-noise ratio equalization on the initial precoding matrix based on the transformation matrix to obtain the target precoding matrix includes: Obtain the signal-to-noise ratio (SNR) matrix fed back by each of the receiving devices, and construct the approximate equivalent channel matrix corresponding to each of the receiving devices based on the weighting matrix and SNR matrix corresponding to each of the receiving devices. The approximate equivalent channel matrix corresponding to each of the receiving devices is decomposed by lattice basis reduction to obtain the channel transformation matrix corresponding to each of the receiving devices. The target precoding matrix is ​​obtained based on the channel transformation matrix corresponding to each of the receiving devices and the initial precoding submatrix corresponding to each of the receiving devices.

[0012] Optionally, obtaining the target precoding matrix based on the channel transformation matrix corresponding to each of the receiving devices and the initial precoding submatrix corresponding to each of the receiving devices includes: Traverse all receiving devices. For the current receiving device, multiply the initial precoding submatrix corresponding to the current receiving device in the initial precoding moment with the channel transformation matrix corresponding to the current receiving device, and use the product as the precoding submatrix corresponding to the current receiving device. After the traversal is completed, the target precoding matrix is ​​obtained based on the precoding submatrices corresponding to all receiving devices.

[0013] Optionally, the initial precoding matrix is ​​subjected to signal-to-noise ratio equalization to obtain the target precoding matrix, including: The number of data stream columns corresponding to each of the receiving devices is determined respectively, and the initial precoding submatrix corresponding to each of the receiving devices in the initial precoding moment is determined according to the number of data stream columns corresponding to each of the receiving devices. The initial precoding submatrix corresponding to each of the receiving devices is sequentially decomposed into QR decomposition to obtain the orthogonal matrix corresponding to each of the receiving devices. The target precoding matrix is ​​obtained based on the orthogonal matrix corresponding to each of the receiving devices.

[0014] Optionally, the method further includes: Based on the target precoding matrix, the target data stream to be transmitted is weighted to generate multiple transmission signals, and the multiple transmission signals are sent to each of the receiving devices.

[0015] Secondly, embodiments of this application also provide a Wi-Fi-based multi-user multiple-input multiple-output system, the system comprising: a transmitting end device and multiple receiving end devices; The receiving device is used to feed back the weighting matrix and the signal-to-noise ratio matrix to the transmitting device respectively. The transmitting device is configured to execute the steps of the method described above based on the weighted matrix and signal-to-noise ratio matrix fed back by each of the receiving devices.

[0016] The beneficial effects of this application are: This application provides a precoding method and system for a Wi-Fi-based multi-user multiple-input multiple-output (MIMO) system. In this application, an initial precoding matrix is ​​constructed based on the weighting matrices fed back by all receiving devices. This initial precoding matrix satisfies basic multi-user interference suppression constraints. Then, the initial precoding matrix is ​​subjected to signal-to-noise ratio (SNR) equalization processing to obtain a target precoding matrix. This allows the data stream to be weighted based on the target precoding matrix, thereby eliminating interference between users and balancing the SNR of the data stream, thus improving reception performance.

[0017] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A schematic diagram of the multi-user weighted matrix feedback process in a Wi-Fi system provided by existing technology; Figure 2 A block diagram of the transmitter structure of a Beamformer in a Wi-Fi system provided by existing technology; Figure 3 A schematic diagram of a Wi-Fi-based multi-user multiple-input multiple-output system is provided as an embodiment of this application. Figure 4 A flowchart illustrating a precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system provided in this application embodiment; Figure 5 A flowchart illustrating another precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system provided in this application embodiment; Figure 6 A flowchart illustrating another precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system provided in this application embodiment; Figure 7 A flowchart illustrating another precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system provided in this application embodiment; Figure 8 A schematic diagram of a lattice formed by different basis vectors provided in the embodiments of this application; Figure 9A flowchart illustrating another precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system provided in this application embodiment; Figure 10 A flowchart illustrating another precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system provided in this application embodiment; Figure 11 The diagram illustrates the sensitivity of different precoding schemes provided in the embodiments of this application under ZF equalization. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0021] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0022] First, the technical terms used in this application will be introduced.

[0023] 1. Multiple Input Multiple Output (MIMO) refers to a system where both the transmitting and receiving ends have multiple antennas, utilizing array gain and diversity gain to extend the system's throughput.

[0024] 2. Multi-User Multiple-Input Multiple-Output (MU-MIMO) refers to the simultaneous use of the same frequency domain resources by multiple users. The transmitting device (Beamformer) generates a weighted matrix through precoding to cancel out interference from other users, making full use of the transmitting antenna resources of the transmitting device to achieve the purpose of multiplying the system throughput.

[0025] The background technology involved in this application will be introduced below.

[0026] Starting with 802.11ac, the Wi-Fi protocol introduced MU-MIMO functionality, which utilizes Transmit Beamforming (TxBF) technology. First, the transmitting device sends a Non-Data Packet (NDP) for channel sounding. Second, the receiving device uses the training sequence in the NDP frame to estimate the wireless channel, obtaining the channel coefficient matrix H. It then processes the channel coefficient matrix H to obtain a weighted matrix V, and finally compresses the weighted matrix V and feeds it back to the transmitting device. Each user repeats this process. The multi-user weighted matrix feedback process described in 802.11ac is as follows: Figure 1 As shown.

[0027] Therefore, the most important aspect of MU-MIMO is how to eliminate interference from other users' data streams to the current user (Inter-User Interference, or IUI). The transmitting device uses the weighted matrix V fed back by each user for precoding and sends beamforming frames to eliminate interference from other users.

[0028] Currently, commonly used MU-MIMO precoding techniques include: Zero Forcing (ZF) precoding and Block Diagonalization (BD) precoding. Other methods are mostly unsuitable for Wi-Fi systems due to their high complexity.

[0029] refer to Figure 2 The diagram shown is a structural block diagram of the transmitting device. The core improvement of this application focuses only on the precoding module.

[0030] For example, in a MU-MIMO system, the transmitting device has 4 transmit antennas, the receiving device 1 has 2 receive antennas, the number of data streams is 2, and the feedback weighting matrix is... Receiver 2 has 2 receiving antennas, 2 data streams, and a feedback weighting matrix of... For example, the ZF precoding formula in the Wi-Fi system is shown in formula (1) below: (1) It is understandable that for two receiving devices (only receiving device 1 is discussed here, and receiving device 1 and receiving device 2 are equivalent), all off-diagonal elements are forced to zero, that is, the target precoding matrix is ​​as shown in the following formula (2):

[0031] The formula for BD precoding in a Wi-Fi system is shown in formula (3) below: (3) in, Indicates the solution The null space, i.e. The solved matrix and They are mutually orthogonal.

[0032] For two receiving devices, after BD precoding, the target precoding matrix is ​​as shown in the following formula (4): (4) Compared to ZF precoding, BD precoding retains some off-diagonal element information, which is then processed by the equalizer at the receiving end.

[0033] When the transmitting device uses ZF precoding, for a beamforming frame, the receiving vector Y of the receiving device 1 can be expressed as shown in the following formula (5): (5) Clearly, this data stream does not belong to receiving device 1. It is canceled out. At this point, the approximate equivalent channel matrix can be expressed as shown in the following formula (6): (6) Assuming that the receiving device uses ZF equalization, based on existing derivations, the signal-to-noise ratio (SNR) obtained by the receiving device on each data stream is as shown in the following formula (7): (7) Therefore, when the transmitting device uses ZF precoding, the signal-to-noise ratio and the approximate equivalent channel matrix obtained by the receiving device under ZF equalization for each data stream are... The singular values ​​are relatively correlated, and also correlated with the right matrix. Applying BD precoding to the transmitting device yields similar conclusions, all related to the approximately equivalent channel matrix. The singular values ​​are relatively correlated, only the right matrix is ​​different.

[0034] Because different data streams in a Wi-Fi system use the same modulation and coding scheme (MCS), but the different data streams in the above formula have different signal-to-noise ratios, due to... If the numbers are real numbers arranged in descending order, then the system performance is obviously limited by the data stream with a poor SNR.

[0035] BD precoding outperforms ZF precoding when the receiving device has more than one receiving antenna. This is because, under the condition that the singular values ​​are the same, the right matrix of BD precoding balances the singular values, which is equivalent to improving the SNR of the originally poor data stream.

[0036] Although BD precoding can yield good results, its complexity increases dramatically as the number of antennas on the transmitting device and the number of users increases, as shown in Table 1.

[0037]

[0038] Therefore, the precoding methods of existing MU-MIMO systems are generally based on the fact that the transmitting device knows the complete channel coefficient matrix H, but Wi-Fi transmitting devices can generally only receive the weighting matrix V and signal-to-noise ratio matrix fed back by the user.

[0039] Secondly, since the Wi-Fi system uses the same modulation scheme for all data streams of the same user, the SNR of each data stream must be approximately equal to avoid limiting throughput. Existing precoding schemes all prioritize the complete elimination of interference between users, without considering how to balance the signal-to-noise ratio of the data streams, resulting in poor reception performance of user terminals. To address the aforementioned issues, this application proposes a precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system. The method involves constructing an initial precoding matrix based on the weighted matrix corresponding to each receiving device, and then performing signal-to-noise ratio equalization on the initial precoding matrix to obtain the target precoding matrix. Compared to traditional precoding algorithms, this method can improve receiver sensitivity.

[0040] The following embodiments will be used to explain in detail the structural diagram of the Wi-Fi-based multi-user multiple-input multiple-output system proposed in this application.

[0041] Optionally, refer to Figure 3 The diagram shown is a structural schematic of a Wi-Fi-based multi-user multiple-input multiple-output system provided in this application. Figure 3 As shown, the system includes: a transmitting device and multiple receiving devices, such as receiving device 1, receiving device 2, ..., receiving device n.

[0042] Among them, the receiving end device is used to feed back the weighting matrix and the signal-to-noise ratio matrix to the transmitting end device respectively; The transmitting device is used to execute the precoding method steps provided in the following embodiments based on the weighting matrix and signal-to-noise ratio matrix fed back by each receiving device.

[0043] For example, the sending device is a base station, and the receiving device is a mobile phone, an IoT terminal, etc.

[0044] The receiving device is used to estimate the local channel response and calculate and feed back two key matrices: the weighting matrix and the signal-to-noise ratio (SNR) matrix. The weighting matrix is ​​the conjugate transpose (WH) of the precoding matrix or the codebook index corresponding matrix, which is used to match the precoding at the transmitting end. The signal-to-noise ratio (SNR) matrix is ​​a vector / diagonal matrix of the SNR estimates for each spatial stream.

[0045] The transmitting device is used to calculate the target precoding matrix based on the weighting matrix and signal-to-noise ratio matrix fed back by each receiving device, and to use the target precoding matrix to perform weighted processing on the target data stream to be transmitted to obtain multiple transmission signals. The multiple transmission signals are then transmitted to each receiving device, thereby eliminating interference between users and balancing the signal-to-noise ratio of the data stream to improve receiving performance.

[0046] The following embodiments will be used to explain the specific implementation process and technical effects of the precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system involved in this application.

[0047] Optionally, refer to Figure 4 The diagram shown is a flowchart illustrating the precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system provided in this application, applied to the above-mentioned... Figure 3 The transmitting device in, such as Figure 4 As shown, the method includes: S101, Receive the weighted matrix fed back by multiple receiving devices respectively.

[0048] For example, the weighting matrices fed back by receiving device 1 and receiving device 2 are respectively: , ,in, , .

[0049] In this weighted matrix, each column corresponds to one data stream, i.e., the weighted matrix... Includes: a data stream, a weighted matrix Includes three data streams.

[0050] S102. Construct the initial precoding matrix based on the weighting matrix corresponding to each receiving device.

[0051] The initial precoding matrix refers to the precoding matrix constructed by the transmitting device based solely on the weighted matrices (such as orthogonal codebook matrix, stream alignment matrix, or low-rank projection matrix) fed back by each receiving device, without incorporating the signal-to-noise ratio feedback information from the receiving device for optimization, and which satisfies the basic multi-user interference suppression constraints.

[0052] In one feasible approach, the weighting matrices corresponding to all receiving devices are sequentially concatenated horizontally to construct the initial weighting matrix, i.e., the initial weighting matrix is... ; Then, the pseudo-inverse of the initial weighted matrix is ​​solved to obtain the zero-forcing precoding matrix, also known as the initial precoding matrix, as shown in the following formula (8): (8) S103. Perform signal-to-noise ratio equalization on the initial precoding matrix to obtain the target precoding matrix.

[0053] Among them, signal-to-noise ratio equalization is a process that uses the quantized SINR feedback value as an optimization weight or constraint to perform stream-level fine calibration of the precoder.

[0054] In one feasible approach, the initial precoding matrix can be subjected to signal-to-noise ratio equalization based on a predetermined transformation matrix to obtain the target precoding matrix, thereby achieving signal-to-noise ratio balance for the data stream.

[0055] In summary, this application provides a precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system. In this application, an initial precoding matrix is ​​constructed based on the weighting matrices fed back by all receiving devices. This initial precoding matrix satisfies the basic multi-user interference suppression constraints. Then, the initial precoding matrix is ​​subjected to signal-to-noise ratio equalization processing to obtain a target precoding matrix. This allows the data stream to be weighted based on the target precoding matrix, thereby eliminating interference between users while balancing the signal-to-noise ratio of the data stream and improving reception performance.

[0056] Optionally, step S103 above includes: Determine the transformation matrix, and perform signal-to-noise ratio equalization on the initial precoding matrix based on the transformation matrix to obtain the target precoding matrix.

[0057] In one feasible approach, to ensure all receiving devices achieve the same signal-to-noise ratio (SNR) for the data stream, the Hadamard transform matrix can be used as a transform matrix. The initial precoding matrix is ​​then subjected to SNR equalization based on this transform matrix to obtain the target precoding matrix. Therefore, by multiplying by the Hadamard matrix on top of ZF precoding and utilizing the energy diffusion characteristics of the Hadamard matrix, channel statistical features are projected into the sparse domain, resulting in a uniform distribution of quantization errors and improved feedback accuracy for low SNR users.

[0058] Optionally, refer to Figure 5As shown, the steps described above, including determining the transformation matrix and performing signal-to-noise ratio equalization on the initial precoding matrix based on the transformation matrix to obtain the target precoding matrix, include: S201. Determine the Hada code transformation matrix corresponding to each receiving device, and determine the initial precoding submatrix corresponding to the receiving device in the initial precoding matrix.

[0059] S202. Obtain the target precoding matrix based on the Hada code transformation matrix corresponding to each receiving device and the initial precoding submatrix corresponding to the initial precoding moment.

[0060] In one feasible approach, the Hadamard transformation matrix corresponding to each receiving device is determined separately. For example, the Hadamard transformation matrix corresponding to receiving device 1 is... Then, the initial precoding submatrix corresponding to each receiving device in the initial precoding moments is determined. For example, the initial precoding submatrix corresponding to receiving device 1 in the initial precoding moments is... Then, based on the Hada code transformation matrix corresponding to each receiving device and the initial precoding submatrix corresponding to the initial precoding moment, the target precoding matrix is ​​calculated, thereby realizing the signal-to-noise ratio equalization processing of the initial precoding matrix.

[0061] Optionally, refer to Figure 6 As shown, step S201 above includes: S301. Determine the data stream dimension and the number of data stream columns corresponding to the receiving device.

[0062] S302. Determine the Hada code transformation matrix corresponding to the receiving device based on the data stream dimension of the receiving device.

[0063] S303. Determine the initial precoding submatrix corresponding to the receiving device in the initial precoding matrix based on the number of data stream columns corresponding to the receiving device.

[0064] For example, the data stream dimension corresponding to receiving device 1 is 1 and the number of data stream columns is the first column. Therefore, based on the data stream dimension corresponding to receiving device 1, the Hada code transformation matrix corresponding to receiving device 1 can be obtained as (1x1).

[0065] For example, if the data stream dimension of receiving device 2 is 3 and the number of columns in the data stream is the second to fourth columns, then based on the data stream dimension of receiving device 2, the Hada code transformation matrix corresponding to receiving device 1 can be obtained as 3x3, i.e. .

[0066] Since different receiving devices may receive different numbers of data streams (generally less than or equal to 4), the Hadamard transformation matrices of different orders are given below: .

[0067] Then, based on the number of data stream columns corresponding to the receiving device, the initial precoding submatrix corresponding to the receiving device in the initial precoding matrix is ​​determined. For example, the initial precoding submatrix corresponding to receiving device 2 in the initial precoding matrix is: .

[0068] Optionally, the above steps, based on the Hada code transformation matrix corresponding to each receiving device and the corresponding initial precoding submatrix in the initial precoding moments, yield the target precoding matrix, including: Iterate through all receiving devices. For the current receiving device, multiply the initial precoding submatrix corresponding to the current receiving device in the initial precoding moment with the Hada code matrix corresponding to the current receiving device, and use the product as the precoding submatrix corresponding to the current receiving device. After the traversal is completed, the target precoding matrix is ​​obtained based on the precoding submatrices corresponding to all receiving devices.

[0069] In this embodiment, all receiving devices are traversed. For the current receiving device, such as receiving device 1, the initial precoding submatrix corresponding to the current receiving device in the initial precoding moment is right-multiplied with the Hadamard matrix corresponding to the current receiving device, and the resulting product is used as the precoding submatrix corresponding to the current receiving device.

[0070] For example, the precoding submatrix corresponding to receiver device 1 is: ; The precoding submatrix corresponding to receiver device 2 is: ; After the traversal is complete, the precoding submatrices corresponding to all receiving devices are merged to obtain the target precoding matrix. For example, the precoding submatrices corresponding to receiving device 1 are merged. Precoding sub-matrix corresponding to receiver device 2 After merging, the target precoding matrix is ​​obtained as follows: .

[0071] After ZF_HADA precoding, the equivalent channel obtained by the receiving device is: ; The signal-to-noise ratio of each data stream is:

[0072] Therefore, based on the above precoding method, the signal-to-noise ratio of the data stream of the receiving device has been leveled to be completely consistent, achieving signal-to-noise ratio equalization.

[0073] Optionally, refer to Figure 7 As shown, the above steps determine the transformation matrix, and perform signal-to-noise ratio equalization processing on the initial precoding matrix based on the transformation matrix to obtain the target precoding matrix, including: S401. Obtain the signal-to-noise ratio (SNR) matrix fed back by each receiving device, and construct the approximate equivalent channel matrix corresponding to each receiving device based on the weighting matrix and SNR matrix corresponding to each receiving device.

[0074] S402. Perform lattice basis reduction decomposition on the approximate equivalent channel matrix corresponding to each receiving device to obtain the channel transformation matrix corresponding to each receiving device.

[0075] S403. Based on the channel transformation matrix and the initial precoding submatrix corresponding to each receiving device, the target precoding matrix is ​​obtained.

[0076] It should be noted that lattice theory has been extensively explored and widely applied in the field of digital communication encoding and decoding. A lattice is a discrete geometric structure defined over a finite field, and is a subset of a linear space consisting of n-dimensional vectors. The definition of a complex field lattice is as follows: set up It is a set of linearly independent vectors in an n-dimensional complex integer linear space. L represents the set of complex integers, and the lattice L is composed of this vector group. The set of all linear combinations, i.e.: ; in, The vectors are called a basis of the lattice L. It is a vector composed of complex integer weights.

[0077] For a lattice L, there are many possible bases; any matrix obtained from B through elementary column transformations can be used as its base. The product of multiple elementary transformation matrices is actually equivalent to a unitary modular transformation matrix T, where the elements of matrix T are complex integers and its determinant satisfies... That is, when T is a unitary modular matrix, B and BT produce the same lattice.

[0078] The Lattice Reduction (LR) algorithm optimizes the basis matrix B by performing iterative modulus comparisons and reductions on each column of the generated matrix, making the columns of the transformed matrix BT as orthogonal and of equal modulus as possible.

[0079] The LR algorithm decomposes matrix B into an approximately orthogonal matrix. And a unitary modular matrix T with values ​​of complex integers, i.e. .

[0080] According to the LR algorithm, If the columns of a matrix are as orthogonal and of equal modulus as possible, the singularity of the matrix can be improved. Figure 8 This is an example of the LR algorithm. (dashed line) and (Solid lines) form the same black grid. Depend on Obtained using the LR algorithm. The new basis vectors can be seen. compared to For those that are more orthogonal and have equal moduli, their numerical relationship is as follows: .in, .

[0081] Therefore, this application proposes that: an approximate equivalent channel matrix can be constructed from the initial precoding matrix Wzf after ZF precoding, based on the weighting matrix and signal-to-noise ratio matrix fed back by the receiving device. For the approximate equivalent channel matrix After LR processing, the SNR of the data stream on the receiving device side is processed to be approximately equal.

[0082] In one feasible approach, it can be based on the weighted matrix corresponding to each receiving device. and signal-to-noise ratio matrix Approximate equivalent channel matrices are constructed for each receiving device. For example, the weighted matrix corresponding to each receiving device is calculated. and signal-to-noise ratio matrix Multiplying them together, we construct an approximate equivalent channel matrix. Then, for each receiving device, the approximate equivalent channel matrix is ​​calculated separately. Perform lattice basis reduction decomposition to obtain the channel transformation matrix corresponding to each receiving device. Finally, based on the channel transformation matrix corresponding to each receiving device... The target precoding matrix is ​​calculated by taking the initial precoding submatrix corresponding to each receiving device.

[0083] Optionally, refer to Figure 9 As shown, step S403 above includes: S501. Traverse all receiving devices. For the current receiving device, multiply the initial precoding submatrix corresponding to the current receiving device in the initial precoding moment with the channel transformation matrix corresponding to the current receiving device, and use the product as the precoding submatrix corresponding to the current receiving device.

[0084] S502. After the traversal is completed, the target precoding matrix is ​​obtained based on the precoding sub-matrices corresponding to all receiving devices.

[0085] In this embodiment, all receiving devices are traversed. For the current receiving device, such as receiving device 1, the initial precoding submatrix corresponding to the current receiving device in the initial precoding moment is right-multiplied by the channel transformation matrix corresponding to the current receiving device, and the resulting product is used as the precoding submatrix corresponding to the current receiving device.

[0086] For example, the channel transformation matrix corresponding to receiver device 1 is: ; The channel transformation matrix corresponding to receiver device 2 is: ; The channel transformation matrix corresponding to receiving device 1 is respectively and the channel transformation matrix corresponding to receiver device 2 Perform LR decomposition to obtain the channel transformation matrix. and ;in, , .

[0087] Then, iterate through all receiver devices, and multiply the initial precoding submatrix corresponding to the current receiver device in the initial precoding moment by the channel transform matrix corresponding to that receiver device on the right. Use the product as the precoding submatrix corresponding to the current receiver device, as shown below: The precoding submatrix corresponding to receiver device 1 is: ; The precoding submatrix corresponding to receiver device 2 is: ; After the traversal is complete, the precoding submatrices corresponding to all receiving devices are merged to obtain the target precoding matrix. For example, the precoding submatrices corresponding to receiving device 1 are merged. Precoding sub-matrix corresponding to receiver device 2 After merging, the target precoding matrix is ​​obtained as follows: .

[0088] Optionally, refer to Figure 10 As shown, step S102 above includes: S601. Determine the number of data stream columns corresponding to each receiving device, and determine the initial precoding submatrix corresponding to each receiving device in the initial precoding moment based on the number of data stream columns corresponding to each receiving device.

[0089] S602. Perform QR decomposition on the initial precoding submatrix corresponding to each receiving device in sequence to obtain the orthogonal matrix corresponding to each receiving device.

[0090] S603. Obtain the target precoding matrix based on the orthogonal matrix corresponding to each receiving device.

[0091] It should be noted that this application proposes to further orthogonalize the initial precoding matrix Wzf after ZF precoding, and the singular values ​​after orthogonalization will also tend to be the same. The general orthogonalization scheme is the Gram-Schmidt orthogonalization algorithm. Since the general QR decomposition algorithm also implicitly contains the Gram-Schmidt orthogonalization algorithm or can achieve orthogonalization itself, this embodiment proposes a precoding algorithm (ZF_QR) that can reduce the computational cost of the BD algorithm but achieve the same performance as the traditional BD precoding algorithm.

[0092] For two receiving devices, after ZF_QR precoding: .

[0093] Therefore, after ZF_QR precoding, Beamformee receives a block-like upper triangular matrix, which also helps to balance the signal-to-noise ratio of the data stream. The elements in the lower left corner of the matrix are 0 because the first column of Wzf is normalized, so QR decomposition does not change the values ​​of all elements in the first column.

[0094] In this embodiment, the number of data streams corresponding to each receiving device can also be determined, and the initial precoding submatrix corresponding to each receiving device in the initial precoding moment can be determined according to the number of data streams corresponding to each receiving device. Then, QR decomposition is performed on the initial precoding submatrix corresponding to each receiving device in turn to obtain the orthogonal matrix corresponding to each receiving device. The orthogonal matrices corresponding to each receiving device are combined to calculate the target precoding matrix.

[0095] Optionally, the method further includes: Based on the target precoding matrix, the target data stream to be transmitted is weighted to generate multiple transmission signals, which are then sent to each receiving device.

[0096] In one feasible approach, the target data stream to be transmitted can be weighted based on the target precoding matrix to generate multiple transmit signals, which are then sent to each receiving device. This eliminates interference between users while also balancing the signal-to-noise ratio of the data stream, thereby improving reception performance.

[0097] Optionally, refer to Figure 11 The figure shows the sensitivity diagrams of different precoding schemes under ZF equalization. Figure 11 As shown, the sensitivity of each precoder under ZF equalization at the receiving end device is as follows: ZF_HADA has the best sensitivity, followed by ZF_MLR, ZF_QR is close to BD sensitivity, and ZF is the worst.

[0098] Therefore, the precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system proposed in this application solves the problem that the complexity of BD precoding increases sharply when there are many users and many system antennas. It proposes several precoding schemes with lower complexity than BD and better sensitivity. At the same time, starting from the goal of balancing the signal-to-noise ratio of the data stream, it proposes an improved lattice basis reduction algorithm that makes the modulus approximately equal and applies it to the precoding. This eliminates the need for the transmitter and receiver to pass the lattice basis reduction transformation matrix, making it suitable for practical equipment.

[0099] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0100] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from the spirit or essential characteristics of this application. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within this application. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A precoding method for a Wi-Fi-based multi-user multiple-input multiple-output system, characterized in that, Applied to a transmitting device, the method includes: Receive the weighted matrix fed back from multiple receiving devices; An initial precoding matrix is ​​constructed based on the weighting matrix corresponding to each of the receiving devices. The initial precoding matrix is ​​subjected to signal-to-noise ratio equalization to obtain the target precoding matrix.

2. The method according to claim 1, characterized in that, The step of performing signal-to-noise ratio equalization on the initial precoding matrix to obtain the target precoding matrix includes: Determine the transformation matrix, and perform signal-to-noise ratio equalization on the initial precoding matrix based on the transformation matrix to obtain the target precoding matrix.

3. The method according to claim 2, characterized in that, The process of determining the transformation matrix and performing signal-to-noise ratio equalization on the initial precoding matrix based on the transformation matrix to obtain the target precoding matrix includes: Determine the Hada code transformation matrix corresponding to each of the receiving devices, and determine the initial precoding submatrix corresponding to each receiving device in the initial precoding matrix; The target precoding matrix is ​​obtained based on the Hada code transformation matrix corresponding to each of the receiving devices and the initial precoding submatrix corresponding to the initial precoding moment.

4. The method according to claim 3, characterized in that, The step of determining the Hada code transformation matrix corresponding to each of the receiving devices, and determining the initial precoding submatrix corresponding to each receiving device in the initial precoding matrix, includes: Determine the data stream dimension and the number of data stream columns corresponding to the receiving device; Based on the data stream dimension corresponding to the receiving device, determine the Hada code transformation matrix corresponding to the receiving device; Based on the number of data stream columns corresponding to the receiving device, the initial precoding submatrix corresponding to the receiving device in the initial precoding matrix is ​​determined.

5. The method according to claim 3, characterized in that, The step of obtaining the target precoding matrix based on the Hada code transformation matrix corresponding to each of the receiving devices and the corresponding initial precoding submatrix in the initial precoding moment includes: Iterate through all receiving devices. For the current receiving device, multiply the initial precoding submatrix corresponding to the current receiving device in the initial precoding moment with the Hada code matrix corresponding to the current receiving device, and use the product as the precoding submatrix corresponding to the current receiving device. After the traversal is completed, the target precoding matrix is ​​obtained based on the precoding submatrices corresponding to all receiving devices.

6. The method according to claim 2, characterized in that, The process of determining the transformation matrix and performing signal-to-noise ratio equalization on the initial precoding matrix based on the transformation matrix to obtain the target precoding matrix includes: Obtain the signal-to-noise ratio (SNR) matrix fed back by each of the receiving devices, and construct the approximate equivalent channel matrix corresponding to each of the receiving devices based on the weighting matrix and SNR matrix corresponding to each of the receiving devices. The approximate equivalent channel matrix corresponding to each of the receiving devices is decomposed by lattice basis reduction to obtain the channel transformation matrix corresponding to each of the receiving devices. The target precoding matrix is ​​obtained based on the channel transformation matrix corresponding to each of the receiving devices and the initial precoding submatrix corresponding to each of the receiving devices.

7. The method according to claim 6, characterized in that, The step of obtaining the target precoding matrix based on the channel transformation matrix corresponding to each of the receiving devices and the initial precoding submatrix corresponding to each of the receiving devices includes: Traverse all receiving devices. For the current receiving device, multiply the initial precoding submatrix corresponding to the current receiving device in the initial precoding moment with the channel transformation matrix corresponding to the current receiving device, and use the product as the precoding submatrix corresponding to the current receiving device. After the traversal is completed, the target precoding matrix is ​​obtained based on the precoding submatrices corresponding to all receiving devices.

8. The method according to claim 1, characterized in that, The initial precoding matrix is ​​subjected to signal-to-noise ratio equalization to obtain the target precoding matrix, including: The number of data stream columns corresponding to each of the receiving devices is determined respectively, and the initial precoding submatrix corresponding to each of the receiving devices in the initial precoding moment is determined according to the number of data stream columns corresponding to each of the receiving devices. The initial precoding submatrix corresponding to each of the receiving devices is sequentially decomposed into QR decomposition to obtain the orthogonal matrix corresponding to each of the receiving devices. The target precoding matrix is ​​obtained based on the orthogonal matrix corresponding to each of the receiving devices.

9. The method according to any one of claims 1-8, characterized in that, The method further includes: Based on the target precoding matrix, the target data stream to be transmitted is weighted to generate multiple transmission signals, and the multiple transmission signals are sent to each of the receiving devices.

10. A Wi-Fi-based multi-user multiple-input multiple-output system, characterized in that, The system includes: a transmitting device and multiple receiving devices; The receiving device is used to feed back the weighting matrix and the signal-to-noise ratio matrix to the transmitting device respectively. The transmitting device is configured to perform the method steps of any one of claims 1-9 based on the weighting matrix and signal-to-noise ratio matrix fed back by each of the receiving devices.

Images