CN108462518A - Data transmission method and device based on user's scheduling - Google Patents

Abstract

The embodiment of the present invention provides a data transmission method and device based on user scheduling, the method includes: based on the obtained information of the user terminal equipment, determine the combination of the corresponding data layer from the candidate multi-user pairing set; iteratively determine A combined precoding matrix of the data layers; determining a user scheduling result according to the precoding matrix, and performing data transmission to a corresponding user terminal device according to the user scheduling result. In the embodiment of the present invention, the inversion operation of the matrix is saved, the complexity of the calculation is greatly reduced, and the efficiency and speed of determining the precoding matrix are improved; moreover, the combination of multiple data layers for multi-user pairing can be determined quickly enough The precoding matrix does not need to reduce the system capacity. Therefore, the present invention can improve the efficiency of multi-user scheduling without affecting the system capacity, speed up the response to user terminal equipment as a whole, and improve user experience.

Description

Data transmission method and device based on user scheduling

technical field

The present invention relates to the technical field of wireless communication, in particular, the present invention relates to a data transmission method and device based on user scheduling.

Background technique

In a mobile communication system using a multiple-antenna MIMO (Multiple Input Multiple Output, Multiple Input Multiple Output) technology, channels between different users have a certain degree of spatial isolation. Utilizing the isolation of spatial channels between different users, the same time/frequency/code resources can be used to simultaneously send signals for multiple users. In a MIMO system, these ways of simultaneously sending signals for multiple users using the same time/frequency/code resource are called MU (Multi-User, multi-user) MIMO technology.

According to the MIMO theory, the capacity of the MIMO system under MU-MIMO is proportional to the number of base station antennas and the total number of multi-user receiving antennas. Through MU-MIMO, the system capacity can be significantly improved by utilizing the spatial channel isolation between different users. However, in an actual system, the spatial channels between multiple users using the same resource are not completely isolated, and there is mutual interference, which leads to a reduction in the capacity of the MU-MIMO system. In the downlink system, the multi-user MU-MIMO precoding technology can be used to suppress or eliminate the interference between multiple users, thereby improving the capacity of the MU-MIMO system.

The existing MU-MIMO wireless communication system, such as LTE (Long Term Evolution, long-term evolution)/LTE-A system is a typical multi-antenna wireless communication system, and the number of downlink antenna ports in the traditional LTE/LTE-A system is generally 2 or 4 or 8. In existing MU-MIMO wireless communication systems, MU-MIMO scheduling combined with MU-MIMO precoding technology is one of the key technologies. MU-MIMO scheduling determines the users for paired transmission on each time, frequency, and code resource in MU-MIMO, and determines the MCS (Modulation and Coding Scheme, modulation and coding scheme) for each user and the precoding matrix for each user's MU-MIMO transmission . The data transmission method based on user scheduling in the classic MU-MIMO system usually uses a search algorithm to search for all possible candidate multi-user pairing sets in MU-MIMO. Generally, for each searched MU-MIMO possible candidate multiple User pairing sets need to use MU-MIMO precoding technology to calculate the corresponding precoding matrix; perform user scheduling according to the calculated precoding matrix.

In the existing MU-MIMO precoding matrix calculation, high-complexity operations such as matrix inversion and matrix multiplication are involved. And as the number of antennas and the number of paired layers increase, the computational complexity of the MU-MIMO precoding matrix will increase significantly.

However, wireless communication systems including base station prototypes with 32 antennas, 64 antennas and 128 antennas have appeared, such as the introduction of FD (Full-Dimensional, omnidirectional)-MIMO systems based on AAS (Active Antenna System) systems After the Massive (large-scale) MIMO technology, the number of transmitting antennas in the wireless communication system may reach 16, 32, 64 or 128, etc., or even more. In the AAS MIMO system, the number of antennas of the base station is much larger than the number of antennas of a single terminal. In addition, operators have also proposed that base stations need to be equipped with a large number of antennas. Therefore, the number of base station antennas in actual systems will continue to increase. If the existing multi-user scheduling method is still used, the computational complexity of the MU-MIMO precoding matrix in this method will continue to increase, which will inevitably greatly reduce the efficiency of determining the precoding matrix, resulting in a reduction in the efficiency of multi-user scheduling , it is easy to greatly increase the response delay to the user terminal equipment, and it is easy to reduce the user experience.

Contents of the invention

Aiming at the shortcomings of the existing methods, the present invention proposes a data transmission method and device based on user scheduling to solve the problems in the prior art that the calculation complexity of the precoding matrix is high and the efficiency of determining the precoding matrix is low , so as to reduce the computational complexity of the precoding matrix and improve the efficiency of determining the precoding matrix.

According to the first aspect, an embodiment of the present invention provides a data transmission method based on user scheduling, including:

Based on the acquired information of the user terminal equipment, determine a combination of corresponding data layers from the candidate multi-user pairing set;

Iteratively determining the combined precoding matrix of the data layer;

A user scheduling result is determined according to the precoding matrix, and data transmission is performed to a corresponding user terminal device according to the user scheduling result.

According to the second aspect, an embodiment of the present invention also provides a data transmission device based on user scheduling, including:

A data layer combination determination module, configured to determine the corresponding data layer combination from the candidate multi-user pairing set based on the acquired information of the user terminal equipment;

A precoding matrix determination module, configured to iteratively determine the combined precoding matrix of the data layers;

A scheduling transmission module, configured to determine a user scheduling result according to the precoding matrix, and perform data transmission to a corresponding user terminal device according to the user scheduling result.

In the embodiment of the present invention, the precoding matrices corresponding to the large number of data layer combinations are iteratively determined according to the precoding matrices corresponding to the small number of data layer combinations. In the iterative process, the inversion operation of the matrix is saved, and the multiplication operation of the vector (or matrix) is greatly reduced, the complexity of the calculation is greatly reduced, and the efficiency and speed of determining the precoding matrix are improved; and even if the wireless communication system The number of data layers and antennas of the system is large, and the precoding matrix of the combination of multiple data layers for multi-user pairing can be determined quickly enough without reducing the system capacity. Therefore, the present invention can improve the efficiency of multi-user scheduling without affecting the system capacity, speed up the response to user terminal equipment as a whole, and improve user experience.

Additional aspects and advantages of the invention will be set forth in part in the description which follows, and will become apparent from the description, or may be learned by practice of the invention.

Description of drawings

The above and/or additional aspects and advantages of the present invention will become apparent and easy to understand from the following description of the embodiments in conjunction with the accompanying drawings, wherein:

FIG. 1 is a schematic flowchart of a data transmission method based on user scheduling according to an embodiment of the present invention;

FIG. 2a is a schematic diagram of an example of a precoding matrix selected from a data layer combination selected from an alternative multi-user pairing set through iterations according to an embodiment of the present invention;

Fig. 2b is a schematic diagram of an example of the decomposition step of the second step in Fig. 2a according to an embodiment of the present invention;

Fig. 2c is a schematic diagram of an example of the decomposition step of the fourth step in Fig. 2a according to an embodiment of the present invention;

FIG. 3 is a schematic framework diagram of an internal structure of a data transmission device based on user scheduling according to an embodiment of the present invention;

Fig. 4 is a schematic diagram of an example of a comparative experiment between the existing method for determining a precoding matrix and the method for determining a precoding matrix in the present invention.

Detailed ways

Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

Those skilled in the art will understand that unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the word "comprising" used in the description of the present invention refers to the presence of said features, integers, steps, operations, elements and/or components, but does not exclude the presence or addition of one or more other features, Integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Additionally, "connected" or "coupled" as used herein may include wireless connection or wireless coupling. The expression "and/or" used herein includes all or any elements and all combinations of one or more associated listed items.

Those skilled in the art can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this invention belongs. It should also be understood that terms, such as those defined in commonly used dictionaries, should be understood to have meanings consistent with their meaning in the context of the prior art, and unless specifically defined as herein, are not intended to be idealized or overly Formal meaning to explain.

Those skilled in the art can understand that the "terminal" and "terminal equipment" used here not only include wireless signal receiver equipment, which only has wireless signal receiver equipment without transmission capabilities, but also include receiving and transmitting hardware. A device having receive and transmit hardware capable of bi-directional communication over a bi-directional communication link. Such equipment may include: cellular or other communication equipment, which has a single-line display or a multi-line display or a cellular or other communication equipment without a multi-line display; PCS (Personal Communications Service, personal communication system), which can combine voice, data Processing, facsimile and/or data communication capabilities; PDA (Personal Digital Assistant, Personal Digital Assistant), which may include radio frequency receiver, pager, Internet/Intranet access, web browser, notepad, calendar and/or GPS (Global Positioning System (Global Positioning System) receiver; a conventional laptop and/or palmtop computer or other device having and/or including a radio frequency receiver. As used herein, a "terminal", "terminal device" may be portable, transportable, installed in a vehicle (air, sea, and/or land), or adapted and/or configured to operate locally, and/or In distributed form, the operation operates at any other location on Earth and/or in space. The "terminal" and "terminal equipment" used here can also be communication terminals, Internet terminals, music/video playback terminals, such as PDAs, MIDs (Mobile Internet Devices, mobile Internet devices) and/or with music/video playback terminals. Functional mobile phones, smart TVs, set-top boxes and other devices.

The inventors of the present invention have found that, in order to reduce the complexity of the determination process (such as matrix inversion) of the precoding matrix in the existing multi-user scheduling method, the existing low-complexity method includes lumped matrix method (block matrix method) ), Gaussian elimination method (Gaussian elimination method), or Cholesky decomposition method (square root decomposition method), etc. At present, for the inversion of general high-order matrices, the Cholesky decomposition method is the known algorithm with the lowest complexity and is widely used.

Assuming that the number of base station antennas is N _T , when calculating the pairing of the i-th data layer in the candidate multi-user pairing set, using the most commonly used precoding algorithm of the Cholesky decomposition method requires about 2i(i ² -1)/3+ (6i ² +2i) _NT real number multiplications. It can be seen that as the number of antennas and the number of paired layers increase, the computational complexity of the existing MU-MIMO precoding matrix will increase significantly, and the efficiency of determining the precoding matrix will still be greatly reduced, resulting in reduced multi-user scheduling. Efficiency, it is easy to greatly increase the response delay to the user terminal equipment, and it is easy to reduce the user experience.

The inventors of the present invention have found that currently there are some low-complexity precoding matrix algorithms that sacrifice system capacity, such as reducing the number of pairing times and reducing the number of actual antennas. However, these algorithms not only do not fundamentally reduce the complexity of the precoding algorithm, but also reduce the actual number of antennas used, which is contrary to the development trend of using more antennas to communicate with more user terminal equipment, resulting in A large number of antennas become decorations, which is equivalent to reducing the system capacity of the actual multi-antenna system.

The data transmission device based on user scheduling provided by the present invention includes: multiple sending antennas.

Preferably, the user-scheduling-based data transmission device of the present invention includes a user-scheduling-based data transmission device.

The data transmission device based on user scheduling can be set in the base station.

The flowchart of the data transmission method based on user scheduling provided by the present invention is shown in Figure 1, including: S101 Determine the combination of corresponding data layers from the candidate multi-user pairing set based on the obtained information of the user terminal equipment; S102 Iteratively determine the combined precoding matrix of the data layer; S103: Determine the user scheduling result according to the precoding matrix, and perform data transmission to the corresponding user terminal equipment according to the user scheduling result.

Wherein, the information of the user terminal equipment includes at least one of the following items: channel state information, retransmission feedback information, and service-related information.

The set of candidate multi-user pairings includes all combinations of data layers that can participate in the transmission of at least one user.

It can be easily seen that in the present invention, precoding matrices corresponding to a large number of data layer combinations are iteratively determined according to precoding matrices corresponding to a small number of data layer combinations. In the iterative process, the inversion operation of the matrix is saved, and the multiplication operation of the vector (or matrix) is greatly reduced, the complexity of the calculation is greatly reduced, and the efficiency and speed of determining the precoding matrix are improved; and even if the wireless communication system The number of data layers and antennas of the system is large, and the precoding matrix of the combination of multiple data layers for multi-user pairing can be determined quickly enough without reducing the system capacity. Therefore, the present invention can improve the efficiency of multi-user scheduling without affecting the system capacity, speed up the response to user terminal equipment as a whole, and improve user experience.

The following describes the data transmission method based on user scheduling in the embodiment of the present invention.

Obtain information sent by at least one user terminal device.

Preferably, in some cases, a user terminal equipment needs to occupy a channel to send information involving data transmission using a piece of information. In other cases, a user terminal equipment needs to occupy more than two channels and send information involving data transmission using more than two channels.

The information includes at least one of the following: channel state information, retransmission feedback information, and service-related information.

Wherein, the channel state information includes at least one of the following: channel quality information, spatial channel state information, and rank information.

The channel quality information includes: CQI (Channel Quality Indicator, channel quality indicator).

The rank information includes: RI (rank indicator, rank indication).

The spatial channel state information includes: the precoding matrix corresponding to the PMI (Pre-coding Matrix Index, precoding matrix index) information, the spatial channel state determined by channel reciprocity and/or the precoding matrix corresponding to the precoding matrix index information information.

Specifically, in a system that satisfies the reciprocity condition of the uplink and downlink channels and can pass the uplink channel state information, such as the TD-LTE system, the spatial channel state information may be the spatial channel state information determined by channel reciprocity.

In a system that satisfies the reciprocity condition of uplink and downlink channels at the same time, and includes one or more precoding matrix index PMI feedback, the spatial channel state information may be the precoding matrix corresponding to the precoding matrix index information, or the The spatial channel state information determined by the reciprocity may also be the spatial channel state information obtained by combining the channel reciprocity and the precoding matrix corresponding to the precoding matrix index information.

Preferably, the spatial channel state information may be power-normalized spatial channel state information. The so-called power normalization means that the total power value of the spatial channel is set to 1.

Preferably, the spatial channel state information can be represented in the form of aggregated channel matrix and/or channel vector.

Further, for ease of understanding, the dimension of the aggregated channel matrix may be defined as the number of data layers multiplied by the number of transmit antennas. The characterization methods of other forms of aggregated channel matrix do not affect the use of the method of this patent, only the dimensions of the aggregated channel matrix need to be converted accordingly.

The retransmission feedback information includes: ACK (ACKnowledgment, confirmation answer) and/or NACK (NegativeACKnowledgment, negative answer) information.

The service-related information includes at least one of the following: the service type of the user terminal equipment, the amount of transmitted data, and the buffer (gain) status.

According to the acquired information of at least one user terminal device, a corresponding set of candidate multi-user pairings is determined.

Specifically, based on the acquired channel state information, retransmission feedback information and/or service-related information, use at least one of the following search methods to determine the corresponding combination of data layers from the candidate multi-user pairing set: exhaustive Search method, Greedy search method, Search method by layer, Search method by user.

The set of candidate multi-user pairings includes: all data layer combinations that can participate in at least one user transmission. The combination of data layers determined from the candidate multi-user pairing set is paired with a channel allocated to at least one user terminal equipment; wherein, one data layer corresponds to a channel allocated to a user terminal equipment.

In the embodiment of the present invention, one of the following linear precoding algorithms can be used to determine the precoding matrix of the data layer combination selected from the candidate multi-user pairing set: SLNR (Signal to Leakage and noise ratio, signal-to-leakage-to-noise ratio ) algorithm, MMSE (Minimum Mean Square Error, minimum mean square error) algorithm, ZF (Zero Forcing, zero forcing) algorithm.

Assume that the precoding matrix of at least one user is a precoding matrix conforming to W=f{(X+H ^H H) ^-1 H ^H }; where X is a diagonal matrix whose dimension is equal to that from the candidate multi-user pairing set The number of data layers in the data layer combination determined in , H is the aggregation channel matrix composed of the spatial channel state information corresponding to the user terminal equipment of the data layer combination, the number of rows is the number of data layers in the data layer combination, and the number of columns is equal to The number of transmitting antennas of the base station; f() indicates matrix operation, the superscript H in H ^H indicates the conjugate transposition of the matrix, W indicates the precoding matrix when at least one user is transmitted, and the number of rows is equal to the number of transmitting antennas of the base station , the number of columns is the number of data layers in the determined data layer combination.

Specifically, when X is a zero matrix and f(A)=A, W is a precoding matrix during transmission of at least one user based on a zero-forcing algorithm.

When X is a non-zero diagonal matrix and f(A)=A, W is a precoding matrix when at least one user transmits based on the minimum mean square error algorithm.

When X is a diagonal matrix, f(A)=max eigenvector{A _i }, W is a precoding matrix when at least one user transmits based on the minimum signal-to-leakage ratio algorithm.

Preferably, formulas are used to represent precoding matrices determined by different linear precoding algorithms.

Assuming that the (assigned) channel of user (terminal device) i is H _i , and the (aggregated) channel matrix of all users is H, the formulas of these linear precoding matrices are as follows:

W _ZF＝ H ^H (HH ^H ) ^-1 …………………………………………………………………………………………………………………………………………………………………………

W _MMSE ＝H ^H (HH ^H +σ ² I) ^-1 ＝(σ ² I+H ^H H) ^-1 H ^H ……………… (Formula 2)

In the above formulas (1)-(3), W _ZF represents the precoding matrix based on the ZF algorithm, W _MMSE represents the precoding matrix based on the MMSE algorithm, and W _SLNR,i represents the precoding matrix based on the SLNR algorithm of the i-th user; H ^H represents the conjugate transpose (hermit) of matrix H, A ^-1 represents the inverse (inversion) of matrix A, eigenvector(A) represents the eigenvector (eigenvector) of matrix A, max(.) represents the largest element, I is the unit matrix, and σ is the MMSE factor.

For ease of understanding, the method for determining the precoding matrix based on the MMSE algorithm is taken as an example below to specifically introduce the method for iteratively determining the precoding matrix of the data layer combination.

Based on the above formula (2), it is assumed that the number of base station antennas in the system is N _T . Among the data layer combinations determined from the candidate multi-user pairing set, the spatial channel state information of the rth data layer is a 1*N _T -dimensional channel vector, denoted as h _r , a total of r data layers selected by the search The (aggregation) channel matrix (ie data layer combination) is N _r , and its dimension is r*N _T , and the matrix that needs to be inverted is defined as

_Ar in the above formula (4) represents the inverse matrix of a total of r data layers, and A0 is the set initialization matrix. Then when calculating the precoding matrix when at least one user including the r-th layer transmits, the precoding matrix when at least one user of all r data layers transmits is W _r , which is obtained by the following formula (5):

The precoding vector corresponding to the pth data layer in the above formula (5) when at least one user transmits is W _r,p , p is less than or equal to r, and can be obtained by the following formula (6):

The method for iteratively determining the precoding matrix of the data layer combination selected from the candidate multi-user pairing set includes: (based on) an outer loop iteration (determining the precoding matrix of the data layer combination) method.

Each iteration of the outer loop includes: inner loop 1 and inner loop 2.

Wherein, the number of iterations of the outer loop is equal to the total number of selected data layers; the number of iterations of the inner loop 1 during an iteration of the outer loop is equal to the total number of calculated data layers; an iteration of the outer loop includes an inner loop 2 once.

In each outer loop iteration process, based on the channel vector of the current data layer and the precoding matrix corresponding to the combination of each calculated data layer, the precoding matrix corresponding to the combination of the current data layer and each calculated data layer is determined; until The number of outer loops is equal to the number of corresponding data layers determined from the candidate multi-user pairing set.

Wherein, the calculated data layer is a data layer that has participated in the determination of the precoding matrix.

Preferably, in each iteration of the outer loop, the first processing of the inner loop includes: according to the product of the specified inversion result and the conjugate transpose of the channel vector of the current data layer, iteratively determine the corresponding The product of the inversion result of the current data layer and the conjugate transpose of the channel vector of the current data layer; and then determine the channel vector of the current data layer, the inversion result corresponding to the last calculated data layer and the channel vector of the current data layer. The product of the yoke transpose, multiplied and the inverse of one added.

Among them, the last calculated data layer is the data layer adjacent to the current data layer and the previous data layer; the specified inversion result includes the specified initial inversion result and the inversion result corresponding to the calculated layer.

Specifically, the inversion result corresponding to the data layer has been calculated, and its specific content is related to the calculation of the precoding matrix algorithm when at least one user is assumed to be transmitting. According to the algorithm adopted when determining the precoding matrix, the type of the inversion result corresponding to the calculated data layer is determined. Preferably, when the minimum mean square error MMSE algorithm is used to determine the precoding matrix, the product of the conjugate transpose of the aggregated channel matrix corresponding to each calculated data layer and the aggregated channel matrix corresponding to each calculated data layer is determined matrix, and the inverse matrix of the sum matrix of the diagonal square matrix with the interference noise power value as the diagonal element, as the inversion result corresponding to the calculated data layer. Further, when the number of calculated data layers is 0, the inversion result corresponding to the calculated data layers is the inverse matrix of the diagonal matrix.

The processing process of the inner loop two includes: according to the above-mentioned product and reciprocal involved in the current data layer (that is, determined by the inner loop one), and the inversion result corresponding to the last calculated data layer, and each calculated data layer The product of the conjugate transpose of the channel vector of the current data layer is determined to determine the corresponding inversion result of the current data layer, and the product of the conjugate transpose of the channel vector of each calculated data layer and the current data layer respectively, as the current data layer and each A precoding matrix corresponding to a combination of data layers has been calculated.

Preferably, the iterative process of the inner loop 1 will be described in detail below.

In the first iteration of inner loop 1, according to the product of the specified initial inversion result and the conjugate transpose of the channel vector of the current data layer, determine the inversion result corresponding to the first calculated data layer and the current The product of the conjugate transposes of the channel vectors of the data layer.

In each subsequent iterative process of the inner loop 1, according to the product of the inversion result corresponding to the previous calculated data layer and the conjugate transpose of the channel vector of the current data layer, determine the calculation value corresponding to the calculated data layer The product of the inverse result and the conjugate transpose of the channel vector of the current data layer; until the product of the inverse result corresponding to the last calculated data layer and the conjugate transpose of the channel vector of the current data layer is determined, the inner loop one ends .

Further, for each calculated data layer, according to the product of the inversion result corresponding to the previous calculated data layer and the conjugate transpose of the channel vector of the current data layer, and the pre-stored respective first type of each calculated data layer As an intermediate result, the product of the inversion result corresponding to the calculated data layer and the conjugate transpose of the channel vector of the current data layer is determined.

Among them, the first type of intermediate results of each calculated data layer includes: the product of the inversion result corresponding to the previous calculated data layer and the conjugate transpose of the channel vector of the calculated data layer, and the previous calculated The inverse result corresponding to the data layer is multiplied by the channel vector of the calculated data layer and its conjugate transpose, and then the reciprocal of one is added.

Preferably, for each calculated data layer, the inverse result corresponding to the previous calculated data layer, multiplied by the channel vector of the calculated data layer and its conjugate transpose, and then the reciprocal of 1 can be specifically expressed as : The channel vector of the calculated data layer, the inversion result corresponding to the previous calculated data layer, and the conjugate transpose of the channel vector of the calculated data layer (these three items) are multiplied and the reciprocal of one is added.

More preferably, the above-mentioned product and reciprocal involved in the current data layer determined by the inner loop one are stored as the first-type intermediate results of the current data layer for use in determining the first-type intermediate results of subsequent data layers.

More optimally, the product of the inversion result corresponding to the current data layer determined by the inner loop 2 and the conjugate transpose of each calculated data layer and the channel vector of the current data layer is used as the second The class intermediate results are stored for use when determining the second class intermediate results of the subsequent data layer.

Further, the first-type intermediate results of each calculated data layer are stored when the following precoding matrix is determined: the calculated data layer corresponds to the combination of the calculated data layer before the calculated data layer precoding matrix.

The product of the inversion result corresponding to the last calculated data layer and the conjugate transpose of the channel vectors of the previous calculated data layers, as the second type of intermediate result corresponding to the last calculated data layer, is determined in the final A precoding matrix corresponding to a combination of a calculated data layer and a previous calculated data layer is stored.

Fig. 2a is a schematic diagram of an example of iteratively determining a precoding matrix of a data layer combination selected from a set of candidate multi-user pairings.

In Fig. 2a, in the iterative process of the (i+1)th outer loop, the i-th and (i+1) data layers are respectively the last calculated data layer (the corresponding precoding matrix has been determined), the current data layer ( The corresponding precoding matrix has not been determined), using the inner loop one, iteratively determine the conjugate transposition h of the inversion result A _i ^-1 corresponding to the i-th data layer and the channel vector of the (i+1)-th data layer The product of _i+1 ^H is A _i ^-1 h _i+1 ^H . (i+1) is a positive integer greater than or equal to 1 and less than or equal to r.

Then determine the product of the inversion result corresponding to the i-th data layer and the conjugate transpose of the channel vector of the (i+1)-th data layer, and multiply it by the channel vector h _i+1 of the (i+1)-th data layer reciprocal of plus one

Specifically, step 1 in Figure 2a obtains A ₀ ^-1 h _i+1 ^H , and calculates according to the following formula (6):

A ₀ ^-1 h _i+1 ^H ＝Sigma*h _i+1 ^H ……………………………(Formula 6)

In the above formula (6), A ₀ ^-1 represents the specified initial inversion result; h _i+1 ^H is the (user's) channel vector corresponding to the input (i+1)th data layer; Sigma=1/σ ² , σ ² is the MMSE weight factor.

Inner loop one includes steps two and three in Figure 2a. Step 2 in Figure 2a calculates A _k+1 ^-1 h _i+1 ^H , k+1 is a positive integer greater than or equal to 1 and less than or equal to i; the inner loop needs to iterate i times, which is equal to the total number of calculated data layers .

The first type of intermediate results of each calculated data layer are pre-stored in the first buffer (buffer 1) in FIG. 2a. For the first type of intermediate results of the kth (calculated) data layer, it includes: the inversion result A _k-1 ^-1 corresponding to the (k-1) (previous) data layer and the channel vector of the kth data layer The product A _k-1 ^-1 h _k ^H of the conjugate transpose h _k ^H , and the conjugate transpose h _k ^H of the channel vector of the kth data layer, and the inversion result A corresponding to the (k-1)th data layer _K-1 ^-1 is multiplied by the channel vector h _k of the k-th data layer and then the reciprocal of one is added

For the first type of intermediate results of the k+1th (calculated) data layer, it includes: the conjugate of the inversion result A _k ^-1 corresponding to the kth (previous) data layer and the channel vector of the k+1th data layer The product A _k ^-1 h k ₊ 1 ^H of the transpose h _k+1 ^H , and the conjugate transpose h _k+1 ^H of the channel vector of the k+1th data layer, and the inversion result A corresponding to the kth data layer The reciprocal of _k ^-1 and the channel vector h _k+1 of the k+1th data layer multiplied by one

Preferably, according to the product of the inversion result corresponding to the previous calculated data layer (or the output result of the above step 1 in Figure 2a) and the conjugate transpose of the channel vector of the current data layer, and the first buffer (buffer The respective first-type intermediate results of each calculated data layer pre-stored in the device 1), using the following formula (7) to determine the product of the conjugate transpose of the channel vector corresponding to the calculated data layer:

In the formula (7) is predetermined; the value range of k+1 is a positive integer from 1 to i. According to the above formula (7), the following results can be deduced:

Please note that the current data layer is the (i+1)th layer, and k+1 is less than or equal to i.

When k+1=1, it is the first iteration in the inner loop 1, according to the conjugate transposition h _i+1 of A ₀ ^-1 corresponding to the specified noise as the output result of step 1 and the channel vector of the current data layer The product of ^H A ₀ ^-1 h _i+1 ^H (the first item on the right side of the equal sign in formula 7), the first intermediate result of the pre-stored first data layer and and derivable Determine the conjugate transpose of the inversion result A ₁ ^-1 corresponding to the first data layer and the channel vector of the current data layer the product of Further, the first type of intermediate result of the first data layer is generated and stored when determining the precoding matrix corresponding to the first data layer (that is, when the first data layer is the current data layer).

When k+1=2, it is the second iteration in the inner loop one, according to the inversion result A ₁ ^-1 h _i+1 ^H corresponding to the first (calculated) data layer and the channel vector of the current data layer in common The product A ₁ ^-1 h _i+1 ^H of the yoke transpose h _i+1 ^H , the first-class intermediate result of the pre-stored second data layer and and derivable Determine the conjugate transpose of the inversion result A ₂ ^-1 corresponding to the second data layer and the channel vector of the current data layer the product of Further, the first type of intermediate result of the second data layer is generated and stored when determining the precoding matrix corresponding to the combination of the first and second data layers (that is, when the second data layer is the current data layer).

When k+1=k, it is the kth iteration in the inner loop one, according to the inversion result A _k-1 ^-1 h _i+1 ^H corresponding to the (k-1)th (calculated) data layer and the current data The product A _k-1 ^-1 h _i+1 ^H of the conjugate transpose h _i+1 ^H of the channel vector of the layer, the first-class intermediate result of the pre-stored k-th data layer and and derivable Determine the conjugate transpose of the inversion result A _k ^-1 corresponding to the kth data layer and the channel vector of the current data layer the product of Further, the first type of intermediate result of the kth data layer is generated and stored when determining the precoding matrix corresponding to the combination of the 1st to kth data layers (that is, when the kth data layer is the current data layer).

When k+1=k+1, it is the k+1th iteration in the inner loop one, according to the inversion result A _k ^-1 h _i+1 ^H corresponding to the kth (calculated) data layer and the current data layer The product A _k ^-1 h _i+1 ^H of the conjugate transpose h _i+1 ^H of the channel vector, the first-class intermediate result of the pre-stored (k+1)th data layer and and derivable Determine the conjugate transpose of the inversion result A k+1 ^-1 corresponding to _{the k+1th} data layer and the channel vector of the current data layer the product of Further, the first type of intermediate result of the k+1th data layer is when determining the precoding matrix corresponding to the combination of the 1st to (k+1) data layers (that is, when the k+1th data layer is the current data layer) , generated and stored.

Until k+1=i, it is the i-th (last) iteration in inner loop one, according to the inversion result A i _- ^1-1 h _i+1 ^H corresponding to the i-1th (calculated) data layer and The product A _i-1 ^-1 h _i+1 ^H of the conjugate transpose h _i+1 ^H of the channel vector of the current data layer, the first-class intermediate result of the pre-stored i-th data layer and and derivable Determine the conjugate transpose of the inversion result A _i ^-1 corresponding to the i-th data layer and the channel vector of the current data layer the product of Further, the first type of intermediate result of the i-th data layer is generated and stored when determining the precoding matrix corresponding to the combination of the 1st to i-th data layers (that is, when the i-th data layer is the current data layer).

More preferably, Fig. 2b is a schematic diagram of an example of the decomposition steps of the second step in Fig. 2a, showing that the calculation steps of the polynomial on the right side of the above formula (7) can be divided into steps 1-4. step 1 The calculation result of is a value; in step 2 The calculation result of is a value; in step 3 The result of is a vector; step 4 is to subtract one vector from another vector, which is equivalent to the addition operation of vectors.

It can be seen that in step 2 of Figure 2a and are all involved in the loop iteration calculation as a whole, where as a vector, as numbers rather than vectors or matrices. Therefore, the inner loop 1 of the embodiment of the present invention does not involve matrix (or vector) inversion operation, the multiplication operation of matrix (or vector) is greatly reduced, the calculation complexity is low, the calculation amount is small, and the calculation speed is fast. Higher efficiency.

Step 3 in Figure 2a includes: according to the output of step 2, the conjugate transposition of the inversion result A _i ^-1 corresponding to the last calculated (i-th) data layer and the channel vector of the current (i+1) data layer the product of Use the following formula (8) to determine the result of multiplying the channel vector h i ₊₁ of the current (i+1) data layer with the product:

Furthermore, the multiplied result obtained according to the above formula (8) Determine the result of the multiplication plus one reciprocal

More optimal, the reciprocal and the product of the above as the output of step two Stored in the first buffer (buffer 1) as the first type of intermediate results of the current data layer, it is used for determining the precoding matrix corresponding to the data layer combination that includes the subsequent data layer, specifically for determining the subsequent data layer (all subsequent data layers) data layer) for first-class intermediate results.

The second type of intermediate result of the last calculated data layer is pre-stored in the second buffer (buffer 2) in FIG. 2a. For the second type of intermediate results of the i-th (last calculated) data layer, it includes: the conjugate transpose of the inversion result A _i ^-1 corresponding to the i-th data layer and the channel vector of the first to i data layers ( h ₁ ^H , ..., h _k ^H , h _k+1 ^H , ..., h _i ^H ) product A _i ^{- 1} h ₁ ^H , ..., A _i ^-1 h _k ^H , A _i ^-1 h _k+1 ^H , . . . , A _i ⁻¹ h _i ^H . Further, the product of the inversion result corresponding to the last calculated (i-th) data layer and the conjugate transpose of the channel vectors of the previous calculated data layers is used as the second-type intermediate result corresponding to the i-th data layer, It is generated and stored when determining the precoding matrix corresponding to the combination of the 1-i data layers.

Inner loop two includes step four in Fig. 2a. Step 4 of Figure 2a calculates A _i+1 ^-1 h ₁ ^H , ..., A _i+1 ^-1 h _k ^H , A _i+1 ^{- 1} h _k+1 ^H , ..., A _i+1 ^-1 h _{i +1} In ^H , according to the product and reciprocal of the first type of intermediate results of the current data layer in steps 2 and 3, that is, the conjugate transposition of the inversion result corresponding to the last calculated data layer and the channel vector of the current data layer the product of And the product is multiplied with the channel vector of the current data layer and then the reciprocal of one And the conjugate transposition of the inversion result A _i ^-1 corresponding to the i-th data layer as the last calculated data layer and the channel vector of each calculated data layer and any data layer in the current data layer the product of Using the following formula (9), determine the inversion result A _i+1 ^-1 corresponding to the current data layer, and the conjugate transpose of the channel vector of each calculated data layer and any data layer in the current data layer respectively the product of

In formula (9) is predetermined; the value range of j is a positive integer from 1 to i.

Preferably, when j=any value from 1 to i, according to the product of the inversion result corresponding to the last calculated data layer and the conjugate transpose of the channel vector of the current data layer And the product is multiplied with the channel vector of the current data layer and then the reciprocal of one and A _i ^-1 h ₁ ^H , ..., A _i ^-1 h _k ^H , A _i ^-1 h _k+1 ^H , ..., A _i ^-1 h _i ^H prestored in the second buffer, using the formula (9 ), determine the inversion result corresponding to the current data layer, and the product A _i+1 ^-1 h ₁ ^H ,..., A _i+1 ^-1 h _k of the conjugate transpose of the channel vector of each calculated data layer ^H , A _i+1 ^-1 h _k+1 ^H , ..., A _i+1 ^-1 h _i ^H .

When making j=i+1, according to the product of the inversion result corresponding to the last calculated data layer and the conjugate transpose of the channel vector of the current data layer And the product is multiplied with the channel vector of the current data layer and then the reciprocal of one and the predetermined Using formula (9), the product A _i+1 ^-1 h _i+1 ^H of the inversion result corresponding to the current data layer and the conjugate transpose of the channel vector of the current data layer is determined.

The determined inversion result corresponding to the current data layer, and the conjugate transposition h ₁ ^H , ..., h _k ^H , h _k+1 ^H , ..., The product of h _i+1 ^H A _i+1 ^-1 h ₁ ^H , ..., A _i+1 ^-1 h _k ^H , A _i+1 ^-1 h _k+1 ^H , ..., A _i+1 ^-1 h _i+1 ^H , as the precoding matrix corresponding to the combination of the current data layer and each calculated data layer.

More preferably, Fig. 2c is a schematic diagram of an example of the decomposition steps of the fourth step in Fig. 2a. Fig. 2c shows that the calculation steps of the polynomial on the right side of the above formula (9) can be divided into steps 1-4. step 1 The calculation result of is a value; in step 2 The calculation result of is a value; in step 3 The result of is a vector; step 4 is to subtract one vector from another vector, which is equivalent to the addition operation of vectors. Further, A _{i -1} h _i+1 ^H required by A _i ^{+1 -1} h ⁱ ₊₁ ^H has been obtained in step three.

It can be seen that in step 4 of Figure 2a and They are all involved in the iterative calculation of the loop as a whole, as a vector, as numbers rather than vectors or matrices. Therefore, the inner loop 1 of the embodiment of the present invention does not involve matrix (or vector) inversion operation, the multiplication operation of matrix (or vector) is greatly reduced, the calculation complexity is low, the calculation amount is small, and the calculation speed is fast. Higher efficiency.

Thus, the precoding matrix corresponding to the combination of the current (i+1) data layer and each calculated (1st to i) data layer is obtained, that is, the precoding matrix corresponding to the combination of the 1st to the current (i+1) data layer matrix.

When the data layers selected from the determined user pairing set are 1 to i+1 data layers, the precoding vectors of all users in the (i+1) data layer are obtained to form the corresponding combination of i+1 data layers Precoding matrix; the (user's) precoding vector corresponding to the jth data layer in the precoding matrix corresponding to the combination of i+1 data layers is

More optimally, in the precoding matrix corresponding to the combination of the current data layer and each calculated data layer, the inversion result corresponding to the current data layer, and the conjugate conversion of the channel vectors of each calculated data layer and the current data layer respectively For example, when the current data layer is the i+1th data layer, it is A _i+1 ^-1 h ₁ ^H ,..., A _i+1 ^-1 h _k ^H , A _i+1 ^-1 h _{k+ 1} ^H ,..., A _i+1 ^-1 h _i+1 ^H , as the second-type intermediate results corresponding to the current data layer, are stored in the second-type buffer (buffer 2) for determining the data that includes the subsequent data layer It is used when the data layer combines the corresponding precoding matrix, specifically used when determining the second-type intermediate result of the subsequent data layer, and further used when determining the second-type intermediate result of the subsequent data layer.

Preferably, a memory can be used to store the first and second types of intermediate results.

Preferably, the memory may be a buffer or other types of memory.

Preferably, since the present invention can be implemented in different modules in the base station system, the memory can be different types of memory according to the different implementation modules in the base station system where the present invention is located. Specifically, the present invention can be implemented in modules such as DSP chip modules, MIPS core modules, or FPGAs.

It can be understood that directly using the above-mentioned first-type intermediate results and second-type intermediate results stored in advance can save the steps of instant calculation of the first-type intermediate results and the second-type intermediate results, which is conducive to saving computation and improving calculation speed and efficiency .

Afterwards, a user scheduling result is determined according to the iteratively determined combined precoding matrix of the data layer; data transmission is performed to a corresponding user terminal device according to the user scheduling result. The combination of data layers is selected from a set of candidate multi-user pairings.

Specifically, it may be to use the combined precoding matrix of the data layer determined according to iterations as the precoding in the user scheduling result used for data transmission of at least one user channel (a channel allocated to at least one user terminal equipment) matrix.

It may be a modulation and coding scheme determined by using a corresponding signal-to-interference-noise ratio calculated based on the combined precoding matrix of the data layer determined iteratively, as a user scheduling result used for data transmission of at least one user channel Modulation coding scheme.

The number of MU scheduling data layers calculated based on the precoding matrix of the combination of data layers determined iteratively may be used as the number of MU-MIMO transmission data layers in the user scheduling result.

The paired user channel of MU-MIMO calculated based on the precoding matrix of the combination of data layers determined iteratively may be used as the paired user channel of MU-MIMO transmission in the user scheduling result.

Other transmission modes related to MU-MIMO transmission calculated based on the precoding matrix of the combination of data layers determined iteratively may be used as other transmission modes in the user scheduling result.

According to the precoding matrix, modulation and coding scheme, MU-MIMO transmission data layer number, MU-MIMO transmission paired user channel, and/or other transmission modes in the user scheduling result, the corresponding user terminal equipment (user channel) is sent data transmission.

Taking the precoding matrix determination method based on ZF algorithm as an example, for the combination of data layers selected from the candidate multi-user pairing set, the method of iteratively determining the precoding matrix of the data layer combination is specifically introduced.

Specifically, the difference between the method for determining the precoding matrix based on the ZF algorithm and the method for determining the precoding matrix based on the MMSE algorithm is mainly introduced.

When the zero-breaking algorithm is used to determine the precoding matrix, the product matrix of the conjugate transpose of the aggregated channel matrix corresponding to each calculated data layer and the aggregated channel matrix corresponding to each calculated data layer, and The inverse matrix of the diagonal sum matrix of diagonal elements, as the result of the inversion corresponding to the computed data layer.

The ZF algorithm calculates the precoding matrix formula when at least one user transmits as described in the above formula (1), and repeats as follows:

W _ZF ＝H ^H (HH ^H ) ^-1 ………………………………………………………………………………………………………………………(Formula 1)

In the above formula (1), H ^H represents the conjugate transpose (hermit) of the matrix H, and A ⁻¹ represents the inversion of the matrix A (inversion).

By adding an item σ ² I to the matrix being inverted, such that

W _ZF ≈(σ ² I+H ^H H) ^-1 H ^H ＝W _MMSE ……………………… (Formula 10)

In the above formula (2), σ ² is a very small number, and I is an identity matrix, so that the inverse matrix H ^H H is full rank. At this time, calculating W _ZF is transformed into calculating W _MMSE .

The following specific operation steps are consistent with the above-mentioned method for determining the precoding matrix based on the MMSE algorithm, and will not be repeated here.

Taking the precoding matrix determination method based on the SLNR algorithm as an example, for the combination of data layers selected from the candidate multi-user pairing set, the method of iteratively determining the precoding matrix of the data layer combination is specifically introduced.

Specifically, the difference between the method for determining the precoding matrix based on the SLNR algorithm and the method for determining the precoding matrix based on the MMSE algorithm is mainly introduced.

When the minimum signal-leakage-to-noise ratio SLNR algorithm is used to determine the precoding matrix, the product matrix of the conjugate transpose of the aggregated channel matrix corresponding to each calculated data layer and the aggregated channel matrix corresponding to each calculated data layer, and Use the interference noise power value as the inverse matrix of the sum matrix of the diagonal square matrix of the diagonal elements, as the inverse result corresponding to the calculated data layer.

The SLNR algorithm calculation assumes that the precoding matrix formula when at least one user transmits is as described in the above formula (3), and is repeated as follows:

In formula (3), H ^H represents the conjugate transpose (hermit) of matrix H, A ^-1 represents the inverse (inversion) of matrix A, eigenvector(A) represents the eigenvector of matrix A, and max(.) represents the maximum element, I is the identity matrix, and σ is the MMSE factor.

At this time, for the kth user, there can be

Where norm(.) represents vector normalization. Equation (a) utilizes the definition of SLNR, and equation (b) utilizes the matrix inversion lemma. For the eigenvalue decomposition formula USV ^H =A, when the rank of matrix A is 1, the first column of A is normalized to obtain the first column of matrix U, and the equation (c) takes advantage of this.

It can be seen that SLNR is completely equivalent to MMSE

The following specific operation steps are consistent with the above-mentioned method for determining the precoding matrix based on the MMSE algorithm, and will not be repeated here.

Based on the above data transmission method based on user scheduling, an embodiment of the present invention provides a data transmission device based on user scheduling. The frame diagram of the internal structure of the device is shown in FIG. A coding matrix determination module 302 and a scheduling transmission module 303 .

Wherein, the data layer combination determining module 301 is configured to determine a corresponding data layer combination from the candidate multi-user pairing set based on the acquired information of the user terminal equipment.

The precoding matrix determining module 302 is configured to iteratively determine the combined precoding matrix of the data layers.

The scheduling transmission module 303 is configured to determine the user scheduling result according to the precoding matrix, and perform data transmission to the corresponding user terminal equipment according to the user scheduling result.

Preferably, the information of the user terminal equipment includes at least one of the following: channel state information, retransmission feedback information, and service-related information.

Preferably, the precoding matrix determination module 302 is specifically used in each outer loop iteration process, based on the channel vector of the current data layer and the precoding matrix corresponding to the combination of each calculated data layer, to determine the calculated data layer The precoding matrix corresponding to the combination of the current data layer; until the number of times of the outer loop is greater than the number of corresponding data layers determined from the candidate multi-user pairing set, the outer loop is ended; wherein, the calculated data layer is already participated in The data layer determined by the precoding matrix.

Preferably, the precoding matrix determination module 302 is specifically configured to use the inner loop one to iteratively determine the corresponding The product of the inverse result and the conjugate transpose of the channel vector of the current data layer; then determine the reciprocal of the product multiplied by the channel vector of the current data layer and then add one; the last calculated data layer is adjacent to the current data layer and the previous data layer; using inner loop 2, according to the product and reciprocal involved in the current data layer, and the inversion result corresponding to the last calculated data layer, and the conjugate conversion of the channel vector of each calculated data layer The product of the position, determine the inversion result corresponding to the current data layer, and the product of the conjugate transpose of the channel vector of each calculated data layer and the current data layer respectively, as the combined correspondence of the current data layer and each calculated data layer The precoding matrix of .

Preferably, the precoding matrix determination module 302 is specifically configured to determine the first iteration according to the product of the specified initial inversion result and the conjugate transpose of the channel vector of the current data layer during the first iteration of the inner loop one. The product of the inversion result corresponding to a calculated data layer and the conjugate transpose of the channel vector of the current data layer; in each subsequent iteration in inner loop 1, for each calculated data layer, according to the previous The product of the inverse result corresponding to the calculated data layer and the conjugate transpose of the channel vector of the current data layer is determined, and the product of the inverse result corresponding to the calculated data layer and the conjugate transpose of the channel vector of the current data layer is determined ; Until the product of the inversion result corresponding to the last calculated data layer and the conjugate transpose of the channel vector of the current data layer is determined, the inner loop one is ended.

Preferably, the precoding matrix determination module 302 is specifically configured to use the product of the inversion result corresponding to the previous calculated data layer and the conjugate transpose of the channel vector of the current data layer, and the respective pre-stored calculated data layers The first type of intermediate result determines the product of the inversion result corresponding to the calculated data layer and the conjugate transpose of the channel vector of the current data layer; wherein, the first type of intermediate result of each calculated data layer includes: The product of the inverse result corresponding to the previous calculated data layer and the conjugate transpose of the channel vector of the calculated data layer, and the inverse result corresponding to the previous calculated data layer, and the channel vector of the calculated data layer The reciprocal of multiplied by its conjugate transpose plus one.

Preferably, the precoding matrix determination module 302 is specifically configured to, for the k+1th calculated data layer, according to the inversion result A _k ^-1 corresponding to the kth data layer as the previous calculated data layer, and the current data The conjugate transpose of the channel vector of the i+1th data layer of the layer the product of And the conjugate transpose of the inversion result A _k ^-1 corresponding to the previous calculated data layer and the channel vector of the calculated data layer the product of The inversion result A _k ^-1 corresponding to the previous calculated data layer, the channel vector h _k+1 of the calculated data layer and its conjugate transpose Reciprocal of multiplication plus one Use the above formula (7) to determine the conjugate transposition of the inversion result A _k+1 ^-1 corresponding to the calculated data layer and the channel vector of the current data layer the product of In formula (7) is predetermined; the range of k+1 is 1 to i.

Preferably, the precoding matrix determination module 302 is specifically configured to use and countdown And the conjugate transposition of the inversion result A _i ^-1 corresponding to the i-th data layer as the last calculated data layer and the channel vector of each calculated data layer and any data layer in the current data layer the product of Using the above formula (9), determine the inversion result A _i+1 ^-1 corresponding to the current data layer, and the conjugate transpose of the channel vector of each calculated data layer and any data layer in the current data layer respectively the product of In formula (9) is predetermined; the value range of j is from 1 to i+1.

Preferably, the precoding matrix determination module 302 is also used to store the product and reciprocal involved in the current data layer determined by the inner loop one as the first type of intermediate result of the current data layer, for determining the first type of intermediate results of the subsequent data layer. Used for a class of intermediate results; the product of the inversion result corresponding to the current data layer determined by the inner loop 2, and the conjugate transpose of each calculated data layer and the channel vector of the current data layer is used as the current data layer The intermediate results of the second type are stored for use when determining the intermediate results of the second type in the subsequent data layer.

Preferably, the precoding matrix determining module 302 is further configured to determine the type of the inversion result corresponding to the calculated data layer according to the algorithm adopted when determining the precoding matrix.

Preferably, the precoding matrix determination module 302 is specifically configured to determine that the conjugate transpose of the aggregated channel matrix corresponding to each calculated data layer corresponds to each calculated data layer when the zero-breaking algorithm is used to determine the precoding matrix. The product matrix of the aggregate channel matrix of , and the inverse matrix of the sum matrix of the diagonal square matrix with a small number of diagonal elements, as the inversion result corresponding to the calculated data layer; when using the minimum mean square error algorithm, or When the minimum signal-to-leakage-to-noise ratio algorithm determines the precoding matrix, determine the product matrix of the conjugate transpose of the aggregated channel matrix corresponding to each calculated data layer and the aggregated channel matrix corresponding to each calculated data layer, and the interference noise The power value is used as the inverse matrix of the diagonal square matrix of the diagonal elements and the matrix, as the inverse result corresponding to the calculated data layer.

The methods for realizing the functions of the data layer combination determination module 301, the precoding matrix determination module 302, and the scheduling transmission module 303 may refer to the specific content of the data transmission method based on user scheduling, and will not be repeated here.

Fig. 4 is a schematic diagram of an example of a comparative experiment between the existing method for determining a precoding matrix and the method for determining a precoding matrix in the present invention.

Assuming that the number of base station antennas is N _T , when calculating the pairing of the i-th data layer in the data layer combination selected from the candidate multi-user pairing set, using the most commonly used precoding algorithm of the Cholesky decomposition method requires about 2i(i ² -1)/3+(6i ² +2i)N _T times real number multiplication. There is a 2-3 power relationship between the number of real number multiplications corresponding to the Cholesky decomposition method and the number of data layers.

However, the corresponding complexity of the present invention is approximately (16i-8) N _T times of real number multiplications, and the number of real number multiplications is linearly related to the number of data layers. It can be seen that in the present invention, compared with the technical solution based on the Cholesky decomposition method, the complexity of determining the precoding matrix is greatly reduced.

Fig. 4 shows the comparison of the number of multiplications between the matrices (vectors) of the base station with 64 transmitting antennas and 16 data layers using the Cholesky decomposition method and the method of this solution. In the process of determining the precoding matrix corresponding to the first data layer, this scheme can reduce the complexity by 11.3%. When calculating to a total of 16 data layers, this scheme can reduce the complexity by 78.3%. The effect is very obvious .

Those skilled in the art will appreciate that the present invention includes devices related to performing one or more of the operations described in this application. These devices may be specially designed and fabricated for the required purposes, or they may include known devices found in general purpose computers. These devices have computer programs stored therein that are selectively activated or reconfigured. Such a computer program can be stored in a device (e.g., computer) readable medium, including but not limited to any type of medium suitable for storing electronic instructions and respectively coupled to a bus. Types of disks (including floppy disks, hard disks, CDs, CD-ROMs, and magneto-optical disks), ROM (Read-Only Memory, read-only memory), RAM (Random Access Memory, random memory), EPROM (Erasable Programmable Read-Only Memory, Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory, Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic card or optical card. That is, a readable medium includes any medium that stores or transmits information in a form readable by a device (eg, a computer).

Those skilled in the art will understand that computer program instructions can be used to implement each block in these structural diagrams and/or block diagrams and/or flow diagrams and combinations of blocks in these structural diagrams and/or block diagrams and/or flow diagrams . Those skilled in the art can understand that these computer program instructions can be provided to general-purpose computers, professional computers, or processors of other programmable data processing methods for implementation, so that the computer or processors of other programmable data processing methods can execute the present invention. A scheme specified in a block or blocks of a structure diagram and/or a block diagram and/or a flow diagram of the invention disclosure.

Those skilled in the art can understand that the various operations, methods, and steps, measures, and solutions in the processes discussed in the present invention can be replaced, changed, combined, or deleted. Further, other steps, measures, and schemes in the various operations, methods, and processes that have been discussed in the present invention may also be replaced, changed, rearranged, decomposed, combined, or deleted. Further, steps, measures, and schemes in the prior art that have operations, methods, and processes disclosed in the present invention can also be alternated, changed, rearranged, decomposed, combined, or deleted.

The above descriptions are only part of the embodiments of the present invention. It should be pointed out that those skilled in the art can make some improvements and modifications without departing from the principles of the present invention. It should be regarded as the protection scope of the present invention.

Claims (13)

1. A data transmission method based on user scheduling, characterized in that, comprising: Based on the acquired information of the user terminal equipment, determine a combination of corresponding data layers from the candidate multi-user pairing set; Iteratively determining the combined precoding matrix of the data layer; A user scheduling result is determined according to the precoding matrix, and data transmission is performed to a corresponding user terminal device according to the user scheduling result. 2. The method according to claim 1, wherein the iteratively determining a combined precoding matrix of the data layers comprises: In each outer loop iteration process, based on the channel vector of the current data layer and the precoding matrix corresponding to the combination of each calculated data layer, the precoding matrix corresponding to the combination of each calculated data layer and the current data layer is determined; until The number of times of the outer loop is equal to the number of corresponding data layers determined from the candidate multi-user pairing set, and the outer loop is ended; Wherein, the calculated data layer is a data layer that has participated in the determination of the precoding matrix. 3. The method according to claim 2, wherein, in each outer loop iterative process, based on the channel vector of the current data layer and the precoding matrix corresponding to the combination of each calculated data layer, the current The precoding matrix corresponding to the combination of the data layer and each calculated data layer includes: Each outer loop iteration process includes inner loop one and inner loop two; The processing of the inner loop 1 includes: according to the product of the specified inversion result and the conjugate transpose of the channel vector of the current data layer, iteratively determine the inversion result corresponding to the last calculated data layer and the result of the current data layer. The product of the conjugate transposition of the channel vector; and then determine the channel vector of the current data layer and the reciprocal of the product multiplied by one; the last calculated data layer is adjacent to the current data layer and the previous one the data layer; The processing process of the inner loop 2 includes: according to the product and reciprocal involved in the current data layer, and the inversion result corresponding to the last calculated data layer, and the conjugate transposition of the channel vector of each calculated data layer The product of the inversion result corresponding to the current data layer is determined, and the product of the conjugate transpose of the channel vector of each calculated data layer and the current data layer is determined as the combination of the current data layer and each calculated data layer. precoding matrix. 4. The method according to claim 3, wherein, according to the product of the specified inversion result and the conjugate transpose of the channel vector of the current data layer, iteratively determine the corresponding inversion of the last calculated data layer The product of the inverse result and the conjugate transpose of the channel vector of the current data layer, consisting of: In the first iteration of inner loop 1, according to the product of the specified initial inversion result and the conjugate transpose of the channel vector of the current data layer, determine the inversion result corresponding to the first calculated data layer and the current The product of the conjugate transposes of the channel vectors of the data layer; In each subsequent iterative process of inner loop 1, for each calculated data layer, according to the product of the inversion result corresponding to the previous calculated data layer and the conjugate transpose of the channel vector of the current data layer, determine The product of the inverse result corresponding to the calculated data layer and the conjugate transpose of the channel vector of the current data layer; until the conjugate transpose of the inverse result corresponding to the last calculated data layer and the channel vector of the current data layer is determined Set the product, ending the inner loop one. 5. The method according to claim 4, wherein the calculated data is determined according to the product of the inversion result corresponding to the previous calculated data layer and the conjugate transpose of the channel vector of the current data layer The product of the inversion result corresponding to the layer and the conjugate transpose of the channel vector of the current data layer, including: According to the product of the inversion result corresponding to the previous calculated data layer and the conjugate transpose of the channel vector of the current data layer, and the first-type intermediate results of each pre-stored calculated data layer, the calculated data layer is determined The product of the corresponding inversion result and the conjugate transpose of the channel vector of the current data layer; Among them, the first type of intermediate results of each calculated data layer includes: the product of the inversion result corresponding to the previous calculated data layer and the conjugate transpose of the channel vector of the calculated data layer, and the previous calculated The inverse result corresponding to the data layer is multiplied by the channel vector of the calculated data layer and its conjugate transpose, and then the reciprocal of one is added. 6. The method according to claim 3, further comprising: The product and reciprocal involved in the current data layer determined by the inner loop one are stored as the first type of intermediate results of the current data layer for use when determining the first type of intermediate results of the subsequent data layer; The product of the inversion result corresponding to the current data layer determined by the inner loop 2 and the conjugate transpose of each calculated data layer and the channel vector of the current data layer is stored as the second type of intermediate result of the current data layer , for use in determining intermediate results of the second type for subsequent data layers. 7. The method according to claim 5, wherein the product of the inversion result corresponding to the previous calculated data layer and the conjugate transpose of the channel vector of the current data layer, and each pre-stored calculated The respective first-type intermediate results of the data layer determine the product of the inversion result corresponding to the calculated data layer and the conjugate transpose of the channel vector of the current data layer, including: For the k+1th calculated data layer, according to the inversion result A _k ^-1 corresponding to the kth data layer which is the previous calculated data layer, and the channel vector of the i+1th data layer which is the current data layer yoke transpose the product of And the conjugate transpose of the inversion result A _k ^-1 corresponding to the previous calculated data layer and the channel vector of the calculated data layer the product of The inversion result A _k ^-1 corresponding to the previous calculated data layer, the channel vector h _k+1 of the calculated data layer and its conjugate transpose Reciprocal of multiplication plus one Use the following formula (7) to determine the conjugate transpose of the inversion result A _k+1 ^-1 corresponding to the calculated data layer and the channel vector of the current data layer the product of In the formula (7) is predetermined; the value range of k+1 is a positive integer from 1 to i; And the reciprocal of the multiplication of the channel vector of the current data layer and the product and the addition of one is determined, including: Use the following formula (8) to determine the channel vector h _i+1 of the current data layer, the product of the inverse result corresponding to the k+1th calculated data layer and the conjugate transpose of the channel vector of the current data layer The result of multiplying Determine the reciprocal of the result of the multiplication plus one 8. The method according to claim 7, characterized in that, according to the product and reciprocal involved in the current data layer, and the inversion result corresponding to the last calculated data layer, respectively with the calculated data layer The product of the conjugate transpose of the channel vector determines the inversion result corresponding to the current data layer, and the product of the conjugate transpose of the channel vector of each calculated data layer and the current data layer respectively, including: According to the product of the i+1th data layer as the current data layer and countdown And the conjugate transposition of the inversion result A _i ^-1 corresponding to the i-th data layer as the last calculated data layer and the channel vector of each calculated data layer and any data layer in the current data layer the product of Using the following formula (9), determine the inversion result A _i+1 ^-1 corresponding to the current data layer, and the conjugate transpose of each calculated data layer and the channel vector of any data layer in the current data layer the product of In formula (9) is predetermined; the value range of j is a positive integer from 1 to i+1. 9. The method according to claim 8, further comprising: When k+1=i, the conjugate transposition of the inversion result A _i ^-1 corresponding to the last calculated data layer determined by formula (7) and the channel vector of the current data layer the product of and the multiplied result determined based on formula (8) reciprocal of plus one It is stored as the first type of intermediate result of the current data layer for use when determining the first type of intermediate result of the subsequent data layer; The inversion result A _i+1 ^-1 corresponding to the current data layer determined by the formula (9) is respectively conjugated with the channel vector of each calculated data layer and the current data layer. the product of Where j=1,...,i+1, stored as the second-type intermediate results of the current data layer, for use when determining the second-type intermediate results of the subsequent data layer. 10. The method according to any one of claims 1-9, further comprising: According to the algorithm adopted when determining the precoding matrix, the type of the inversion result corresponding to the calculated data layer is determined. 11. The method according to claim 10, wherein, according to the algorithm adopted when determining the precoding matrix, determining the type of the inversion result corresponding to the calculated data layer includes: When the zero-breaking algorithm is used to determine the precoding matrix, the product matrix of the conjugate transpose of the aggregated channel matrix corresponding to each calculated data layer and the aggregated channel matrix corresponding to each calculated data layer, and Measure the inverse matrix of the diagonal square matrix and matrix of diagonal elements, as the corresponding inversion result of the calculated data layer; When the minimum mean square error algorithm or the minimum signal-leakage-to-noise ratio algorithm is used to determine the precoding matrix, it is determined that the conjugate transpose of the aggregated channel matrix corresponding to each calculated data layer and the corresponding The product matrix of the aggregation channel matrix and the inverse matrix of the sum matrix of the diagonal square matrix with the interference noise power value as the diagonal element are used as the inversion result corresponding to the calculated data layer. 12. The method according to any one of claims 1-11, wherein the information of the user terminal equipment includes at least one of the following: Channel state information, retransmission feedback information, and service-related information. 13. A data transmission device based on user scheduling, characterized in that it comprises: A data layer combination determination module, configured to determine the corresponding data layer combination from the candidate multi-user pairing set based on the acquired information of the user terminal equipment; A precoding matrix determination module, configured to iteratively determine the combined precoding matrix of the data layers; A scheduling transmission module, configured to determine a user scheduling result according to the precoding matrix, and perform data transmission to a corresponding user terminal device according to the user scheduling result.