CN104935365A - Precoding matrix construction, selection and scheduling method used for multiple-input-multiple-output transmission system and equipment thereof - Google Patents

Abstract

Embodiments of the invention provide a precoding matrix construction, selection and scheduling method used for a multiple-input-multiple-output transmission system and equipment thereof. The construction method comprises the following steps that a plurality of first precoding matrixes are determined so as to form a first precoding codebook; the plurality of first precoding matrixes determination may comprise basic wave-beam universal set determination and each basic wave beam is determined by a downward inclination angle and an azimuth angle; the basic wave-beam universal set is divided into a plurality of basic wave-beam subsets; each basic wave-beam subset in the plurality of basic wave-beam subsets is used to construct the related first precoding matrix. In some embodiments, at least one item of the basic wave-beam universal set, the plurality of basic wave-beam subsets and the plurality of first precoding matrixes is determined in advance or parts of the above items are determined through receiving an indication from a network node.

Description

Method and apparatus for constructing, selecting and scheduling precoding matrix for MIMO transmission system

Technical Field

Embodiments of the present invention relate to systems for communication using multiple-input multiple-output (MIMO), and more particularly, to a multi-user transmission scheme in a three-dimensional (3D) MIMO system based on dual polarization channels and using dual codebooks.

Background

The ever-increasing demand for data rates has driven the development of communication technologies. Increased transmission bandwidth is a straightforward means of increasing data rates, for example, in 3GPP LTE, carrier aggregation is utilized to provide communication bandwidths of up to 100 MHz. However, due to the limited nature of time and frequency resources, improving the efficiency of resource utilization has been a continuing goal in communication technology. Multi-user multiple input multiple output (MU-MIMO) technology can exploit new degrees of freedom with airspace and can effectively improve system throughput by serving multiple users simultaneously on the same time-frequency resource block. The research field of conventional MU-MIMO is limited to two-dimensional (2D) horizontal planes. Thus the downtilt angle is unique for each beam. The introduction of three-dimensional (3D) MIMO will more fully utilize spatial resources by adaptively changing the downtilt angle of each beam according to the location of the target user.

The implementation of MIMO techniques will depend on the utilization of Channel State Information (CSI), which typically requires the receiving end to estimate the channel state and feed it back to the transmitting end for calculating the appropriate precoding or beamforming parameters at the transmitting end. There have been some known studies in order to reduce the amount of feedback. For example, in european patent application EP2556638a1([1]), filed 4/6/2010, entitled "Codebook design and structure for multi-granular feedback," a precoding method based on a dual Codebook W = W1 × W2 is proposed for the case of high Rank (Rank) and low Rank, where W1 and W2 are a long-term precoding matrix (i.e., a precoding matrix calculated using long-term statistics of channel state information) and a short-term precoding matrix (i.e., a precoding matrix calculated using short-term statistics of channel state information), respectively. In addition, chinese patent application CN201110443683.4([2]), which was filed 12/27/2011 and entitled "a dual codebook-based multi-user adaptive feedback method", also mentions a dual codebook-based precoding method, and provides a corresponding multi-user transmission scheme. And a paper ([3]) entitled "estimated Feedback schedule for3D Multiuser MIMO based on Kronecker Product Codebook" published in Personal Index and Mobile Radio Communications (PIMRC)2013IEEE international conference, studied the 3D MU-MIMO system, and proposed a limited Feedback 3D beamforming method according to the Best partner Cluster (BCC) principle, in which a Kronecker Product Codebook is proposed. However, in none of these prior arts, it is mentioned how to perform effective precoding to support 3D-MIMO in case of dual-polarized antenna, and how to feed back effectively to improve the pairing rate of multiple users, thereby improving the spectrum utilization efficiency and improving the throughput of the system.

In fig. 1, an example of a system operation flow employing precoding in the prior art is shown. Where steps 101 and 102 are performed in the UE segment. In step 101, a User Equipment (UE) receives downlink reference signals, estimates long-term CSI statistics on the basis of the downlink reference signals, and calculates a long-term precoding matrix that is the best matched and the least matched for the UE using the long-term CSI statistics, e.g., obtains an index PMI1 of the long-term precoding matrix, and feeds back the index to a Base Station (BS). In step 102, the UE obtains an estimate of the composite channel from the long-term precoding matrix and the current short-term channel, selects the short-term precoding matrix based on the estimate of the composite channel (e.g., selects the appropriate index PMI2), and calculates a Channel Quality Indicator (CQI), which is fed back to the base station along with PMI 2. Based on the feedback of the UE, the base station performs the operations in steps 103-105. In step 103, the base station performs user clustering and pairing operations according to the long-term feedback value PMI 1; in step 104, the base station implements user scheduling according to the feedback value CQI; in step 105, the base station determines long-term and short-term precoding matrices used in downlink transmission according to the feedback values PMI1 and PMI2, and then determines the overall MU-MIMO adaptive transmission scheme.

Although a dual codebook based precoding scheme is proposed in the related art, for example, in the above-mentioned example, the degree of freedom of spatial resources is not fully utilized, and particularly, spatial resources in the vertical domain are not fully utilized. These schemes thus have difficulty supporting multi-user transmission in situations where multiple users are distributed on different floors of the same building.

Therefore, there is a need to utilize a new MU-MIMO transmission scheme to solve the above-mentioned problems, thereby more efficiently utilizing spatial resources.

Disclosure of Invention

Aiming at the problems in the prior art, the embodiment of the invention provides a MIMO precoding method and a corresponding multi-user transmission scheme with limited feedback.

According to an aspect of the present invention, a method for constructing a precoding matrix for a mimo transmission system is provided, which includes determining a plurality of first precoding matrices to form a first precoding codebook; wherein each of said first precoding matrices is associated with one of a plurality of basic beam subsets obtained from a full set of basic beams by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domain.

According to one embodiment of the invention, the method wherein the full set of basic beams is arranged in the form of a uniform planar array, wherein the beams of the same row have the same downtilt angle and the beams of the same column have the same azimuth angle; and the full set of basic beams is generated by a uniform planar antenna array.

According to a further embodiment of the invention, at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predetermined or determined at least in part by receiving an indication from a network node.

According to yet another alternative embodiment of the present invention, in the method said determining a plurality of first precoding matrices comprises determining a full set of said basic beams, wherein each of said basic beams is determined by a downtilt and an azimuth; dividing the full set of basic beams into a plurality of basic beam subsets; and constructing an associated first precoding matrix using each of the plurality of subsets of base beams.

According to one embodiment of the invention, in the method $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ Associating each of the first precoding matrices with one of a plurality of basic beam subsets, wherein W₁(k) Denotes the k-th first precoding matrix, X_kRepresents the kth basic beam subset and k is a non-negative integer value.

According to an embodiment of the invention, the method further comprises determining a plurality of second precoding matrices for constituting the dual codebook matrix.

According to another aspect of the present invention, there is provided a method for selecting precoding information for a mimo transmission system, including selecting a best matched basic beam and a least matched basic beam from a full set of basic beams according to long-term statistics of channel state information; wherein the full set of basic beams is divided into a plurality of basic beam subsets by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domains;

and selecting, from the plurality of basic beam subsets, a best matched basic beam subset and a least matched basic beam subset, the best matched basic beam subset comprising the selected best matched basic beam, and the least matched basic beam subset comprising the selected least matched basic beam; and selecting a corresponding first precoding matrix from a plurality of first precoding matrices according to the selected most matched basic beam subset and/or the most unmatched basic beam subset.

According to one embodiment of the invention, the method wherein the full set of basic beams is arranged in the form of a uniform planar array UPA, wherein the beams of the same row have the same downtilt and the beams of the same column have the same azimuth; and the full set of basic beams is generated by a uniform planar antenna array.

According to a further embodiment of the invention, at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predetermined or determined at least in part by receiving an indication from a network node.

According to an embodiment of the present invention, the selecting a corresponding first precoding matrix from a plurality of first precoding matrices according to the selected most matched basic beam subset and/or the most unmatched basic beam subset comprises: and obtaining the index of the corresponding first precoding matrix in the plurality of precoding matrices according to the selected index of the most matched basic beam subset and/or the most unmatched basic beam subset.

According to an embodiment of the present invention, the selecting a corresponding first precoding matrix from a plurality of first precoding matrices according to the selected most matched basic beam subset and/or the most unmatched basic beam subset comprises: according to the most matched basic beam subset and/or the most unmatched basic beam subset X_kAccording to $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ A respective first precoding matrix W selected from a plurality of first precoding matrices₁(k)。

According to another embodiment of the present invention, the selecting the best matched basic beam and the least mismatched basic beam from the full set of basic beams according to the long-term statistics of the channel state information in the method comprises: obtaining a characteristic vector of a channel correlation matrix according to the long-term statistics of the channel state information; selecting the best matched base beam and the least mismatched base beam based at least in part on the distance of the principal eigenvector of the eigenvectors from each base beam in the full set of base beams; and, in case the divided plurality of basic beam subsets partially overlap, selecting the most matched basic beam subset and the least unmatched basic beam subset at least partly according to a second one of the eigenvectors.

According to some embodiments of the invention, the method further comprises selecting a second precoding matrix for selecting a beam from the given basic beam subset based on the short-time channel state information.

In a further embodiment of the invention, the method further comprises selecting the cluster index CI by one of:

-selecting a CI equal to an index of the selected respective first precoding matrix; or

-arranging said plurality of basic beam subsets or said plurality of first pre-coding matrices associated with said plurality of basic beam subsets in the form of a two-dimensional uniform planar array, wherein the basic beams involved in each row of said uniform planar array have the same or similar downtilt angles and the basic beams involved in each column of said uniform planar array have the same or similar azimuth angles;

dividing the uniform planar array into a plurality of clusters that are uniform in a horizontal domain and non-uniform in a vertical domain, wherein at least one cluster comprises more than one basic beam subset or more than one first precoding matrix; and

selecting a first CI and a second CI, respectively, such that the first CI indicated cluster includes the selected best matching basic beam subset or a first precoding matrix associated with the best matching basic beam subset, and the second CI indicated cluster includes the selected least matching basic beam subset or a first precoding matrix associated with the least matching basic beam subset.

According to some embodiments of the invention, the method further comprises feeding back at least the selected precoding matrix index and the cluster index to a base station.

According to still another aspect of the present invention, there is provided a scheduling method for a mimo transmission system, including selecting a primary user to be scheduled at least partially according to a channel quality fed back by the user and a cluster index CI; and determining paired users scheduled simultaneously with the primary user; and when the paired users are determined to exist, scheduling the main user and the paired users simultaneously on the same time-frequency resource at least partially according to the channel quality fed back by the users, the first precoding matrix and the CI; wherein the first precoding matrix is selected from a plurality of first precoding matrices, each of which is associated with one basic beam subset, the plurality of basic beam subsets being obtained from a full set of basic beams by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domains.

According to some embodiments of the invention, the paired user is determined in the method by comparing the CI fed back by the other user with the CI fed back by the primary user, wherein the CI fed back by the other user and the primary user is selected by the other user and the primary user, respectively, by one of:

-selecting a CI equal to an index of the selected respective first precoding matrix; or

-arranging said plurality of basic beam subsets or said plurality of first pre-coding matrices associated with said plurality of basic beam subsets in the form of a two-dimensional uniform planar array, wherein the basic beams involved in each row of said uniform planar array have the same or similar downtilt angles and the basic beams involved in each column of said uniform planar array have the same or similar azimuth angles; dividing the uniform planar array into a plurality of clusters that are uniform in a horizontal domain and non-uniform in a vertical domain, wherein at least one cluster comprises more than one basic beam subset or more than one first precoding matrix; and

selecting a first CI and a second CI, respectively, such that the first CI indicated cluster includes the selected best matching basic beam subset or a first precoding matrix associated with the best matching basic beam subset, and the second CI indicated cluster includes the selected least matching basic beam subset or a first precoding matrix associated with the least matching basic beam subset.

According to a further embodiment of the invention, in the method the full set of basic beams is arranged in the form of a uniform planar array, wherein the beams of the same row have the same downtilt and the beams of the same column have the same azimuth; and, the full set of basic beams is generated by a uniform planar antenna array.

In some embodiments of the invention, at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices in the method is predetermined or determined at least in part by receiving an indication from a network node.

According to some embodiments of the invention, the method comprises $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ Associating each of the first precoding matrices with one of a plurality of basic beam subsets, wherein W₁(k) Denotes the k-th first precoding matrix, X_kRepresents the kth basic beam subset and k is a non-negative integer value.

According to still another aspect of the present invention, there is provided a precoding matrix constructing apparatus for a multiple-input multiple-output transmission system, including:

a first precoding matrix determining device for determining a plurality of first precoding matrices to form a first precoding codebook; wherein each of said first precoding matrices is associated with one of a plurality of basic beam subsets obtained from a full set of basic beams by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domain.

According to some embodiments of the invention, the full set of basic beams is arranged in the form of a uniform planar array, wherein the beams of the same row have the same downtilt angle and the beams of the same column have the same azimuth angle; and the full set of basic beams is generated by a uniform planar antenna array.

According to a further embodiment of the invention, at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predetermined or determined at least partly by receiving an indication from a network node, and the first precoding matrix determining means further comprises storing means for storing a predefined definition of at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices, and/or receiving means for determining at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices at least partly by receiving an indication from a network node.

According to still another embodiment of the present invention, the first precoding matrix determining apparatus further includes: beam defining means for determining a full set of said base beams, wherein each of said base beams is determined by a downtilt angle and an azimuth angle; subset dividing means for dividing the full set of basic beams into a plurality of basic beam subsets, and first precoding matrix constructing means for constructing an associated first precoding matrix with each of the plurality of basic beam subsets.

According to yet another embodiment of the invention, by $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ Associating each of the first precoding matrices with one of a plurality of basic beam subsets, wherein W₁(k) Denotes the k-th first precoding matrix, X_kRepresents the kth basic beam subset and k is a non-negative integer value.

According to some embodiments of the present invention, the apparatus for constructing a precoding matrix for a mimo transmission system further comprises a second precoding matrix determining device for determining a plurality of second precoding matrices for constructing a dual codebook matrix.

According to still another aspect of the present invention, there is provided a precoding information selecting apparatus for a multiple-input multiple-output transmission system, including: beam selection means for selecting a basic beam that is most matched with the current channel and a basic beam that is most unmatched with the current channel from the full set of basic beams according to long-term statistics of the channel state information; wherein the full set of basic beams is divided into a plurality of basic beam subsets by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domains;

beam subset selection means for selecting, from said plurality of basic beam subsets, a best matched basic beam subset and a least matched basic beam subset, said best matched basic beam subset comprising said selected best matched basic beam and said least matched basic beam subset comprising said selected least matched basic beam; and a first precoding matrix selecting device, configured to select a corresponding first precoding matrix from the plurality of first precoding matrices according to the selected most matched basic beam subset and/or the most unmatched basic beam subset.

According to one embodiment of the invention, the full set of basic beams is arranged in the form of a uniform planar array UPA, wherein the beams of the same row have the same downtilt and the beams of the same column have the same azimuth; and the full set of basic beams is generated by a uniform planar antenna array.

According to a further embodiment of the invention, at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predetermined or determined at least in part by receiving an indication from a network node.

In an embodiment of the present invention, the first precoding matrix selecting means is configured to obtain an index of the corresponding first precoding matrix in the plurality of precoding matrices according to the selected index of the most matched basic beam subset and/or the most unmatched basic beam subset.

In a further embodiment of the invention, the first precoding matrix selection means is arranged for selecting the most matched basic beam subset and/or the most unmatched basic beam subset X based on said most matched basic beam subset and/or said most unmatched basic beam subset_kAccording to $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ A respective first precoding matrix W selected from a plurality of first precoding matrices₁(k)。

According to some embodiments of the present invention, the beam selecting means is configured to obtain an eigenvector of a channel correlation matrix according to long-term statistics of the channel state information, select the best matched basic beam and the least matched basic beam according to at least a part of distances between a main eigenvector in the eigenvector and each basic beam in the full set of basic beams, and select the best matched basic beam subset and the least matched basic beam subset according to at least a part of a second eigenvector in the eigenvector in a case that the divided multiple basic beam subsets are partially overlapped.

According to some embodiments of the present invention, the apparatus for selecting precoding information for a mimo transmission system further comprises a second precoding matrix selecting device for selecting a second precoding matrix for selecting a beam from a given basic beam subset according to the short-term channel state information.

According to still another embodiment of the present invention, the apparatus for selecting precoding information for a multiple-input multiple-output transmission system further comprises Cluster Index (CI) selecting means for selecting a cluster index CI by one of:

-selecting a CI equal to an index of the selected respective first precoding matrix; or

-arranging said plurality of basic beam subsets or said plurality of first pre-coding matrices associated with said plurality of basic beam subsets in the form of a two-dimensional uniform planar array, wherein the basic beams involved in each row of said uniform planar array have the same or similar downtilt angles and the basic beams involved in each column of said uniform planar array have the same or similar azimuth angles;

dividing the uniform planar array into a plurality of clusters that are uniform in a horizontal domain and non-uniform in a vertical domain, wherein at least one cluster comprises more than one basic beam subset or more than one first precoding matrix; and

selecting a first CI and a second CI, respectively, such that the first CI indicated cluster includes the selected best matching basic beam subset or a first precoding matrix associated with the best matching basic beam subset, and the second CI indicated cluster includes the selected least matching basic beam subset or a first precoding matrix associated with the least matching basic beam subset.

According to another embodiment of the present invention, the apparatus further comprises a feedback device for feeding back at least the selected precoding matrix index and the cluster index to a base station.

According to yet another aspect of the present invention, there is provided a scheduling apparatus for a mimo transmission system, comprising user selection means for selecting a primary user to be scheduled and determining a paired user to be scheduled simultaneously with the primary user, at least partly according to a channel quality fed back by the user and a cluster index CI fed back by the user; and; the scheduling device is used for simultaneously scheduling the main user and the paired users on the same time-frequency resource at least partially according to the channel quality fed back by the users, a first precoding matrix and the CI under the condition that the paired users exist; wherein the first precoding matrix is selected from a plurality of first precoding matrices, each of which is associated with one basic beam subset, the plurality of basic beam subsets being obtained from a full set of basic beams by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domains.

According to some embodiments of the invention, the user selection means determines the paired user by comparing CIs of other user feedback with CIs of the primary user feedback, wherein CIs of the other user and the primary user feedback are selected by the other user and the primary user, respectively, by one of:

-selecting a CI equal to an index of the selected respective first precoding matrix; or

-arranging said plurality of basic beam subsets or said plurality of first pre-coding matrices associated with said plurality of basic beam subsets in the form of a two-dimensional uniform planar array, wherein the basic beams involved in each row of said uniform planar array have the same or similar downtilt angles and the basic beams involved in each column of said uniform planar array have the same or similar azimuth angles;

dividing the uniform planar array into a plurality of clusters that are uniform in a horizontal domain and non-uniform in a vertical domain, wherein at least one cluster comprises more than one basic beam subset or more than one first precoding matrix; and

selecting a first CI and a second CI, respectively, such that the first CI indicated cluster includes the selected best matching basic beam subset or a first precoding matrix associated with the best matching basic beam subset, and the second CI indicated cluster includes the selected least matching basic beam subset or a first precoding matrix associated with the least matching basic beam subset.

According to a further embodiment of the invention, wherein the full set of basic beams is arranged in the form of a uniform planar array, wherein the beams of the same row have the same downtilt angle and the beams of the same column have the same azimuth angle; and, the full set of basic beams is generated by a uniform planar antenna array.

According to an embodiment of the invention, at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predetermined or determined at least in part by receiving an indication from a network node.

In some embodiments of the invention, wherein the plurality of first precoding matrices pass $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ Associating each of the first precoding matrices with one of a plurality of basic beam subsets, wherein W₁(k) Denotes the k-th first precoding matrix, X_kRepresents the kth basic beam subset and k is a non-negative integer value.

Drawings

Embodiments of the invention will be described in more detail below with reference to the accompanying drawings, wherein like reference numerals refer to like parts. The drawings are not necessarily to scale, schematically illustrating exemplary embodiments. In the drawings, there is shown in the drawings,

FIG. 1 is an exemplary flow chart of system operation with precoding in the prior art;

FIG. 2 is an exemplary diagram of a single cell in a system employing 3D MU-MIMO;

FIG. 3 shows a flow diagram of a precoding construction method according to an example embodiment of the invention;

FIG. 4 is a schematic diagram of a given transmit beam;

an example of a dual polarized uniform planar antenna array is shown in fig. 5;

FIG. 6 is a schematic diagram of a full set of basic beams represented in the form of a UPA;

figures 7A-7D are schematic diagrams of a method of dividing a full set of basic beams into a plurality of basic beam subsets according to an embodiment of the present invention;

fig. 8 shows a schematic position diagram of a subset of basic beams in the full set of basic beams;

FIG. 9A is a schematic flow diagram of a precoding matrix selection method according to an embodiment of the invention;

fig. 9B illustrates an example of mapping of a cluster index and a first precoding matrix index according to an embodiment of the present invention;

FIG. 10 depicts a flowchart of an exemplary method of scheduling, according to an embodiment of the invention;

FIG. 11 illustrates an exemplary block diagram of an apparatus for precoding matrix construction according to an embodiment of the present invention;

fig. 12 is an exemplary structural diagram of an apparatus for selecting a precoding matrix according to an embodiment of the present invention;

fig. 13 is a schematic structural diagram of an apparatus for MU-MIMO scheduling according to an embodiment of the present invention; and

FIG. 14A illustrates sector throughput of a system using a method according to an embodiment of the invention; and

fig. 14B shows the cell throughput of a system using a method according to an embodiment of the invention.

Detailed Description

Fig. 2 shows an example of a single cell in a system employing 3D MU-MIMO, where a Base Station (BS) is located in the center of the cell and serves a total of K User Equipments (UEs), which may be assumed for simplicity to be evenly distributed throughout the cell. In this example, only three UEs are shown for clarity. Where UE1 and UE2 are located on different floors within the same 8-storey (the first floor being the ground) building. It should be noted that the cells shown in fig. 2 are merely illustrative and may in fact have coverage areas of other shapes.

In the scenario shown in fig. 2, for UE1 and UE2 located on different floors of the same building, traditional two-dimensional MIMO (2D-MIMO) sometimes cannot spatially distinguish them, resulting in that they often cannot be scheduled with the same time-frequency resource at the same time, i.e., MU-MIMO cannot be employed. This limits the efficiency of resource utilization. To solve this problem, 3D-MIMO is adopted in the embodiments of the present invention, and an effective precoding construction method, a precoding selection method, a scheduling method, and a corresponding apparatus are proposed to fully utilize horizontal and vertical spatial resources.

For the purpose of illustration, embodiments of the invention are described only in the context of 3GPP LTE-a, but as will be appreciated by those skilled in the art, embodiments of the invention are in no way limited to this application scenario in practice. Embodiments of the present invention may also be applied in other wireless communication systems, present and future, as long as they are compatible with the features of the present invention.

To facilitate understanding of the precoding method according to the embodiment of the present invention, a downlink is taken as an example below, and a transmission model is first described.

Transmission model

In the downlink, the receiving apparatus estimates a channel state using a downlink pilot signal and feeds back an estimated Channel Quality Indicator (CQI) to the base station. The base station then uses the feedback value for scheduling. For ease of explanation, a simplified model is used here as an example, where the base station serves two users simultaneously using MU-MIMO and the number of layer mappings is 1. Assuming that the two users served simultaneously at time t are user u and user s, the signal received by user u at time t can be expressed as:

y_u，t=H_u，tW_u，tx_u，t+H_u，tW_s，tx_s,t+n_u，t(1) wherein x_u，t，x_s，tIs a symbol transmitted by the base station that satisfies the following power constraint:W_u，tand W_s，tPrecoding matrices for user u and user s at time t, respectively; h_u，tIs the received signal y of user u_u，tThe channel experienced; n is_u，tComplex gaussian noise is represented, which follows a normal distribution:

further assuming that a Minimum Mean Square Error (MMSE) receiver is employed at the receiving end, the signal-to-interference-and-noise ratio (SINR) of the received signal obtained at user u can be represented by the following equation:

<math> <mrow> <msub> <mi>SINR</mi> <mi>u</mi> </msub> <mrow> <mo>(</mo> <msub> <mi>W</mi> <mi>u</mi> </msub> <mo>,</mo> <msub> <mi>W</mi> <mi>s</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <msubsup> <mi>f</mi> <mi>u</mi> <mrow> <mrow> <mo>(</mo> <mi>u</mi> <mo>)</mo> </mrow> <mi>H</mi> </mrow> </msubsup> <msup> <mrow> <mo>(</mo> <msubsup> <mi>f</mi> <mi>u</mi> <mrow> <mo>(</mo> <mi>s</mi> <mo>)</mo> </mrow> </msubsup> <msubsup> <mi>f</mi> <mi>u</mi> <mrow> <mrow> <mo>(</mo> <mi>s</mi> <mo>)</mo> </mrow> <mi>H</mi> </mrow> </msubsup> <mo>+</mo> <mfrac> <msup> <mrow> <mn>2</mn> <mi>&sigma;</mi> </mrow> <mn>2</mn> </msup> <mi>P</mi> </mfrac> <mi>I</mi> <mo>)</mo> </mrow> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <msubsup> <mi>f</mi> <mi>u</mi> <mrow> <mo>(</mo> <mi>u</mi> <mo>)</mo> </mrow> </msubsup> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow> </math>

whereinAs can be seen from the above formula, in order to maximize SINR and minimize inter-user interference, the precoding matrix W_uAnd W_sIs very critical. In one embodiment of the invention, dual codebook based precoding is used, i.e. precoding matrices W for users u and s_uAnd W_sCan all be expressed as W₁×W₂Form (1), wherein W₁Is a long-term precoding matrix, and W₂Is a short-term precoding matrix.

It should be noted that although some parameters are simplified herein for simplicity, such as the number of users is 2, the rank is 1, a dual codebook is employed, an MMSE receiver is employed, etc., in other embodiments, different parameters may be employed, such as a greater number of users, a higher rank, single codebook precoding, etc., and thus embodiments of the present invention are not limited thereto. Various possible setup parameters are not listed here, as these parameters do not affect the understanding of the embodiments of the present invention.

The following respectively describes a precoding construction method, a precoding selection method, a scheduling method, and corresponding devices of embodiments of the present invention.

Construction of precoding matrices

According to the embodiment of the invention, the precoding matrix constructing method comprises the steps of determining a plurality of first precoding matrixes to form a first precoding codebook; wherein each of said first precoding matrices is associated with one of a plurality of basic beam subsets obtained from a full set of basic beams by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domain.

Fig. 3 shows a flowchart of a method for precoding construction according to an exemplary embodiment of the present invention. As shown in fig. 3, the method comprises a step 301 of determining a total set of basic beams into which the total set of basic beams is divided into a plurality of basic beam subsets X_k302 and determining to utilize each of the basic beam subsets X_kConstructed associated first precoding matrix W₁(k) Step 303.

In step 301, each of the full set of determined base beams is determined by a downtilt and an azimuth, according to an embodiment of the present invention. The downtilt angle γ for a given transmit beam (grey arrow direction) is schematically shown in fig. 4_tAnd azimuth angle theta_t。

According to an embodiment of the invention, the full set of basic beams is generated by a uniform plane antenna array, e.g. a dual polarized uniform plane antenna array (UPA) topology, which has been adopted by the 3GPP as recommended by the 3D channel model for base station antenna arrays. An example of a dual polarized uniform planar antenna array is shown in fig. 5. In this example, in the horizontal direction there isFor dual-polarized antenna, there areFor dual polarized antennas, they form a uniform planar antenna array. Each beam produced by the antenna array is determined by the downtilt and azimuth angles, so that the entire space can also be divided using the downtilt and azimuth angles and thereby define a full set of basic beams.

According to one embodiment of the invention, in step 301, the full set of fundamental beams is determined by the following operations. In [0, pi ]]Select N between_vA down tilt angle, and at [0, 2 π]Selection of N_hAn azimuth, then combining these values can coverThe beam directions, which constitute the full set of basic beams.

According to one embodiment of the invention, the Kronecker Product Codebook (KPC) in document [3] mentioned in the background section herein can be selected as the full set of basic beams.

According to another embodiment of the invention, the full set of basic beams is predefined, so that it can be determined by accessing predefined content in step 301.

According to a further embodiment of the invention, the full set of basic beams can be determined in step 301 by receiving predefined information about the full set of basic beams from a network node (e.g. a base station).

According to an embodiment of the present invention, the determined full set of basic beams may be arranged in the form of a UPA as shown in fig. 6. In the example shown in fig. 6, beams on the same row share the same downtilt angle, but have different azimuth angles as shown on the left side of fig. 6; while beams on the same column share the same azimuth but have different downtilts as shown on the right in fig. 6. Each elementary beam c_iCan be expressed as N_TxVector of X1, where N_TxI is 0 to the number of antennas used to generate the beamIs an integer of (1).

According to one embodiment of the invention, the full set of basic beams is divided into a plurality of basic beam subsets X, step 302_kWherein K =0, 1, ·, K,k is the number of basic beam subsets. According to an embodiment of the present invention, the plurality of basic beam subsets may be obtained by dividing the full set of basic beams into uniform non-overlapping subsets, as shown in fig. 7A. In the example of fig. 7A, the full set of elementary beams is arranged in a 4 row 32 column uniform planar array, with the elementary beams of each row having the same downtilt angle and the elementary beams of each column having the same azimuth angle. The array is uniformly divided into K =16 non-overlapping basic beam subsets, i.e. X₀To X₁₅. Each basic beam subset X_kHave similar or similar downtilt and azimuth characteristics.

According to another embodiment of the present invention, the plurality of basic beam subsets may be obtained by dividing the full set of basic beams into uniform but partially overlapping subsets, as shown in fig. 7B-7D. In fig. 7B, the divided basic beam subset partially overlaps with an adjacent basic beam subset in the horizontal domain. E.g., subset X in FIG. 7B₁And X₂With overlapping portions C₄、C₅、C₃₆、C₃₇. In fig. 7C, the divided subset of basic beams partially overlaps with the adjacent basic beam subset in the vertical domain. E.g., subset X in FIG. 7C₀=[C₀、C₁、C₂、C₃、C₃₂、C₃₃、C₃₄、C₃₅]And X₈=[C₃₂、C₃₃、C₃₄、C₃₅、C₆₄、C₆₅、C₆₆、C₆₇]With overlapping portions C₃₂-C₃₅. In fig. 7D, the divided basic beam subsets are partially overlapped with the adjacent basic beam subsets in both the horizontal and vertical domains. The method of obtaining overlapping basic beam subsets shown in fig. 7B-7D enables to generate more basic beam subsets, i.e. more X_k. For example, in fig. 7B-7D, 32, and 64 basic beam subsets are obtained by dividing the same full set of basic beams as in fig. 7A, respectively, while 16 basic beam subsets are obtained according to the method of fig. 7A.

According to a further embodiment of the invention, the above-described operations need not be performed in step 302, but instead the subset X of basic beams resulting from the above-described operations_kMay be predefined and stored, in which case the basic beam subset can be obtained by accessing predefined information in step 302.

According to another embodiment of the invention, the subset X of elementary beams obtained as described above_kMay be predefined and stored (e.g. in the network node or in both the network node and the user), e.g. subsets of basic beams obtained according to different splitting methods may be predefined, respectively. In this case, the basic beam subset X can be determined in step 302 by receiving further signaling from the network node indicating the basic beam subset, e.g. signaling indicating which partitioning method resulting basic beam subset should be applied_k。

According to one embodiment of the invention, in step 303, the method comprises $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ To construct each of the plurality of basic beam subsets X_kThe associated precoding matrix W₁(k) Where K =0, 1, …, K-1. Basic beam subset X_kIs determined in step 302, each X_kFrom N in the horizontal direction_bhOne beam and N in the vertical direction_bvThe number of beams that form, i.e., $X_k=\left[\begin{array}{cccc}{c}_i& c_{i+1}& ...& c_{i+N_b-1}\end{array}\right],N_b=N_{\mathrm{bv}}{N}_{\mathrm{bh}}$ . In the example shown in FIG. 7A, N_bh=4，N_bvAnd (2). Fig. 8 shows a schematic position diagram of a subset of basic beams in the full set of basic beams.

According to another embodiment of the present invention, the precoding matrix constructed according to the above steps may be predefined and stored, and thus, step 303 can determine the precoding matrix by accessing predefined information without performing the above operation in such an embodiment.

According to a further embodiment of the present invention, the flow of the precoding construction method further comprises a step 304 for constructing a second precoding matrix W₂In one embodiment, W₂In the form of <math> <mrow> <msub> <mi>W</mi> <mn>2</mn> </msub> <mo>=</mo> <mfenced open='[' close=']'> <mtable> <mtr> <mtd> <msubsup> <mi>e</mi> <mi>i</mi> <mrow> <mo>(</mo> <msub> <mi>N</mi> <mi>b</mi> </msub> <mo>)</mo> </mrow> </msubsup> </mtd> </mtr> <mtr> <mtd> <msubsup> <mi>&alpha;e</mi> <mi>i</mi> <mrow> <mo>(</mo> <msub> <mi>N</mi> <mi>b</mi> </msub> <mo>)</mo> </mrow> </msubsup> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> </mrow> </math> Wherein e_i ^(Nb)For selecting a matrix for use from the basic beam subset X_kN in (1)_bSelecting one basic beam from the basic beams; according to one embodiment of the invention, e_i ^(Nb)Is N_bx1, where only 1 element is 1 and the rest are zero. And alpha is a phase factor, in some embodiments, a finite phase value is defined, e.g., alpha = [1 j-1-j)]. Having 4 phase values at α and e_i ^(Nb)With N_bIn one possible example of values, W₂Has 4N_bA possible representation of, and thus constructed W₂Having a codebook size of 4N_b. In case alpha has another number of phases, W₂The codebook size can be calculated accordingly, and will not be described herein.

In some embodiments, the second precoding matrix W2 that conforms to the above construction principle may also be predefined, and thus, can be determined by accessing predefined information in step 304.

As can be appreciated by those skilled in the art, some steps in the flow chart shown in fig. 3 may be omitted in some embodiments. For example, when the first domain coding matrix generated using the above-mentioned rules has been predefined, step 301 and 302 may be omitted, and the precoding matrix may be determined in step 303 by accessing predefined information as described above. In embodiments employing only a single codebook, for example, step 304 may be omitted.

Selection of precoding matrices

After the construction of the precoding matrix is obtained according to the method in fig. 3, a device serving as a receiving end of MIMO transmission (the receiving end device is a user equipment when MIMO is used for downlink transmission of wireless communication) needs to estimate a channel state, and then feeds back an index of a precoding matrix selected from the constructed precoding matrix and other information to a transmitting end so that the transmitting end determines an appropriate transmission parameter. According to an embodiment of the present invention, the method for selecting a precoding matrix at a receiving end includes operations as shown in fig. 9A, in which:

in step 901, selecting the best matched basic beam and the least matched basic beam from the full set of basic beams according to the long-term statistics of the channel state information; wherein the full set of basic beams is divided into a plurality of basic beam subsets;

at step 902, selecting a best matching base beam subset and a least matching base beam subset from the plurality of base beam subsets such that the best matching base beam subset comprises the selected best matching base beam and the least matching base beam subset comprises the selected least matching base beam;

in step 903, a selected first precoding matrix is obtained according to the selected best matching basic beam subset and/or least matching basic beam subset.

According to an embodiment of the present invention, the full set of basic beams from which the best matched basic beam and the least matched basic beam are selected in step 901 may be predefined and arranged in the form of a Uniform Planar Array (UPA), for example, in the form as shown in fig. 6, wherein the beams of the same row have the same downtilt angle and the beams of the same column have the same azimuth angle; according to an embodiment of the invention, the full set of basic beams is generated by a dual polarized uniform plane antenna array, for example as shown in fig. 5.

According to an embodiment of the invention, the full set of basic beams is divided into said plurality of basic beam subsets by one of the following methods:

-dividing the full set of basic beams into uniform non-overlapping subsets to obtain the plurality of basic beam subsets, e.g. as shown in fig. 7A;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain to obtain the plurality of basic beam subsets, e.g. as shown in fig. 7B;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain to obtain the plurality of basic beam subsets, e.g. as shown in fig. 7C;

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domain to obtain the plurality of basic beam subsets, e.g. as shown in fig. 7D.

According to an embodiment of the present invention, in step 901, the best matching base beam and the least matching base beam are selected by the following operations. First, a correlation matrix of a channel is calculated according to long-term statistics of channel state information,andfor the correlation matrixSingular Value Decomposition (SVD) operation is carried out to obtain a main eigenvector v₁. Calculating v₁And all the basic beams (c in fig. 6)_i) Then the distance from v is selected₁Nearest beam c_iAnd the farthest beam c_j. According to an embodiment of the invention, selected c_iAnd c_jRespectively satisfy the following formula:

<math> <mrow> <mi>i</mi> <mo>=</mo> <mi>arg</mi> <munder> <mi>max</mi> <mrow> <mi>i</mi> <mo>&Element;</mo> <mo>[</mo> <mn>1,2</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <msub> <mi>N</mi> <mi>v</mi> </msub> <msub> <mi>N</mi> <mi>h</mi> </msub> <mo>]</mo> </mrow> </munder> <mfrac> <mrow> <mo>|</mo> <msubsup> <mi>v</mi> <mn>1</mn> <mi>H</mi> </msubsup> <msub> <mi>c</mi> <mi>i</mi> </msub> <mo>|</mo> </mrow> <mrow> <mo>|</mo> <mo>|</mo> <msub> <mi>v</mi> <mn>1</mn> </msub> <mo>|</mo> <mo>|</mo> <mo>&CenterDot;</mo> <mo>|</mo> <mo>|</mo> <msub> <mi>c</mi> <mi>i</mi> </msub> <mo>|</mo> <mo>|</mo> </mrow> </mfrac> <mo>,</mo> <mi>j</mi> <mo>=</mo> <mi>arg</mi> <munder> <mi>min</mi> <mrow> <mi>i</mi> <mo>&Element;</mo> <mo>[</mo> <mn>1,2</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <msub> <mi>N</mi> <mi>v</mi> </msub> <msub> <mi>N</mi> <mi>h</mi> </msub> <mo>]</mo> </mrow> </munder> <mfrac> <mrow> <mo>|</mo> <msubsup> <mi>v</mi> <mi>I</mi> <mi>H</mi> </msubsup> <msub> <mi>c</mi> <mi>j</mi> </msub> <mo>|</mo> </mrow> <mrow> <mo>|</mo> <mo>|</mo> <msub> <mi>v</mi> <mn>1</mn> </msub> <mo>|</mo> <mo>|</mo> <mo>&CenterDot;</mo> <mo>|</mo> <mo>|</mo> <msub> <mi>c</mi> <mi>j</mi> </msub> <mo>|</mo> <mo>|</mo> </mrow> </mfrac> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>3</mn> <mo>)</mo> </mrow> <mo>.</mo> </mrow> </math>

since each basic beam is comprised in one basic beam subset, in step 902, its corresponding basic beam subset can be determined from the selected best matching basic beam and the least matching basic beam, respectively. For example if the basic beam c in fig. 7A₃Selected as the best matching basic beam, then the corresponding basic beam subset is X₀. If the basic beam subsets are partially overlapping, as is the case for example in fig. 7B-7D, each basic beam selected is comprised in a plurality of basic beam subsets, in which case the basic beam subset to be selected cannot be uniquely determined by the best basic beam selected. According to one embodiment of the invention, in this case the second eigenvector v of the correlation matrix of the channel may be utilized₂To further determine a final selected basic beam subset from the plurality of basic beam subsets to which the selected beam corresponds. For example, comparing beams in a plurality of basic beam subsets with a second eigenvector v₂The distance of (c).

In step 903, a selected first precoding matrix is further determined from the selected basic beam subset. According to an embodiment of the invention, the basic beam subset X_kAnd a corresponding first precoding matrix W₁(k) The relationship of (1) is: $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right].$ thus, according to one embodiment of the invention, the index of the first precoding matrix is directly determined by the index of the selected basic beam subset.

The method of fig. 9A further includes additional steps, such as the step shown in the dashed box in fig. 9A, according to an embodiment of the present invention. Wherein the second precoding matrix is selected based on the short-time channel state information at step 904. According to an embodiment of the invention, the selected second precoding matrix finally determines the transmission beam and maximizes the power of the received signal, e.g. the second precoding matrix is selected according to the following principle:

${\mathrm{PMI}2}_u=\arg \underset{i=[1,...,4n_b]}{\max }\left({\left|\right|{\tilde{H}}_u{W}_2^{\left(i\right)}\left|\right|}^2\right),$ wherein ${\tilde{H}}_u=H_u{W}_1^{\left({\mathrm{PMI}1}_u\right)}$ (4)。

The method of fig. 9A further includes step 905, in which CQI is calculated, according to one embodiment of the present invention. According to the embodiment of the present invention, the selected long-term precoding matrix, the short-term precoding matrix, and the short-term precoding matrix of the possible interfering user are considered in the calculation of CQI, for example, the CQI for user u may be calculated as shown in the following formula:

${\mathrm{CQI}}_u=\underset{j=[1,...,{4N}_b]}{\min }\left\{{\mathrm{SINR}}_u(W_1^{\left({\mathrm{PMI}1}_u\right)}{W}_2^{\left({\mathrm{PMI}2}_u\right)},W_1^{\left({\mathrm{PMI}1}_s\right)}{W}_2^{\left(j\right)})\right\}---\left(5\right)$

the calculation of the above equation assumes that the interfering user uses the least matching long-term precoding matrix selected by user u to minimize interference, and at the same time, considers all the short-term precoding matrices that the interfering user may select, and then selects the lowest CQI as its conservatively estimated CQI. However, in other embodiments, different CQI calculation strategies may be employed, for example, instead of selecting the worst CQI, the average CQI is selected as the final CQI estimate.

The method of FIG. 9A further includes step 906, wherein a Cluster Index (CI) is selected, in accordance with one embodiment of the present invention. According to an embodiment of the present invention, the CI may be obtained by making the selected CI equal to the index of the selected first precoding matrix.

According to another embodiment of the present invention, when the total set of employed basic beams is large, or when the number of basic beam subsets is large, in order to improve the user pairing rate of MU-MIMO, one cluster may be made to correspond to a plurality of precoding matrices, i.e., to a plurality of basic beam subsets; in this case, CI is obtained by: arranging a plurality of basic beam subsets or a plurality of first pre-coding matrices associated with the plurality of basic beam subsets in the form of a two-dimensional (2D) uniform planar array, wherein basic beams involved in each row of the uniform planar array have the same or similar downtilt angles, and basic beams involved in each column of the uniform planar array have the same or similar azimuth angles; dividing the uniform planar array into a plurality of clusters that are uniform in a horizontal domain and non-uniform in a vertical domain, wherein at least one cluster comprises more than one basic beam subset or more than one first precoding matrix; the first CI and the second CI are respectively selected such that the cluster indicated by the first CI includes the selected best matching basic beam subset or the first precoding matrix associated with the best matching basic beam subset, and the cluster indicated by the second CI includes the selected least matching basic beam subset or the first precoding matrix associated with the least matching basic beam subset. The benefit of this choice of CI is to ensure a reliable pairing rate and less inter-user interference.

According to an embodiment of the invention, the mapping of clusters to basic beam subsets or the mapping of clusters to first coding matrices may be predefined, i.e. which basic beam subsets or first precoding matrices a given cluster is mapped to are predefined. The CI is directly derived from the selected basic beam subset or the first precoding matrix according to the predefined mapping in step 906.

According to another embodiment of the invention, a mapping of clusters to a subset of basic beams, or a mapping of clusters to a first coding matrix, is obtained by means of information from a network node, and the CI is derived from the selected subset of basic beams or the first precoding matrix in step 906 on the basis of the obtained mapping information.

An example of mapping of clusters to the first precoding matrix is shown in fig. 9B. As shown in fig. 9B, the first precoding matrix is divided into a plurality of clusters that are uniform in a horizontal domain and non-uniform in a vertical domain, wherein at least one cluster includes more than one basic beam subset or more than one first precoding matrix. According to one embodiment of the present invention, the plurality of first precoding matrices arranged in the uniform planar array correspond to basic beams having the same or similar downtilt angles at each row and basic beams having the same or similar azimuth angles at each column. Such a mapping will therefore result in a similar number of azimuths, and a different number of downtilts, being covered in each cluster. According to the results of practical system experiments and tests, it can be found that the distribution of the azimuth angles actually adopted in the system is relatively uniform, while the distribution of the declination angles is not uniformly distributed, so that the distribution can be more suitable for the characteristics, and the utilization rate of each cluster is improved. The method is particularly suitable for the situation that the number of basic beam subsets is large. In consideration of the fact that in order to reduce the size of the short-time codebook, the long-time precoding codebook contains fewer basic beams, which results in more careful division of the full set of basic beams and more basic beam subsets. In this case, if the index of the long precoding matrix PMI1 is still used as the basis for scheduling and pairing of users, the success rate of pairing will not be too low only under the condition of a large number of users. Thus, in this example, proximally located PMIs 1 are grouped into one CI, so that the number of CIs is reduced to increase the pairing rate in a manner that reduces the set of partners. After the PMI1 is arranged in the UPA topology, considering the limitation of the user distribution in the vertical domain and the nonuniformity of the vertical beam pair space division, a uniform merging criterion is followed in the horizontal direction, and according to the sparsity of the beam downtilt angle in the vertical direction, for example, 1: 3 the non-uniform merging criterion performs the partitioning of the clusters. The CI computed according to the example method can maximize user pairing success rate while ensuring that interference between paired users is as small as possible.

According to an embodiment of the present invention, the method of fig. 9A further includes a step 907 of feeding back the index of the selected precoding matrix (e.g., the index of the selected first precoding matrix and the second precoding matrix), the cluster index, and Channel Quality Information (CQI) to the base station. It should be noted that the periods of these information feedbacks may not be the same, and therefore, they are not necessarily fed back at the same time. For example, the first precoding matrix index and CI may have the same feedback period P1, and the second precoding matrix index and CQI may have another feedback period P2.

The above is an introduction of the construction, selection, and feedback method of the precoding matrix according to the embodiment of the present invention. The following describes a method for scheduling by a transmitting end in an embodiment of the present invention based on obtaining feedback from a receiving end.

Scheduling for MU-MIMO

In fig. 10, a flow diagram of an exemplary method of scheduling according to an embodiment of the present invention is shown. In this example, the scheduling method includes the operations of:

in step 1001, selecting a primary user to be scheduled and determining a paired user capable of being scheduled simultaneously with the primary user, at least partially according to information related to channel quality and a cluster index fed back by the user;

in step 1002, when it is determined that the paired user exists, the primary user and the paired user are scheduled simultaneously on the same time-frequency resource at least partially according to the information related to the channel quality, the first precoding matrix, and the cluster index, which are fed back by the user.

According to an embodiment of the invention, the fed-back first precoding matrix referred to in step 1001 is selected from a plurality of first precoding matrices, each of which is associated with one basic beam subset. In one embodiment, the plurality of first precoding matrices are passed $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ Is associated with the plurality of basic beam subsets, wherein W₁(k) Denotes the k-th first precoding matrix, X_kRepresents the kth basic beam subset and k is a non-negative integer value.

In one embodiment, the basic beam subset is a plurality of basic beam subsets X obtained by uniformly dividing the full set of basic beams_kFor example, one of the methods described with respect to fig. 7A-7D.

According to one embodiment of the invention, the full set of basic beams is arranged in the form of a Uniform Planar Array (UPA) in which beams of the same row have the same downtilt and beams of the same column have the same azimuth; and the full set of basic beams is generated by a uniform planar antenna array.

According to yet another embodiment of the invention, wherein at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predefined or determined at least in part by receiving an indication from a network node.

According to an embodiment of the present invention, the cluster index fed back by the user is selected by the user according to the method described with respect to step 906 in fig. 9A, for example. The CI includes a belonging set CI (first CI) and a best interference partner set CI (second CI) corresponding to a best matching base beam subset and a least matching base beam subset, respectively.

According to the embodiment of the present invention, in step 1001, a base station (base station) sorts CQIs fed back by each user, selects a maximum CQI, and considers a UE corresponding to the maximum CQI as a primary user, assuming that the user is u.

According to the embodiment of the invention, in step 1001, after the primary user u is determined, the user paired with the primary user is determined by comparing the CI fed back by the primary user with the CI fed back by other users. For example, if the first CI and the second CI fed back by the main user u are (CIi)_u，CIj_u) Then if the CI fed back by another user happens to be (CIj)_u，CIi_u) Then the base station considers the user as a potential pairing user of the primary user.

According to an embodiment of the present invention, if a plurality of potential paired users are obtained by comparing CIs, step 1001 further includes determining one paired user scheduled simultaneously with the primary user by comparing channel quality indicators fed back by the plurality of potential paired users. For example, the potential paired user with the largest CQI may be selected as the final paired user.

According to the embodiment of the invention, under the condition that the paired users are found in the step 1001, in a step 1002, the base station simultaneously schedules the main user and the paired users on the same time-frequency resource to carry out MU-MIMO transmission; in step 1001, if no paired user is found, in step 1002, the base station randomly selects one user from the best interference partner set determined by the secondary CI of the primary user, and schedules the user at the same time with the primary user, or schedules only the primary user, so as to avoid excessive interference.

Embodiments of apparatus corresponding to the above-described method are described below with reference to fig. 11-13.

An exemplary structure of a construction apparatus 1100 of a precoding matrix for a multiple-input multiple-output transmission system is shown in fig. 11.

According to some embodiments of the present invention, the constructing apparatus 1100 comprises a first precoding matrix determining means 110 for determining a plurality of first precoding matrices to construct a first precoding codebook;

wherein each of said first precoding matrices is associated with one of a plurality of basic beam subsets obtained from a full set of basic beams by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domains

According to an embodiment of the present invention, a precoding matrix codebook obtained according to the method for constructing a precoding matrix illustrated in fig. 3 may be predefined and stored at a transmitting end and a receiving end. In this case, for determining the precoding matrix codebook, the first precoding matrix determining means 110 does not need a functional unit for performing the operations shown in fig. 3, but can be implemented by the storage means 1101 and the means for controlling access 1102.

According to another embodiment of the present invention, a plurality of precoding matrix codebooks obtained according to the method for constructing precoding matrices as described in fig. 3 may be predefined, and in order to determine the precoding matrix codebook, at the receiving end, the first precoding matrix determining apparatus 110 includes a receiving apparatus 1103 for receiving an indication from a transmitting end (e.g. a network node, a base station) as to which precoding matrix codebook should be applied, thereby determining the precoding matrix codebook.

According to a further embodiment of the present invention, the first precoding matrix determining means 110 alternatively comprises functional units for performing the operations shown in fig. 3, for example, in this case, in order to determine a precoding matrix codebook, the first precoding matrix determining means 110 comprises the following means:

beam defining means 1111 for determining a full set of basic beams, wherein each basic beam is determined by a down dip and an azimuth;

subset dividing means 1112 for dividing said full set of basic beams into a plurality of basic beam subsets X_k，

A first precoding matrix constructing means 1113 for utilizing each basic beam subset X_kConstructing an associated first precoding matrix W₁(k) All constructed precoding matrixes form a first precoding codebook W₁。

According to an embodiment of the present invention, the full set of basic beams determined in the beam defining means 1111 is arranged in the form of a Uniform Planar Array (UPA) in which the beams of the same row have the same downtilt angle and the beams of the same column have the same azimuth angle; and the full set of basic beams is generated by a uniform planar antenna array.

According to an embodiment of the invention, the beam defining means 1111 determines the full set of basic beams by performing the operations described with respect to step 301 in fig. 3. For example, by applying a voltage at [0, π]Select N between_vA down tilt angle, and at [0, 2 π]Selection of N_hAn azimuth angle, combining these values to coverIndividual beam directions, thereby forming a full set of basic beams; or the full set of the basic beams is predefined, so that the beam defining means 1111 can determine the full set of the basic beams by accessing predefined contents; alternatively, the beam defining means 1111 may be able to determine the basic beam by receiving predefined information on the full set of basic beams from a network node (e.g. base station) (or by the receiving means 1103)The complete set of (a).

According to the embodiment of the present invention, the subset division device 1112 obtains the plurality of basic beam subsets X by one of the following methods_k：

-dividing the full set of basic beams into uniform non-overlapping subsets to obtain the plurality of basic beam subsets;

-dividing the full set of basic beams into uniform, horizontally partially overlapping subsets to obtain the plurality of basic beam subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain to obtain the plurality of basic beam subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domain to obtain the plurality of basic beam subsets.

Likewise, in embodiments where a plurality of basic beam subsets are predefined, the subset partitioning means 1112 may also determine the plurality of basic beam subsets by accessing predefined content and/or by receiving corresponding information from the network node.

According to still another embodiment of the present invention, the first precoding matrix constructing means 1113 is constructed by $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ To construct each basic beam subset X of the plurality of basic beam subsets determined by the subset dividing means 1112_kAssociated first precoding matrix W₁(k)。

According to another embodiment of the invention, the constructing device 1100 further comprises a second precoding matrix constructing means 1114 for constructing a second precoding matrix W₂E.g. W₂In the form of <math> <mrow> <msub> <mi>W</mi> <mn>2</mn> </msub> <mo>=</mo> <mfenced open='[' close=']'> <mtable> <mtr> <mtd> <msubsup> <mi>e</mi> <mi>i</mi> <mrow> <mo>(</mo> <msub> <mi>N</mi> <mi>b</mi> </msub> <mo>)</mo> </mrow> </msubsup> </mtd> </mtr> <mtr> <mtd> <msubsup> <mi>&alpha;e</mi> <mi>i</mi> <mrow> <mo>(</mo> <msub> <mi>N</mi> <mi>b</mi> </msub> <mo>)</mo> </mrow> </msubsup> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> </mrow> </math> Wherein e_i ^(Nb)To select a matrix, which is N_bX1 vector with only 1 element being 1 and the remainder being 0, for use in deriving from the basic beam subset X_kSelecting a basic beam; and alpha is a phase factor having a finite number of values.

It should be noted that each of the devices shown in the schematic structure of fig. 11 is a logical unit, and thus, they may be implemented by hardware, software, firmware, or a combination thereof; in addition, the functions of several of the devices shown in the figures may be performed by a single device in some embodiments, or the functions of one of the devices shown in the figures may be performed by multiple devices in some embodiments. Furthermore, in some embodiments, some of the means shown in fig. 11 may be omitted, for example, in some embodiments, the first precoding matrix determining means 110 may comprise only means 1100, 1101, or only 1102, or only means 1111, 1113; and/or other devices not shown in the figures may also be included in some embodiments.

Fig. 12 is an exemplary structure of an apparatus 1200 for selecting a precoding matrix. In this example, the apparatus comprises:

a beam selecting unit 1201, configured to select a best-matched basic beam and a least-matched basic beam from the full set of basic beams according to long-term statistics of the channel state information; wherein the full set of basic beams is divided into a plurality of basic beam subsets;

beam subset selecting means 1202 for selecting, from the plurality of basic beam subsets, a best matching basic beam subset and a least matching basic beam subset, the best matching basic beam subset comprising the selected best matching basic beam and the least matching basic beam subset comprising the selected least matching basic beam;

first precoding matrix selecting means 1203, configured to obtain a selected first precoding matrix according to the selected most matched basic beam subset and/or least matched basic beam subset.

According to one embodiment of the invention, the full set of basic beams from which the best matching and the least matching basic beams are selected is predefined. In some embodiments, the full set of basic beams is arranged in the form of a Uniform Planar Array (UPA) in which beams of the same row have the same downtilt and beams of the same column have the same azimuth; and the full set of basic beams is generated by a uniform planar antenna array.

According to yet another embodiment of the invention, wherein the full set of basic beams is divided into the plurality of basic beam subsets by one of the following methods:

-dividing the full set of basic beams into uniform non-overlapping subsets to obtain the plurality of basic beam subsets;

-dividing the full set of basic beams into uniform horizontally overlapping subsets to obtain the plurality of basic beam subsets;

-dividing the full set of basic beams into uniform subsets overlapping in the vertical domain to obtain the plurality of basic beam subsets;

-dividing the full set of basic beams into uniform subsets overlapping in both the vertical and horizontal domain to obtain the plurality of basic beam subsets.

According to one embodiment of the invention, the plurality of basic beam subsets obtained by the above method is predefined.

According to an embodiment of the present invention, beam selection apparatus 1201 is configured to perform the functions described in conjunction with step 901 in fig. 9A; according to an embodiment, the apparatus 1201 obtains eigenvectors of the channel correlation matrix according to long-term statistics of the channel state information, and selects the best matching base beam and the least matching base beam at least partially according to distances between a main eigenvector in the eigenvectors and each base beam in the full set of base beams, and the beam subset selecting apparatus 1202 is configured to perform the functions described in conjunction with step 902 in fig. 9A; according to an embodiment, the apparatus 1202 selects the best matching basic beam subset and the least matching basic beam subset based at least in part on a second eigenvector of the eigenvectors and the selected best matching basic beam and the least matching basic beam in case of an overlap of the divided basic beam subsets.

According to another embodiment of the present invention, the first precoding matrix selection means 1203 is configured to perform the functions described in connection with step 903 in fig. 9A; according to an embodiment, the means 903 derives the index of the selected first precoding matrix from the index of the most matched basic beam subset and/or the least matched basic beam subset selected by the beam subset selecting means 1202.

According to another embodiment of the present invention, the apparatus further comprises a second precoding matrix selecting device 1204 configured to perform the functions described in conjunction with step 904 in fig. 9A, and select a second precoding matrix according to the short-time channel state information; the second precoding matrix is used to select one basic beam from the selected best matching basic beam subset to be suitable for the instantaneous channel state.

According to another embodiment of the present invention, the apparatus further comprises CQI calculating means 1205 configured to perform the operations described in conjunction with step 905 in fig. 9A. According to another embodiment of the invention, the apparatus further comprises Cluster Index (CI) selection means 1206 configured to perform the operations of step 906 of the method described in connection with fig. 9A, the CI being selected by one of the following operations:

-selecting a CI such that the selected CI equals the index of the selected first precoding matrix; or,

-arranging the plurality of basic beam subsets or a first precoding matrix associated with the plurality of basic beam subsets in the form of a two-dimensional (2D) uniform planar array, wherein each row of the uniform planar array relates to basic beams having the same downtilt angle and each column of the uniform planar array relates to basic beams having the same azimuth angle; dividing the uniform planar array into a plurality of clusters that are uniform in a horizontal domain (row direction) and non-uniform in a vertical domain (column direction), wherein at least one cluster includes a plurality of basic beam subsets or includes a plurality of first precoding matrices; selecting a first CI and a second CI, respectively, such that the cluster indicated by the first CI includes the selected best matching base beam subset or a first precoding matrix associated with the best matching base beam subset, and the cluster indicated by the second CI includes the selected least matching base beam subset or the first precoding matrix associated with the least matching base beam subset. The definition of CI is helpful to improve the pairing rate in MU-MIMO scheduling.

According to another embodiment of the present invention, the apparatus further comprises a feedback device 1207, configured to perform the operations described in conjunction with step 907 in fig. 9A, and feed back the index of the selected precoding matrix, the cluster index and the channel quality information to the base station. It should be noted that the feedback periods of the first precoding matrix index, the second precoding matrix index, the CQI, and the CI may be different, and thus the feedback of these information may occur at different times. In addition, the feedback means may further comprise means for pre-processing the information to be fed back, such as coding and modulation means, as will be appreciated by a person skilled in the art.

It should also be noted that the various devices shown in the schematic configuration of FIG. 12 are logical units, and thus, they may be implemented in hardware, software, firmware, or a combination thereof; in addition, the functions of several of the devices shown in the figures may be performed by a single device in some embodiments, or the functions of one of the devices shown in the figures may be performed by multiple devices in some embodiments. Moreover, in certain embodiments, certain devices shown in fig. 12 may be omitted, and/or other devices not shown in the figures may be included.

Fig. 13 is a schematic block diagram of an apparatus 1300 for MU-MIMO scheduling, as shown, the apparatus 1300 comprising:

user selection means 1301 for selecting a primary user to be scheduled and determining a paired user capable of being scheduled simultaneously with the primary user, at least partially according to information on channel quality and cluster index fed back by the user; and

a scheduling device 1302, configured to, when it is determined that the paired user exists, schedule the primary user and the paired user simultaneously on the same time-frequency resource at least partially according to information, which is fed back by the user and is related to channel quality, a first precoding matrix, and a cluster index.

According toIn the embodiment of the invention, the first precoding matrix fed back by the user is selected from a plurality of first precoding matrices, each first precoding matrix W in the plurality of first precoding matrices₁(k) With a basic beam subset X_kThe association may be, for example, $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ the basic beam subset is a plurality of basic beam subsets X obtained by uniformly dividing the full set of basic beams (e.g., according to any of the methods of fig. 7A-7D)_kOne of them.

According to one embodiment of the invention, the full set of basic beams is arranged in the form of a Uniform Planar Array (UPA) in which the beams of the same row have the same downtilt and the beams of the same column have the same azimuth; and the full set of basic beams is generated by a uniform planar antenna array.

An embodiment of the invention wherein at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predefined and known to the sender and the receiver or the receiver determines the at least one at least in part by receiving an indication from the sender.

User selection means 1301 is configured to perform the operations described in conjunction with step 1001 in fig. 10, in accordance with an embodiment of the present invention. For example, in one embodiment, the apparatus 1301 first determines a primary user through CQI fed back by the user; determining the paired user by comparing CIs of other user feedback to the CIs of the primary user feedback. In another embodiment, the user selection means 1301 is further configured to determine a plurality of paired users according to the CI fed back by the user, and determine a final paired user scheduled simultaneously with the primary user from the plurality of paired users by comparing channel quality indicators fed back by the plurality of paired users.

According to an embodiment of the present invention, the cluster index fed back by the user is selected by the user according to the selection method described in conjunction with step 906 in fig. 9A, for example, by: dividing the uniform planar array into a plurality of clusters that are uniform in a horizontal domain and non-uniform in a vertical domain, wherein at least one cluster comprises more than one basic beam subset or more than one first precoding matrix; the first CI and the second CI are respectively selected such that the cluster indicated by the first CI includes the selected best matching basic beam subset or the first precoding matrix associated with the best matching basic beam subset, and the cluster indicated by the second CI includes the selected least matching basic beam subset or the first precoding matrix associated with the least matching basic beam subset.

According to one embodiment, the CI fed back by the user may be equal to the index of the first precoding matrix, e.g. having the second CI equal to the index of the first precoding matrix with which the selected least matching basic beam subset is associated. According to another embodiment of the present invention, the cluster indicated by the CI fed back by the user may include a plurality of first precoding matrices, and the mapping relationship of the cluster to the first precoding matrices may be as shown in fig. 9B, where each gray box represents a PMI 1. Considering that in order to reduce the size of the short-term codebook, the long-term precoding codebook often contains fewer basic beams, which results in more detailed division of the full set of the basic beams and more basic beam subsets, if the index of the long-term precoding matrix PMI1 is still used as a user scheduling and pairing basis, the pairing success rate will not be too low only under the condition of more users. Thus, in this embodiment, proximally located PMIs 1 are grouped into one CI, so that the number of CIs is reduced to increase the pairing rate in a manner that reduces the set of partners. Arranging the PMI1 into a UPA topology, considering the limitation of user distribution in the vertical domain and the nonuniformity of the vertical beam pair space division, a uniform merging criterion can be followed in the horizontal direction, and the sparsity in the vertical direction according to the downward inclination angle of the beam is, for example, 1: 3 the non-uniform merging criterion performs the partitioning of the clusters. The CI calculated according to the method of the embodiment can maximize the success rate of user pairing while ensuring that the interference between the paired users is as small as possible.

The scheduling means 1302 is configured to perform the operations described in connection with step 1002 in fig. 10 according to an embodiment of the present invention. For example, when a paired user is found, the master user and the paired user are scheduled on the same time-frequency resource at the same time, and MU-MIMO transmission is performed; and under the condition that no paired user is found, randomly selecting one user from the optimal interference partner set, and scheduling the user and the master user at the same time, or scheduling the master user only to avoid excessive interference.

Although some embodiments of the present invention are described in the context of a downlink for wireless communications, it should be understood that embodiments of the present invention may be applied to an uplink as well. The precoding construction method in the embodiment of the invention can be used at a transmitting side and a receiving side of MIMO transmission, the precoding selection method can be used at the receiving side of the MIMO transmission, and the scheduling method can be used at the transmitting side of the MIMO transmission. Furthermore, both the sender and the receiver as MIMO transmission may be, for example, a base station of a cellular network, a pico base station, a femto base station, a relay node, a UE, other possible nodes, and any node in other wireless networks.

It should also be noted that the various devices shown in the schematic configuration of FIG. 13 are logical units, and thus, they may be implemented in hardware, software, firmware, or a combination thereof; in addition, the functions of several of the devices shown in the figures may be performed by a single device in some embodiments, or the functions of one of the devices shown in the figures may be performed by multiple devices in some embodiments. Furthermore, in certain embodiments, certain devices shown in fig. 13 may be omitted, and/or other devices not shown in the figures may be included, such as devices for control, code modulation, and/or transmission.

In fig. 14A, the throughput obtained in a 120 degree sector for3D MU-MIMO scheduling using an embodiment method of the present invention is shown. The results of different partitioning methods (the methods in fig. 7A-D) using the basic beam subset are shown. The main simulation parameters used are as follows:

height of the base station: 25m

Cell radius: 500 m;

base station transmission power: 46dBm

A base station antenna: a uniform planar array UPA of 8X 8; the horizontal spacing is half wavelength, and the vertical spacing is 2 times wavelength;

codebook: 3D: KPC, wherein the lid N_v=16，N_h=32；2D：DFT

Beam subset partitioning: uniform and non-overlapping: 4x 16; horizontal overlapping: 4x 32; vertical overlap 8x 16; horizontal and vertical overlap: 16x 32;

long-term feedback period: 100 TTI;

short-term feedback period: 1 TTI;

height of the user: 3(n-1) +1.5m, wherein n is taken from [1, 8 ];

the moving speed of the user is as follows: 3 km/h.

In the figure, the throughput in the 2D MU-MIMO case is also shown as a comparison, and results are given for the number of system users of 10 and 50, respectively. From the results in the figure, no matter which basic beam subset partitioning method in fig. 7A-7D is adopted, the system throughput is significantly improved compared to the 2D case. Also, in case of fewer users (10), the different partitioning methods of the basic beam subsets do not differ much, whereas in case of more users (50), partitioning into partially overlapping basic beam subsets can lead to higher throughput.

The result of the comparison of the throughput in the entire cell is shown in fig. 14B, and similar to fig. 14A, the result shows that the throughput can be significantly improved according to the method of the embodiment of the present invention from the viewpoint of the entire cell.

Aspects of embodiments of the present invention have been described above with reference to the accompanying drawings. It should be noted that the present invention also covers any conceivable combination of method steps and operations described above, as well as any conceivable combination of nodes, devices, modules or units described above, as long as the above-described method and structural arrangement concepts are applicable.

In general, the respective functional blocks or units of the method and network node described above may be implemented by any known means, respectively in hardware, software, firmware or a combination thereof, if it is only adapted to perform the described functions of the respective part. The mentioned method steps may be implemented in individual functional blocks or by individual devices or one or more of the method steps may be implemented in a single functional block or by a single device.

In general, any method steps are suitable to be implemented as software or by hardware without changing the idea of the invention. Such software may be independent of software code and may be specified using any known or future developed programming language, such as, for example, Java, C + +, C, and assembler, as long as the functionality defined by the method steps is preserved. Such hardware may be independent of hardware type and may be implemented using any known or future developed hardware technology or any mixture of such technologies, such as MOS (metal oxide semiconductor), CMOS (complementary MOS), BiMOS (bipolar MOS), BiCMOS (bipolar CMOS), ECL (emitter coupled logic), TTL (transistor-transistor logic), ASIC (application specific IC (integrated circuit)) components, FPGA (field programmable gate array) components, CPLD (complex programmable logic device) components, DSP (digital signal processor) components, or the like. Also, this does not exclude the possibility that the functionality of the device/apparatus or module is not implemented in hardware but in software in a (software) module, such as a computer program or a computer program product comprising executable software code portions for execution/running on one or more processors or processing systems. Software in the sense of this description comprises both software code, such as a computer program product comprising code means or portions or a computer program or for performing the respective functions, and software (or a computer program product) embodied on a tangible medium, such as a computer readable (storage) medium (having stored thereon a respective data structure or code means/portions) or possibly embodied in a signal or in a chip during processing thereof.

Even though the invention and/or exemplary embodiments have been described above with reference to examples according to the accompanying drawings, it is to be understood that they are not limited thereto. Indeed, it will be apparent to those skilled in the art that the present invention may be modified in numerous ways without departing from the scope of the inventive concept as disclosed herein.

Claims (40)

1. A method for constructing a precoding matrix for a mimo transmission system, comprising:

determining a plurality of first precoding matrixes to form a first precoding codebook;

wherein each of said first precoding matrices is associated with one of a plurality of basic beam subsets obtained from a full set of basic beams by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domain.

2. The method of claim 1, wherein the full set of basic beams is arranged in the form of a uniform planar array, wherein beams of a same row have a same downtilt angle and beams of a same column have a same azimuth angle; and

the full set of basic beams is generated by a uniform planar antenna array.

3. The method according to claim 1, wherein at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predetermined or determined at least in part by receiving an indication from a network node.

4. The method of claim 1, wherein the determining a plurality of first precoding matrices comprises:

determining a full set of said base beams, wherein each of said base beams is determined by a downtilt angle and an azimuth angle;

dividing the full set of basic beams into a plurality of basic beam subsets; and

constructing an associated first precoding matrix using each of the plurality of subsets of base beams.

5. The method of claim 1, wherein the improvement is achieved by $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ Associating each of the first precoding matrices with one of a plurality of basic beam subsets,

wherein W₁(k) Denotes the k-th first precoding matrix, X_kRepresents the kth basic beam subset and k is a non-negative integer value.

6. The method of any of claims 1-5, further comprising:

a plurality of second precoding matrices are determined for constructing a dual codebook matrix.

7. A method for selecting precoding information for a multiple-input multiple-output transmission system, comprising:

selecting a basic beam which is most matched with the current channel and a basic beam which is not matched with the current channel from the full set of basic beams according to long-term statistics of the channel state information; wherein the full set of basic beams is divided into a plurality of basic beam subsets by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domains;

selecting, from the plurality of basic beam subsets, a best matched basic beam subset and a least matched basic beam subset, the best matched basic beam subset comprising the selected best matched basic beam, and the least matched basic beam subset comprising the selected least matched basic beam; and

selecting a corresponding first precoding matrix from a plurality of first precoding matrices according to the selected most matched basic beam subset and/or the most unmatched basic beam subset.

8. The method of claim 7, wherein the full set of basic beams is arranged in the form of a Uniform Planar Array (UPA) with beams of the same row having the same downtilt and beams of the same column having the same azimuth; and the full set of basic beams is generated by a uniform planar antenna array.

9. The method according to claim 7, wherein at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predetermined or determined at least in part by receiving an indication from a network node.

10. The method according to claim 7, wherein selecting a respective first precoding matrix from a plurality of first precoding matrices in dependence on the selected best matching subset of basic beams and/or the best non-matching subset of basic beams comprises:

and obtaining the index of the corresponding first precoding matrix in the plurality of precoding matrices according to the selected index of the most matched basic beam subset and/or the most unmatched basic beam subset.

11. The method according to claim 7, wherein selecting a respective first precoding matrix from a plurality of first precoding matrices in dependence on the selected best matching subset of basic beams and/or the best non-matching subset of basic beams comprises:

according to the most matched basic beam subset and/or the most unmatched basic beam subset X_kAccording to $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ A respective first precoding matrix W selected from a plurality of first precoding matrices₁(k)。

12. The method of claim 7, wherein selecting a best matched basic beam and a least matched basic beam from the full set of basic beams according to long-term statistics of channel state information comprises:

obtaining the eigenvector of the channel correlation matrix according to the long-term statistics of the channel state information,

selecting the best matched base beam and the least mismatched base beam based at least in part on the distance of the principal eigenvector in the eigenvector from each base beam in the full set of base beams, an

In case the divided plurality of basic beam subsets partially overlap, selecting the most matched basic beam subset and the least unmatched basic beam subset at least partly according to a second one of the eigenvectors.

13. The method of claim 7, further comprising:

a second precoding matrix is selected for selecting a beam from the given subset of basic beams based on the short-time channel state information.

14. The method according to any one of claims 7-13, further comprising:

the cluster index CI is selected by one of the following operations:

-selecting a CI equal to an index of the selected respective first precoding matrix; or

-arranging said plurality of basic beam subsets or said plurality of first pre-coding matrices associated with said plurality of basic beam subsets in the form of a two-dimensional uniform planar array, wherein the basic beams involved in each row of said uniform planar array have the same or similar downtilt angles and the basic beams involved in each column of said uniform planar array have the same or similar azimuth angles;

dividing the uniform planar array into a plurality of clusters that are uniform in a horizontal domain and non-uniform in a vertical domain, wherein at least one cluster comprises more than one basic beam subset or more than one first precoding matrix; and

selecting a first CI and a second CI, respectively, such that the first CI indicated cluster includes the selected best matching basic beam subset or a first precoding matrix associated with the best matching basic beam subset, and the second CI indicated cluster includes the selected least matching basic beam subset or a first precoding matrix associated with the least matching basic beam subset.

15. The method of claim 14, further comprising:

feeding back at least the index of the selected precoding matrix and the cluster index to a base station.

16. A scheduling method for a multiple-input multiple-output transmission system, comprising:

selecting a master user to be scheduled and determining a paired user to be scheduled simultaneously with the master user at least partially according to the channel quality fed back by the user and the cluster index CI; and

when the paired users are determined to exist, simultaneously scheduling the main user and the paired users on the same time-frequency resource at least partially according to the channel quality fed back by the users, a first precoding matrix and the CI;

wherein the first precoding matrix is selected from a plurality of first precoding matrices, each of which is associated with one basic beam subset, the plurality of basic beam subsets being obtained from a full set of basic beams by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domains.

17. The method of claim 16, wherein determining paired users that are scheduled concurrently with the primary user comprises:

determining the paired user by comparing the CI of the other user feedback with the CI of the primary user feedback, wherein the CI of the other user and the CI of the primary user feedback are selected by the other user and the primary user, respectively, by one of:

-selecting a CI equal to an index of the selected respective first precoding matrix; or

-arranging said plurality of basic beam subsets or said plurality of first pre-coding matrices associated with said plurality of basic beam subsets in the form of a two-dimensional uniform planar array, wherein the basic beams involved in each row of said uniform planar array have the same or similar downtilt angles and the basic beams involved in each column of said uniform planar array have the same or similar azimuth angles; dividing the uniform planar array into a plurality of clusters that are uniform in a horizontal domain and non-uniform in a vertical domain, wherein at least one cluster comprises more than one basic beam subset or more than one first precoding matrix; and

selecting a first CI and a second CI, respectively, such that the first CI indicated cluster includes the selected best matching basic beam subset or a first precoding matrix associated with the best matching basic beam subset, and the second CI indicated cluster includes the selected least matching basic beam subset or a first precoding matrix associated with the least matching basic beam subset.

18. The method of claim 16, wherein the full set of basic beams is arranged in the form of a uniform planar array, wherein beams of a same row have a same downtilt angle and beams of a same column have a same azimuth angle; and the number of the first and second groups,

the full set of basic beams is generated by a uniform planar antenna array.

19. The method according to claim 16, wherein at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predetermined or determined at least in part by receiving an indication from a network node.

20. The method of claim 16, wherein the treatment is by $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ Associating each of the first precoding matrices with one of a plurality of basic beam subsets,

wherein W₁(k) Denotes the k-th first precoding matrix, X_kRepresents the kth basic beam subset and k is a non-negative integer value.

21. A constructing apparatus of a precoding matrix for a multiple-input multiple-output transmission system, comprising:

a first precoding matrix determining device for determining a plurality of first precoding matrices to form a first precoding codebook;

wherein each of said first precoding matrices is associated with one of a plurality of basic beam subsets obtained from a full set of basic beams by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domain.

22. The apparatus of claim 21, wherein the full set of basic beams is arranged in the form of a uniform planar array, wherein beams of a same row have a same downtilt angle and beams of a same column have a same azimuth angle; and

the full set of basic beams is generated by a uniform planar antenna array.

23. The apparatus according to claim 21, wherein at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predetermined or determined at least in part by receiving an indication from a network node, and

the first precoding matrix determination means further includes:

a storage means for storing a predefined definition of at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices, and/or

Means for determining at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices at least in part by receiving an indication from a network node.

24. The apparatus of claim 21, wherein the first precoding matrix determining means further comprises:

beam defining means for determining a full set of said base beams, wherein each of said base beams is determined by a downtilt angle and an azimuth angle;

subset dividing means for dividing said full set of basic beams into a plurality of basic beam subsets, an

First precoding matrix constructing means for constructing an associated first precoding matrix using each of the plurality of basic beam subsets.

25. The apparatus of claim 21, wherein the treatment is by $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ Associating each of the first precoding matrices with one of a plurality of basic beam subsets,

wherein W₁(k) Denotes the k-th first precoding matrix, X_kRepresents the kth basic beam subset and k is a non-negative integer value.

26. The apparatus of any of claims 21-25, further comprising:

second precoding matrix determining means for determining a plurality of second precoding matrices for constituting a dual codebook matrix.

27. A selection apparatus of precoding information for a multiple-input multiple-output transmission system includes,

beam selection means for selecting a best matched basic beam and a least matched basic beam from the full set of basic beams according to long-term statistics of channel state information; wherein the full set of basic beams is divided into a plurality of basic beam subsets by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domains;

beam subset selection means for selecting, from said plurality of basic beam subsets, a best matched basic beam subset and a least matched basic beam subset, said best matched basic beam subset comprising said selected best matched basic beam and said least matched basic beam subset comprising said selected least matched basic beam; and

and a first precoding matrix selecting device, configured to select a corresponding first precoding matrix from the plurality of first precoding matrices according to the selected most matched basic beam subset and/or the most unmatched basic beam subset.

28. The apparatus of claim 27, wherein the full set of basic beams is arranged in the form of a Uniform Planar Array (UPA) in which beams of a same row have a same downtilt and beams of a same column have a same azimuth; and is

The full set of basic beams is generated by a uniform planar antenna array.

29. The apparatus according to claim 27, wherein at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predetermined or determined at least in part by receiving an indication from a network node.

30. The apparatus according to claim 27, wherein the first precoding matrix selecting means is configured to derive an index of the corresponding first precoding matrix among the plurality of precoding matrices according to the selected index of the most matched basic beam subset and/or the most unmatched basic beam subset.

31. The apparatus according to claim 27, wherein the first precoding matrix selection means is adapted to select the first precoding matrix based on the best matching basic beam subset and/or the least matching basic beam subset X_kAccording to $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ A respective first precoding matrix W selected from a plurality of first precoding matrices₁(k)。

32. The apparatus of claim 27, wherein the beam selection means is for:

obtaining the eigenvector of the channel correlation matrix according to the long-term statistics of the channel state information,

selecting the best matched base beam and the least mismatched base beam based at least in part on the distance of the principal eigenvector in the eigenvector from each base beam in the full set of base beams, an

In case the divided plurality of basic beam subsets partially overlap, selecting the most matched basic beam subset and the least unmatched basic beam subset at least partly according to a second one of the eigenvectors.

33. The apparatus of claim 27, further comprising:

second precoding matrix selection means for selecting a second precoding matrix for selecting a beam from a given subset of basic beams, based on the short time channel state information.

34. The apparatus of any of claims 27-33, further comprising:

cluster Index (CI) selection means for selecting a cluster index CI by one of:

-selecting a CI equal to an index of the selected respective first precoding matrix; or

-arranging said plurality of basic beam subsets or said plurality of first pre-coding matrices associated with said plurality of basic beam subsets in the form of a two-dimensional uniform planar array, wherein the basic beams involved in each row of said uniform planar array have the same or similar downtilt angles and the basic beams involved in each column of said uniform planar array have the same or similar azimuth angles;

dividing the uniform planar array into a plurality of clusters that are uniform in a horizontal domain and non-uniform in a vertical domain, wherein at least one cluster comprises more than one basic beam subset or more than one first precoding matrix; and

selecting a first CI and a second CI, respectively, such that the first CI indicated cluster includes the selected best matching basic beam subset or a first precoding matrix associated with the best matching basic beam subset, and the second CI indicated cluster includes the selected least matching basic beam subset or a first precoding matrix associated with the least matching basic beam subset.

35. The apparatus of claim 34, further comprising:

a feedback device, configured to feed back at least the selected precoding matrix index and the cluster index to a base station.

36. A scheduling apparatus for a multiple-input multiple-output transmission system, comprising,

the user selection device is used for selecting a master user to be scheduled and determining a paired user which is scheduled simultaneously with the master user at least partially according to the channel quality fed back by the user and the cluster index CI fed back by the user; and;

a scheduling device, configured to, when it is determined that the paired user exists, schedule the primary user and the paired user simultaneously on the same time-frequency resource at least partially according to a channel quality fed back by the user, a first precoding matrix, and the CI;

wherein the first precoding matrix is selected from a plurality of first precoding matrices, each of which is associated with one basic beam subset, the plurality of basic beam subsets being obtained from a full set of basic beams by one of:

-dividing the full set of basic beams into uniform non-overlapping subsets;

-dividing the full set of basic beams into uniform subsets that partially overlap in the horizontal domain;

-dividing the full set of basic beams into uniform subsets that partially overlap in the vertical domain; and

-dividing the full set of basic beams into uniform subsets that partially overlap in both the vertical and horizontal domains.

37. The apparatus of claim 36, wherein the user selection means determines the paired user by comparing CIs of other user feedback with CIs of the primary user feedback, wherein CIs of the other user and the primary user feedback are selected by the other user and the primary user, respectively, by one of:

-selecting a CI equal to an index of the selected respective first precoding matrix; or

-arranging said plurality of basic beam subsets or said plurality of first pre-coding matrices associated with said plurality of basic beam subsets in the form of a two-dimensional uniform planar array, wherein the basic beams involved in each row of said uniform planar array have the same or similar downtilt angles and the basic beams involved in each column of said uniform planar array have the same or similar azimuth angles;

dividing the uniform planar array into a plurality of clusters that are uniform in a horizontal domain and non-uniform in a vertical domain, wherein at least one cluster comprises more than one basic beam subset or more than one first precoding matrix; and

selecting a first CI and a second CI, respectively, such that the first CI indicated cluster includes the selected best matching basic beam subset or a first precoding matrix associated with the best matching basic beam subset, and the second CI indicated cluster includes the selected least matching basic beam subset or a first precoding matrix associated with the least matching basic beam subset.

38. The apparatus of claim 36, wherein the full set of basic beams is arranged in the form of a uniform planar array, wherein beams of a same row have a same downtilt angle and beams of a same column have a same azimuth angle; and the number of the first and second groups,

the full set of basic beams is generated by a uniform planar antenna array.

39. The apparatus according to claim 36, wherein at least one of the full set of basic beams, the plurality of basic beam subsets and the plurality of first precoding matrices is predetermined or determined at least in part by receiving an indication from a network node.

40. The method of claim 36, wherein the plurality of first precoding matrices pass $W_1\left(k\right)=\left[\begin{array}{cc}{X}_k& 0\\ 0& X_k\end{array}\right]$ Associating each of the first precoding matrices with one of a plurality of basic beam subsets,

wherein W₁(k) Denotes the k-th first precoding matrix, X_kRepresents the kth basic beam subset and k is a non-negative integer value.