CN101599789B

CN101599789B - Measurement method of channel matrix rank as well as device and terminal thereof

Info

Publication number: CN101599789B
Application number: CN200910088476A
Authority: CN
Inventors: 许百成; 牟秀红; 冯心睿
Original assignee: Beijing T3G Technology Co Ltd
Current assignee: Beijing T3G Technology Co Ltd
Priority date: 2009-07-02
Filing date: 2009-07-02
Publication date: 2012-10-24
Anticipated expiration: 2029-07-02
Also published as: CN101599789A

Abstract

The invention provides a measurement method of channel matrix rank as well as a device and a terminal thereof; wherein the method comprises: calculating according to the related information received from a base station by a terminal antenna of a wireless communication system to obtain a channel matrix H to generate a hermitian matrix A; using the hermitian matrix A to construct an equitation set with an eigenvalue of nonnegative real number; calculating the root of the equitation set so as to obtain an eigenvalue of the channel matrix H. The invention can be used for the terminal of the wireless communication system having four transmitting and four receiving antennae and can measure the rank of a four-order matrix, therefore, the base station can determine the number of data flow to be parallelly transmitted according to the current situation of the channel matrix rank so as to increase the number of plural data flow transmitted simultaneously at the same frequency. The method of the invention is also applicable to the terminal of wireless communication system having two transmitting and two receiving antennae, four transmitting and two receiving antennae and two transmitting and four receiving antennae.

Description

Method, device and terminal for measuring channel matrix rank

Technical Field

The invention relates to the technical field of wireless communication, in particular to a method, a device and a terminal for measuring a channel matrix rank of a 3GPP LTE (Long term Evolution) system terminal.

Background

In order to increase the data transmission rate in the wireless communication system, a Multiple-Input Multiple-Output (MIMO) technology is introduced in a future wireless communication system, i.e., an LTE system, and the MIMO technology increases the transmission rate according to the uncorrelated characteristics of transmission channels among Multiple antennas. This means that the MIMO technology can simultaneously transmit a plurality of parallel data streams at the same frequency, and the parallel data streams are distinguished by the irrelevance of the channel matrix, so the number of data streams that can be simultaneously transmitted in parallel in the MIMO system depends on the number of linearly independent vectors in the channel matrix, that is, the rank of the channel matrix.

In the MIMO technique, the relationship between the transmitted signal, the received signal and the channel matrix can be simply expressed by the following formula:

Y_R×1＝H_R×TX_T×1

in the formula, R represents the number of receiving antennas, and T represents the number of transmitting antennas. H is the channel matrix, X is the transmitted signal vector and Y is the received signal vector.

In a wireless channel, the channel condition changes in real time, so that in the case of transmission using MIMO technology, a terminal must feed back information of a channel matrix rank to a base station at all times, so that the base station can determine the number of data streams to be transmitted in parallel according to the condition of the current channel matrix rank. In the system, this information is called RI (rank index).

The present wireless communication system terminal generally adopts a single transmitting antenna and a single receiving antenna to transmit data, and this transceiving mode can only transmit one data stream at most. However, how to accurately transmit more data streams becomes an urgent problem for a terminal using multiple antennas.

Disclosure of Invention

The invention aims to provide a method, a device and a terminal for measuring channel matrix rank, which are used for increasing the number of parallel data streams transmitted at the same frequency at the same time. The method for measuring the channel matrix rank comprises the following steps:

generating a Hermite matrix A according to a channel matrix H obtained by calculating relevant information from a base station and received by a terminal antenna of a wireless communication system;

constructing an equation set with eigenvalues of nonnegative real numbers by using the Hermite matrix A;

calculating the root of the equation set so as to obtain the eigenvalue of the channel matrix H;

and judging the rank of the channel matrix H according to the eigenvalue so that the base station determines the number of data streams sent to the terminal according to the rank.

The antenna configuration of the wireless communication system specifically includes: four-transmission and four-receiving or two-transmission and two-receiving or two-transmission and four-receiving or four-transmission and two-receiving.

When the antenna of the wireless communication system is four-transmission and four-reception, the step of generating the hermitian matrix a specifically includes:

A_4×4＝H_4×4 ^H×H_4×4，

wherein H_4×4 ^HIs H_4×4The conjugate transpose matrix of (a);

when the antenna of the wireless communication system is configured as two-transmission two-reception or two-transmission four-reception, the step of generating the hermitian matrix a specifically includes: from said channel matrix H, a hermitian matrix A is calculated which is of smaller dimension, i.e. if R < T, A_2×2＝H_R×T×H_R×T ^H；

Otherwise, A_2×2＝H_R×T ^H×H_R×T；

Wherein,

r is the number of receiving antennas;

and T is the number of the transmitting antennas.

When the antenna of the wireless communication system is configured to four-transmit and four-receive, the step of constructing the equation set with the eigenvalues of the non-negative real numbers specifically includes: a system of one-element quartic equations is constructed using | λ I-a | ═ 0, where I is a 4 × 4 unit matrix, assuming a_4×4Comprises the following steps:

then | λ I-a | ═ 0 is specifically:

namely, the unary quartic equation set is: lambda [ alpha ]⁴+a₃λ³+a₂λ²+a₁λ+a₀0, wherein each time term a of the equation₀～a₃Calculating according to a conventional determinant;

when the antenna of the wireless communication system is configured as two-transmission two-reception or two-transmission four-reception, the step of constructing the equation set with the non-negative real number eigenvalue specifically includes: a quadratic equation of unity is constructed using λ I-a | ═ 0, where I is a 2 × 2 unit matrix, assuming a_2×2Comprises the following steps:

then | λ I-a | ═ 0 is specifically:

i.e. the quadratic equation of one unit is lambda²+bλ+c＝0，

Wherein, b ═ r₁₁+r₂₂)，c＝(r₁₁r₂₂-r₁₂r₂₁)。

When the antenna of the wireless communication system is configured to transmit four, the step of obtaining the eigenvalue of the channel matrix H specifically includes:

using a Fisher-Tropsch solution to a unitary quadratic equation lambda⁴+a₃λ³+a₂λ²+a₁λ+a₀Solving for 0 to obtain four eigenvalues of the matrix A as lambda₀，λ₁，λ₂，λ₃；

When the antenna of the wireless communication system is configured as two-transmission two-reception or two-transmission four-reception, the step of obtaining the eigenvalue of the channel matrix H specifically includes:

solving two characteristic values lambda 'of the matrix A according to the Widah theorem'₀，λ′₁，

Wherein,

when the antenna of the wireless communication system is configured to transmit four, the step of determining the rank of the channel matrix H specifically includes:

four eigenvalues lambda₀，λ₁，λ₂，λ₃The lambda is obtained by sorting from big to small_i0，λ_i1，λ_i2，λ_i3The rank is determined as follows:

if it is not

The rank of the channel matrix H is 1;

otherwise, if

The rank of the channel matrix H is 2;

otherwise, if

The rank of the channel matrix H is 3;

otherwise, the rank of the channel matrix H is 4;

when the antenna of the wireless communication system is configured as two-transmission two-reception or two-transmission four-reception, the step of determining the rank of the channel matrix H specifically includes:

if it is not

The rank of the channel matrix H is 1;

otherwise, the rank of the channel matrix H is 2;

wherein Δ is a condition number threshold of the hermitian matrix a.

And calculating the corresponding Hermite matrix by using the channel matrix, further constructing a unitary four/quadratic equation by using the Hermite matrix, solving the root of the equation, namely the characteristic value of the channel matrix, by using a Fisher algorithm/Wedah theorem, and finally judging the rank of the matrix by using the characteristic value.

The invention also provides a device for measuring the channel matrix rank, which comprises:

the Hermite matrix generation module is used for generating a Hermite matrix A according to a channel matrix H obtained by calculating relevant information from a base station and received by a terminal antenna of a wireless communication system;

the equation generation module is used for generating an equation set with eigenvalues of non-negative real numbers by utilizing the Hermite matrix A;

the eigenvalue calculation module is used for calculating the eigenvalue of the channel matrix H according to the equation set;

and the matrix rank judging module is used for judging the rank of the channel matrix H according to the eigenvalue so that the base station determines the number of data streams sent to the terminal according to the rank.

When the antenna of the wireless communication system is configured to four-transmission, the method for calculating the characteristic value of the channel matrix H is a Fisher-Rayleigh solution method;

when the antenna of the wireless communication system is configured to be two-transmission and two-reception or two-transmission and four-reception, the method for calculating the characteristic value of the channel matrix H is the Wedad theorem.

The present invention also provides a terminal, including: a measurement apparatus of a channel matrix rank, the measurement apparatus comprising:

Compared with the prior art, the invention has the following beneficial effects:

the invention constructs a matrix with characteristic values which are all non-negative real numbers by utilizing a channel matrix, and obtains the rank of the matrix by solving the matrix. The invention can be used for a terminal of a wireless communication system with four transmitting and four receiving antennas, and can measure the rank of the four-order matrix, so that a base station can determine the number of parallel transmitting data streams according to the condition of the current channel matrix rank, thereby improving the number of simultaneously transmitting a plurality of parallel data streams at the same frequency. The method of the invention is also suitable for the terminal of the wireless communication system with two-sending and two-receiving, four-sending and two-receiving and two-sending and four-receiving antennas.

Drawings

FIG. 1 is a flow chart of a method for measuring channel matrix rank according to the present invention;

fig. 2 is a schematic diagram of a channel matrix rank measurement apparatus according to the present invention.

Detailed Description

The rank of the matrix is equal to the number of eigenvalues of the matrix other than 0, and therefore the rank of the matrix can be calculated by only calculating the eigenvalues of the matrix. And the calculation of the eigenvalues can be converted into a problem that solves the complex equation unary to the degree N equation | λ I-H | ═ 0. It is noted that the root equations for the unitary quartic and subquartic general real (not complex) algebraic equations exist, for example, the Fischer-Tropsch method can be used for the unitary quartic equation, the Kardan formula can be used for the unitary cubic equation, and the well-known Weddar theorem can be used for the unitary quadratic equation. There is no fixed root equation for algebraic equations of more than four times. The three equation solutions mentioned above are for real numbers, while the channel matrix is a complex matrix, and the root equation cannot be directly applied.

The invention is based on the matrix H and A ═ H^HThe principle that H has the same rank and the eigenvalues of a are all non-negative real numbers can convert the calculation of the H rank of the channel matrix into the solution of a unary N-degree equation | λ I-a | ═ 0 root of real numbers.

The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.

Referring to fig. 1, fig. 1 is a flowchart of a channel matrix rank measurement method of the present invention, including the steps of:

step 1, generating a Hermite matrix A according to a channel matrix H obtained by calculating relevant information from a base station and received by a terminal antenna of a wireless communication system;

step 2, constructing an equation set with a characteristic value of a non-negative real number by using the Hermite matrix;

step 3, calculating the root of the equation set so as to obtain the characteristic value of the channel matrix H;

and 4, judging the rank of the channel matrix H according to the eigenvalue so that the base station determines the number of data streams sent to the terminal according to the rank.

After obtaining the rank of the channel matrix H, the base station may determine the number of data streams for parallel transmission according to the rank of the channel matrix H.

The following describes in detail the implementation process of the above steps for four cases of two-transmission two-reception, four-transmission two-reception, two-transmission four-reception, and four-transmission four-reception configured for MIMO transmit-receive antennas in the LTE system. Here, m-sending and n-receiving means: m is the number of transmitting antennas at the base station side, and n is the number of receiving antennas at the terminal side.

Step 1:

for the four-transmit-four-receive case, the following operation is performed:

hypothetical channel matrix

Then A is_4×4＝H_4×4 ^H×H_4×4' of course, matrix A_4×4H can also be utilized_4×4×H_4×4 ^HTo obtain i.e. A_4×4＝H_4×4×H_4×4 ^H。

This (·)^HRepresenting the conjugate transpose of the matrix (·).

Let matrix A_4×4The material is composed of the following elements:

for three situations, namely two-sending and two-receiving, four-sending and two-receiving and two-sending and four-receiving, the operation is carried out according to the following modes:

hypothetical channel matrix

If R < T, then A_2×2＝H_R×T×H_R×T ^H

Otherwise, A_2×2＝H_R×T ^H×H_R×T

This (·)^HDenotes the conjugate transpose of the matrix (·), min (R, T) ═ 2.

Let matrix A_2×2The material is composed of the following elements:

step 2:

for the four-transmit-four-receive case, the following operation is performed:

a one-element system of quartiles is constructed from 0. Wherein, I is a 4 × 4 unit array, specifically:

i.e. lambda⁴+a₃λ³+a₂λ²+a₁λ+a₀0, wherein each time term a of the equation₀～a₃Can be obtained by conventional determinant calculation.

a system of one-dimensional quadratic equations is constructed from λ I-a | ═ 0. Wherein, I is a unit array of 2 multiplied by 2, which is specifically as follows:

i.e. lambda²+ b λ + c ═ 0, where b ═ r₁₁+r₂₂)，c＝(r₁₁r₂₂-r₁₂r₂₁)

And step 3:

for the four-transmit-four-receive case, the following operation is performed:

calculating equation lambda by using Fisher's solution⁴+a₃λ³+a₂λ²+a₁λ+a₀Four roots of 0. The method specifically comprises the following two steps:

in the first step, a balancing parameter α is calculated by solving a one-dimensional cubic equation.

The root of the following cubic equation is calculated.

y^{3} - a_{2} y^{2} + (a_{3} a_{1} - {4 a}_{0}) y + a_{0} ({4 a}_{2} - a_{3}^{2}) - a_{1}^{2} = 0

For simplicity of expression, the above formula is rewritten as:

y³+b₂y²+b₁y+b₀＝0

wherein, b₂＝-a₂，b₁＝(a₃a₁-4a₀)，

b_{0} = a_{0} ({4 a}_{2} - a_{3}^{2}) - a_{1}^{2} .

Calculating variables p and q:

p = b_{1} - \frac{b_{2}^{2}}{3}

q = b_{0} + \frac{{2 b}_{2}^{3}}{27} - \frac{b_{2} b_{1}}{3}

calculating variables u, v:

u = \sqrt[3]{\frac{- q + \sqrt{q^{3} + \frac{{4 p}^{3}}{27}}}{2}}

v = \sqrt[3]{\frac{- q - \sqrt{q^{3} + \frac{{4 p}^{3}}{27}}}{2}}

u and v are each the cubic root of a real number.

y_{0} = u + v - \frac{b_{2}}{3},

y_{1} = ue + {ve}^{2} - \frac{b_{2}}{3},

y_{2} = ve + {ue}^{2} - \frac{b_{2}}{3}

Wherein,

and secondly, constructing two linear equations.

Any real root of the above-mentioned unitary cubic equation is substituted into the following two equations:

four roots of two unary quadratic equations which can be solved by using the Weddar theorem are respectively lambda₀，λ₁，λ₂，λ₃。

For the three conditions of two-transmitter and two-receiver, four-transmitter and two-receiver and two-transmitter and four-receiver, the characteristic value lambda 'can be directly solved according to the Werdan theorem'₀、λ′₁：

And 4, step 4: when the rank is judged in practical application, since the signal is inevitably affected by noise, the situation that some (several) eigenvalues are just 0 does not occur, and the judgment of the rank by using the condition number is more effective.

For the four-transmit-four-receive case, the following operation is performed:

the four eigenvalues are sequenced from big to small to obtain lambda_i0，λ_i1，λ_i2，λ_i3The rank is determined as follows:

if it is not

The rank of the matrix a is 1, that is, the rank of the channel matrix H is 1;

otherwise, if

The rank of the matrix a is 2, i.e. the rank of the channel matrix H is 2;

otherwise, if

The rank of the matrix a is 3, i.e. the rank of the channel matrix H is 3;

otherwise, the rank of matrix a is 4, i.e. the rank of channel matrix H is 4.

if it is not

The rank of the matrix a is 1, that is, the rank of the channel matrix H is 1;

otherwise, the rank of the matrix a is 2, i.e. the rank of the channel matrix H is 2.

Where Δ is the condition number threshold of matrix a. A higher condition number indicates a more unstable matrix, i.e. closer to the singular matrix, the threshold should be a larger value.

The present invention also provides a device for measuring a channel matrix rank, referring to fig. 2, where fig. 2 is a schematic diagram of the device for measuring a channel matrix rank of the present invention, and the device includes:

the system comprises an equation generation module, a real-time computation module and a real-time computation module, wherein the equation generation module is used for constructing an equation set with eigenvalues of non-negative real numbers by using a Hermite matrix A;

and the matrix rank judgment module is used for judging the rank of the channel matrix H according to the eigenvalue so that the base station determines the number of data streams sent to the terminal according to the rank.

The specific implementation method of each module corresponds to the steps 1-4 above.

In summary, the invention generates a new matrix with non-negative real number characteristic roots by using a channel matrix, constructs a one-element quartic equation or a one-element quadratic equation according to the generated new matrix, and then determines the rank of the channel matrix according to the coefficients of the equation. Therefore, the base station can determine the number of the data streams transmitted by the terminal in parallel according to the rank determined by the unitary quartic equation or the unitary quadratic equation, and the number of the parallel data streams transmitted at the same time is increased. The invention is not only suitable for the terminal of the wireless communication system with four-transmitting and four-receiving antennas, but also suitable for the conditions of four-transmitting and two-receiving, two-transmitting and four-receiving, and two-transmitting and two-receiving, and has wider application range.

The foregoing is only a preferred embodiment of the present invention, and it should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be construed as the protection scope of the present invention.

Claims

1. A method for measuring channel matrix rank is characterized by comprising the following steps:

judging the rank of the channel matrix H according to the eigenvalue so that a base station determines the number of data streams sent to the terminal according to the rank;

the antenna configuration of the wireless communication system specifically includes: four-transmission four-receiving or two-transmission two-receiving or two-transmission four-receiving or four-transmission two-receiving;

A_4×4＝H_4×4 ^H×H_4×4，

wherein H_4×4 ^HIs H_4×4The conjugate transpose matrix of (a);

Otherwise, A_2×2＝H_R×T ^H×H_R×T；

Wherein,

r is the number of receiving antennas;

t is the number of the transmitting antennas;

then | λ I-a | ═ 0 is specifically:

then | λ I-a | ═ 0 is specifically:

i.e. the quadratic equation of one unit is lambda²+bλ+c＝0，

Wherein, b ═ r₁₁+r₂₂)，c＝(r₁₁r₂₂-r₁₂r₂₁)；

Wherein,

if it is not

The rank of the channel matrix H is 1;

otherwise, if

The rank of the channel matrix H is 2;

otherwise, ifThe rank of the channel matrix H is 3;

otherwise, the rank of the channel matrix H is 4;

if it is not

The rank of the channel matrix H is 1;

otherwise, the rank of the channel matrix H is 2;

wherein Δ is a condition number threshold of the hermitian matrix a;

2. An apparatus for measuring a channel matrix rank, comprising:

a matrix rank judgment module, configured to judge a rank of the channel matrix H according to the eigenvalue, so that the base station determines, according to the rank, the number of data streams sent to the terminal;

when the antenna of the wireless communication system is four-transmission and four-reception, the hermitian matrix generating module generates the hermitian matrix A by adopting the following method:

A_4×4＝H_4×4 ^H×H_4×4，

wherein H_4×4 ^HIs H_4×4The conjugate transpose matrix of (a);

when the antenna of the wireless communication system is configured to be two-transmission two-reception or two-transmission four-reception or four-transmission two-reception, the hermitian matrix generating module generates the hermitian matrix A by adopting the following method:

from said channel matrix H, a hermitian matrix A is calculated which is of smaller dimension, i.e. if R < T, A_2×2＝H_R×T×H_R×T ^H；

Otherwise, A_2×2＝H_R×T ^H×H_R×T；

Wherein,

r is the number of receiving antennas;

t is the number of the transmitting antennas;

when the antenna of the wireless communication system is configured to four-transmission and four-reception, the equation generating module generates an equation set with eigenvalues of nonnegative and real numbers by adopting the following method:

a system of one-element quartic equations is constructed using | λ I-a | ═ 0, where I is a 4 × 4 unit matrix, assuming a_4×4Comprises the following steps:

then | λ I-a | ═ 0 is specifically:

when the antenna of the wireless communication system is configured to be two-transmission and two-reception or two-transmission and four-reception, the equation generating module generates an equation set with eigenvalues of non-negative real numbers by adopting the following method:

a quadratic equation of unity is constructed using λ I-a | ═ 0, where I is a 2 × 2 unit matrix, assuming a_2×2Comprises the following steps:

then | λ I-a | ═ 0 is specifically:

i.e. the quadratic equation of one unit is lambda²+bλ+c＝0，

Wherein, b ═ r₁₁+r₂₂)，c＝(r₁₁r₂₂-r₁₂r₂₁)；

When the antenna of the wireless communication system is configured to four-transmission and four-reception, the eigenvalue calculation module calculates the eigenvalue of the channel matrix H by adopting the following method:

When the antenna of the wireless communication system is configured as two-transmission two-reception or two-transmission four-reception, the eigenvalue calculation module calculates the eigenvalue of the channel matrix H by adopting the following method:

Wherein,

when the antenna of the wireless communication system is configured to four times for four transmissions, the matrix rank judgment module adopts the following method to judge the rank of the channel matrix H:

if it is not

The rank of the channel matrix H is 1;

otherwise, if

Then the channel is describedThe rank of matrix H is 2;

otherwise, if

The rank of the channel matrix H is 3;

otherwise, the rank of the channel matrix H is 4;

when the antenna of the wireless communication system is configured to be two-transmission and two-reception or two-transmission and four-reception, the matrix rank judgment module adopts the following method to judge the rank of the channel matrix H:

if it is not

The rank of the channel matrix H is 1;

otherwise, the rank of the channel matrix H is 2;

wherein Δ is a condition number threshold of the hermitian matrix a;

3. The measurement apparatus according to claim 2, wherein when the antenna of the wireless communication system is configured to four transmission times, the method of calculating the eigenvalues of the channel matrix H is a fischer-tropsch method;