WO2006098379A1

WO2006098379A1 - Adaptive modulation method based on multi-user pre-coding

Info

Publication number: WO2006098379A1
Application number: PCT/JP2006/305154
Authority: WO
Inventors: Qiang Wu; Jifeng Li
Original assignee: Matsushita Electric Industrial Co., Ltd.
Priority date: 2005-03-16
Filing date: 2006-03-15
Publication date: 2006-09-21
Also published as: CN1835425A

Abstract

In this method, a base station acquires the channel matrix and noise power of each of user terminals; uses them and a predetermined pre-coding scheme to calculate an equivalent signal-to-noise ratio of each user terminal; decides, based on it, an adaptive modulation parameter table; decides, based on the equivalent signal-to-noise ratio of each user terminal, a power distribution by use of the water filling principal; calculates, from the noise power and power distribution value of each user terminal, a new signal-to-noise ratio of each user terminal; decides, based on it, a corresponding modulation scheme from the adaptive modulation parameter table; distributes, by use of a similar scheme to the signal-to-noise ratio, a power anew to the user terminals belonging to the same modulation scheme; acquires an ultimate distribution power of each user terminal; and transmits data on the basis of the ultimate distribution power and modulation scheme.

Description

Adaptive modulation method based on multi-user precoding

Technical field

The present invention relates to precoding and adaptive modulation in a multiuser MIMO communication system.

Background art

[0002] Multi-input multi-output (MIMO) technology is a major achievement in the field of wireless mobile communications. MIMO technology uses multiple antennas for both data transmission and reception. As a result of various studies, it has become possible to increase channel capacity by using MIMO technology, and it has become possible to improve channel reliability and reduce error rate. MIMO technology has a tremendous potential for increasing the capacity of wireless communication systems, and is a key technology used in next-generation mobile communications. In particular, multi-user MIMO communication is one of the technologies that are currently attracting attention.

FIG. 1 is a block diagram showing a configuration of a commonly used single user MIMO system. In this system, MIMO transmitters (transmitters that use the MIMO communication method) use M transmit antennas, and MIMO receivers (receiver devices that use the MIMO communication method) use N receive antennas for signals. Send and receive. In the MIMO transmission apparatus, transmission data is first processed by an SZP conversion unit 701 and divided into M data streams, and each data stream is transmitted from a corresponding transmission antenna 702. In the MIMO receiver, first, signals are received by N reception antennas 703, and then channel estimation section 704 performs channel estimation based on the received signals to estimate current channel characteristic matrix H. The detection unit 705 detects the received signal using this channel characteristic matrix H! /, And demodulates the information bits transmitted by the MIMO transmitter.

[0004] On the other hand, FIG. 2 is a diagram showing an outline of a multiuser MIMO system. In FIG. 2, the base station transmits signals Sl and S2 simultaneously to user terminal (communication terminal) # 1 and user terminal # 2. However, the desired signal S2 of user terminal # 2 interferes with user terminal # 1, and the desired signal S1 of user terminal # 1 also interferes with user terminal # 2. Yes. Also, user terminal # 1 cannot obtain the channel characteristics of user terminal # 2, and user terminal # 2 cannot obtain the channel characteristics of user terminal # 1 as well.

Disclosure of the invention

Problems to be solved by the invention

[0005] There are the following four precoding methods used in a multi-user MIMO communication system! /.

1. Reverse channel precoding method

2. Tomlinson-Harashima precoding method

3. Precoding method based on lattice reduction

4. AMC precoding method

These methods have the following characteristics.

[0006] It should be noted that the mathematical model of the multiuser MIMO system is expressed as the following equation (1).

[Number 1]

X = Hs + n… ^ 1 In equation (1), n represents white Gaussian noise with mean value zero and square error σ ^{2 at} the receiving antenna, X is the signal vector of the receiving antenna, s is after precoding The transmission signal of represents the channel matrix.

[0007] First, the reverse channel 'precoding method will be described. FIG. 3 is a diagram for explaining the reverse channel precoding method. In FIG. 3, the data d first undergoes processing of the channel inverse matrix ^H_1 , and after passing through the channel H, ^H_1H = I, and the interference is completely eliminated. However, at the cost of this, the transmission power must be increased to eliminate interference, and conversely, if the transmission power is not changed, noise will increase.

[0008] Next, the Tomlinson-Harashima precoding method will be described. FIG. 4 is a diagram for explaining the Toml inson-Harashima precoding method. The Tomlinson-Harashima precoding method decomposes the channel matrix H using LQ decomposition, which is a modification of QR decomposition, as shown in Equation (2) below. [Equation 2]

H = SF ^H (2) In equation (2), S is a lower triangular matrix and F is a vertical matrix (F ^H F = I). And C is defined by the following equation (3).

[Equation 3]

C = G HF = G S (3) In Eq. (3), G means Eq. (4) below. C is a lower triangular matrix with a diagonal element of 1.

Picture

G iA «.., 5… (4)

[0009] The idea of this method is that if the interference is known, it can be reduced in advance by the MIMO transmitter. This is the case for the “CI” process in Figure 4. The “CI” processing is mathematically equivalent to a lower triangular matrix having zero diagonal elements. For the first user! There is no interference !, but second

It is necessary to reduce the interference of the first user, and for the Kth user, it is necessary to reduce the interference of the first to K1 users. Also, the “MOD” process in FIG. 4 represents a MOD calculation, and thus the transmission power can be reduced. When the Kth user obtains his data, the received signal is restored by multiplying the received signal by the Kth diagonal element g of G.

K

Then, a “demodulation” process is performed and output.

[0010] Although this precoding method is superior to the method of obtaining an inverse matrix in terms of performance in order to reduce transmission power, a gain control value g is set for each user for each frame data.

K

Therefore, if the overhead of the transmission signal increases, there is a problem! is there.

Next, a precoding method based on lattice reduction will be described. The precoding method based on lattice reduction performs pseudo-inverse matrix decomposition on H as shown in Equation (5) below.

[Equation 5]

H _red = “T ... (5) Here, each column of H is an approximate orthogonal basis vector. Each element of T and T ^{_ 1}

red

Is an integer and the determinant is ± 1. Then, using the characteristics of T, precoding is performed by combining MOD operations according to the following equation (6).

[Equation 6] s = H _red T {T ~ ^l d modr) (6) In equation (6), d is the symbol before precoding, τ is the boundary value, and the real part of the coordinate point Both imaginary parts are located between 0.5 and 0.5.

[0012] Next, I will explain the AMC precoding method. The basic principle of the AMC precoding method is to change the modulation and code formats to adapt to the channel conditions within the limits of the system, and the channel conditions are estimated from the feedback information. In AMC systems, users typically use higher-order modulation schemes and code rates under desirable channel conditions, and sometimes lower-order modulation schemes and coding methods when desired and not under channel conditions. Is used.

[0013] An object of the present invention is to provide an adaptive modulation method based on multiuser precoding to improve downlink error tolerance in a multiuser trap system.

Means for solving the problem

[0014] Adaptive modulation based on multi-user precoding provided by one aspect of the present invention The method includes: obtaining a channel matrix and noise power of each communication terminal; and calculating an equivalent signal-to-noise ratio of each communication terminal using the channel matrix, the noise power, and a predetermined precoding method. Determining an adaptive modulation parameter table using the equivalent signal-to-noise ratio; determining a power distribution value of each communication terminal according to a water injection theorem based on the equivalent signal-to-noise ratio; and the noise Calculating a signal-to-noise ratio of each communication terminal using the power and the power distribution value, and determining the modulation method for each communication terminal based on the signal-to-noise ratio; According to the same method as the noise-to-noise ratio, power is allocated again to communication terminals that use the same modulation method, and the final distribution power of each communication terminal is allocated. And a step of transmitting data addressed to each communication terminal based on the final distributed power and the modulation method.

The invention's effect

[0015] According to the present invention, it is possible to improve error resilience in a multiuser MIMO system.

Brief Description of Drawings

[0016] [Figure 1] Diagram showing the configuration of a single-user MIMO system

[Figure 2] Diagram showing an overview of a multi-user MIMO system

[Figure 3] Diagram for explaining the reverse channel precoding method

[Fig.4] Diagram for explaining the Tomlinson-Harashima precoding method

FIG. 5 is a diagram showing a main configuration of a MIMO system according to an embodiment of the present invention.

FIG. 6 is a flowchart showing the procedure of an adaptive modulation method based on multi-user precoding according to the present embodiment.

[Fig.7] Diagram showing M-QAM modulation constellation

[Fig.8] Diagram showing performance comparison of each precoding method

[Fig.9] Performance comparison of precoding methods with and without adaptive modulation

BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. Here, the same reference numerals are assigned to a plurality of configurations having similar functions, and Each code is followed by a different branch number to distinguish each other.

FIG. 5 is a diagram showing a main configuration of the MIMO system according to one embodiment of the present invention.

[0019] The MIMO system according to the present embodiment normally includes a base station device 100 and a plurality (N in this case) of user terminal devices (communication terminal devices) 200-1 to 200-N which are mobile stations. Prepare.

[0020] Base station apparatus 100 includes a data source 101 such as a memory storing data of each user, adaptive modulation units 102-1 to 102-N, precoding units 103-1 to 103-N, and a transmission antenna. 104—1 to 104—N. There are N adaptive modulation sections 102—1 to 102—N, precoding sections 103—1 to 103—N, and transmission antennas 104—1 to 104—N corresponding to N user terminal devices. .

The user terminal device 200-1 includes a reception antenna 201-1, a channel Z noise estimation unit 202-1, and a demodulation unit 203-1. Similarly, the user terminal device 200-N also includes a reception antenna 201-N, a channel Z noise estimation unit 202-N, and a demodulation unit 203-N.

[0022] Since the configuration of each user terminal device is the same, the generic term user terminal device 200 with the branch number omitted is basically used below. Further, the generic name in which the branch number is omitted is also used for the configuration in the user terminal device 200.

[0023] Furthermore, in the configuration within base station apparatus 100, a generic term in which branch numbers are omitted is used for the same configuration.

First, the function of each unit in the user terminal device 200 will be described.

The reception antenna 201 receives a signal from the base station apparatus 100. The channel Z noise estimation unit 202 performs channel estimation using the pilot included in the received signal, performs line fluctuation compensation based on the obtained channel estimation value, and demodulates the signal after compensation. Output to. Demodulation section 203 also demodulates the received data from the signal output from channel Z noise estimation section 202. The obtained demodulated data is output to another configuration (not shown). The channel Z noise estimator 202 also estimates the noise power of the received signal along with the channel estimation value, and sends this information (channel matrix and noise power that also includes the channel estimation value power) via the feedback channel CH1 and the like. Feedback to base station apparatus 100. More than Data reception and feedback processing are repeated.

Next, functions of each unit in base station apparatus 100 will be described.

[0027] Base station apparatus 100 uses user terminal apparatuses 200-1 to 200-N based on feedback information respectively notified via feedback channels CH1 to CHN! /, Channel matrix H and noise. Electric power can be acquired.

[0028] Based on the fed back channel matrix H and noise power, adaptive modulation section 102 determines the format of adaptive modulation of user terminal apparatus 200 according to the method described later, and transmits transmission of user terminal apparatus 200. Determine the power. Further, when the modulation method and transmission power of the user terminal device 200 are determined, the adaptive modulation unit 102 acquires data to be transmitted from the data source 101 to the user terminal device 200, modulates this data, and then performs a precoding unit. Output to 103. Precoding section 103 transmits data addressed to user terminal apparatus 200 via transmission antenna 104 according to the modulation scheme and transmission power finally determined according to the method described later.

FIG. 6 is a flowchart showing a procedure of an adaptive modulation method based on multi-user precoding according to the present embodiment.

[0030] In step ST3010, base station apparatus 100 acquires the channel matrix of user terminal apparatus 200. For time division multiplexing (TDD) systems, the channel matrix obtained by estimating the uplink can be used directly for downlink precoding due to the uplink and downlink symmetry. For the frequency division multiplexing (FDD) system, the user terminal device 200 needs to feed-knock channel information to the base station device 100. Further, the channel Z noise estimation unit 202 of the user terminal device 200 estimates the noise power of the received signal, and notifies the base station device 100 of the estimated noise power via the feedback channel CH1 and the like. In this way, base station apparatus 100 acquires channel matrix H and noise power of user terminal apparatus 200.

[0031] In step ST3020, base station apparatus 100, based on channel matrix H and noise power of user terminal apparatus 200, performs an equivalent signal-to-noise ratio (equivalent SNR) of each user according to a predetermined precoding method. Calculate For example, if the reverse channel precoding method is used as the precoding method, the transmission channel before precoding is applied. If the symbol is d and the transmission symbol after precoding is s, the following equation (7) is obtained.

[Equation 7] s = H ~ ^l d (7) Assuming that the transmission symbol d before precoding has already been properly normalized, the following relationship (8) exists.

[Equation 8]

E (dd ^H ) ^ I (8) Based on the equation (8), the following equation (9) is obtained.

[Equation 9]

E (ss ^H ) ^ H- ^x H- ^H (9) In Equation (8), the average transmit power of the i-th antenna is the square of the 2-norm of the i-th row from 1 to H. It is shown that this corresponds to an increase in power. Assuming that the noise power of the i-th user is the 2-norm square of the i-th row of σ Η ^_1 , calculate the equivalent signal-to-noise ratio of the i-th user by the following equation (10) Can do.

[Equation 10]

When the equivalent signal-to-noise ratio for each user is determined, the flow proceeds to step ST3030, where possible modulation schemes are determined according to each user's equivalent signal-to-noise ratio, and an adaptive modulation meter table is formed. To do. Usually, for example, when the signal-to-noise ratio is less than 5 dB QPSK is selected as the modulation method, 8PSK is selected when the signal-to-noise ratio is 5 dB or more and less than lOdB, and 16QAM is selected when the signal-to-noise ratio is 10 dB or more. Different modulation schemes can be selected depending on the noise ratio. When the modulation method is determined, there is a parameter table corresponding to each modulation method such as transmission power.

[0033] Next, in step ST3040, after obtaining the equivalent signal-to-noise ratio of each user in the base station apparatus 100, the water injection (Wa ter Filling) theorem determines the power distribution value of each user terminal device 200.

The water injection theorem is a method for obtaining the maximum channel capacity by allocating power to each user for n independent parallel channels. For example, assume that the total power is limited by the following equation (11).

[Equation 11]

In this case, if the gain ^ of each channel is σ ² and the noise power of each channel is σ ² , the signal-to-noise ratio of each channel is expressed by the following equation (12).

[Equation 12]

Therefore, the channel capacity is expressed by the following equation (13).

¾13]

C ... (1 3)

Here, based on the Lagrange multiplier method, an auxiliary function expressed by the following equation (14) is introduced.

[Equation 14]

In Equation (14), L is a Lagrange multiplier. And let the partial derivative be the following formula (15).

[Equation 15]

Then, it is possible to obtain the following expression (16).

[Equation 16] = _w —… (1 6)

λ In equation (16), u is a constant, and its value is the value shown in equation (17) below.

[Equation 17]

Also, 満たし satisfies Eq. (10), which is a power constraint, and Ρ is 0 or more.

In practice, precoding goes through a linear transformation, so when receiving each user, Is completely removed. Therefore, from the viewpoint of the MIMO receiver (user), it is a parallel channel with no interchannel interference. According to the water injection theorem, a channel with a high signal-to-noise ratio should be allocated higher power to obtain greater capacity. Therefore, it is possible to allocate power based on the equivalent signal-to-noise ratio of each user by the water injection theorem. If the total transmission power is NXP, that is, the number of antennas, and the average transmission power of each transmission antenna is P, the distribution of power to the i-th user is as shown in Equation (18) below.

[Equation 18]

P)… (1 8)

In equation (18), SNR is the equivalent signal-to-noise ratio of the i-th user. If x + is defined as x + = max (x, 0), λ satisfies the following equation (19).

[Equation 19]

[0036] In step ST3050, since the noise power of each user is known, in this case, a new signal-to-noise ratio SNR 'for each user is calculated based on the power distribution value described above. Based on this SNR ′, the adaptive modulation parameter table power is selected and acquired by the required modulation method. Then, the power is redistributed to the same modulation method using a method based on the principle of signal-to-noise ratio or a similar method.

[0037] For example, assuming that the number of transmission antennas of base station apparatus 100 is Ν, each user terminal has one antenna, and the number of user terminals is also Ν, Η is a matrix of N X Ν.

[0038] When the modulation scheme is determined, power is allocated again with the same modulation scheme as a set. . For example, k users are all 16QAM, and the power distributed to these users is P (l), P (2),..., P (k), respectively. The approximate signal-to-noise ratios of these 16 users all satisfy the lower limit using 16QAM. In the process of power allocation, users with higher signal-to-noise ratios are allocated higher power and have a greater achievable capacity. However, in adaptive modulation, there are only a few types of modulation schemes that can be selected. Furthermore, in a set of modulation schemes of the same type, the allocated power P (l), P (2), ..., P (k) is also different, and the higher the signal-to-noise ratio, the more allocated power Is also expensive. In this way, the performance of a certain user is good (the allocated power is high!), While the performance of a certain user is poor (the allocated power is relatively low). This leads to performance imbalance. Therefore, the total power of this set is redistributed within a set of similar modulation schemes. In doing so, it shall follow the principle of equal signal-to-noise ratio described below.

The principle of the equal signal to noise ratio is as follows. Assume that the total allocated power of k users corresponding to the same modulation method is expressed by the following equation (20).

[Equation 20]

PM- _Q AM = P (… (2 0)

The principle of equal signal-to-noise ratio is that P (i) 'newly allocated satisfies the condition of the following equation (21).

[Equation 21] 尸尸 ( ² ) '— _ 尸 ()' ...

twenty two ? (2 1) and ₂ l

That is, in the equation (21), the relationship of the following equation (22) is satisfied.

[Number 22]

This can be viewed as a combination of adaptive modulation and power control. Adaptive modulation means assigning more bits to users with a high signal-to-noise ratio by the water injection theorem. However, once the modulation scheme is determined, power is distributed fairly among users of the same modulation scheme. In this way, in the same modulation system, the received signal-to-noise ratio of each user is basically the same, improving the overall error resilience and avoiding the occurrence of artificial “far-far effects”.

[0040] In step ST3060, after obtaining the modulation scheme and power distribution value of each user, precoding is performed by the multiuser precoding method. In this case, as the precoding method, the reverse channel precoding method, Tomlinson-Harashima multiuser MIMO precoding method, precoding method based on lattice reduction, AMC precoding method, or the like can be used.

[0041] Finally, in step ST3070, the acquired data is transmitted from antennas 104-1 to 104-N based on the power distribution value and the corresponding modulation scheme V.

FIG. 7 is a diagram showing constellation in the case of the M-QAM modulation scheme, particularly 16QAM.

FIG. 8 and FIG. 9 are diagrams showing simulation results. In this simulation, the base station apparatus 100 has four antennas, four user terminals, and each user terminal uses one antenna.

FIG. 8 is a diagram showing a performance comparison between the respective precoding methods. Here, 16QAM, 4

Take X 4 antenna as an example. A comparison of the inverse channel (ZF) method, the Tomlinson-Harashima precoding (THP) method, and the lattice reduction (LR) method is shown. As is clear from this figure, the Lattice reduction precoding method is superior to the other two.

[0045] Fig. 9 shows the performance comparison of each precoding method with and without adaptive modulation when the throughput is 12 bits, that is, the average of each user is 3 bits. The horizontal SNR represents the ratio between the average transmission power of precoding and the average noise power of the MIMO receiver. Figure 9 shows a comparison of ZF performance, ZF-AM performance, and LR-AM performance under conditions using 8PSK without adaptive modulation. Here, the ZF AM performance plot shows the performance when using the adaptive modulation of the present invention, that is, the modulation method also selects the medium power of 16QAM or QPSK, the number of users is 4, and the total number of bits is 12. The performance and LR-AM performance plots show the performance with the adaptive modulation and lattice reduction precoding method of the present invention! /. As is apparent from this figure, the performance is greatly improved when the adaptive modulation of the present invention is used.

[0046] The embodiments of the present invention have been described above.

Note that what has been described above is merely an example of a specific embodiment of the present invention. The scope of protection of the present invention is not limited to this, and all modifications and substitutions easily conceivable within the technical scope disclosed in the present invention by any person skilled in the art are also included in the scope of the present invention. It is what That is, the adaptive modulation method and MIMO transmission apparatus based on multi-user precoding according to the present invention are not limited to the above embodiments, and can be implemented with various modifications.

[0048] Also, the MIMO transmission apparatus according to the present invention can be mounted on a communication terminal apparatus and a base station apparatus in a mobile communication system, and thereby a communication terminal apparatus having the same operational effects as described above, A base station apparatus and a mobile communication system can be provided.

[0049] Here, the power described with reference to an example in which the present invention is configured by nodeware can also be realized by software. For example, the algorithm of the adaptive modulation method based on multi-user precoding according to the present invention is described in a programming language, and the program is stored in a memory and executed by an information processing means. The same function as that of the MIMO transmission apparatus can be realized.

[0050] This specification is based on Chinese Patent Application No. 200510056303.6 filed on Mar. 16, 2005. All this content is included here.

Industrial applicability

[0051] Adaptive modulation method and MIM based on multi-user * precoding according to the present invention o The transmission device can be applied to applications such as communication terminal devices and base station devices in mobile communication systems.

Claims

The scope of the claims

[1] obtaining a channel matrix and noise power of each communication terminal;

Calculating an equivalent signal-to-noise ratio of each communication terminal using the channel matrix, the noise power, and a predetermined precoding method;

Determining an adaptive modulation parameter table using the equivalent signal-to-noise ratio;

Determining a power distribution value of each communication terminal according to the water injection theorem based on the equivalent signal-to-noise ratio;

Calculating a signal-to-noise ratio of each communication terminal using the noise power and the power distribution value, and determining a modulation scheme for each communication terminal based on the signal-to-noise ratio; ,

In accordance with the same method as the signal-to-noise ratio, a step of allocating power again to communication terminals using the same modulation method to obtain a final distribution power of each communication terminal; and Based on the modulation method! /, Transmitting data addressed to each communication terminal,

An adaptive modulation method based on multi-user precoding comprising:

[2] In time division multiplexing, the channel matrix is obtained based on the symmetry of the uplink and downlink.

The adaptive modulation method based on multi-user precoding according to claim 1.

[3] In the frequency division multiplexing method, the channel matrix is acquired based on feedback information of each communication terminal.

[4] The predetermined precoding method is an inverse channel precoding method, and an equivalent signal-to-noise ratio of each communication terminal is calculated according to the following equation (1).

[Number 1]

However,

σ ^: Noise power of i-th communication terminal

β: inverse matrix of channel matrix 2 ^{_ 1} row i 2-norm square

[5] The water injection theorem obtains the maximum channel capacity by distributing power to each communication terminal for n independent parallel channels.

[6] For communication terminals using the same modulation scheme, when the total allocated power of k communication terminals of the same modulation scheme is expressed by the following equation (2), P (i) Distribute power again so that 'satisfies the condition of the following equation (3) and the following equation (4):

[Equation 2]

[Equation 3]

[Equation 4]

The adaptive modulation method based on multi-user precoding according to claim 1, wherein the modulation scheme includes at least one of 16QAM, 8PSK, or QPSK.

Means for obtaining the channel matrix and noise power of each communication terminal;

Means for calculating an equivalent signal-to-noise ratio of each communication terminal using the channel matrix, the noise power, and a predetermined precoding method;

Means for determining an adaptive modulation parameter table using the equivalent signal-to-noise ratio; means for determining a power distribution value for each communication terminal according to a water injection method based on the equivalent signal-to-noise ratio;

Means for calculating a signal-to-noise ratio of each communication terminal using the noise power and the power distribution value, and determining a modulation scheme for each communication terminal based on the signal-to-noise ratio; ,

In accordance with the same method as the signal-to-noise ratio, means for allocating power again to communication terminals using the same modulation method, and obtaining the final distribution power of each communication terminal;

Based on the final allocated power and the modulation method, means for transmitting data addressed to each communication terminal;

A MIMO transmitter comprising: