WO2008152540A2

WO2008152540A2 - Method and apparatus for antenna selection in a multi-antenna system

Info

Publication number: WO2008152540A2
Application number: PCT/IB2008/052167
Authority: WO
Inventors: Ni Ma; Qi Zhou; Suoming Pu; Gang Wu; Xun Fan; Xiaobo Zhang
Original assignee: Koninklijke Philips Electronics NV
Current assignee: Koninklijke Philips NV
Priority date: 2007-06-15
Filing date: 2008-06-04
Publication date: 2008-12-18
Anticipated expiration: 2009-12-15
Also published as: WO2008152540A3; TW200901652A

Abstract

The present invention proposes an antenna selection method and apparatus for a multi-antenna system, by means which transmit antennas and quantized vectors of a codebook can be selected, based on the channel information about the received signal in combination with vector quantization technology, so as to decrease the overhead for feedback over the uplink while obtaining a multi-user system capacity.

Description

METHOD AND APPARATUS FOR ANTENNA SELECTION IN A MULTI-ANTENNA SYSTEM

Field of the Invention The invention relates to a wireless communication technology, and more particularly, to a method and apparatus for antenna selection in a multi-antenna system.

Background of the Invention

Compared with Single (antenna) Input Single (antenna) Output (SISO, Single Input Single Output) wireless communication systems, Multiple (antenna) Input Multiple (antenna)

Output (MIMO, Multiple Input Multiple Output) wireless communication systems can provide greater channel capacity, and significantly increase spectrum efficiency and decrease error rate in wireless communication connection.

A multi-user MIMO system can obtain not only a large capacity by using MIMO processing, but also the advantage of Space Division Multiple Access (SDMA).

An exemplary system comprises M_t transmit antennas and M_r receive antennas, where the channels may be represented by a matrix H of M _r xM _t dimensions. Using pre-coding technology, a received signal may be represented as:

Y = HFS + W ( 1 ) where F is a pre-coded matrix of M _r x M dimensions, S is a codeword of M x T dimensions to be transmitted, and W is a noise matrix. The Pre-coded parameter M satisfies the condition that M < M_t . The channel matrix H and the pre-coded matrix F satisfy the function F = f (H) .

Basically, the relationship between the input and the output represented by equation (1) covers most MIMO systems with feedback, including beam-forming, pre-coded spatial multiplexing and antenna selection technologies (i.e., selecting M antennas for signal transmission from M_t antennas), etc. The pre-coded matrix F consists of M columns of the

M x M₁ matrix.

Although a multi-antenna system may increase the capacity and reliability of wireless communication, the cost related to the physical size, power consumption and hardware induced by the multi-antenna system and multiple radio frequency links is very high. Antenna selection technology as an alternative solution may combine (?) many advantages of MIMO systems with low cost and low complexity. The key point of antenna selection technology lies in the selection of the most appropriate antennas from transmit antennas by using feedback information received from the receiver to achieve the greatest equivalent Signal to Noise Ratio (SNR).

Using the Channel Quality Indicator (CQI), a base station (Node B) can customize transmission waveforms by optimizing the MIMO system, so as to provide a larger link capacity and throughput, and increase the system capacity by sharing spatial channels with a plurality of users. Unfortunately, the requirement for feedback information increases typically with the product of the number of transmit antennas, the number of receive antennas, the delay spread, and the number of users, while the capacity increases only linearly. Therefore, some people have started to study the possibility of using Vector Quantization (VQ) technology in a communication system with feedback, so as to quantize CQI before being transmitted via a feedback channel with limited data rate, thereby making full use of the limited bandwidth sources of the feedback channels. This requires the design of a codebook to represent the basic degrees of freedom of the channels, and the codebook should satisfy the requirement of channel models and the design of receivers. Figure 1 shows a general example of VQ processing. Firstly, the columns of the channel matrix H are stacked into a complex vector h_vec of M_r - M_t dimensions, where M_r and M_t represent the numbers of receive antennas and transmit antennas, respectively. Then, the vector h_vec is quantized by using a VQ algorithm.

The feedback information defines a selection function /(where F = /(H) ), therefore,/ is mapped to the codebook as follows: ξ = [F₁, F₂,-, F_N ] where all possible pre-coded matrixes are specified. The value of N is defined so as to satisfy the condition that N = 2 for an integer B. The selected matrix is transferred from one receiver to the multi-antenna transmitter as feedback information with B bits. Though the multi-user MIMO system is advantageous in many aspects, Node B has to receive sufficiently precise CQI from each of the User Equipments (UE) to obtain the multi-user capacity. As the number of transmit antennas increases, the overhead of feedback signaling will increase greatly.

Limited feedback information is used in a multi-antenna system in some prior arts. This type of multi-user MIMO system using limited Finite-rate Feedback is essentially an Interference-limited system, in which the level of available CQI at the transmitter will affect the multi-path gain of the downlink in the multi-user MIMO system.

As described above, it is particularly important to provide precise channel feedbacks for a multi-user MIMO system. On the other hand, however, the heavy feedback signaling overhead caused by the precise CQI requirement will affect the uplink capacity severely.

Summary of the Invention

One of the objects of the invention is to propose a method and apparatus for antenna selection in a multi-antenna system, which decreases the feedback signaling overhead of a multi-user MIMO system by a combination with VQ. A method for antenna selection used in a multi-antenna system according to the present invention, comprising the steps of: a. receiving a signal from a multi-antenna transmitter; b. calculating the channel information about the received signal; and c. selecting transmit antennas and codebook elements, based on the calculation, and feeding the index information about the selected transmit antennas and the index information about the corresponding codebook back to the transmitter.

An apparatus for antenna selection used in a multi-antenna system according to the present invention, including: a radio frequency unit for receiving a signal from a multi-antenna transmitter; a channel calculation unit for calculating the channel information about the received signal; and an antenna selection unit for selecting transmit antennas and codebook elements, based on the calculation of the channel calculation unit, and feeding the index information about the selected transmit antennas and the index information about the corresponding codebook back to the transmitter.

Compared with the prior arts, using the method and apparatus for antenna selection in the multi-antenna system of the present invention, UE can select the most appropriate antennas from all the antennas to decrease the number of different dimensions of the codebook, thereby decreasing the overhead for feedback over the uplink, improving the system performance and stability, and decreasing the complexity of some configurations, while obtaining a multi-user system capacity. Other objects and attainments together with a better understanding of the invention will become apparent and appreciated by referring to the following descriptions and claims taken in conjunction with the accompanying drawings.

Brief Description of the Drawings The invention will be described with reference to the accompanying drawings, in which:

Figure 1 shows a flowchart for a general example of VQ processing; Figure 2 shows a codebook of a vector of M dimensions;

Figure 3 shows a process for codebook processing using an antenna selection method of the present invention; Figure 4 shows the principle of a multi-user MIMO system using an antenna selection method of the present invention; and

Figure 5 shows the structure of main functional modules of a UE using an antenna selection method of the present invention.

Throughout the above drawings, same reference numerals will be understood to refer to same, similar or corresponding features or functions.

Detailed Description

A typical channel reporting technology based on multi-user MIMO makes use of a codebook having M-dimension vectors, providing an index of B bits and a real-valued CQI to the transmitter via each UE's feedback channel. As illustrated in Figure 2, the codebook includes quantized vectors of N unit lengths, where JV = 2^B • All the quantized vectors have

M dimensions, where M is the number of transmit antennas at Node B. For simplicity, the i^Λ dimension of the codebook is represented as CB(M,i,N), where the length of the channel quantized vectors is N, and there are M dimensions in total. A UE feeds back the quantization index B together with an estimate of a real-valued lower-bound of its Signal Interference and Noise Ratio (SINR), the SINR depending on the channel amplitude and directional quantization error.

In one of the schemes of the present invention, as illustrated in Figure 3, the method for Transmit Antenna Selection (TAS) is fully used to decrease the quantized vectors of the channel. The TAS update period, i.e., the period in which a valid antenna selection is not ^ changed, is typically long enough to keep the number of reduced codebook dimensions unchanged during a plurality of Transmit Time Intervals (TTI), such that the signaling for indicating the group of antennas selected by each UE can be well controlled. Basically, Node B is responsible for informing the UE of the corresponding TAS update period, and the UE is responsible for updating the TAS result accordingly.

As illustrated in Figure 4, the UE will not feed back all the codebook elements, even if their indexes can indicate the best channel information of different transmit antennas. Instead, the UE calculates the channel information, such as the SNR value of the MIMO channel, and selects the most appropriate M(i) ' transmit antennas from the M transmit antennas, that is those which can provide the greatest equivalent received SNR. After the TAS processing, the^ UE feeds back to Node B the quantization index of the M(i) '-dimension codebook together with the transmit antenna index selected by the UE for subsequent data communication.

Accordingly, as illustrated in Figure 5, a UE using an antenna selection method according to the present invention mainly includes a radio frequency processing unit 10 and a base band processing unit 20, with the base band processing unit 20 including a TAS unit 2010 and a channel information calculation unit 202. The channel information calculation unit 202 calculates the channel information about the received signals, such as the SNR value of the MIMO channel, and the TAS unit 201 selects the most appropriate M(i) ' from the M transmit antennas, based on the calculation result of the channel information calculation unit 202, such that these transmit antennas are able to provide the maximum equivalent received SNR. The5 TAS unit 201 feeds the quantization index of the M(i) '-dimension codebook , as well as the index of the selected transmit antennas for subsequent data communication, back to Node B via the radio frequency processing unit 10. To simplify the description, the other modules of the UE, which are the same as or similar to those in the prior arts, are not shown in Figure 5.

It is obvious that decreasing the number of dimensions of the codebook by introducing0 TAS technology is very helpful for decreasing the overhead of the feedback signaling over the uplink. Taking a system with N pieces of channel information already established and M transmit antennas as an example, the existing multi-user MIMO channel report technology requires M x log₂ = M x B bits to be fed back. But in the solution of the present invention, it is assumed that each UE selects half of the transmit antennas, i.e. only

M J /2

Sx + 1°§2^M bits need to be fed back, with the additional index bit ^B:

2 being used to indicate which group of antennas is selected by the UE. For a four-transmit-antenna system using a codebook of 256 unit length vectors, the number of bits which may be saved by each UE is:

BxM -Bx- - (flog₂ ^"'² ]) = 8x4-8x 2-([log⁶ ₂]) = 29 (2)

It should be noted that such an antenna selection rate is very low, i.e., the antenna selection result obtained by a specific UE may remain unchanged during a long period of data r r^Mn transmission, such as at least several TTIs. Therefore, the additional index bit ^⁰S₂" needs not to be sent during the whole data transmission period, and this will result in a more efficient uplink feedback signaling. In the example of the equation (2), for most rounds of signaling transmission, each round may save up to 3 bits of feedback overhead.

The TAS update period can be obtained directly from section 7.1.1.4.3 of the 3GPP LTE Physical Layer Specification Standard, but this section only requires that the update interval for the selected pre-coded vector should be short enough to track the instantaneous channel variations. In the present invention, the coarse estimate of the TAS update period is done with reference to the time -varying characteristics in the Doppler Shift domain. In other words, if the channel response does not change during the time period To, the TAS result needs not to be changed either. Because an advanced signal processing ability can generally overcome the problems caused by channel variations, the above case is only an extreme example of TAS technology. The relationship between the Doppler propagation f_d and To is approximately:

9

T ¹ O

16C

, _ 2πf_c where J_d — represents the Doppler shift, c represents the light speed, and f_c c represents the carrier frequency.

In the above exemplary four-transmit-antenna system using the codebook of 256 unit length vectors, the saved signaling overhead in different environments can be calculated, as in the approximation shown in Table 1. Here, the TTI duration and the carrier frequency can be determined based on the 3GPP LTE Physical Layer Specification Standard, and the assumed value for each TTI in a frame is used in the uplink signaling feedback transmission.

Table 1

It may be seen from Table 1 that the TAS method of the present invention combined with VQ is more advantageous in low speed environments. Since TAS technology does not depend only on the time-varying characteristics of the wireless channel, the improvement in capacity due to the present invention may be greater actually.

As shown above, the antenna selection method and apparatus in a multi-antenna system of the present invention can reduce the overhead for uplink feedback and improve system performance and stability, while providing multi-user capacity.

It should be noted that the above described embodiments are given for describing rather than limiting the present invention, and the person skilled in the art should understand that the antenna selection method and apparatus for the multi-antenna system disclosed in the present invention may be modified without departing from the scope of the invention based on the contents of the present invention. Thus, the protective scope of the present invention is defined by the accompanying claims. In addition, any of the reference numerals in the claims should not be interpreted as a limitation to the claims.

Claims

What is claimed is:

1. A method for antenna selection used in a multi- antenna system, comprising the steps of: a. receiving a signal from a multi-antenna transmitter; b. calculating the channel information about the received signal; and c. selecting transmit antennas and codebook elements, based on the calculation, and feeding the index information about the selected transmit antennas and the index information about the corresponding codebook back to the transmitter.

2. The method according to claim 1, wherein calculating the channel information in step b comprises: calculating the signal to noise ratio of the channel.

3. The method according to claim 2, wherein the selection performed on the transmit antennas in step c is based on the criterion that the selected transmit antennas can provide a maximum equivalent received signal to noise ratio.

4. The method according to claims 1, 2, or 3, wherein selecting the codebook elements in step c comprises: selecting vectors, having a number of dimensions corresponding to the selected transmit antennas, from quantized vectors of the codebook.

5. The method according to claims 1, 2, or 3, wherein if the channel response remains unchanged during a period of communication, the index information about the selected transmit antennas is fed back only once during the period in step c.

6. The method according to claims 1, 2, or 3, wherein the selected transmit antennas are used by the transmitter for subsequent signal transmission.

7. An apparatus for antenna selection used in a multi-antenna system, including: a radio frequency unit for receiving a signal from a multi-antenna transmitter; a channel calculation unit for calculating the channel information about the received signal; and an antenna selection unit for selecting transmit antennas and codebook elements, based on the calculation of the channel calculation unit, and feeding the index information about the selected transmit antennas and the index information about the corresponding codebook back to the transmitter.

8. The apparatus according to claim 7, wherein calculating the channel information by the channel calculation unit comprises: calculating the signal to noise ratio of the channel .

9. The apparatus according to claim 8, wherein the selection performed on the transmit antennas by the antenna selection unit is based on the criterion that the selected transmit antennas can provide a maximum equivalent received signal to noise ratio.

10. The apparatus according to claims 7, 8, or 9, wherein selecting the codebook elements by the antenna selection unit comprises: selecting vectors, having a number of dimensions corresponding to the selected transmit antennas, from quantized vectors of the codebook.

11. The apparatus according to claims 7, 8, or 9, wherein if the channel response remains unchanged during a period of communication, the antenna selection unit feeds back the index information about the selected transmit antennas only once during the period.

10