WO2008066468A2

WO2008066468A2 - Method and arrangement for selection of mimo transmission mode

Info

Publication number: WO2008066468A2
Application number: PCT/SE2007/050725
Authority: WO
Inventors: Bo Göransson; Reihaneh Malekafzaliardakani; Ari Kangas
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 2006-11-30
Filing date: 2007-10-09
Publication date: 2008-06-05
Also published as: WO2008066468A3

Abstract

The present invention relates to a method in a receiver unit for selection of MIMO transmission mode in a radio telecommunication network. The method comprises a first stage comprising the steps of - estimating a channel quality parameter (10); - comparing said channel quality parameter with a first threshold value (20); - depending on if the estimated channel quality parameter is above or below said threshold, selecting if multiplexing gain will be utilized or not; and a second stage comprising - determining the optimal subset of transmitted streams if multiplexing gain is selected in said first stage (401-403). The invention furthermore relates to a user equipment and a radio base station for carrying out said method.

Description

Method and Arrangement in a Telecommunication System

FIELD OF THE INVENTION

The present invention relates to selection of MIMO

(Multiple Input Multiple Output) transmission mode.

BACKGROUND

MIMO (Multiple Input Multiple Output) transmission techniques are seen as the main way to improve the transmission rate in today's and tomorrows wireless networks. By using the spatial multiplexing property, the transmission rate of a system can grow linearly with the number of antennas deployed at the different network nodes and at the terminal side.

The multiple antenna advantage can be utilized in a number of ways. Which way or transmission technique to use depends on several things including antenna arrangements, channel properties and operating regime to mention a few. In other words, there is no one transmission scheme that fit all scenarios. To overcome this problem, one can deploy a system optimized for certain parameters that is valid in a particular cell. For example, in large macro cells where coverage is a main obstacle, the antenna gain provided by beamforming is a good candidate. In other scenarios, multi- stream transmission techniques such as PARC (Per Antenna Rate Control) or BLAST (Bell Labs Layered Space Time) may be preferred.

When multi-stream transmission techniques such as PARC or

BLAST are used, it is important to match the number of transmitted streams to the effective channel conditions. This effective channel condition depends on a number of parameters. Firstly, the actual transmission channel will influence the performance to a great deal. The number of transmitted streams should not be more than the effective rank of the channel. The rank is mostly dependent on the number of transmit and receive antennas, but to some extent the noise and interference will also influence the effective channel rank. In low SINR (Signal to Interference-plus-Noise Ratio) conditions, the sub-channel with lowest power will be "hidden" by noise and hence the effective rank will decrease.

This problem can be circumvented by using adaptive transmission techniques. One such technique is S-PARC (Selective PARC) , in which the number of transmitted streams, and used antennas, is decided using feedback from the mobile terminal. However, it turns out that estimating the optimal transmission is far from a trivial task. Further, if a network centric approach is preferred the feedback load from the terminal will be substantial. Here, we refer to network centric when the network is in control of the transmission parameters .

Alternatively, a terminal centric approach may be used. Here, the terminal estimates the optimal transmission strategy and then feedback this information to the network which has to transmit according to this strategy. This approach will likely save feedback capacity. However, since the terminal is not, in general, aware of the available resources nor the situation in the network, this approach is usually avoided in multi user cellular networks.

Different combinations of terminal and network centric approaches have also been suggested. In this case, the terminal "suggests" a preferred transmission strategy, while the actual decision is taken in the network and then forwarded to the terminal together with the data transmission.

Regardless of which approach that is taken, the optimal transmission strategy has to be estimated. It turns out that for many cases this is a computational expensive procedure. The terminal should estimate the number of antennas, which antennas (or beams) together with transmission parameters such as modulation and coding format to use.

The existing selection criteria are as follows:

1. Maximum Capacity: For every subset of transmit antennas p e P compute the channel capacity by

then select the subset with the largest C_p . See R. W. Heath,

S. Sandhu, and A. Paularj , "Antenna selection for spatial multiplexing systems with linear receivers", IEEE Commun. Lett., vol.5, no.4, pp.142-144, April 2001.

2. Maximum Transmission data rate: In this scheme, first the outputs SINR of each subset of transmit antennas are calculated, then the corresponding transmission rate is obtained, for example through a lookup table which is prepared in advance. The subset which maximizes the sum data rate will be chosen. Lookup table maps the SINR to the corresponding data rate by choosing the modulation scheme and coding rate. See S. J. Grant, K. J. Molnar, L. Krasny, "System-Level Performance Gains of Selective Per-Antenna- Rate-Control (S-PARC)". IEEE VTC spring 2005.

For example, consider the Maximum Transmission data rate where the base station has four transmitter antennas, Tx, while the terminal has two receiver antennas, Rx. In this case, the terminal should decide (in general) from which transmitter antenna or pair of antennas the data should be transmitted and with what transmission rate. Assuming now that the terminal will maximize the Mutual Information (MI) in order to find the optimal transmission strategy, the MI for ten different antenna combinations (six pairs and four single antennas) have to be estimated. For each antenna combination it has to calculate the MI for many combinations of modulation format and coding rate. Note that for many wideband systems, this has to be done for each frequency, e.g. in an OFDM system this should optimally be done per sub- carrier, but in practice this will be done by e.g. each resource block (i.e. group of sub-carriers). In all, this will be a very challenging task since it has to be done for every TTI (Transmission Time Interval,) .

SUMMARY

A problem experienced with previously known techniques is thus that a computational expensive procedure is required in the receiving unit in order to select an appropriate transmission mode. It is therefore an object of the present invention to provide a method that reduces the computational load in the receiving unit. The present invention therefore relates to a method in a receiving unit for selection of transmission mode. More specifically, the invention relates to a method in a receiver unit for selection of MIMO transmission mode in a radio telecommunication network. The method comprises a first stage comprising the steps of - estimating a channel quality parameter;

- comparing said channel quality parameter with a first threshold value; - depending on if the estimated channel quality parameter is above or below said threshold, selecting if multiplexing gain will be utilized or not; and a second stage comprising the step of - determining the optimal subset of transmitted streams if multiplexing gain is selected in said first stage.

Hereby, the second stage of the method is only carried out in cases where the estimated channel quality parameter exceeds the threshold value such that more than one transmitted stream is selected, which reduces the number of calculations required considerably.

In cases where the estimated channel quality parameter is below a threshold value such that a single transmitted stream is selected, the second stage of the method is not required. Instead, for the single stream mode, the transmitted stream having the highest estimated channel quality in the first stage could be chosen, or alternatively, the data stream could be sent using transmit diversity, where the data stream (or coded versions thereof) is transmitted from more than one antenna at a transmitter unit .

It should be noted that in this application the term "antenna" is used as a more general term then just the physical antenna. The term "antenna" can also be interpreted as a configuration or array of antennas, where the antennas are combined with different weights (beam forming) . In the example given in the foregoing, the four transmitter antennas can thus be seen as four "physical" antennas or as a set of e.g. four pre-defined beams. The invention is also applicable for applications with pre-coders or distributed antennas. Pre-coding includes the techniques of closed-loop beam forming and directional beam forming and is considered e.g. in the LTE concept.

The present invention further relates to a user equipment adapted for MIMO transmission, having a receiver unit comprising

- means for estimation a channel quality parameter; means for comparing said channel quality parameter with one or more threshold values;

- means for selecting if multiplexing gain should be utilized or not depending on if the estimated channel quality parameter is above or below said threshold; means for determining the optimal subset of transmitted streams if multiplexing gain is selected.

The invention further relates to a radio base having a receiver unit comprising the same means as presented for the user equipment.

The main advantage with this invention is thus to reduce the computational load in the mobile terminal. The method disclosed here can also be used to reduce the feedback load in the system, since the receiver makes the selection of the mode, which means that only parameters associated by this mode has to be fed back to the transmitter side. For example, if the receiver makes the selection according to the invention that one stream should be transmitted, only the channel quality for this stream has to be fed back. If, instead, this choice should be made in the transmitter, channel qualities for each transmission mode have to be fed back.

BRIEF DESCRIPTION OF THE DRAWINGS

The foregoing and other objects, features and advantages of the invention will be apparent from the following detailed description of preferred embodiments as illustrated in the drawings.

Figure 1 shows a flowchart of the selection algorithm of one embodiment of the transmission control scheme;

Figure 2 shows a flowchart of the selection algorithm of an alternative embodiment of the transmission control scheme;

Figure 3a shows a user equipment having a receiver unit according to the invention.

Figure 3b shows a base station having a receiver unit according to the invention.

Figure 4 illustrates the performance of the proposed transmission scheme.

DETAILED DESCRIPTION

The two prior art schemes mentioned above are optimal with respect to maximizing the system throughput. However, the present invention is aiming for suboptimal transmission schemes that involve less computation (low complexity) and yet provide system throughput close to optimal schemes. The invention proposes a scheme with a two-stage evaluation. In the first stage the evaluation is based on a quality measure that is independent of what is transmitted, e.g. SNR (Signal to Noise Ratio) , and in the second stage, the evaluation is based on a channel quality parameter to be used for link adaptation, e.g. SINR. It should be noted that other parameters that serves the same purpose can be used in both stages. For example, MI (Mutual Information) can be used as a measure of the channel quality parameter in the second stage.

In the following, an exemplified embodiment of the invention with reference to Fig. 1 will be described. In this example that refers to an outdoor cellular network with two antennas allocated at the terminal and four antennas at a base station, BS, the antennas are represented by physical antennas. It is however possible to apply the invention in a situation where each "antenna" is represented by e.g. a configuration of antennas in which the antennas are combined with different weights, so called beam-forming. Moreover, the invention can be applied in connection with pre-coding, which includes the techniques of closed-loop beam-forming and directional beam-forming or distributed antennas.

In the exemplified outdoor cellular network mentioned above, with two antennas at the terminal and four antennas at the base station, there are two different possible modes in the downlink:

• mode 1: only one transmit antenna is selected and there is no multiplexing gain.

• mode 2: a subset of two transmit antennas is selected so it delivers the multiplexing gain.

In this transmission control scheme the operating mode is first identified by calculating a received channel quality parameter, e.g. the SNR according to step 10, and thereafter through comparing the output average SNR with a predetermined threshold value, see step 20. If the average SNR is below the threshold, mode 1 will be chosen whereby no multiplexing gain is utilized. If, on the other hand the SNR exceeds this threshold, mode 2, i.e. multiplexing gain, as described above will be selected. If mode 2 is selected then the channel quality parameter, e.g. SINR, is used to choose the best subset. The best subset is the one which maximizes the sum SINR. Thus, if the SNR exceed this threshold an antenna mode comprising two transmitter antennas will be chosen. Using this information, the terminal can calculate the MI for the appropriate number of antennas. Hence, in this case the number of necessary calculations has been reduced to approximately half. In this case: from 10 to 6 or 4 depending on the threshold.

The selection procedure at the receiver is as follows:

Step 10: Calculate the average received SNR as:

\/i.j : i = l...,n_r ;j = \,..., n, ,

where P₁ , n, and n_r are total transmit power and number of transmit and receive antennas, respectively.

Since no multiplexing gain is utilized in mode 1, at step 301 the antenna with highest instantaneous SNR is selected according to the following equation:

At step 302, the information is fed back to the transmitter unit.

Furthermore, for mode 1, the receiver unit could calculate a channel quality measure such as SINR or MI for both alternatives above, and then inform the transmitter unit which transmission mode that is the most appropriate, and also which channel quality measure value, e.g. SINR value, the calculation resulted in. Alternatively, a preferred transmission format including e.g. a suitable modulation format and coding rate can be fed back to the transmitter. Since in mode 2, the multiplexing gain is utilized, the SINR for all possible subsets is calculated, see step 401. In a MMSE (Minimum Mean Square Error) receiver, which is a linear receiver that maximizes the output signal-to-interference- plus noise ratio (SINR) for any values of SNR, the output SINRs are calculated as

SINR_m

At step 402, the subset with maximum sum SINR is selected. Step 403 provides feedback of the information to the transmitter.

As illustrated in Fig. 4, simulation results show that the performance of the proposed transmission scheme, maximum sum SINR selection, is reasonably close to the gain achieved by the optimal scheme, i.e. maximum transmission data rate.

It should be noted that the invention is not limited to a situation where only one or two receiving antennas is available at the receiving unit. Obviously, there can be three or more antennas available at the receiving unit, e.g. a user equipment. For such situation, it is suitable to provide a set of a plurality of thresholds values, whereby the number of threshold levels in a set should be equal to the number of receiver antennas at the receiver unit minus one. Hereby the lowest threshold defines whether one or two transmission antennas should be selected; a second threshold defines whether two or three transmission antennas should be selected and so on. Fig. 2 shows an alternative embodiment of the invention, where steps 10 and 20 are identical to the embodiment shown in Fig. 1. Step 303 provides an alternative to previously described step 301 in Fig. 1. According to step 303, if mode 1 is selected at step 20, the data stream is transmitted using transmit diversity where the data stream (or coded versions thereof) is transmitted from more than one antenna at a transmitter unit. The following step 302 is identical to the embodiment shown in Fig. 1.

According to the embodiment illustrated in Fig. 2, it is assumed that a set of threshold values is utilized to determine the number of transmit antennas to be used in mode 2. Thus, if mode 2 is selected at step 20, i.e. the multiplexing gain path, the SNR is compared with further threshold values in order to determine the number of antennas to be used. Thereafter, at step 405, the SINR for all possible combinations within the selected number of streams is calculated. The following steps 402 and 403 are identical to the embodiment shown in Fig. 1.

It is obvious that combinations of the embodiments described with reference to Fig. 1 and 2 are possible.

Thus, according to the invention, a threshold value identifies if multiple transmit antennas should be used or not, and possibly further threshold values indicate the number of transmit antennas to be used as described in the foregoing. It is thus obvious that selecting the right threshold value or values has significant effect on the performance of the MIMO system. Selecting a low threshold results in increasing the complexity, sometimes without achieving high multiplexing gain, on the other hand, a large threshold can cause loosing multiplexing gain due to not using the multiple antennas. The threshold value or values can be selected based on the channel capacity. The channel capacity of a MIMO system with identity covariance matrix (i.e. spatially white transmit covariance) is given by:

At high SNR the capacity is approximated as

. SNR '^ r ., 1 C₁₁ * «_llώl log, + ∑ E[IOg₂ λ; J ,

>h ..i

where /J_mjn = min(«_r ,/?, ) . At low SNR we have

e .

Considering the fact that the MIMO technique provides multiplexing gain at high SNR regime, the threshold can be selected as the SNR in which the difference between the channel capacity and the high SNR approximation of capacity exceeds the difference between the channel capacity and the low SNR approximation of capacity, C-C_h>C_l-C .

Furthermore, different threshold values or sets of threshold values can be utilized under different conditions, dependent on e.g. channel characteristics and/or channel properties, or estimated transmission parameters such as SNR, time dispersion and frequency selectivity. Yet another alternative is to derive the threshold value or set of threshold values from signaled transmission parameters such as power, code rate, transport block size, modulation order etc.

As mentioned above, the channel type may have an effect on the threshold, which means that the threshold can be selected adaptively with respect to channel type. For example, in a situation with two receiver antennas and two transmitter antennas, the threshold for a Rayleigh fading channel could be set to 5dB, while for a 3GPP channel model the threshold could be set to 7dB. The 3GPP channel model referred to in this example is a suburban macro-cell channel, which is modeled using 3GPP spatial channel model for MIMO simulations, see 3GPP TR 25.996 V6.1.0, Spatial channel model for Multiple Input Multiple Output (MIMO) simulations. The skilled person will be able to find the most suitable threshold levels by performing tests under various conditions.

Fig. 3a illustrates schematically a user equipment 100 provided with a receiver unit 101. Said receiver unit comprising means 110 for estimating a channel quality parameter according to the invention, e.g. the SNR value. The receiver unit furthermore comprises means 120 for comparing said channel quality parameter with the threshold value as described in the foregoing and means 130 for selecting if multiplexing gain is to be utilized or not dependent on the outcome of the comparison with the threshold. If the multiplexing path is utilized, stage 2 of the method according to the invention is carried out by using means 140 for determining the optimal subset of transmitted streams.

Fig. 3b illustrated schematically a base station 200 provided with the same receiver unit as described in the foregoing.

For the case where the receiver unit is comprised in a User Equipment such as a mobile terminal, the receiver unit can according to one embodiment receive information regarding which threshold value or values to be utilized from the network, e.g. a base station. Alternatively, the receiver unit determines itself for each TTI which threshold value or values to be utilized.

The invention is not to be limited to the disclosed embodiments, but is intended to cover various modifications within the scope of the appended claims.

Claims

1. A method in a receiver unit for selection of MIMO transmission mode in a radio telecommunication network, characterized by a first stage comprising the steps of

- estimating a channel quality parameter (10);

- comparing said channel quality parameter with a first threshold value (20) ; depending on if the estimated channel quality parameter is above or below said threshold, selecting if multiplexing gain will be utilized or not; and a second stage comprising determining the optimal subset of transmitted streams if multiplexing gain is selected in said first stage (401-403) .

2. A method according to claim 1, wherein said channel quality parameter is compared to a set of threshold values (404), such that a first threshold defines whether one or two transmitted streams should be selected; a second threshold defines whether two or three transmitted streams should be selected and so on.

3. A method according to claims 1 or 2, wherein different threshold values or sets of threshold values can be utilized under different conditions.

4. A method according to claim 3, wherein the threshold value or set of threshold values is selected with regard to channel characteristics and/or channel properties .

5. A method according to claim 3, wherein the threshold value or set of threshold values is selected with regard to estimated transmission parameters.

6. A method according to claim 3, wherein the threshold value or set of threshold values is selected with regard to signaled transmission parameters.

7. A method according to any of claims 1 to 6, wherein the SNR (Signal to Noise Ratio) for the transmission channel is used as the channel quality parameter in the first stage of the method (10) .

8. A method according to any of claims 1-7, wherein if the estimated channel quality parameter is below the threshold such that no multiplexing gain is utilized, the antenna or antenna configuration with highest instantaneous SNR is selected (301) .

9. A method according to any of claims 1-7, wherein if the estimated channel quality parameter is below the threshold such that no multiplexing gain is utilized, data is transmitted with transmit diversity from more than one antenna or antenna configuration at a transmitter unit (303) .

10. A method according to any of claims 1 to 7, wherein the optimal subset of transmitted streams is determined in the second stage of the method by calculating a channel quality parameter for each possible subset of transmitted streams (401) .

11. A method according to claim 10, wherein SINR (Signal to Interference-plus-Noise Ratio) is used as said channel quality parameter.

12. A method according to claim 10, wherein MI (Mutual Information) is used as said channel quality parameter.

13. A method according to any of the preceding claims, wherein said receiver unit is comprised in a User Equipment (100) .

14. A method according to claim 13, wherein the receiver unit receives information from the network regarding which threshold value or values to be utilized.

15. A method according to claim 13, wherein the receiver unit determines for each TTI which threshold value or values to be utilized.

16. A method according to any of claims 1-12, wherein said receiver unit is comprised in a radio base station (200) .

17. A user equipment (100) adapted for MIMO transmission, characterized in a receiver unit (101) comprising means (110) for estimating a channel quality parameter; - means (120) for comparing said channel quality parameter with one or more threshold values;

- means (130) for selecting if multiplexing gain should be utilized or not depending on if the estimated channel quality parameter is above or below said threshold; - means (140) for determining the optimal subset of transmitted streams if multiplexing gain is selected.

18. A user equipment according to claim 17, adapted to perform the method according to any of claims 2-15

19. A radio base station (200) adapted for MIMO transmission, characterized in a receiver unit (101) comprising

- means (110) for estimation a channel quality parameter;

- means (120) for comparing said channel quality parameter with one or more threshold values;

20. A radio base station according to claim 20, adapted to perform the method according to any of claims 2- 12.