WO2008156389A1

WO2008156389A1 - Adaptive antenna transmission and antenna calibration

Info

Publication number: WO2008156389A1
Application number: PCT/SE2007/000612
Authority: WO
Inventors: Henrik Asplund; Jonas Medbo; Mathias Riback
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 2007-06-21
Filing date: 2007-06-21
Publication date: 2008-12-24
Also published as: US8548399B2; JP2010531571A; EP2160853A4; EP2160853B1; EP2160853A1; JP5102355B2; US20100190456A1

Abstract

The present invention relates to a method for adapting transmission parameters in a transmitter Tx in communication with at least one antenna 27; 38; 47; 57; 77, 88. The method comprises: transmitting a signal from the transmitter Tx measuring at least one reflection coefficient S11 of said signal for each antenna 27; 38; 47; 57; 77, 88, and adapting the transmission parameters based on the measured reflection coefficient S11. The antenna could be a transmit antenna and/ or a receive antenna. In a preferred embodiment antenna calibration may be performed by measuring propagation time for each individual transmitter chain 481, 482; 781, 782, and compensate for differences in time delay between the individual transmitter chains 481, 482; 781, 782 to assure that the signal is transmitted simultaneously from all transmit antennas.

Description

ADAPTIVE ANTENNA TRANSMISSION AND ANTENNA CALIBRATION

Technical field

The present invention relates to a method for adaptive antenna transmission and a method for antenna calibration.

Background

It is well known that adapting transmission parameters to the current channel conditions improves the performance of a wireless communication system. Parameters that can be adapted are for example the power allocation, both in the frequency domain and on different antennas in a multi antenna system, as well as modulation, coding, etc.

With channel state information (CSI) at the transmitter it is possible to allocate the available power to the parts of the frequency spectrum that have good channel conditions, i.e. not waste the power on frequencies that are in deep fade for the moment. On the parts of the spectrum that have good channel conditions it is also advantageous to use higher order modulation and lower coding rate.

In future mobile systems larger bandwidths will be used and therefore broadband antennas. The gain of these broadband antennas are however not equal over the whole frequency range. In a handheld unit the gain on different frequencies will also change when the antenna interacts with the user. In a system where the CSI is reported by a receiving unit to a transmitting unit, the antenna gain will be incorporated in the reported CSI.

In systems with multiple transmit antennas it is also known that the capacity of the system is improved when the transmission parameters are adapted to the current channel conditions for each antenna. Once again not to waste power on an antenna that is in a deep fade or for some other reason have poor channel conditions. An optimal power allocation can only be achieved when the transmitter has full CSI. Without CSI at the transmitter it is not possible to adapt to the current channel conditions at the transmitter and the best strategy is to transmit equal power, same modulation format and same coding rate over the whole frequency band and at all antennas, as illustrated in figure 5.

Summary

An object of the present invention is to provide a method to adapt the transmission parameters of a transmitter without the need to obtain information regarding the channel condition.

By measuring the reflection coefficient(s) at the transmitter the relative antenna gain can be estimated on different frequencies and/ or different antennas if more than one antenna is provided. It is then possible to adapt the transmission parameters without any CSI at the transmitter based on this frequency/ antenna dependent reflection coefficient.

When measuring the reflection coefficient it is also possible to measure the propagation time from the antenna port to the actual antenna, i.e. the delay introduced by antenna feeders etc. In a system with multiple antennas it is then possible to compensate for the different time delay in the different transmitter chains, assuring that the signal is transmitted simultaneously from all antennas. It is possible to ensure that the signals will be transmitted with a known amplitude and phase. This is often called antenna calibration.

An advantage with the present invention is that an improved performance of a wireless communication system is obtained compared to not adapting power allocation, modulation format, coding rate etc., although the adaptation is not optimal.

Another advantage with the present invention is that the interaction between a user and a handheld mobile unit are taken into consideration, thus providing an improved performance. Additional objects and advantages will become apparent for the skilled person from the detailed description of the preferred embodiments.

Brief description of the drawings

Fig. 1 shows an illustration of a communication channel.

Fig. 2 shows the transmitter chain of the communication channel provided with a first embodiment of the invention implemented in the transmitter chain.

Fig. 3 shows a communication channel provided with a second embodiment of the invention implemented in the receiver chain.

Fig. 4 shows a plot of measured reflection coefficients used in the present invention.

Fig. 5 shows a graph illustrating uniform power allocation in an OFDM system according to prior art.

Figs. 6a - 6c show graphs illustrating power allocation in an OFDM system according to the present invention.

Fig. 7a shows the transmitter chain of the communication channel provided with a third embodiment of the invention implemented in the transmitter chain.

Fig. 7b shows a graph illustrating power allocation between transmit antennas for the transmitter chain in figure 7a.

Figs. 8a and 8b show allocation of power for each transmit antenna in figure 7a as a function of frequency.

Fig. 9a shows the transmitter chain of the communication channel provided with a fourth embodiment of the invention implemented in the transmitter chain. Fig. 9b shows a graph illustrating power allocation between antennas elements for the transmit antenna in figure 9a.

Fig. 10 shows a telecommunication system according to the present invention.

Fig. 11 shows a communication channel provided with a fifth embodiment of the invention implemented in the receiver chain and in the transmitter chain.

Detailed description

Figure 1 shows the different parts of a communication channel 10 interconnecting a signal source 11 and devices, such as speaker, display memory, etc. in a receiving unit 12. The communication channel 10 comprises the transmitter chain 13, the wireless radio channel 14, and the receiver chain 15. The transmitter chain 13 comprises in this example a transmitter T_x, a transmit distribution network 16 and three transmit antennas 17, each having one or more antenna elements (not shown). The receiver chain 15 comprises in this example two receive antennas 18, a receive distribution network 19 and a receiver R_x, which comprises means to process received data and forward information to the devices in the receiving unit 12.

Each transmit antenna 17 is connected to the signal source 11 via an individual transmitter chain including the transmitter T_x and a part of the transmit distribution network 16. Each receive antenna 18 is connected to the receiver R_x via an individual receive chain including the receiver R_x and a part of the receive distribution network 19.

Traditionally, information regarding the channel condition has been determined by the receiver unit and reported back to the transmitter as indicated by the dashed line denoted CSI. The CSI includes information regarding the complete communication channel, whereby the transmitter T_x adapts the transmission parameters based on the CSI.

If the transmitter chain 13 comprises multiple transmit antennas 17 and employs beam forming, or some other form of precoding, it may also be necessary to have calibrated antennas, i.e. adjust the individual transmitter chain to make sure that the signal is transmitted from the transmit antennas simultaneously and with known amplitude and phase. This is normally achieved by requesting calibration measurements reports from user equipment communicating through the transmit antennas, and thereafter estimating parameters to compensate for the RF chain impairments, as described in reference [1] and [2].

Figure 2 shows a transmitter chain 20 of a communication channel provided with a first embodiment of an adaptation circuit 21. The transmitter chain 20 is provided with a transmitter T_x connected to a signal source 11 , and a distribution network 26 connected to the transmitter T_x and an antenna port of one transmit antenna 27 having five antenna elements 22 arranged on an antenna panel 23. A beam forming network 24, which is integrated in the antenna 27, is supplied with a signal from the transmitter T_x, and is connected to each antenna element 22. The beam forming network 24 may comprise phase shifters and means to adapt the amplitude of the signal being fed to each antenna element 22. A directional coupler 25 is used to determine the reflection coefficient together with the adaptation circuit 21, which preferably comprises a receiver capable of comparing the transmitted reference signal with the reflection of the same signal - thereby calculating Sn. A system for determining the reflection coefficient is described in reference [3]. The reference signal could either be the ordinary transmission from the transmitter T_x, or a signal generated in the adaptation circuit and transmitted from the transmitter during idle periods of the transmitter

The inventive concept relies on the ability to measure the reflection coefficient, usually denoted S_n , of a signal at each antenna using the adaptation circuit 21. S₁₁ is in this embodiment measured as a function of frequency, i.e. S₁₁(Z) , and the relative antenna gain can be estimated on different frequencies based on the reflection coefficient. It is then possible to adapt the transmission parameters, without the feedback of channel state information CSI, at the transmitter based on the measured frequency- dependent reflection coefficient. Only the characteristics of the T_x chain 20 will be taken into consideration when adapting the transmission properties, which will result in a sub-optimal adaptation compared to the prior art adaptation with CSI. However, the sub-optimal improvement will still provide an improvement compared to not adapting power allocation, modulation format, coding rate, etc.

The reflection coefficient S₁₁ (/) is a measure of how much of the transmitted power that is reflected by the antenna (and other parts of the transmission chain) . The power that is not reflected can be assumed to be transmitted by the antenna. Some parts will be burnt in the internal load of the antenna but the fraction of energy lost in the internal load is often small and does not have a frequency-dependence, or a very slight frequency-dependence, and will therefore not affect the optimal power allocation vs. frequency. The part of the transmit power that actually is transmitted, i.e. the transfer function, H(Z) can thus be expressed as:

|H(/)|² = i-|S_u(/)|² (i)

If S₁₁(Z) is measured at the transmitter for the Tx chain, as illustrated in figure 2, a sub-optimal power allocation can be calculated without any CSI at the transmitter by performing "water filling", as described in more detail below. The reflection coefficient S₁₁ may also be measured for the Rx chain in a similar fashion, as illustrated in connection with figure 3. In the following illustrative examples, the reflection coefficient for the TX chain S₁* (continuous thin line in figure 4) and the relative coefficient for the Rx chain S™ (dotted line in figure 4) have been measured, and the corresponding transfer functions H_1x and H_n , respectively, have been calculated using equation (1) and used to illustrate the inventive concept.

Figure 3 shows a receiver chain 30 of a communication channel provided with an adaptation circuit 31. The receiver chain 30 is provided with a receiver R_x connected to devices in a receiving unit 42, and a distribution network 39 connected to the receiver R_x and an antenna port of one receive antenna 38 having three antenna elements 32. A signal transmitted from the transmitter T_x is received by the antenna elements 32 and forwarded to the receiver R_x through the distribution network 39. A directional coupler 33 is used to determine the reflection coefficient together with the adaptation circuit 31, which preferably comprises a receiver capable of comparing the transmitted reference signal with the reflection of the same signal - thereby calculating S₁I, see reference [3]. The adaptation circuit 31 generates a weak signal which is transmitted to the receiver antenna 38 and is reflected by the antenna elements 32 of the receive antenna 38, and the reflection coefficient may be determined. The weak signal generated by the adaptation circuit 31 is preferably selected to be transmitted when the receiver R_x is not receiving any signals from the transmitter T_x, or the magnitude of the weak signal is much lower than the magnitude of the received signal to avoid distortion of the received signal and/ or interference towards other units communicating on the same frequencies. In a time-division duplex (TDD) unit, measurements of the reflection coefficient(s) may be done during the time slot at which the unit is transmitting provided the receiver chain and transmitter chain are calibrated. The unit could then make use of the transmitted data as reference without the need for a special reference signal to be transmitted. The adaptation circuit 31 is, in this embodiment, arranged in the receiver unit 42, such as a base station or a mobile unit, and information regarding channel condition is signaled to the transmitter T_x in order to adapt the transmission parameters accordingly. This signaling would typically require much less frequent signaling compared to full CSI feedback. The arrangement to measure the antenna reflection coeffϊcient(s) at the receiving unit may be useful in the case that the receiver should signal its preference for e.g. a certain frequency band (subcarrier allocation in OFDM) but does not yet have any received data upon which it can estimate the channel conditions. Such a situation could occur e.g. during random access or when pilot symbols are not transmitted across the entire available frequency band. By measuring the reflection coefficients at the receiver, the receiving unit can predict what frequencies that would be more likely to support good channel conditions. Most importantly, the use of frequencies where the receiving antenna currently is poorly matched can be avoided.

The invention will be illustrated using an OFDM (Orthogonal Frequency Division Muliplex) system since the system operates in the frequency domain. However, the invention is not limited to OFDM system and may be implemented in other telecommunication systems, such as WCDMA.

Figure 4 shows a plot of reflection coefficients S_n in relative power [dB] as a function of subcarrier index [n] of an OFDM system. A measured reflection coefficient Sf₁ , for a transmitter chain as described in connection with figure 2 is illustrated by a thin continuous line, and a measured reflection coefficient S_n for a receiver chain as described in connection with figure 3 is illustrated by a dotted line. A calculated combined reflection coefficient

S₁*^*1* |² = l - (l-|s£|²)fl-|δ£|²] is also indicated by a fat continuous line. The

Relative power in figure 4 is in relation to transmitted power from the transmitter for Sf₁ , and in relation to incoming power to the receiver antenna for S* .

Figure 5 shows a power allocation graph illustrating transmission parameter adaptations for an OFDM system having 64 sub-carriers, wherein each bar represents a sub-carrier. The solid part of each bar represents the inverted transfer function of both the transmitter chain and the receiver chain [H_1x * H_n. ) . Note that the transfer function of the radio channel is omitted. A uniform power distribution has been applied according to prior art, as mentioned above. The uniform power allocation is illustrated as Tx power in the graph on top of the solid part of each bar.

As an example, the power allocation on subcarrier n in an OFDM system with the known transmission function H(f) can be calculated as:

where λ is chosen such that

N-X

∑P{n) = P_tot , (3) o

where Pm is the total transmit power of the transmitter T_x and N is the number of subcarriers of the OFDM system. An illustration of the waterfilling concept is presented in figures 6a-6c.

Figure 6a shows a graph illustrating power allocation in an OFDM system according to the present invention, wherein the transfer function of the transmitter chain H_1x and the transfer function of the receiver chain H_n. are both known to an adaptation circuit, such as the system described in connection with figure 11.

The solid parts of each bar represent the inverted transfer function of both the transmitter chain and the receiver chain [H_1x * H_n. j as described in connection with figure 5. "Waterfilling" has been applied to allocate Tx power to the sub-carriers n. The frequencies represented in the sub-carriers arranged between 28 and 38 are predicted to have the best conditions for the transmission, and thus most transmit power has been allocated to these sub-carriers. Frequencies represented in sub-carriers arranged below 10 and above 55 are predicted to have the worst conditions for the transmission, and therefore no transmit power has been allocated to these sub-carriers.

If the transfer function of the receiver chain is not known, an adaptation based on the transmitter chain may be performed. Figure 6b shows a graph illustrating power allocation in an OFDM system according to the present invention, wherein only the transfer function of the transmitter chain H_1x is known to an adaptation circuit, such as the system described in connection with figure 2.

The solid parts of each bar represent the inverted transfer function of the transmitter chain [H_1x) . "Waterfilling" has been applied to allocate Tx power to the sub-carriers n based on only the transmitter transfer function. The frequencies represented in the sub-carriers arranged between 33 and 41 are predicted to have the best conditions for the transmission, and thus most transmit power has been allocated to these sub-carriers. Frequencies represented in sub-carriers arranged below 14 and above 59 are predicted to have the worst conditions for the transmission, and therefore no (or very little) transmit power has been allocated to these sub-carriers.

Figure 6c illustrates the impact of the transmit power allocation determined in figure 6b, based on the transfer function of the transmitter chain, in relation to the inverted transfer function of both the transmitter chain and the receiver chain [H_1x * H_n ) . A comparison between the prior art power distribution presented in figure 5 and figure 6c indicate that a major part of the available transmit power in the transmitter is allocated to sub-carriers having low reflection coefficients. However, it should be noted that the power allocation is not as good as the power allocation illustrated in figure 6a, since knowledge of the receiver reflection coefficient(s) will provide an even more improved power allocation compared to prior art (figure 5) .

For systems with multiple transmit antennas the average transmit coefficient can be calculated for each transmit antenna. This mean value can then be used to perform waterfilling across the transmit antennas provided a first data stream is supported over a first antenna and orthogonal to a second data stream on a second antenna or for example choose which antenna to transmit on if transmit selection diversity is used. As an example the mean transfer function for transmit antenna number one Htxi in an OFDM system can be calculated as

If Sij is not measured as a function of frequency but rather as the mean value over the whole frequency band this part of the invention is still applicable.

Fig. 7a shows a transmitter chain 40 of the communication channel provided with a third embodiment of the invention implemented in the transmitter chain. Antenna ports of two transmit antennas 47i and 472, each having a single antenna element 22 are connected to a transmitter T_x using a distribution network. A signal source 11 is connected to the transmitter T_x and directional couplers 45 are used to determine a reflection coefficient together with an adaptation circuit 41 for each transmit antenna 471 and 472 (commonly denoted as 47). The adaptation circuit 41 calculates, or measures, in this embodiment the mean value of the reflection coefficient for each antenna as mentioned above. This results in a calculated mean transfer function for each antenna. Information regarding the reflection coefficients and/ or transfer functions are used to control the transmitter T_x to generate the desired transmit power allocation.

Fig. 7b shows a graph illustrating power allocation between transmit antennas 47i and 472 for the transmitter chain 40 in figure 7a. The inverse mean transfer function for each antenna is illustrated by the solid part of the bars, and waterfilling over the antennas results in the power allocation as indicated provided the datastream on each antenna is orthogonal against each other. The major portion of the transmit power is directed to antenna 2. It should be noted that the adaptation circuit could be provided with means to select which antenna to transmit on in dependency of the measured reflection coefficient. In this example, antenna 2 should be selected and antenna 1 is not used until the measured reflection coefficients for the antennas indicate better transmission properties for antenna 1.

If the reflection coefficient for each antenna in figure 7a is measured as a function of frequency, the mean transfer function is calculated using equation (4). The power allocated to each antenna (as indicated in figure 7b) may be allocated across the frequency range for each antenna. This is illustrated in figures 8a and 8b, which show allocation of power for each transmit antenna in figure 7a as a function of frequency. Waterfilling across the frequency is used in combination with waterfilling across the antennas.

Fig. 9a shows a transmitter chain 50 of the communication channel provided with a fourth embodiment of the invention implemented in the transmitter chain. A transmitter T_x, connected to a signal source 11 supply signals to three antenna ports of an antenna 57 comprising three antenna elements 52i, 522, 523, commonly denoted 52, each connected to one of the three antenna ports. A directional coupler 55 is used to determine the reflection coefficient for each antenna element together with an adaptation circuit 51. The adaptation circuit 51 measures the reflection coefficient for each antenna element 52, either as a mean value or as a function of frequency, and calculates a transfer function for each antenna element. Information regarding reflection coefficient and/ or transfer function is used to control the transmitter T_x to adapt the transmission parameters, such as allocate the transmit power.

Fig. 9b shows a graph illustrating power allocation between antennas elements 52 for the transmit antenna 57 in figure 9a. Waterfilling across the antenna elements has been applied, but it is naturally possible to select only one or two of the antenna elements to transmit the signal from the signal source. Fig. 10 shows a telecommunication system according to the present invention. A base station 60 including an antenna tower 61 provided with a base station antenna arrangement 62, and base station equipment BSE, which includes transmitter, receiver, and adaptation circuit as described in the preferred embodiments. A mobile unit 63 is within the coverage area of the base station and communicates with the BSE via the base station antenna arrangement 62 as indicated by 64. Furthermore, an optional signal 65 may be transmitted from the mobile unit 63 to the BSE if the mobile unit is equipped with the resources to measure the reflection coefficient of the receive chain as described in connection with figures 3 and 11.

Fig. 11 shows a communication channel provided with a fifth embodiment of the invention implemented in the transmitter chain 70 and in the receiver chain 80.

The transmitter chain 70 comprises in this embodiment a transmitter T_x, connected to a signal source 11 , and feeding signals to antenna ports of two antennas 771 and 772 (commonly denoted 77) through a distribution network 76. Each antenna is provided with five antenna elements 22. Directional couplers 751 and 752 are used together with an adaptation circuit 71 to determine the reflection coefficient Sn for each antenna 771, 772, each having an individual transmit chain. The adaptation circuit 71 may be configured to calculate a signal indicative of suitable transmit power allocation, beamforming weights, modulation, coding, etc which is forwarded to the transmitter, as indicated by connection 72 based on the determined reflection coefficient for the transmit chain 70 and the determined reflection coefficient for the receive chain 80.

The receiver chain 80 comprises in this embodiment a receiver R_x that receives signals from antenna ports of two receive antennas 881 and 882 (commonly denoted 88) through a distribution network 89. Each antenna is provided with three antenna elements 22. Directional couplers 831 and 832 are used together with a circuit 81 to determine the reflection coefficient(s) for each receiver chain. The circuit may be configured to calculate the transfer function(s) H_n. based on the determined reflection coefficient(s) and thereafter transmit information regarding channel condition back to the adaptation circuit 71 in a suitable way, e.g. wireless signaling over the radio channel. The receiver is connected to devices in a receiving unit 92.

As an example, the effect of waterfilling of the available transmit power is illustrated in connection with figure 6a as described earlier. Other types of suitable transmit power allocation, beamforming weights, antenna selection, modulation and coding may be performed as a result of the calculated transfer function for both the transmitter chain 70 and the receiver chain 80.

The transmitter chain described in connection with figures 2, 7a, 9a and 11 may be implemented in a base station and/ or a mobile unit in a communication system as illustrated in connection with figure 10. The receiver chain described in connection with figures 3 and 11 may also be implemented in a base station and/ or a mobile unit in a communication system as illustrated in connection with figure 10.

The needed update rate for the Su measurements is different for a mobile unit and a base station. At the base station Sn is not likely changed at a high rate and therefore the measurements can be updated at a slow rate.

This is because the connections to the antenna and the environment around the antenna are almost static. At the mobile unit, on the other hand, Su change rather fast as the user interacts with the antenna. Therefore Sn measurements have to be updated at a higher rate. A typical update rate at the mobile unit is once per second, or higher.

Calibration

In a system with multiple transmit antennas, such as described in connection with figures 7a and 11, it is possible to perform antenna calibration with or without the adaptation of the transmission parameters based on the measured reflection coefficient Sn as described earlier. The antenna calibration is performed by compensating for the different time delay in the available transmitter chains, in order to assure that the signal is transmitted simultaneously from all antennas, and also to ensure that the amplitude and the phase on each antenna is equal or at least known. If the system employs beamforming or some other form of precoding it is sometimes necessary to have calibrated antennas. The present invention provides means to perform antenna calibration without any feedback from the receiver.

The different time delays are determined by measuring the propagation time from the antenna port to the actual antenna, i.e. the delay introduced by antenna feeders etc. This time delay may be deduced from the Su(f) measure by e.g. performing an inverse Fourier transform of S₁₁(I] giving an equivalent impulse response sn(τ). The time delay i_peak is visible as a peak in sn(τ) that will correspond to the propagation delay from the transmitter Tx to the reflection point at the antenna and back to the adaptation circuit 71. From this delay the time delay from the transmitter to the antenna may be determined by dividing τ_peak by 2. The measurement is performed on the individual transmitter chain for each antenna, as described in more detail below.

In figure 7a, each transmit antenna 47i and 472 receives a signal from the transmitter T_x via an individual transmitter chain 48i and 482, respectively. The adaptation circuit 41 may be configured to only measure the propagation time for each transmitter chain 48i and 482, or it may be configured to measure the propagation time together with the functionality to determine the reflection coefficients of each transmit antenna 471 and 472.

In figure 11, each transmit antenna 771 and 772 receives a signal from the transmitter T_x via an individual transmitter chain 781 and 782, respectively. The adaptation circuit 71 may be configured to only measure the propagation time for each transmitter chain 78i and 782, and also be configured to receive information regarding reflection coefficient(s) and /or channel condition from the circuit 81 in the receiver chain 80 to adapt the transmission parameters only based on the determined transfer function H_n of the receiver chain. However, it is also possible to configure the adaptation circuit 71 to measure the propagation time together with the functionality to determine the reflection coefficients of each transmit antenna 771 and 772.

The main advantage of the invention is the possibility to adapt the transmission parameters, such as antenna selection, power allocation, beamforming weights, modulation and coding rate, without any CSI at the transmitter. The adaptation will be suboptimal but nevertheless provide an improvement over the traditional equal power/ modulation/ coding rate allocation. In a handheld unit the antenna gain on different antennas and on different frequencies will change as the user interacts with the antennas. With this invention these effects are taken into consideration in the waterfilling solution.

For systems with multiple transmit antennas it is possible to adapt the transmission parameters across the transmit antennas without CSI at the transmitter. This is particularly useful at the mobile station since one can avoid transmitting on an antenna that is attenuated by the user. As the interaction between a user and the mobile station antenna easily can result in more than 1OdB attenuation significant gains can be achieved.

If the transfer function of the receive chain is not available to the adaptation circuit in the transmitter unit when transmission parameters are adapted, a default transfer function H%^fau" of the receive chain may be used in combination with the transfer function H_1x of the transmitter chain. The default transfer function is preferably stored in the adaptation circuit and is preferably established based on a number of measured reflection coefficients from standard receiver units. This is most useful when the variations in reflection characteristics among different units are expected to be limited. It is even possible to implement the present invention in a system using CSI to adapt the transmission parameters. The information regarding the transfer function of the transmitter chain may be used in the time period between the updated CSI is received by the transmitter unit since the changes of the transfer function in the transmitter normally are faster than the CSI has a possibility to forward to the transmitter.

The described embodiments have illustrated the invention to emphasize certain aspects, and it should be noted that it is obvious for a skilled person in the art to combine them to obtain a desired functionality.

The relative power used on the y-axis in figures 5, 6a-6c, 7b, 8a-8b,and 9b should be considered to be in relation to a fictitious power level used to illustrate the relative power levels between sub-carriers, antennas, or frequencies.

Abbreviations

CSI Channel State Information

Htx, Hrx Transfer function for Tx chain and Rx chain

OFDM Orthogonal Frequency Division Multiplex

R_x Receiver

Si₁ Reflection Coefficient T_x Transmitter

WCDMA Wideband Code Division Multiple Access

References

1 3GGP Rl-071048, "The Need for Measurement Report Mechanism

Supporting NodeB RF Front End Calibration", Ericsson.

2. 3GGP R 1-071602, "Absence of Array Calibration - Impact on

Precoding Performance", Ericsson. 3. Agilent AN 1287-2, "Exploring the Architectures of Network Analyzers", Agilent Technologies.

Claims

1. A method for adapting transmission parameters in a transmitter (T_x) in communication with at least one antenna (27; 38; 47; 57; 77, 88), said method comprising: - transmitting a signal from the transmitter (T_x), and measuring at least one reflection coefficient (Si₁) of said signal for each antenna (27; 38; 47; 57; 77, 88), c h a r a c t e r i z e d i n that said method further comprises: adapting the transmission parameters based on the measured reflection coefficient (Sn).

2. The method according to claim 1, wherein the adaptation of the transmission parameter comprises any combination of: antenna selection, power allocation, beamforming weights, modulation, or coding.

3. The method according to claim 2, wherein the power allocation is performed in the frequency domain.

4. The method according to claim 2 or 3, wherein the power allocation is performed between different transmit antennas (47; 77) in a multi antenna system.

5. The method according to any of claims 1-4, wherein said reflection coefficient is measured as a function of frequency.

6. The method according claim 5, wherein the at least one antenna is selected to be at least one transmit antenna (27; 47; 57; 77) to which said transmitter (T_x) is connected, said reflection coefficient is measured for each transmit antenna (27; 47; 57; 77) and the adaptation of transmission parameters comprises waterfilling across the at least one transmit antenna (27; 47; 57; 77) to calculate power allocation in frequency.

7. The method according to any of claims 1-6, wherein the at least one antenna is selected to be multiple transmit antennas (47; 77) to which said transmitter (T_x) is connected, and said reflection coefficient for each transmit antenna (47; 77) is determined as an average value over a frequency band.

8. The method according to claim 7, wherein each transmit antenna having one or more antenna elements (22), the average value is calculated from the reflection coefficients (Sn) measured for each antenna element (22).

9. The method according to claim 7, wherein the average value is measured for each transmit antenna (47).

10. The method according to any of claims 7-9, wherein the adaptation of transmission parameters comprises waterfilling across the transmit antennas (47).

11. The method according to any of claims 6-9, wherein transmit selection diversity is used and the adaptation of the transmission parameters comprises choosing which transmit antenna (47) to transmit on.

12. The method according to claim 5, wherein the at least one antenna is selected to be at least one receive antenna (38; 88) in a receiving unit (42;

92), said method further comprises: predicting channel conditions from the transmitter to the receiver unit based on measured reflection coefficient s) in the receiving unit (42; 92), and - signaling information regarding reflection coefficients to the transmitter (T_x) in order to adapt the transmission parameters.

13. The method according to claim 12, wherein said information regarding channel conditions comprises frequencies suitable for transmission and/ or frequencies unsuitable for transmission.

14. The method according to any of claims 12 or 13, wherein the reflection coefficient(s) is /are measured relative a signal generated in the adaptation unit (31; 81) and transmitted to the receive antenna (38; 88).

15 The method according to any of claims 12 or 13, wherein the receiver unit is a time-division duplex (TDD) unit, and the reflection coefficient(s) is/ are measured during a time slot at which the TDD unit is transmitting.

16. The method according to any of claims 1-11, wherein the at least one antenna is selected to be multiple transmit antennas (47; 77) to which said transmitter (T_x) is connected, said method further comprises: determining the time delay for each individual transmitter chain (481,

482; 78i, 782) without any information from a receiving unit (92), and compensating for differences in time delay between the individual transmitter chains (48i, 482; 78]., 782) to assure that the signal is transmitted simultaneously from all transmit antennas (47; 77).

17. The method according to claim 16, wherein the time delay for each individual transmitter chain (481, 482; 781, 782) is obtained by: measuring propagation time in each individual transmitter chain (48i, 482; 78i, 782) from antenna port in the transmitter (T_x) to each transmit antenna (47; 77), and calculating time delay to compensate for differences in time delay between the individual transmitter chains (48i, 482; 78i, 782).

18. The method according to any of claims 1-17, wherein said method further comprises updating the measured reflection coefficient(s) at regular intervals.

19. The method according to claim 18, wherein the updating rate is higher when said transmitter (T_x) and said at least one antenna are arranged in a mobile unit compared to when arranged in a base station.

20. The method according to claim 19, wherein the updating rate is once per second, or higher, for a mobile unit.

21. A method for calibrating a transmit system having multiple transmit antennas (47; 77) connected to a transmitter (T_x), said method comprises: transmitting a signal from the transmitter (T_x) to each transmit antenna (47; 77) via an individual transmitter chain (48i, 482; 78i, 782), c h a r a c t e r i z e d i n that said method further comprises: determining the time delay for each individual transmitter chain without any information from a receiving unit (92), and compensating for differences in time delay between the individual transmitter chains (481, 482; 781, 782) to assure that the signal is transmitted simultaneously from all transmit antennas (47; 77).

22. The method according to claim 20, wherein the determination of the time delay comprises: measuring propagation time for each individual transmitter chain (48i, 48₂; 78i, 78₂), and calculating time delay for each individual transmitter chain based on the measured propagation time.

23. The method according to any of claims 21-22, wherein said transmit system is selected to be arranged in a base station and/ or in a mobile unit in a wireless communication system.

24. A node comprising at least one antenna, said node being configured to measure at least one reflection coefficient of a signal for each antenna, said signal being transmitted from a transmitter (T_x), c h a r a c t e r i z e d i n that said node further is provided with a means (21; 31; 41; 51; 61; 71) to adapt the transmission parameters in said transmitter (T_x) according to any of claims 1-5.

25. The node according to claim 24, wherein the at least one antenna comprises at least one transmit antenna (27; 47; 57; 67; 77), said node comprises the transmitter (T_x) being connected to said at least one transmit antenna, said node further being provided with means (21; 41; 51; 61; 71, 81) to adapt the transmission parameters in said transmitter (T_x) according to any of claims 6-11.

26. The node according to claim 24, wherein the at least one antenna comprises at least one receive antenna (38; 68), said node is a receiving unit (42; 92) provided with means (31; 81) to predict channel conditions and to signal information regarding channel conditions to the transmitter according to any of claims 12-15.

27. The node according to any of claims 24-25, wherein the transmitter (T_x) is connected to multiple transmit antennas (47; 77), said node further comprises means (41) to measure propagation time in each individual transmitter chain (48i, 482; 78_1} 782), and means to calculate time delay to compensate for difference in time delay between the individual transmitter chains (48_1? 48₂; 78i, 78₂).

28. The node according to any of claims 24-27, wherein said node is provided with means to regularly update the measured reflection coefficient.

29. The node according to any of claims 24-28, wherein the node is a base station (60).

30. The node according to any of claims 24-28, wherein the node is a mobile unit (63).

31. A wireless communication system comprising at least one node according to any of claims 24-30.

32. A node comprising a transmitter (T_x) connected to multiple transmit antennas (47; 77), said node is configured to transmit a signal from the transmitter (T_x) to each transmit antenna (47; 77) via an individual transmitter chain (48i, 482; 78i, 782), c h a r a c t e r i z e d i n that said node further is provided with: - means (41; 71) to determine the time delay for each individual transmitter chain (481, 482; 781, 782) without any information from a receiving unit (92), and means to compensate for differences in time delay between the individual transmitter chains (48i, 482; 78i, 782) to assure that the signal is transmitted simultaneously from all transmit antennas (47; 77).

33. The node according to claim 31, wherein said means to determine the time delay comprises: means to measure propagation time for each individual transmitter chain (481, 482; 781, 782) from antenna port in the transmitter (T_x) to each transmit antenna (47; 77), and means to calculate time delay for each individual transmitter chain (48i, 482; 78i, 782) based on the measured propagation time.

34. The node according to any of claims 32 or 33, wherein said node is a base station (60) and / or a mobile unit (63) in a wireless communication system.

35. A wireless communication system comprising at least one node according to any of claims 32-34.