Calibration method for an adaptive antenna system

SWEDEN (12) PATENT (13) C2 in) 522 563 (19) SE ₍₅₁₎

International class ⁷

H04B 7/04

<img file="SE522563C2_D0001.tif" />

PATENT AND REGISTRATION (45) (41) (22) (24) (62) (66) (86) (83)

Patent filed Application widely available The patent application was submitted on expiration date

National application number International filing date Filing date for European patent application Deposit of microorganism

2004-02-17

2001-08-02

2000-02-01

2000-02-01 (21) Patent Application Number 0000316-0

Application received as:

eg Swedish patent application completed international patent application with number □ converted European patent application with number (30) Priority information (73) (72) (74) (54) (56) (57)

PATENT HOLDER Telefonaktiebolaget LM Ericsson (publ),

126 25 Stockholm SE

INVENTOR Leonard Rexberg, Hässelby SE, Magnus Johansson, Uppsala SE, Eberhard Pauck, Hässelby SE, Karsten Kruger, Arden DK

OMBUD Aros Patent AB

NAME Calibration method for an adaptive group antenna

CALLED PUBLICATIONS: - - SUMMARY:

A method and system are shown to obtain calibration with a minimum of hardware realization. The method and system are applied to base stations having a realization with a fully adaptive antenna. The calibration takes into account irregularities from the source generator via channels, upconverter mixers, power amplifiers and filters. A proposed illustrative embodiment of the set-up and routine provides both transmission path and reception path calibration. An arrangement according to the invention uses the internal radio base station radio parts by sequentially switching signal paths from transmitter to receiver and thus does not require any external equipment. The necessary hardware switching is then combined with the duplex filters conveniently located near the base station. The process can be easily visualized by means of a basic flow chart illustrating the three main steps which are a first step of receiving calibration that is repeated for all available reception frequencies, a second step of pre-calibration on transmission, and a third step of transmission calibration that is repeated for all available transmission frequencies.

<img file="SE522563C2_D0002.tif" />

The numbers in brackets indicate international identification code, INID code. The letter within the pinch indicates the internal document code.

522 563

SUMMARY

A method and system are shown to obtain calibration with a minimum of hardware realization. The method and system are applied to base stations having a realization with a fully adaptive antenna. The calibration takes into account irregularities from the source generator via channels, upconverter mixers, power amplifiers and filters. A proposed illustrative embodiment of the set-up and routine provides both transmission path and reception path calibration. An arrangement in accordance to the invention uses the internal radio base station radio parts by sequentially switching signal paths from transmitter to receiver and thus does not require any external equipment. The necessary hardware switching is then combined with the duplex filters conveniently located near the base station. The process can be easily visualized by means of a basic flow chart illustrating the three main steps which are a first step of receiving calibration that is repeated for all available reception frequencies, a second step of pre-calibration on transmission, and a third step of transmission calibration that is repeated for all available transmission frequencies.

(Fig. 11)

522,563 DDV ηΐ.; ··; · Μϋ

TECHNICAL FIELD

The present invention relates to the calibration of transmission and reception paths for an adaptive array antenna comprising multiple antenna branches in a radio base station arrangement.

BACKGROUND

Adaptive antennas have shown impressive potential for increasing the capacity of cellular networks. Extensive theoretical work as well as field trials substantiate this claim. When using adaptive antenna beams, it is desirable to control the beams of the antenna continuously and at the same time independently control the zero points. Control is achieved by adjusting the relative phase and amplitude between the antenna branches in such a way that constructive or destructive interference will occur for beam direction and zero position, respectively. That is, in contrast to a switched beam realization where the best beam is selected, such a solution has to do with actual realization with continuous tracking and active zeroing of interference. This calls for an accurate phase and amplitude adjustment for all hardware and software including the signal path from the source to the antenna connection.

Today, the capacity of a cellular network is limited by the ability to withstand channel interference. A certain level of protection for the undesired signal power received by the mobile phone must be maintained below the desired signal power. This is ensured by keeping base stations transmitting at the same frequency at a certain minimum geometric distance from each other. In this way, other base stations broadcasting at the same frequency will not interfere with each other's mobiles. On the other hand, base stations transmitting within other frequency groups can be placed in an interleaved manner between these first mentioned base stations. This approach can be repeated until all available frequencies are occupied, but in such a way that they do not interfere with each other. The distances between the base stations

522₂563 pDU ΠΙ-Ηΐ-Ιϋ;

and the total number of frequencies available results in a performance value for the capacity of a cellular network.

It is obvious that if the power from interfering base stations received by mobiles (and also received power at the base station) could be reduced in any other way than physically relocating the base stations, capacity could be increased in the network. Frequencies could then be reused more often in a given physical area to serve more subscribers in this way. Limitation of transmitted power can be achieved by means of an antenna having a narrower aperture angle. In addition to this, the antenna beam must be able to continuously monitor the mobile in its working sector. There are mainly two realizations: the switched beam solution and the fully adaptive solution. In both cases, a group of antenna elements is used, but in the latter case each element can be individually controlled, while in the former case only one of a plurality of beams can be selected. In the former case, preset beams are used, while in the latter a single beam is used, but which continuously follows the mobile. This is referred to in this document as a fully adaptive antenna.

A necessary condition is that each antenna path is calibrated so that possible irregularities can be taken into account and compensated for. Both uplink and downlink must normally be calibrated. However, the problem is more pronounced in the downlink direction (base station to mobile) while in the uplink (mobile to base station), steel forming algorithms can be made to auto-correct for phase errors.

The most widely used realization of adaptive antennas is to use switched beams. The beam former is conveniently placed at the top of the mast near the antenna array, and does not entail any major phase errors. A connection at the input of the beam former provides the proper excitation to all the antenna elements simultaneously. This gives a ray in a certain direction. Selecting another input connection gives a beam in yet another direction. So depending on, for example, the strongest received signal, the correct beam is selected and in this way no calibration is needed since only

522₃ 563 MQ means a single RF path to the antenna. However, the above switched beam realization does not give the freedom of controlling the beam to any position, nor the zeroing of interfering signals.

The most obvious solution for calibrating an antenna system would be to input a signal at a suitable location (in the transmission path) and then follow the signal as it propagates through the system. The signal must be detected at some point close to the antenna and then compared with the input signal. At first glance, this needs additional hardware and detectors to be realized. Usually, an external source is applied to the equipment and input to output signals are then compared. The difference between the two will then provide the necessary phase and amplitude to be compensated for. The procedure in the receiving direction would be similar to the case of the transmission. It is obvious that the approach needs additional generator and detector equipment and it is difficult to do such calibration when the base station is in operation.

Usually, calibration is done on the laboratory bench prior to installation, or by calibrating each part of the RF chain separately. Calibration of a group antenna only at the factory is not satisfactory with regard to use at one location. Calibration must be done during operation in some special way and monitored continuously. This is not normally fully supported today. If installed, one way would be to use an external separate generator to measure the signal path both in transmission branches and reception branches. This requires additional receivers tuned to the correct frequency along with some software realization to calculate amplitude and phase deviations. Furthermore, signal input and output points must be realized in the base station.

Consequently, today the calibration of group antennas is a problem without a proper solution. Therefore, there is a need for a procedure to calibrate on-site group antennas with a minimum of hardware requirements and this applies to base stations that have realization with fully adaptive antenna.

522 563 - ::: - Γ: ·:.: ··: ·:. ': O:' ··· ': · / -

SUMMATION

The present presentation proposes a method and system for obtaining calibration with a minimum of hardware realization and applies to base stations which have realization with fully adaptive antenna. It takes into account irregularities from the generator source via channels, upconverters, mixers, power amplifiers and filters. A proposed illustrative embodiment of the arrangement and routine provides both calibration at the transmission path and the reception path. The arrangement according to the invention uses the radio parts of the internal base station by sequentially switching signal paths from transmitter to receiver and thus does not need any external equipment. The necessary switches are then combined with the duplex filters suitably near the base station. The method can simply be visualized through a basic flow chart showing the three main steps, which is a first step with a reception calibration repeated for all available reception frequencies, a second step with pre-transmission calibration, and a third step with a transmission calibration repeated for all available transmission frequencies.

A method according to the present invention is determined by the independent claim 1 and the dependent claims 2-4. A system for calibrating an apadaptive array antenna is established by the independent claim 5 and further embodiments are determined by the dependent claims 6 to 9.

BRIEF DESCRIPTION OF THE DRAWINGS The invention, together with further objects and advantages thereof, can best be understood by reference to the following description, together with the accompanying drawings, in which:

FIG. 1 is a concentrated sketch of a base station with four RF antenna branches and includes a transmitter, a receiver and a duplex filter;

522 £63

FIG. 2 illustrates an RX calibration wherein the receiver branches are fed through one of the transmitters tuned to receive frequency (RX);

FIG. 3 shows a signal path during RX calibration;

FIG. 4 shows the RX calibration block diagram including a signal correlation matrix calculation;

FIG. 5 illustrates a pre-calibration for TX which achieves calibration of reception branches but at a transmission frequency with a transmitter tuned to transmission frequency and the signal divided into all receiver branches;

FIG. 6 illustrates the signal path at TX pre-calibration with duplex files bypassed in this operation when the receiver branches are calibrated at TX frequency to obtain values ​​to subtract from the final transmit calibration results;

FIG. 7 illustrates TX calibration when each transmitter is tuned to transmit frequency and fed back to its respective antenna receiver branch;

FIG. 8 shows the signal path during TX calibration when the transmit frequency signal is feedback from the transmitter to each receiver,

FIG. 9 shows the layout of the TX calibration block diagram for TX calibration with each transmitter feedback to its respective radio branch and the covariance matrix for calculating the relative phase and amplitude of the branches indicated;

FIG. 10 shows signal correlation from all antenna branches compiled in a matrix for storage, as well

522 563 H in Q

FIG. 11 is a basic flow chart illustrating the present method for calibrating an adaptive array antenna.

DETAILED DESCRIPTION

In accordance with the present invention, the proposed solution to the calibration problem is to use existing transmitters and receivers in the base station to perform calibration. The transmitters will be fed back to the receivers, thus eliminating the need for extra hardware. Assuming that each antenna branch has a realization with a complete set of hardware equipment (Figure 1), calibration is performed in three steps.

Step 1: First, the calibration of the receiver branch is performed by tuning one of the transmitters to a reception frequency and the transmitting signal supplied at equal power by a power divider to all the receiving branches which in turn is terminated by reception channels. Comparisons of received signals on a channel-to-channel basis provide the sought differences between reception branches at a reception frequency. (See Figures 2 to 4). The transmission portion of the DPX duplex filters is bypassed in this step.

Step 2: A pre-calibration at transmission is performed in which step 1 is repeated, but now with the transmitters tuned to an ordinary transmission frequency. Received signals in the receivers are compared, providing calibration data for the receiving branches at transmission frequency. This will be needed for calibration of the transmission branches. (See Figures 5 and 6). The DPX duplex filters are all bypassed in this step.

Step 3: Transmission branch calibration is finally performed by connecting each transmission branch to one of the receiving branches on a one-to-one basis. Comparison of the signals received by the receivers (at transmission frequency) provides calibration data for the transmission branches. Note that the reception branches have already been calibrated for transmission frequency in step 2. (See Figures 7 to 9). The DPX receiving portion of the duplex filter is bypassed in this step.

522 563

However, it should be noted that in this calibration, antenna supply cables between duplex filters, DPX, and antenna elements are not included in the present calibration set discussed.

Calculation of calibration data is generally done using a main building block for adaptive antenna algorithms, the covariance matrix (See also Figure 10). A subroutine for the covariance matrix is ​​largely an integral part of all algorithms that have to do with adaptive beamforming. Therefore, a minimum added software realization is necessary for this calibration procedure.

The covariance matrix, which is of the same order of magnitude as the number of antenna elements, consists of signal correlation factors between all antenna branches. Therefore, for example, by dividing the intrinsic correlation for one signal path with the cross-correlation for another signal path, the path difference between these two specific branches will be obtained. Both amplitude and phase difference will be obtained in this way.

Figure 1 shows a concentrated sketch of an illustrative embodiment of a base station comprising four antennas. The four antennas are connected to a receiver RX1-RX4 and transmitter TX1-TX4 via a respective duplex filter, DPX. Further, as well known to those skilled in the art, the receiver / transmitter portions may include other radio equipment such as filters, amplifiers, channels, and mixers.

Figure 2 illustrates a receiver calibration. The present four reception branches are all supplied via a signal divider from one of the transmitters (TX1) tuned to reception frequency. Thus, the receiving portion of each duplex filter is included and will therefore be calibrated. As indicated in Figure 2, the four antennas are conveniently disconnected from the duplex filter DPX during the RX path calibration sequence to avoid external signals being detected by the receivers. The arrangement of the receiver 52¾ 563 path calibration is further illustrated in Figure 3, which illustrates in more detail the signal path. In Figure 3, XSW indicates a transmitter switch, RSW indicates a receiver switch, DPX is as before the duplex filter and ASW indicates an antenna switch.

Figure 4 illustrates calibration of the RX path using a simple block diagram indicating a controller CU used to provide receivers RX1 to RXn with signal from a selected transmitter, in this case TX1. From the received calibration signals, a covariance matrix is ​​calculated.

Pre-calibration at transmission is then illustrated in Figure 5. This allows calibration of the receiving branches but at transmission frequency. A transmitter, TX1, is tuned to transmit frequency and the signal is shared to all receiving branches to obtain a calibration of the receiver branches to be used for subtraction from the final transmit calibration results. Note that all duplex filters DPX are bypassed in this operation. As indicated in Figure 5, the four antennas are conveniently disconnected from the duplex filter DPX even during the pre-calibration sequence of TX to avoid strong external signals being detected by the receivers. The arrangement for pre-calibration for transmission is further illustrated in Figure 6, which illustrates in more detail the signal path.

Furthermore, Figure 7 illustrates transmission calibration when each transmitter TX1 - TX4 is tuned to transmission frequency and fed back to its respective antenna limit receiver RX1 - RX4. The total signal path is calibrated and transmit path calibration is performed using the pre-calibration transmission data for the receive threshold portion. As also indicated in Figure 7, the four antennas are conveniently disconnected from the duplex filter DPX during the TX path calibration sequence to avoid strong external signals being detected by the receivers. Like Figures 3 and 6, Figure 8 illustrates in more detail the signal path during transmission calibration. The signal on the transmission frequency is fed back from the transmitter to each receiver. The final transmit branch calibration is obtained as the residual value for amplitude and phase after

522_g 563 Η Η'Π. ΡΊ subtraction of the pre-calibration at transmission received from the receiver branch (at transmission frequency) in the pre-calibration step 2 discussed at transmission.

Similar to Figure 4, Figure 9 illustrates a schematic block diagram of the TX calibration in step 3. Each transmitter is fed back to its respective receiving radio branch. The covariance matrix is ​​indicated for calculating relative phase and amplitude of branches.

Correlation factors between signals from all antenna branches are compiled into a storage matrix as illustrated in Figure 10. The correlation procedure provides a certain degree of mean value formation and noise reduction for more accurate results. Calculation of correlation between signals is briefly summarized below.

The main steps of the present calibration method according to the invention are also indicated in the simplified flow diagram of Figure 11.

Calibration realization for digital signal processor software

In an illustrative embodiment, signal correlation will be recorded for the receiving and transmitting antenna branches and transmitted to the controller (CU) and / or a digital signal processing unit (DSP). It need not be necessary to calculate the actual phase and amplitude differences between branches, but rather to use correlation calculations directly. The way in which this can be realized is described below.

Calibration - correlation

By calculating the correlation between signals, the phase difference between antenna branches is easily detected. The amplitude difference is also detectable, but may need some explanation. Consider two antenna branches RX1 and RX2 with phase and amplitude imbalance as:

RX1 = a, -e^yes‘

RX2 = a₂-e^j“²

522 563 .· .··. .: .··. .: .:

U

The complete correlation matrix between these two signals is calculated as:

p_2] = [RXl · RX2 = a, ä₂ · E^{j (a |}'“²⁾] p "= RX1RXT p₁₂ = P * 21 = RXT-RX2 p₂₂ = RX2-RX2

Kovariansmatrisjustering

The expression above may be associated with the covariance matrix of the group antenna and the signals incident to it at an angle θί. Consider a two-element group antenna with element spacing I, where a (for simplicity's) signal occurs. The covariance matrix with antenna port voltages (VI, V2) can be written:

V! = RXl · e⁷²·’¹^®^

V₂ = RX2 · e^{} 2, tcos (}®'⁾ v₂.y /

V₂-V₂_

RX2 RXT-e^j'²’^{Ilcos (}®ⁱⁿ⁾'

RX2RX2 vy v, -v₂

RXl · RXl

RXl · RX2 · e-^j-²”^{LCOS (}®‘⁾ from which if properly calibrated we will obtain:

Calib

-j-2Klcos (0i) _e j ^ nlcosfa) 1

As can be seen above, calibration could be achieved by multiplying each element by the corresponding correlation factor according to:

522 ^ 63 caiib

RX1RX1

LRX1RX2 ^ 21

RX2RX1

RX2RX2

<img file="SE522563C2_D0003.tif" />

or

<td> ' 1</td><td>c '</td><td> 1</td><td>• C * -Ί2</td><td> 1</td><td>c '</td><td>Arrow p '-k</td>

<td>Pl.</td><td></td><td>PL2</td><td></td><td> 1</td><td></td><td>P12</td>

<td> 1</td><td>C_2J</td><td> 1</td><td> C * -22</td><td>Arrow</td><td>Arrow p '^ 21</td><td>Arrow p ^ 22</td>

<td>_p21</td><td></td><td>P22</td><td></td><td></td><td>_ P₂1</td><td>P22</td>

Accordingly, it is readily apparent that the actual amplitude and phase differences do not necessarily have to be calculated. It is sufficient to calculate the correlation between all the antenna ports for a particular calibration signal. It is only a matter of multiplying each of the matrix elements with the inverse of the correlation matrix elements.

Internal base station calibration

The same procedure as described above can be used for the internal calibration of the base station, and thus only switch between one of the signals (RX1 or RX2) with, for example, TX1 - TX4. Thus, for calibration step 1, we have:

<td>RX1 = kjTX1</td><td>RX1 · ΤΧΪ = k, TXl · TXT</td>

<td>RX2 = k₂TX1 =&gt;</td><td>RX2 TXT = k₂TXl TXT</td>

<td>RX3 = k₃TXL</td><td>RX3 TXT = k₃TXl · ΤΧΪ</td>

<td>RX4 = k₄TXL</td><td>RX4 · ΤΧΪ = k₄TXl · ΤΧΪ</td>

Which gives:

, _r,. ,. . RXI TXT => calibration data k, = -__¹ TX1 TX 1

RX2TX1

ΤΧ1ΤΧΪ

Similar expressions are obtained for step 2 and step 3 of the calibration procedure.

The multiplications in the above expression should be seen as correlation operations instead of just multiplications. The result will be a scalar rather than a vector. TX1 is interpreted as the signal transmitted to everyone

522₁₂563 receiving branches RX1 - RX4. The present notation should be regarded as a correlation operation in the above equation rather than a pure multiplication.

It will be appreciated by those skilled in the art that various modifications and modifications may be made to the present invention without departing from the scope thereof as defined by the appended claims.

522 563 ::: -: 1::: "v \