WO2017118468A1

WO2017118468A1 - Signal processing for active antenna array

Info

Publication number: WO2017118468A1
Application number: PCT/EP2016/050011
Authority: WO
Inventors: Leonard Rexberg; Farshid Ghasemzadeh
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 2016-01-04
Filing date: 2016-01-04
Publication date: 2017-07-13

Abstract

A transmitter arrangement and a corresponding method for an active antenna array are provided, where the active antenna array comprises a plurality of antenna elements being associated with a plurality of active components. The active antenna array is further associated with a beamformer, W, having a number of input and output ports. The method comprises obtaining a plurality, N, of signals, Xn, destined for a respective input port of the beamformer W. The method further comprises deriving a number, P, where P≥N, of linear combinations of the signals Xn, thus obtaining P multiplexed signals. The method further comprises performing CFR on the number, P, of multiplexed signals, thus obtaining P adapted signals. The method further comprises re-establishing, from the number P of adapted signals, the number N of signals, X'n, to be provided to the respective input ports of the beamformer W.

Description

• SIGNAL PROCESSING FOR ACTIVE ANTENNA ARRAY

TECHNICAL FIELD

The invention relates to signal processing in association with active antenna arrays, in particular to so-called massive MIMO. BACKGROUND

An active antenna array has a number of antenna branches comprising active elements or components, such as power amplifiers. These active components are typically associated with certain characteristics or restrictions in regard of their operation. In case of power amplifiers, these are associated e.g. with a peak output power, which is a value typically indicating a maximum possible instantaneous peak amplitude that the power amplifier can support and still stay fairly in its linear range.

In order to reduce the risk for causing exhaustive intermodulation by leaving the linear range of the power amplifiers, and further in order to improve the efficiency of the power amplifier operation, some type of Crest Factor Reduction, CFR, e.g.

clipping functionality, is typically applied to each amplifier/antenna branch. By clipping signal peaks above a certain level before feeding the signal to a power amplifier, the amplifier may operate in its linear range, and be more efficiently used due to the reduction of the Peak to Average Power Ratio of the signal. Typically, filtering is also involved in applying CFR in order to avoid additional spectral emissions coming from the clipping itself. Further, channel filters may be applied in order to maintain unwanted emissions below a required level, such as e.g. below a regulatory requirement.

The existing technology to perform clipping for a manifold of active array antenna branches is to apply a clipping unit (and a channel filter) to each one of the amplifiers in the array. Today, the number of antenna branches is typically about 2-4, or possibly 8 branches. However, the number of active antenna branches in active antenna arrays is expected to increase at a fast pace, far beyond 8 antennas, in the future. It should also be noted that the number of active antenna branches may well, and is expected to, exceed the number of actual MIMO layers or beams being transmitted. Certain aspects of an envisioned use of a multitude of antennas for transmission are sometimes referred to as "Massive MIMO". . When the antenna branches are as few as today, the applying of one clipping unit per power amplifier/antenna branch is a limited problem. However, as the number of antenna branches increases, the large number of clipping units and filters will become a problem, e.g. in terms of cost, complexity and power consumption. SUMMARY

It is desired to find an improved solution for signal processing, such as CFR and channel filtering, for active antenna arrays. This is achieved by embodiments described herein and defined in the appended set of claims. An advantage of embodiments described herein is that signals to a whole array of antenna branches may be processed (CFR, filtering) with a smaller number (N) of processing units than the actual number of active antenna branches (M). Further, digital CFR and channel filtering is enabled also for active antenna array architectures applying purely analog beamforming. Embodiments of the invention are applicable for analog, digital, as well as hybrid beamforming. Being able to achieve a clipped signal in analog beamforming together with active devices in each branch makes it possible to increase the average power into the different Power Amplifiers (PAs) with optimum efficiency or avoid power back-off. This has previously only been possible when having the signals in digital form to each amplifier in the antenna array. Clipping (CFR) has then been performed in each antenna branch in the digital domain. Using embodiments as presented here, each power amplifier in an active antenna array may be run at a high efficiency also in case of analog beamforming.

In addition, applying a limited number (N) of digital filtering chains (one per baseband port or beam) would not only give proper performance in terms of unwanted emission, but also result in significant reduction in the number of digital filtering chains as compared to prior art solutions. Such a reduction will affect the complexity, cost and current consumption associated with active antenna arrays.

According to a first aspect, a method is provided, which is to be performed by a transmitter arrangement. The method is for processing signals for an active antenna array, where the active antenna array comprises a plurality of antenna elements, being associated with a plurality of active components. The active antenna array is further associated with a beamformer, W, having a number of input and output ports. The method comprises obtaining a plurality, N, of signals, Xn, destined for a respective input port of the beamformer W. The method further comprises deriving a number, P, where P≥N, of linear combinations of the signals Xn, thus obtaining P multiplexed signals. The method further comprises performing CFR on the number, P, of multiplexed signals, thus obtaining P adapted signals. The method further comprises re-establishing, from the number P of adapted signals, the number N of signals, X'n, to be provided to the respective input ports of the beamformer W.

According to a second aspect, a transmitter arrangement for an active antenna array is provided, where the active antenna array comprises a plurality of antenna elements, being associated with a plurality of active components. The active antenna array is further associated with a beamformer, W, having a number of input and output ports. The transmitter arrangement is configured to obtain a plurality, N, of signals, Xn, destined for a respective input port of the beamformer W, and further to derive a number, P, where P≥N, of linear combinations of the signals Xn, thus obtaining P multiplexed signals. The transmitter arrangement is further configured to perform CFR on the number, P, of multiplexed signals, thus obtaining P adapted signals; and to re-establish, from the number P of adapted signals, the number N of signals, X'n, to be provided to the respective input ports of the beamformer W. According to a third aspect, a network node is provided, comprising the transmitter arrangement according to the second aspect.

According to a fourth aspect, a computer program is provided, comprising

instructions which, when executed on at least one processor, cause the at least one processor to carry out the method according to the first aspect. According to a fifth aspect, a carrier is provided, containing the computer program of the fourth aspect.

BRIEF DESCRIPTION OF DRAWINGS

The foregoing and other objects, features, and advantages of the technology disclosed herein will be apparent from the following more particular description of embodiments as illustrated in the accompanying drawings. The drawings are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the technology disclosed herein.

Figure 1 illustrates standard placement of functional CFR and channel filtering, according to the prior art. Figure 2. shows a proposed new architecture for CFR and channel filtering

placement and computation, according to an exemplifying embodiment.

Figure 3 is a flow chart showing a method according to an exemplifying embodiment.

Figure 4 shows an architecture according to an exemplifying embodiment.

Figure 5 illustrates equivalent signal addition to give clipped and digitally filtered signals in the analog domain at RF.

Figure 6 illustrates equivalent signal addition to give clipped filtered signals at digital baseband.

Figure 7 illustrates an embodiment where the expansion matrix is based on a beamformer W. Figure 8 is CCDF diagram for 1 ) an actual clipped signal (dashed line) resulting from an exemplifying embodiment, 2) an ideally clipped signal (solid line), and 3) No clipping (dotted line)

Figure 9a illustrates a typical channel filter for a 20 MHz LTE carrier at sampling frequency=30.72 MHz (8x3.84 MHz). Figure 9b illustrates a simplified CFR and channel filter architecture when only one (N=1 ) digital beam port is available. As only the envelope is affected, one-and-for-all clipping may be performed to the one only signal. The clipped and filtered signal is then distributed to all of the active branches with exactly the same statistical signal level distribution. Figures 10a-c are schematic block diagrams illustrating different implementations of a transmitter arrangement according to exemplifying embodiments. DETAILED DESCRIPTION

Today, the approach for signal processing for active antenna arrays is to apply CFR and channel filtering functionality to each of the antenna branches separately, and in digital domain. Therefore, digital beamforming is applied, and each active antenna branch has a DAC(Digital to Analog Converter) associated to it. Considering the expected increase in the number of antenna branches, and that each CFR (clipping) stage might itself get rather involved, this approach will most probably cause an overload of digital signal processing. Having a CFR unit per antenna branch or transceiver chain in a massive MIMO context would further result in excessive current consumption and cost.

Furthermore, for applications with analog beamforming, which is beneficial in many aspects, the approach of performing CFR and channel filtering in each antenna branch would require implementation of analog CFR. Analog CFR, as opposed to analog pre-distortion, is so far not mentioned in the literature and would be difficult to achieve. So, there exists an inherent problem with analog beamforming when it comes to CFR. At this moment, there exists no approach as of how to implement analog CFR. In other words, CFR or Clipping, when it comes to analog beamforming with amplification, is up to now an unknown art. Herein, however, a method for performing CFR for active antenna arrays, which (method) is applicable for analog, digital as well as hybrid beamforming, will be disclosed.

Herein, the terms CFR and clipping will be used interchangeably as meaning a procedure for reducing the dynamics of a signal subjected to the CFR or clipping, by reducing the "crests" or "peaks" of the signal. Another term which is sometimes used is Peak-to-Average reduction. In addition to the possibility to significantly reduce the number of active CFR operations, embodiments herein also cover a solution to reduce the number of performed digital channel filtering operations, as compared to the case where a channel filter is implemented in each antenna branch. The digital channel filtering is a necessity to reduce the level of unwanted emissions to the regulatory levels of unwanted emissions. Similar to discussion for CFR, the reduced number of channel filtering chains would result in benefits such as reduced digital signal processing, implementation complexity, cost and power consumption. Embodiments herein will be described in a context of an active antenna array comprising a plurality of antenna elements, being associated with a plurality of active components. The active components will be exemplified by power amplifiers. The active antenna array is further associated with a beamformer, W, having a number of input and output ports. The beamformer W may be, or be referred to as, a precoder, a Beam Forming Network, BFN, an Array feed network, an Array Antenna feed network, a signal/power distribution network, a signal processing device, or similar, performing a matrix multiplication, or corresponding, on signals fed to the input ports. The number of output ports of the beamformer W may be assumed to be equal to or larger than the number of input ports. In a typical case, the number of output ports is much larger than the number of input ports. The beamformer W may be analog, digital or a hybrid. Figure 2 shows an active antenna array associated with an analog beamformer W having N input ports and M output ports. In exemplifying

embodiments described herein, the CFR, and also the channel filtering, is performed on a set of signals before being fed to a beamformer W. In case of analog

beamforming, the signals are provided to the beamformer W via a respective Digital to Analog Converter, DAC. This is illustrated in figure 2, where the CFR is illustrated as a block 201 . The operation and function of this CFR block 201 will be exemplified in different embodiments described below. A method embodiment performed by a transmitter arrangement will be described with reference to figure 3, which is a flow chart illustrating the different actions of the embodiment. The transmitter arrangement may be defined e.g. as comprising only a CFR-part of an implementation, such as the block 201 in figure 2, or, could

alternatively be defined as further comprising e.g. the beamformer W, the active antenna array, the DACs and/or the channel filters (cf. e.g. figure 2).

The method embodiment illustrated in figure 3 comprises obtaining 301 a plurality, N, of signals, Xn, which are destined for a respective input port of the beamformer W. The method further comprises deriving 302 a number, P, where P≥N, of linear combinations of the signals Xn, thus obtaining P multiplexed signals. These P signals could also be referred to e.g. as intermediate signals. The method further comprises performing 303 CFR on the number, P, of multiplexed signals, thus obtaining P adapted signals. The method further comprises re-establishing 304 the number N of signals from the number P of adapted signals. The N re-established signals will here be denoted X'n, to indicate that they are not identical to the previously obtained N signals Xn, since the signals now have been subjected to CFR. The N signals X'n are to be provided to the respective input ports of the beamformer W. In case the beamformer W is analog, the signals X'n will be subjected to digital to analog conversion before being fed to the input ports of the beamformer W.

By that the signals Xn are "destined for a respective input port of the beamformer W", is meant that these are signals which eventually, after processing, are to be fed to the input ports of the beamformer W. The re-establishing 304 of the N signals X'n may be described as comprising demultiplexing of the number P of adapted signals, and re-combining of the plurality N of signals. It could also be described as performing the reverse, or inverse, of the deriving 302 of linear combinations of the signals Xn. The number P should, in preferred embodiments, be lower than, or equal to, the number of output ports of the beamformer W. Given that the beamformer W has M output ports, the number P should then be: N≤P≤M. Different values of P will be described further below.

The deriving 302 of the number P of signals may comprise multiplying the plurality N of signals with a matrix, A, of size NxP. In other words, performing a matrix

multiplication (cf. XA). This could alternatively be referred to as applying a matrix A of size NxP to the plurality N of signals. Given that the deriving comprises multiplication with a matrix A of size NxP, the re-establishing could comprise multiplying with (or applying) a matrix being an inverse of the matrix A to the number P of adapted signals. The matrix being an inverse could either be a matrix A^"1, being a true inverse of the matrix A (in case A is invertible). Alternatively, the matrix being an inverse could be a matrix A⁹, being a so-called generalized inverse of the matrix A. A generalized inverse is an approximation of an inverse, which may be used in cases where a matrix is not invertible. One example of a generalized inverse is a so-called Pseudo Inverse, which is typically denoted A^†. A Pseudo Inverse could be described as an inverse of a matrix in a Least Mean Square sense. A Pseudo Inverse, A^†, multiplying the matrix A (i.e. A^†-A) gives the identity matrix.

Standard text books give that a Pseudo Inverse can be computed as: A^†=(A^H-A)^"1 -A^H, where A^H is the Hermitian matrix (transpose + conjugate).

When inserting the above into the identity equation, it can be seen that:

A^†-A=[(A^H-A)^"1 -A^H]-A=I This holds under certain conditions, for example that the product A^H-A is in fact invertible.

The matrix A could be referred to as the "expansion matrix" in embodiments described herein. As a general rule, the expansion matrix A should be selected such that a generalized inverse or Pseudo Inverse actually exists for A, e.g. such that A^H-A is, in fact, invertible. One example of such a matrix would be a subset of the FFT matrix. This choice can be compared with the beamformer, or beamforming matrix, W, which often is chosen as being the full FFT matrix.

The largest reduction of the number of CFR operations, as compared to prior art solutions as the one illustrated in figure 1 , will be achieved for embodiments where P=N, i.e. where the matrix A is chosen to be an NxN matrix. That is, at least for cases where the number of output ports, M, is larger than the number of excited, or used, input ports N, which is a typical case. For example, assuming a 32 branch antenna array being fed, i.e. excited by, only 4 beam ports (input ports of beamformer W) (cf. N=4; M=32). By applying a 4x4 (NxP) expansion matrix, A, and performing and calculating CFR for only these P=4 multiplexed signals, the amount of signal processing would be greatly reduced, as compared to performing and calculating CFR for the 32 branches, as in prior art solutions.

The expansion matrix A could be selected based on the beamforming matrix W, e.g. as described for the FFT matrix above. According to one embodiment, the expansion matrix A could be selected as a replica of the beamformer W. This would entail that

P=M, i.e. the number of multiplexed intermediate signals would be equal to the number of output ports of the beamformer W. In this case, the number of calculated and performed CFRs would be the same as for the prior art solution illustrated in figure 1 . However, the benefits of this solution would comprise being able to perform (digital) CFR together with analog beamforming, and further that the CFR would be performed on signals closely resembling the signals output from the beamformer W (although in the digital domain).

Although the focus of the description so far has been on CFR, a further feature associated with embodiments herein is that the channel filtering may be performed before deriving the P linear combinations of the obtained N signals Xn, e.g. before applying the expansion matrix A to the obtained N signals Xn. This is illustrated e.g. in figures 2 and 4. This feature is indeed beneficial, since it enables significantly reducing the number of required channel filters, as compared to prior art solutions where the channel filtering was performed at each antenna branch. Figure 9a shows an example of a typical channel filter for a 20 MHz OFDM signal.

An analog beamformer W, which may also be referred to as a beamforming matrix W, may be exemplified by the well-known Butler matrix. A Butler matrix typically has the same number of input ports as output ports. However, the number, N, used herein for denoting e.g. input ports, should not necessarily be regarded as representing the total number of input ports of the beamformer W, but rather the number of excited input ports. Exciting one input port of a Butler matrix defines a specific antenna beam, while exciting another port will define yet another antenna beam. Note that Butler matrices outline a very special case of analog beam formers, but should nevertheless act as a good example and is not by any means limiting the embodiments described herein.

Figure 4 illustrates a transmitter arrangement according to an exemplifying embodiment. In figure 4, a possible implementation 401 of the CFR block 201 illustrated in figure 2 can be seen. (The dashed outline 401 is a reference to the CFR block 201 of figure 2). The arrangement in figure 4 comprises an active antenna array with a number, M, of antenna branches. The M branches are fed by M output ports of an analog beamformer W, 402. The beamformer 402 has a number N of input ports that are excited by input signals. A set of N signals, X, are subjected to channel filters 403, and input to an expansion matrix A 404, generating a plurality P of linear combinations of the signals X, the linear combination signals also being referred to as multiplexed signals. The P multiplexed signals are subjected to CFR

405 and input to a Pseudo Inverse, A^†, 406, being the Pseudo Inverse of the expansion matrix A 404. The application of the Pseudo Inverse, A^†, 406, re- establishes the N signals in a clipped version, Xc. The clipped signals Xc are contverted from digital to analog form by a set of DACs, and are input to a respective input port of the beamformer W, 402. The M signals which are output from the beamformer W will then, given (thanks to) the previously performed CFR, have appropriate properties for being fed to the power amplifiers of the antenna branches. In other words, the CFR (401 ) performed according to an embodiment of the invention results in that the power amplifiers will be run in their linear range in the antenna branches.

Below, in order to increase the understanding of embodiments herein, some reasoning will be outlined in a step-by-step manner.

As previously mentioned, a standard clipping and channel filtering situation is depicted in figure 1 , with one CFR and channel filtering block for each amplifier in the antenna branches. The structure in figure 1 could be compared to a proposed new configuration, as the one depicted e.g. in figure 2 where a combined CFR block is placed at digital domain (either in baseband or radio) before the beamformer W.

As can be seen, the clipping blocks "CFR" in the antenna branches in figure 1 are replaced by the CFR block 201 in Figure 2. Typically, the number of output ports (also denoted antenna ports) of a beamforming network, such as beamformer W are usually much larger than the number of input ports (also denoted beam forming ports); that is in a typical case, M»N. This would be a standard situation, but it should be noted that embodiments herein are not limited to this situation.

Having a closer look at the notation of the signals, the clipped (CFRed) signals may be denoted and represented as the original signals (X and Y) with a small part that is added to the signals, where the small part may be denoted ΔΧ and ΔΥ. This notation is used in figures 5 and 6. Now, expressions need to be found, as how to compute the ΔΧ signal manifold (illustrated in figure 6) in order to actually obtain the ΔΥ signal manifold (illustrated in figure 5). In other words, how should the signals be processed by a CFR block as the block 201 illustrated in figure 2, in order to achieve an adequately clipped signal for the power amplifiers in the antenna branches. Letting Xc denote the clipped signals at the input side of the beamformer W, and letting Yc denote the clipped signals in the antenna branches, i.e. at the output side of the beamformer W. Using the above described notation:

Yc=CFR{Y}=Y+AY Eq. 1

Xc=X+AX Eq. 2 Now, we may directly compare these two configurations as to say that the output signal manifold, i.e. the clipped signal Yc, should be equal in the two cases illustrated in figures 1 and 2, at least from a Least Mean Squares perspective. This statement leads to two equations which may be used in order to extract what kind of operation that may be performed e.g. by the CFR block 201 in figure 2. From figure 5, we have:

Yc=W-X+AY Eq. 3

From figure 5 we also have:

Yc=W Xc=W (X+AX) Eq. 4

The preceding two equations give us: Xc=W^†Yc=W^† CFR{W X} Eq. 5

Here we have introduced the Pseudo Inverse W^†, which when multiplying the beam forming matrix W gives the identity matrix. So, it may be viewed as an inversion of the beamforming matrix W, but in the strict case it is only an inversion in a Least Mean Square sense. As previously mentioned the Pseudo Inverse can be computed as:

W^†=(W^H-W)^"1 -W^H Eq. 6

And as stated before, when inserted into the identity equation we see that: W^†-W=[(W^H-W)^"1 -W^H]-W=I Eq. 7

This holds under certain conditions, such as that the product W^H-W should be invertible. One embodiment illustrating the insights above is depicted in figure 7. Returning to the expansion matrix, A, described above, the embodiment in figure 7 represents the special case where the expansion matrix A is a replica of the beamformer W. As previously described, the inventors have realized that the expansion matrix can be a matrix (A) which is smaller than a matrix (W) associated with the beamformer W. The advantage of choosing a more general size expansion matrix than W is obvious, since it makes it possible to reduce the number of CFR operations, as previously described.

In order to demonstrate the operation of embodiments described herein, an exemplifying implementation has been made. For the example, an 8x8 square active array antenna which has 64 RF amplifiers has been defined. The active antenna array is fed by two input ports of an analog beamforming matrix (beamformer) W, each input port corresponding to a beam. We let the two beams be defined in different directions in order to make a good example. In order to simplify, a raw clipping function is used for CFR, which simply limits the envelop of a signal that is fed through it. In a real application, such a clipping function would give rise to extra emissions outside the carrier bandwidth, but it is used here as a non-limiting example.

The two input signals (X) to be fed to the input ports of the beamformer W should be clipped (Xc) such that, when being split through the beamformer W, the result becomes the desired clipped signals (Yc) in each active antenna branch. In a special case, the Complementary Cumulative Distribution Function, CCDF, which describes the signal amplitude statistics, for a demonstrating example of one of the branches in this exemplifying active antenna array is found as the dashed line depicted in figure 8. In the example, it was aimed at a Peak to Average Ratio, PAR,-level of about PARraw = 5.2 dB, which is illustrated by the solid line representing "ideal clipping". The CCDF curve in figure 8 shows a very steep characteristic and it is as it should be, given that a simple raw clipping algorithm is used, for simplicity. As a reference, figure 8 also comprises a curve representing no clipping (dotted line). The dotted line represents an undipped OFDM-signal for the corresponding antenna branch. The dotted line shows the typical CCDF following a Rayleigh distributed envelop that is so typical for these kinds of signals. The PAR at 10-4 probability can be seen to be about 9.5 dB. Thus, for an embodiment as proposed herein being imposed on the signals provided to the two beam ports in the example, the CCDF of the resulting branch signal envelope statistics looks like the dashed, "Actual clipped signal" curve in figure 8. The dashed curve shows a slight increase of the PAR level in relation to the solid "ideally clipped"signal, but still a very good PAR improvement can be seen, as compared to the un-clipped signal. The PAR for this particular branch is observed to be around 6.9 dB.

This shows that although only two input ports were used to feed 64 antenna branches with active amplifier devices in each branch, a considerable reduction in PAR (>2.5 dB) can be observed in each of the 64 branches when applying an embodiment of the invention.

As a special case, feeding of only one input port of a beamformer or beamforming network may be considered. In such a case, clipping gets quite simple for all antenna branches, e.g. since the difference between the different signal branches is only lying in the phase part. That is, the amplitude envelop will be the same for all antenna branches. Therefore, the input signal may be clipped and fed to the beamformer W "as is". As the only object of the beamformer W in this special case is to apply phases in a proper way in order to produce beams in different directions. It should be mentioned that the phase relation between the I- and Q-components will propagate unaffected through the beamformer W. Note that this is only true if one-only signal is fed through the beamformer W (only one beam port is excited). If more than one input port of the beamformer W is excited, the matrix compensation method as proposed above applies. Note further that there is no need to change the

embodiments of the method for cases where only one input port would be excited, since the embodiments can handle also this case. However, things could be made simpler in cases where it is only desired to form one beam (where only one input port of a beamforming network is fed). The case for one-only input port (beam port) is shown in figure 9b.

The method embodiments and techniques described above may be implemented in a wireless communication network, e.g. in transmitter arrangements, which may be comprised in one or more network nodes, such as e.g. radio access nodes, such as eNBs. The methods could be implemented in a distributed manner, e.g. a plurality of nodes or entities could each perform a part of the actions e.g. at different locations in the network. For example, one or more embodiments could be implemented in a so- called cloud solution, or a "Centralized RAN" or "Split Architecture", where e.g. an eNB is divided into 2 or more separate nodes. Correspondingly, the network could be configured such that actions of the method embodiments are performed e.g. partly in a radio access node and partly in a core network node. The distributed case could be described as that the method is performed by a transmitter arrangement or a network node operable in the communication network, but that the transmitter arrangement or the network node could be distributed in the network, and not necessarily be comprised in one physical unit e.g. close to an antenna. However, a possible embodiment would be that the transmitter arrangement is located close to the antenna array and possibly comprising the antenna array.

An exemplifying embodiment of a transmitter arrangement is illustrated in a general manner in figure 10a. The transmitter arrangement 1000 is configured to perform at least one of the method embodiments described above, e.g. with reference to figure 3. The transmitter arrangement 1000 is associated with the same technical features, objects and advantages as the previously described method embodiments. The transmitter arrangement will be described in brief in order to avoid unnecessary repetition. The transmitter arrangement may be implemented and/or described as follows:

The transmitter arrangement 1000 comprises processing circuitry 1001 , and one or more communication interfaces 1002. The transmitter arrangement is configured for an active antenna array, which is associated with a plurality of antenna elements being associated with a plurality of active components, such as power amplifiers. The active antenna array is further associated with a beamformer, W, having a number of input and output ports. As previously mentioned, the beamformer W may be, or be referred to as, a precoder, a Beam Forming Network, an Array feed network, an Array Antenna feed network, a signal/power distribution network, a signal processing device, or similar, performing a matrix multiplication, or corresponding, on signals fed to the input ports. The transmitter arrangement may or may not be defined as comprising the actual antenna array or antenna elements, active components, etc. In the case the transmitter arrangement is defined as not comprising the antenna array and/or the associated components, the transmitter arrangement is at least connectable to the antenna array to be linearized, including the relevant components. The processing circuitry may be composed of one or more parts which may be comprised in one or more nodes in the communication network, but is here illustrated as one entity.

The processing circuitry 1001 is configured to cause the transmitter arrangement 1000 to obtain a plurality, N, of signals, Xn, destined for a respective input port of the beamformer W. The processing circuitry 1001 is further configured to cause the transmitter arrangement 1000 to derive a number, P, where P≥N, of linear

combinations of the signals Xn, thus obtaining P multiplexed signals; and further to perform CFR on the number, P, of multiplexed signals, thus obtaining P adapted signals. The processing circuitry 1001 is further configured to cause the transmitter arrangement 1000 to re-establish, from the number P of adapted signals, the number N of signals, X'n, to be provided to the respective input ports of the beamformer W. The one or more communication interfaces 1002, which may also be denoted e.g. Input/Output (I/O) interfaces, may include an interface for obtaining signals and information from one or more other entities.

The processing circuitry 1001 could, as illustrated in figure 10b, comprise one or more processing means, such as a processor 1003, and a memory 1004 for storing or holding instructions. The memory would then comprise instructions, e.g. in form of a computer program 1005, which when executed by the one or more processing means 903 causes the network node or arrangement 1000 to perform the actions described above. The processing circuitry 1001 may, be composed of one or more parts and be comprised in, or distributed over, one or more nodes in the

communication network, but is here illustrated as one entity.

An alternative implementation of the processing circuitry 1001 is shown in figure 10c. The processing circuitry comprises an obtaining unit 1006, configured to cause the transmitter arrangement to obtain a plurality, N, of signals, Xn, destined for a respective input port of the beamformer W. The processing circuitry further

comprises a deriving unit 1007, configured to cause the transmitter arrangement to derive a number, P, where P≥N, of linear combinations of the signals Xn, thus obtaining P multiplexed signals. The processing circuitry further comprises a CFR unit 1008, configured to cause the transmitter arrangement to perform CFR on the number, P, of multiplexed signals, thus obtaining P adapted signals. The processing circuitry further comprises a re-establishing unit 1009, configured to cause the transmitter arrangement to re-establish, from the number P of adapted signals, the number N of signals, X'n, to be provided to the respective input ports of the beamformer W. The processing circuitry may further comprise a filtering unit 1010, configured to cause the transmitter arrangement to apply a channel filter to each of the plurality, N, of obtained signals. The transmitter arrangements described above could be configured for the different method embodiments described herein. The steps, functions, procedures, modules, units and/or blocks described herein may be implemented in hardware using any conventional technology, such as discrete circuit or integrated circuit technology, including both general-purpose electronic circuitry and application-specific circuitry.

Particular examples include one or more suitably configured digital signal processors and other known electronic circuits, e.g. discrete logic gates interconnected to perform a specialized function, or Application Specific Integrated Circuits (ASICs).

At least some of the steps, functions, procedures, modules, units and/or blocks described above may be implemented in software, such as a computer program, for execution by suitable processing circuitry including one or more processing units. The software could be carried by a carrier, such as an electronic signal, an optical signal, a radio signal, or a computer readable storage medium before and/or during the use of the computer program e.g. in one or more nodes of the wireless communication network. The processing circuitry described above may be

implemented in a so-called cloud solution, referring to that the implementation may be distributed, and may be referred to e.g. as being located in a so-called virtual node or a virtual machine.

The flow diagram or diagrams presented herein may be regarded as a computer flow diagram or diagrams, when performed by one or more processors. A corresponding arrangement or apparatus may be defined as a group of function modules, where each step performed by a processor corresponds to a function module. In this case, the function modules are implemented as one or more computer programs running on one or more processors.

Examples of processing circuitry includes, but is not limited to, one or more microprocessors, one or more Digital Signal Processors, DSPs, one or more Central Processing Units, CPUs, and/or any suitable programmable logic circuitry such as one or more Field Programmable Gate Arrays, FPGAs, or one or more

Programmable Logic Controllers, PLCs. That is, the units or modules in the arrangements in the communication network described above could be implemented by a combination of analog and digital circuits in one or more locations, and/or one or more processors configured with software and/or firmware, e.g. stored in a memory. One or more of these processors, as well as the other digital hardware, may be included in a single application-specific integrated circuitry, ASIC, or several processors and various digital hardware may be distributed among several separate components, whether individually packaged or assembled into a system-on-a-chip, SoC.

It should also be understood that it may be possible to re-use the general processing capabilities of any conventional device or unit in which the proposed technology is implemented. It may also be possible to re-use existing software, e.g. by

reprogramming of the existing software or by adding new software components. The embodiments described above are merely given as examples, and it should be understood that the proposed technology is not limited thereto. It will be understood by those skilled in the art that various modifications, combinations and changes may be made to the embodiments without departing from the present scope. In particular, different part solutions in the different embodiments can be combined in other configurations, where technically possible.

When using the word "comprise" or "comprising" it shall be interpreted as non- limiting, i.e. meaning "consist at least of.

It should also be noted that in some alternate implementations, the functions/acts noted in the blocks may occur out of the order noted in the flowcharts. For example, two blocks shown in succession may in fact be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Moreover, the functionality of a given block of the flowcharts and/or block diagrams may be separated into multiple blocks and/or the functionality of two or more blocks of the flowcharts and/or block diagrams may be at least partially integrated. Finally, other blocks may be added/inserted between the 5 blocks that are illustrated, and/or blocks/operations may be omitted without departing from the scope of inventive concepts.

It is to be understood that the choice of interacting units, as well as the naming of the units within this disclosure are only for exemplifying purpose, and nodes suitable to execute any of the methods described above may be configured in a plurality of l o alternative ways in order to be able to execute the suggested procedure actions.

It should also be noted that the units described in this disclosure are to be regarded as logical entities and not with necessity as separate physical entities.

ABBREVIATIONS

PAR Peak to Average Ratio

15 PA Power Amplifier

BFN Beam Forming Network

RF Radio Frequency

LMS Least Mean Squares

BF Beam Forming [Matrix]

20 CFR Crest Factor Reduction

MIMO Multiple Input Multiple Output

DAC Digital-to-Analog Converter

FFT Fast Fourier Transform

Claims

1 . Method to be performed by a transmitter arrangement; the method being for processing signals for an active antenna array, the active antenna array comprising a plurality of antenna elements, being associated with a plurality of active components, the active antenna array further being associated with a beamformer, W, having a number of input and output ports; the method comprising:

-obtaining a plurality, N, of signals, Xn, destined for a respective input port of the beamformer W:

-deriving a number, P, where P≥N, of linear combinations of the signals Xn, thus obtaining P multiplexed signals;

-performing CFR on the number, P, of multiplexed signals, thus obtaining P adapted signals;

-re-establishing, from the number P of adapted signals, the number N of signals, X'n, to be provided to the respective input ports of the beamformer W.

2. Method according to claim 1 , wherein the re-establishing comprises

demultiplexing the number P of adapted signals, and re-combining the plurality N of signals.

3. The method according to claim 1 or 2, wherein the beamformer W is an

analog, a digital or a hybrid beamformer.

4. The method according to any of claims 1 -3, wherein the number P is lower than, or equal to, the number of output ports of the beamformer W.

5. The method according to any of claims 1 -4, wherein the deriving of the

number P of signals comprises multiplying the plurality N of signals with a matrix, A, of size NxP.

6. The method according to claim 5, wherein the re-establishing comprises:

-multiplying the number P of adapted signals with: -a matrix A⁹, being a generalized inverse of the matrix A;

-a matrix A^†, being the pseudo-inverse of the matrix A; or

-a matrix A^"1, being the inverse of the matrix A.

7. The method according to claim 5 or 6, wherein the matrix A is based on the beamformer W.

8. The method according to any of claims 5-7, wherein the matrix A is a replica of the beamformer W.

9. The method according to any of the preceding claims, wherein the number,

N, of signals, X'n, are provided to the input ports of the beamformer W via Digital to Analog Converters, DACs.

10. The method according to any of the preceding claims, further comprising:

-applying a channel filter to each of the plurality, N, of obtained signals.

1 1 . The method according to any of the preceding claims, wherein the active components are power amplifiers.

12. A transmitter arrangement for an active antenna array, the active antenna array comprising a plurality of antenna elements, being associated with a plurality of active components, the active antenna array further being associated with a beamformer, W, having a number of input and output ports; the transmitter arrangement being configured to:

-obtain a plurality, N, of signals, Xn, destined for a respective input port of the beamformer W;

-derive a number, P, where P≥N, of linear combinations of the signals Xn, thus obtaining P multiplexed signals;

-perform CFR on the number, P, of multiplexed signals, thus obtaining P adapted signals; and to

-re-establish, from the number P of adapted signals, the number N of signals, X'n, to be provided to the respective input ports of the beamformer W.

13. The transmitter arrangement according to claim 12, wherein the reestablishing comprises demultiplexing the number P of adapted signals, and re-combining the plurality N of signals.

14. The transmitter arrangement according to claim 12 or 13, wherein the

beamformer W is an analog, a digital or a hybrid beamformer.

15. The transmitter arrangement according to any of claims 12-14, wherein the number P is lower than, or equal to, the number of output ports of the beamformer W.

16. The transmitter arrangement according to any of claims 12-15, wherein the deriving of the number P of signals comprises multiplying the plurality N of signals with a matrix, A, of size NxP.

17. The transmitter arrangement according to claim 16, wherein the reestablishing comprises:

-multiplying the number P of adapted signals with:

-a matrix A⁹, being a generalized inverse of the matrix A;

-a matrix A^†, being the pseudo-inverse of the matrix A; or

-a matrix A^"1, being the inverse of the matrix A.

18. The transmitter arrangement according to claim 16 or 17, wherein the

matrix A is based on the beamformer W.

19. The transmitter arrangement according to any of claims 16-18, wherein the matrix A is a replica of the beamformer W.

20. The transmitter arrangement according to any of claims 12-19, wherein the number, N, of signals, X'n, are provided to the input ports of the beamformer W via Digital to Analog Converters, DACs.

21 . The transmitter arrangement according to any of claims 12-20, being further configured to:

-apply a channel filter to each of the plurality, N, of obtained signals.

22. The transmitter arrangement according to any of claims 12-21 , wherein the active components are power amplifiers.

23. A network node comprising the transmitter arrangement according to any of claims 12-22.

24. Computer program (1005), comprising instructions which, when executed on at least one processor, cause the at least one processor to carry out the method according to any of claims 1 -1 1 .

25. A carrier containing the computer program of the preceding claim, wherein the carrier is one of an electronic signal, optical signal, radio signal, or computer readable storage medium.