Method and network node for beamformed communication with a plurality of user equipment

Abstract

Disclosed is a method performed by a network node (130) of a wireless communication network (100) for beamformed communication with a plurality of User Equipment, UE, (140, 145). The network node (130) comprises a plurality of antenna elements, each antenna element arranged for handling an antenna signal to be transmitted or received over a frequency band comprising a plurality of sub- bands. The method describes different ways to provide a deliberate beam squint to a beam transmitted or received by the network node (130) by applying an inter- element time delay difference to antenna signals of the plurality of antenna elements. A beam occurring this way will have different frequencies pointing in different directions due to the beam squint. Thereby, one beam could be used for reaching UEs arranged over a wider angular range, wherein the UEs (140, 145) are mapped to different frequencies depending on where in the beam squint the UEs (14, 145) are situated. The method is especially useful in uplink.

Description

METHOD AND NETWORK NODE FOR BEAMFORMED COMMUNICATION WITH A PLURALITY OF USER EQUIPMENT TECHNICAL FIELD

[0001] The present disclosure relates generally to methods and network nodes for beamformed communication with a plurality of user equipment (UE). The present disclosure further relates to computer programs and carriers corresponding to the above methods and nodes. BACKGROUND

[0002] In wireless communication systems, bitrate demand continues to increase. Lower frequency spectrum fills up, so higher frequency spectrum is taken into use. In 5thGeneration (5G) wireless communication systems, also called New Radio (NR), a new radio frequency range (FR) is introduced, called FR2, having an FR of 24250 MHz – 52600 MHz, which is planned to be used in addition to FR1 having an FR of 410 – 7125 MHz. In 5G, beamforming is introduced both to increase capacity and coverage. In FR2, beamforming is mainly used to combat the higher pathloss that occurs when increasing the radio frequency. Beamforming and beam steering is made by coherently combining radio frequency (RF) signals from a plurality of antenna elements. For example, a network node, also called a base station, may have 128 or even 256 antenna elements. By using individual phase shifts for antenna signals of the antenna elements, individual phase shift for the individual antenna element, the desired beam is formed. This technology contributes to a mitigation of the above listed problems, by means of a radically increased beam gain, restoring the rated equivalent isotropic radiated power (EIRP) rating of mmWave network nodes to usable levels.

[0003] Beamforming (BF) can be obtained in many ways. A popular, low complexity way of doing beamforming is analogue BF. Here, the signals to / from the antenna elements are beamformed in the RF domain, close to the antenna elements. The rest of the signal chain is common to all or a portion of the antenna elements. What happens then is that all data to be transmitted isconverted into a time domain stream early, before sent to the radio Application Specific Integrated Circuits (ASICs) and antenna elements (and similarly converted to frequency domain late in the receive direction). Since one set of beam weights is applied during the duration of one Orthogonal Frequency Division Multiplexing (OFDM) symbol, the beam is therefore spatially fixed for all data during one OFDM symbol. Although the beam may have peaks in multiple directions, the data stream will be transmitted through one beam pattern, which obviously limits the possibility to simultaneously transmit data to multiple User Equipment (UE), also called wireless communication device. Put otherwise, this generates problems when it would be advantageous to direct different data streams in different directions, by frequency selective scheduling. In addition, problems are created when a UE is trying to find the network node for initial access. Present FR2 Active Antenna Systems (AAS) use beam sweeping or wider initial beams to combat these issues, but this adds cost in terms of coverage, latency and / or capacity.

[0004] Frequency domain digital beamforming, on the contrary, uses late Inverse Fast Fourier Transform (IFFT) processing to transform complex OFDM symbols to data streams in time, with each user accessing all antenna elements independently, thereby allowing frequency selective beam forming. This does however require per antenna element IFFT processing, which adds complexity and cost.

[0005] A compromise between analog beamforming and frequency domain digital beamforming is time domain digital beamforming. There the beamforming is made per carrier instead of the full frequency bandwidth (BW) as in the analog beamforming case. Time domain digital beamforming can also be made across a wider BW and is then less costly than per carrier beamforming.

[0006] Future 5G network nodes for both FR1 and FR2 will contain more antenna elements to enable higher system capacity. There is also a wish to improve FR2 performance by moving in the direction of more flexible beamforming. In FR1, the complexity added when expanding the array with moreantenna elements together with full frequency domain digital beamforming is considered too high.

[0007] Hybrid beamforming is a mix of frequency domain digital, time domain digital and analog beamforming. It allows trading performance versus complexity of the base station to get a good compromise. It would be of interest to use as much analog beamforming as possible to reduce complexity, however, as stated above, the more analog beamforming that is used, the fewer UEs can be reached at the same time.

[0008] Consequently, there is a need of an improved signal processing in network nodes that would make it possible to communicate with lots of UEs simultaneously when using beamforming, without increasing complexity of the network node. SUMMARY

[0009] It is an object of embodiments of the invention to address at least some of the problems and issues outlined above. It is an object of embodiments of the invention to be able to communicate with more UEs simultaneously compared to prior art when using beamforming, without increasing complexity of the network node. It is an object of embodiments to make it possible to communicate with more UEs simultaneously compared to prior art methods and network nodes, when using the same amount of analog beamforming. Put it the other way around, an object of embodiments is to be able to use an increased part of analog beamforming in the beamforming performed at the network node compared to prior art and still reach the same amount of UEs. It is possible to achieve at least of one of these objects by using methods and network nodes as defined in the attached independent claims.

[0010] According to one aspect, a method is provided that is performed by a network node of a wireless communication network for beamformed communication with a plurality of UEs. The network node comprises a plurality of antenna elements. Each antenna element is arranged for handling an antenna signal to be transmitted or received over a frequency band comprising a plurality ofsub-bands. The method comprises determining, depending on directions between the network node and each of the plurality of UEs, a first direction in which to direct a beam formed by transmitting or receiving the antenna signals by the plurality of antenna elements. The method further comprises obtaining phase shifts or delays to add to the antenna signals for the beam to get the determined first direction. The method further comprises determining an inter-element time delay difference for controlling time delays to add to the antenna signals so that the antenna signals of neighboring of the plurality of antenna elements are mutually delayed with the inter-element time delay difference. I.e. the signal on one element is later by the amount of the inter-element time delay difference with respect to the signal on a neighboring element. The time delays being added in order to squint the beam formed by the transmitted antenna signals or the received antenna signals so that any sub-band of the squinted beam, when the squinted beam is directed in the first direction, reaches the plurality of UEs. The method further comprises determining, based on information on the directions between the network node and each of the plurality of UEs and the determined first direction, to allocate to each of the plurality of UEs, a sub-band of the plurality of sub-bands for communication with the network node, wherein the respective sub-band is allocated depending on a position of the direction between the network node and the respective UE in the squinted beam arising from application of the determined inter-element time delay difference to the antenna signals. The method further comprises sending an instruction to each of the plurality of UEs to use the allocated sub-band for communication with the network node. The method further comprises transmitting to the plurality of UEs, or receiving from the plurality of UEs, the antenna signals via the plurality of antenna elements applying time delays to the antenna signals according to the determined inter-element time delay difference and applying the determined phase shifts or delays hereby achieving the squinted beam directed in the first direction.

[0011] According to another aspect, a network node is provided that is configured to operate in a wireless communication network and configured for beamformed communication with a plurality of UEs. The network node comprises a plurality of antenna elements, where each antenna element is arranged forhandling an antenna signal to be transmitted or received over a frequency band comprising a plurality of sub-bands. The network node comprises a processing circuitry and a memory. Said memory contains instructions executable by said processing circuitry, whereby the network node is operative for determining, depending on directions between the network node and each of the plurality of UEs, a first direction in which to direct a beam formed by transmitting or receiving the antenna signals by the plurality of antenna elements, and obtaining phase shifts or delays to add to the antenna signals for the beam to get the determined first direction. The network node is further operative for determining an inter- element time delay difference for controlling time delays to add to the antenna signals so that the antenna signals of neighboring of the plurality of antenna elements are mutually delayed with the inter-element time delay difference, the time delays being added in order to squint the beam formed by the transmitted antenna signals or the received antenna signals so that any sub-band of the squinted beam, when the squinted beam is directed in the first direction, reaches the plurality of UEs. The network node is further operative for determining, based on information on the directions between the network node and each of the plurality of UEs and the determined first direction, to allocate to each of the plurality of UEs, a sub-band of the plurality of sub-bands for communication with the network node, wherein the respective sub-band is allocated depending on a position of the direction between the network node and the respective UE in the squinted beam arising from application of the determined inter-element time delay difference to the antenna signals. The network node is further operative for sending an instruction to each of the plurality of UEs to use the allocated sub-band for communication with the network node, and transmitting to the plurality of UEs or receiving from the plurality of UEs, the antenna signals via the plurality of antenna elements applying time delays to the antenna signals according to the determined inter-element time delay difference and applying the determined phase shifts or delays hereby achieving the squinted beam directed in the first direction.

[0012] According to other aspects, computer programs and carriers are also provided, the details of which will be described in the claims and the detailed description.

[0013] Further possible features and benefits of this solution will become apparent from the detailed description below. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The solution will now be described in more detail by means of exemplary embodiments and with reference to the accompanying drawings, in which:

[0015] Fig.1 is a schematic diagram of a wireless communication network in which the present invention may be used.

[0016] Fig.2 is a flow chart illustrating a method performed by a network node, according to possible embodiments.

[0017] Fig.3 is a schematic block diagram of a radio block chain and various ways of doing beamforming and port expansion with such a radio block.

[0018] Fig.4 is a block diagram of an antenna array of a network node, which antenna array may be used with the present invention.

[0019] Fig, 5a is a Cartesian coordinate diagram showing beam squint at 4.2 GHz for a beam transmitted at a center frequency of 4 GHz and with 400 MHz bandwidth, the diagram having beam steering on the x-axis and beam squint on the y-axis.

[0020] Fig, 5b is a Cartesian coordinate diagram showing beam squint for 60 degrees beam steering for the beam of fig.5a, the diagram having carrier frequency on the x-axis and beam squint on the y-axis.

[0021] Fig.6a is a Cartesian coordinate diagram showing antenna signals of two antenna elements of a network node, the diagram having carrier frequency on the x-axis and phase on the y-axis, the antenna signals having a frequency-dependent inter-element phase difference, i.e. the signal on one element is phase shifted by the frequency-dependent inter-element phase difference with respect to the phase of the signal on the other element.

[0022] Fig.6b is a Cartesian coordinate diagram showing the achieved beam in elevation direction when the two antenna signals of fig.6a are transmitted wirelessly from the two antenna elements, using the frequency-dependent phase difference of fig.6a, the diagram having carrier frequency on the x-axis and elevation beam direction on the y-axis.

[0023] Fig.7 is a schematic block diagram of a radio transmitter with two parallel ports.

[0024] Fig.8a is a Cartesian coordinate diagram showing four component carriers (CC), the diagram having carrier frequency on the x-axis and inter-port phase difference on the y-axis.

[0025] Fig.8b is a Cartesian coordinate diagram showing the four component carriers of fig.8a, the diagram having carrier frequency on the x-axis and elevation beam direction on the y-axis.

[0026] Fig.9 is a schematic block diagram of signal treatment in a network node for downlink communication according to an embodiment.

[0027] Fig.10 is a schematic block diagram of an implementation of the downlink communication of fig.9.

[0028] Fig.11 is a schematic block diagram of signal treatment in a network node for uplink communication according to an embodiment.

[0029] Fig.12 is a schematic block diagram of an implementation of the uplink communication of fig.11.

[0030] Fig.13 is a Cartesian coordinate diagram showing two elevation beams, the diagram having carrier frequency on the x-axis and elevation phase on the y- axis.

[0031] Fig.14 is a block diagram illustrating a network node in more detail, according to further possible embodiments.DETAILED DESCRIPTION

[0032] Fig.1 shows a wireless communication network 100 comprising a radio access network (RAN) node, also called network node 130, which is in, or is adapted for, wireless communication with one or more wireless communication devices, also called wireless devices, or UEs 140, 145. The network node 130 provides radio access in a cell 150 covering a geographical area.

[0033] The wireless communication network 100 may be any kind of wireless communication network that can provide radio access to wireless devices. Example of such wireless communication networks are networks based on Global System for Mobile communication (GSM), Enhanced Data Rates for GSM Evolution (EDGE), Universal Mobile Telecommunications System (UMTS), Code Division Multiple Access 2000 (CDMA 2000), Long Term Evolution (LTE), LTE Advanced, Wireless Local Area Networks (WLAN), Worldwide Interoperability for Microwave Access (WiMAX), WiMAX Advanced, as well as fifth generation (5G) wireless communication networks based on technology such as New Radio (NR), and any possible future sixth generation (6G) wireless communication network.

[0034] The network node 130 may be any kind of network node that can provide wireless access to the UEs 140, 145 alone or in combination with another network node. Examples of network nodes 130 are a base station (BS), a radio BS, a base transceiver station, a BS controller, a network controller, a Node B (NB), an evolved Node B (eNB), a gNodeB (gNB), a Multi-cell / multicast Coordination Entity, a relay node, an access point (AP), a radio AP, a remote radio unit (RRU), a remote radio head (RRH) and a multi-standard BS (MSR BS).

[0035] The UEs 140, 145 may be any type of device capable of wirelessly communicating with a network node 130 using radio signals. For example, the UE may be a machine type UE or a UE capable of machine to machine (M2M) communication, a sensor, a tablet, a mobile terminal, a smart phone, a laptop embedded equipped (LEE), a laptop mounted equipment (LME), a USB dongle, a Customer Premises Equipment (CPE), an Internet of Things (IoT) device, etc.

[0036] As mentioned, embodiments of the present invention concern beamforming at a network node 130 of signals to be communicated with one or more UEs 140, 145 in order to provide a network node of rather low complexity at the same time as being able to communicate with many UEs simultaneously.

[0037] An interesting beamforming approach to reduce complexity when the network node has many antenna elements is to do part of elevation beamforming in analog domain and the rest of the beamforming in digital domain, including azimuth beamforming and the part of the elevation beamforming that was not done in analog domain. This allows to address a lot of UEs simultaneously, since most UEs reside in same elevation plane but are distributed in azimuth direction. Embodiments of the present invention are especially advantageous when used as such, but the invention is not limited to such parting between elevation beamforming and azimuth beamforming.

[0038] As mentioned in the background, analog beamforming has low complexity and therefore provides a cost-efficient solution. However, the more analog beamforming that is used in comparison to digital beamforming, the more the instantaneous service area shrinks, i.e., the area in which UEs can be served at the same time. So, in order to reach a larger service area and continue using the same amount of analog beamforming, the communication in the cell needs to be time multiplexed, which reduces capacity. In particular, this is a problem for uplink (UL) communication, where UEs ideally should be allowed to transmit with high duty cycle for good performance.

[0039] Embodiments of the invention proposes to modify the analog beamforming by applying certain time delays to the antenna signals, in addition to the phase shifts that are applied to the antenna signals to direct the beam in a certain direction. The time delays are selected so that there is an inter-element time delay difference, i.e., a certain time delay difference between antenna signals of neighboring antenna elements of the network node. By applying such inter-element time delay differences, a beam direction is generated that changes across the frequency range of the radio frequency band over which the signalsare communicated. UEs that are situated in different beam direction can then be frequency multiplexed across the radio frequency band depending on in which direction from the network node they are situated. In other words, a UE is allocated a sub-band of the frequency band, a sub-band that points towards the UE. This means that a desired service area can be covered although extensive analog beamforming is used. This is particularly useful for elevation beamforming, i.e., for reaching UEs at different elevation levels. However, it may also be used for beamforming in azimuth direction. This is also particularly useful in the UL, where the UEs are power limited and cannot utilize the full bandwidth anyway. According to an embodiment, by changing the phase shifts and the inter-element time delay difference, the beam can quickly be switched between pointing the full beam in one direction and generating an elevation ramp versus frequency.

[0040] Fig.2, in conjunction with fig.1, describes a method performed by a network node 130 of a wireless communication network 100 for beamformed communication with a plurality of UEs 140, 145. The network node 130 comprises a plurality of antenna elements. Each antenna element is arranged for handling an antenna signal to be transmitted or received over a frequency band comprising a plurality of sub-bands. The method comprises determining 202, depending on directions between the network node 130 and each of the plurality of UEs 140, 145, a first direction in which to direct a beam formed by transmitting or receiving the antenna signals by the plurality of antenna elements. The method further comprises obtaining 204 phase shifts or delays to add to the antenna signals for the beam to get the determined 202 first direction. The method further comprises determining 206 an inter-element time delay difference for controlling time delays to add to the antenna signals so that the antenna signals of neighboring of the plurality of antenna elements are mutually delayed with the inter-element time delay difference. The time delays being added in order to squint the transmit beam formed by the transmitted antenna signals or the receive beam for the received antenna signals so that any sub-band of the squinted beam, when the squinted beam is directed in the first direction, reaches the plurality of UEs 140, 145. The method further comprises determining 208, based on information on the directionsbetween the network node 130 and each of the plurality of UEs 140, 145 and the determined first direction, to allocate to each of the plurality of UEs 140, 145, a sub-band of the plurality of sub-bands for communication with the network node 130, wherein the respective sub-band is allocated depending on a position of the direction between the network node and the respective UE 140, 145 in the squinted beam arising from application of the determined inter-element time delay difference to the antenna signals. The method further comprises sending 210 an instruction to each of the plurality of UEs 140, 145 to use the allocated sub-band for communication with the network node 130. The method further comprises transmitting 212 to the plurality of UEs 140, 145 or receiving from the plurality of UEs 140, 145, the antenna signals via the plurality of antenna elements applying time delays to the antenna signals according to the determined inter-element time delay difference and applying the determined phase shifts or delays hereby achieving the squinted beam directed in the first direction.

[0041] By such a method, a deliberate squint of the transmit or receivebeam can be achieved, whereby a certain area can be covered where UEs that are desired to be reached are situated. The squint can thus be controlled such that a desired (e.g., maximum) number of UEs can be reached by different frequencies in a squinted beam pattern even though they are located in different directions, where a particular frequency or sub- band is directed, in the squinted beam pattern, to each of the UEs. By allocating the UEs different sub-bands depending on where they are situated in the squinted beam, the UEs can be reached simultaneously, e.g., in one and the same OFDM symbol time. In other words, for a particular UE, a sub-band is chosen that in the squinted beam is directed towards that UE.

[0042] An “antenna element” here means an antenna unit configured to handle an individual antenna signal that can be given an individual phase shift or delay, i.e., the antenna signal to the antenna element can be phase-shifted or delayed independently of any delay or phase shift applied to an antenna signal of anotherantenna element. The antenna element can be just one single antenna element, or the antenna element can comprise a plurality of antenna sub-elements. The plurality of antenna sub-elements may receive one and the same antenna signal or antenna signals that are given the same phase shift or delay. Alternatively, individual phase shifts can be given to antenna signals of the antenna sub- elements so that the antenna element having a plurality of antenna sub-elements provides some beam steering by itself. An antenna element having a plurality of sub-elements may be called a subarray, A purpose for setting up such a subarray is to achieve a stronger signal. The directions between the network node and each of the plurality of UEs can be determined by either UL measurements or based on feedback from a UE that has measured on a DL signal. The first direction may be determined as the spatial mean of the directions between the network node and each of the plurality of UEs. The phase shifts or delays to add to the antenna signals for the beam to get the determined 202 first direction can be obtained by being determined when to be used or determined in advance for different directions and selected from a codebook when to be used. Neighboring antenna element are to be seen as antenna elements that are adjacent to each other. This means that antenna signals of two antenna elements that are situated adjacent to each other are mutually delayed with the inter-element time delay difference. The inter-element time delay difference signifies that time delays added to the antenna signals are different for the different antenna signals so that neighboring antenna signals, i.e., antenna signals to neighboring antenna elements, are delayed in relation to one another with the inter-element time delay difference. There is one antenna signal at each of the plurality of antenna elements. Different time delays are added to the different antenna signals so that there is a difference in time delay, called inter-element time delay difference, between neighboring antennas that will result in a squinted beam with a beam angle that covers the geographical positions of the plurality of UEs that are to be reached by the squinted beam. The UEs are then allocated a sub-band for communication depending on where the UE is positioned in the squinted beam. The inter-element time delay difference also depends on the distance between the adjacent antenna elements, the longer distance, the longer inter-element time delay difference. The time delay differencebetween antenna signals of two elements that are not adjacent each other will then automatically be longer than the inter-element time delay difference. That the time delays are added to squint the beam so that any sub-band of the squinted beam reaches the plurality of UEs means that the sub-bands will be pointing in different directions, i.e., have their main lobes in different directions and all together the sub-bands will cover an area or transmission range in which the plurality of UEs are situated. The respective sub-band is allocated to the respective UE so that the allocated sub-band matches the direction between the network node and the respective UE in the squinted beam. In the transmitting 212 to the plurality of UEs 140, 145 or the receiving from the plurality of UEs 140, 145 of the antenna signals via the plurality of antenna element, the time delays are preferably applied when the antenna signals are in time domain.

[0043] According to an embodiment, the inter-element time delay difference is determined 206 to be at least one period time of a center frequency of the frequency band greater than any minimum inter-element time delay difference required for directing the beam in the first direction. The larger squint that you like to achieve, i.e., the larger angle that is needed by the squinted beam to reach the plurality of UEs with the squinted beam, the larger time delay difference is needed. The more period times of the center frequency of the frequency band the mutual time delay difference is, the greater the squint. Hereby, and by selecting the time delay difference from one time period of the center frequency and upwards, the network node can easily select the squint that is needed to reach the desired UEs.

[0044] According to another embodiment, the inter-element time delay difference is determined 206 based on a distance between neighboring antenna elements of the plurality of antenna elements, bandwidth of the frequency band and a requested transmission angle range or reception angle range for reaching the plurality of UEs. The inter-element time delay difference is determined based on the distance between neighboring antenna elements and the bandwidth in order to achieve a beam squint that would result in a beam covering the requested transmission angle range. According to an embodiment, the inter-element time delay difference may also be determined based on the center frequency of thefrequency band, i.e., the radio frequency. The distance between neighboring antenna elements is related to the center frequency, as it is normally selected based on the wavelength.

[0045] According to another embodiment, the antenna signals transmitted 212 to the plurality of UEs 140, 145 or received from the plurality of UEs 140, 145 are time-domain beamformed by individually controllable phase shifters.

[0046] According to another embodiment, the squint of the beam is formed in elevation direction. As mentioned above, the invention is especially useful in elevation direction where the UEs are not spread out as much as in azimuth direction.

[0047] According to yet another embodiment, the inter-element time delay difference is determined 206 so that time delay difference between the two of the plurality of antenna elements that are furthest away from one another is only a fraction of a duration of a cyclic prefix used for the antenna signals. By keeping the longest time delay difference between antenna signals of the network node considerably smaller than a time span, or duration, of the cyclic prefix, it can be assured that any problems with inter-symbol interference is avoided. Suitable values for the time delay difference between the two of the plurality of antenna elements that are furthest away from one another may be less than 50 % of the duration of the cyclic prefix, less than 30 % or less than 10 % of the duration of the cyclic prefix. If the delay plus other signal dispersion (e.g., from multipath characteristics of the radio channel) is greater that the cyclic prefix, inter-symbol interference may occur.

[0048] According to yet another embodiment, referring to figs.9 and 11, the plurality of antenna elements are divided into a plurality of antenna element groups 420, 422, 424, where each antenna element group comprises unique antenna elements Y and each group consisting of neighboring of the plurality of antenna elements. Each antenna element group 420, 422, 424 has a port 430, 432, 434 where signals in uplink are output for processing and signals in downlink are input for transmission, each antenna element group 420, 422, 424 thus having one portsignal P1, P2, P3. Further, for each antenna element group 420, 422, 424, the time delays τ according to the determined inter-element time delay difference are applied to the antenna signals of the antenna elements of that antenna element group 420, 422, 424, in uplink prior to combining the antenna signals from the antenna elements into the port signal P1, P2, P3, or, in downlink after splitting the port signal P1, P2, P3 to the antenna elements. Hereby a preferably same squinted beam pattern of each antenna element group is created. Further, according to the embodiment, in uplink, each of the port signals P1, P2, P3 are split into one port signal for each sub-band 440, 442, and for each sub-band set of split port signals P1-SB1, P2-SB1, P3-SB1; P1-SB2, P2-SB2, P3-SB2 phase shifts according to an inter-port phase shift difference are applied and the split and phase-shifted port signals of one sub-band 440, 442 are combined into a sub-band signal SB1, SB2. In downlink, for each sub-band 440, 442, the sub-band signal SB1, SB2 is split into one signal for each port, SB1-P1, SB1-P2, SB1-P3 ,SB2-P1, SB2-P2, SB2-P3, the phase shifts according to the inter-port phase shift difference is applied to each split sub- band signal and the split and phase shifted sub-band signals are fed to the ports 430, 432, 434 of each antenna element group 420, 422, 424, and at each port 430, 432, 434, the port set of split and phase shifted sub-band signals SB1-P1, SB2-P1, SB1-P2, SB2-P2, SB1-P3, SB2-P3 are combined into the port signals P1, P2, P3. Hereby, a beam directed to any of the plurality of UEs is formed. Inter-port phase shift difference (or inter-port phase difference, inter-antenna elelement group phase shift difference, inter-group phase shift difference) here means a difference in phase between the signals of the ports of neighboring antenna element groups. In the figures, only two time delays per group of three antenna elements, and only two phase shifters for the three ports are shown, since the desired differences in delay or phase between elements or ports can be created with one signal not being delayed or shifted.

[0049] The phase shift difference for one sub-band is to be selected so that it matches the direction of the squinted beam for the one sub-band. The resulting beam pattern will thereby be a combination of the squinted beam pattern of the antenna element groups and the mostly non-squinted beam pattern associated with the inter-antenna element group phase shift difference applied individually for each sub-band. When the inter-group phase shift difference for an individual sub-band is selected so as to correspond to a beam pattern direction which matches the direction of the squinted beam pattern of the groups for that sub- band, the patterns will add constructively but producing a resulting beam pattern that will be less frequency-dependent within the sub-band. By the solution of the above embodiment, different beams can be generated for the different sub-bands giving the beams varying elevation direction. Compared to a case when the same inter-element time delay difference is applied across all the plurality of antenna elements, an advantage is that total squint across the whole bandwidth will be approximately the same, but there will be considerably less squint within each sub- band, thus directing the sub-bands in different directions, but with a more constant direction within each sub-band. The selection of inter-group phase shift differences to achieve a particular beam pattern direction may be made in conventional ways known in the art of beam steering by phase shift of antenna signals to antenna elements.

[0050] According to another embodiment, the applying of time delays to the antenna signals according to the determined inter-element time delay difference is performed when the antenna signals are in baseband frequency. Applying the time delays in baseband frequency creates the requested beam squint. An advantage of implementing the delay in baseband is that the ratio between the required delay and the minimum time period of the signal is much smaller at baseband frequency than at RF frequency. This makes the delay implementation easier. Further, it may be easier to control a delay applied at baseband frequency than at RF. For example, at baseband, the delay can be applied in the digital domain.

[0051] According to yet another embodiment, the applying of time delays to the antenna signals according to the determined inter-element time delay difference isperformed when the antenna signals are in radio frequency. Applying the time delays in radio frequency (RF) creates the requested beam squint. As mentioned above, this may be a bit trickier than applying the time delays in baseband. However, for some architectures, i.e., implementations, applying in baseband may not fit the architecture, e.g., not be supported by ASICs. Further, it may also be more cost-efficient to apply the time delays in RF than in baseband. An inter- element time delay difference applied in RF can also create a nominal phase shift at the center frequency of the RF frequency band. However, this can be compensated for in the phase shifting.

[0052] In the following paragraph, different embodiments will be described, which embodiments can be combined with each other or with any of the other embodiments described in this disclosure. According to one embodiment, for the plurality of UEs, each residing in different directions from the plurality of antenna elements of the network node, controlling squint of a beam from the plurality of antenna elements so that for each of the plurality of UEs, the squinted beam is directed towards that UE for a particular sub-band of a communicated signal. According to another embodiment, the method further comprises simultaneously communicating with the two or more UEs through the beam, using for each UE a sub-band for which the squinted beam is directed towards that UE. According to another embodiment, directions to the plurality of UEs are determined and the squint is controlled based on the determined directions. According to another embodiment, the squint is controlled by applying inter-element time delays, e.g. true time delays, to achieve desired inter-element time delay differences. The inter-element time delays may be applied at baseband frequency. The inter- element time delays may be applied at radio frequency. The inter-element time delay differences are greater, or much greater than one or two periods of the center radio frequency. According to an embodiment, a method is provided comprising: determining two different directions to two different UEs, directing a beam to a direction being the average of the two different directions, for example by phase shift or true time delay, applying additional true time delay to squint the beam so that for opposite ends of a signal bandwidth, the beam is directed to thefirst and second of the two UEs, respectively. The latter meaning that a total squint angle is applied equal to or greater than the difference between the two directions.

[0053] Fig.3shows an AAS radio block chain diagram and various ways of doing beamforming and port expansion with such a radio block. One radio block chain comprises a Layer 1 beamforming (L1BF) unit 102, a Digital Front End (DFE) unit or chain 104, a transceiver (TRX) 106, an Analog beamforming (ABF) unit 108, a Power Amplifier / Low Noise Amplifier (PA / LNA) 110, a filter unit 112 and an antenna element 114. The #-numbers in the fig.3 symbolizes different kinds of beamforming. The uppermost part of fig.3 shows the radio block chain with the #-numbers below to show where in the radio block chain that kind of beamforming and thus port expansion takes place. The parts of fig.3 below the uppermost part show the expansion for the #Nth kind of beamforming, where N = 1, 2…5. #1 illustrates frequency domain beamforming, which is the most versatile kind of beamforming. However, as shown in the second uppermost part of fig.3, after “Using #1”, one signal is divided into X number of antenna signals, where X = the number of antenna elements in the network node or part of node, already after the L1BF unit 102. Consequently, sole frequency domain beamforming becomes costly as lots of resources are needed. When beamforming is only performed in the digital domain, the ABF unit 108 may be omitted, or the weights of the ABF may be set to 1. #2 and #3 illustrate two different kinds of time domain beamforming. Time domain digital beamforming can be made either per carrier as shown by #2 or for the full bandwidth late in the DFE unit 104 as shown by #3. #4 illustrates analog beamforming. Analog beamforming is equivalent to #3 from a beamforming point of view, but less flexible since no dedicated digital signal processing can be made per branch. On the other hand, as shown in the part of fig.3 after “Using #4” only one TRX 106 and ABF unit 108 is needed at analog beamforming whereas one TRX and ABF unit per antenna element is needed for the #3 kind of time domain beamforming. Finally, #5 shows an antenna subarray that uses a fixed connection of several antenna elements. Here no variable beamforming takes place within the subarray.

[0054] The invention will now be illustrated by an example. Fig.4 shows a large AAS antenna array 180 example with 24x16 dual polarized antenna sub- elements, each antenna sub-element being illustrated with an X. In this example, each antenna element 185, 186 comprises 3 sub-elements. Each antenna element is then illustrated with the three Xs that are within one rectangle of fig.4. Such an antenna element may also be called a sub-array. In this example, the antenna sub-elements are spaced in elevation direction with an element spacing z1 of 2 / 3 λ, i.e., 2 / 3 of the wavelength, i.e.2 / 3 * c / f0, where f0 is the center radio frequency and c is the speed of light. This means that the distance in elevation direction between two neighboring sub-arrays is 2 λ. The antenna elements are spaced apart in azimuth direction with element spacing y1 of 1 / 2 λ. Such spacing gives a grating lobe free service area of 30x120 degrees, if using full digital frequency domain beamforming. This requires 256 baseband ports (128 for each polarization direction), which becomes very complex if combined with a wide bandwidth. This would be like using the set-up of “Using #1” in fig.3 for all antenna elements, i.e., 256 parallel paths as in “Using #1”, in combination with #5 for the sub-elements.

[0055] To simplify the implementation of the example, we might consider doing 2x port expansion for the full bandwidth (BW), as in #3 or #4, i.e., full BW time domain beamforming or analog beamforming, shown in fig.3, combined with 2x port expansion per carrier in #2, i.e., per carrier time domain beamforming. Remaining #1 frequency domain beamforming now only needs to handle 64 ports.

[0056] Analog beamforming (#4) is normally implemented using phase shifters. For a person skilled in the art, it is well known how to derive phase shifts for individual antenna signals in order to create a planar wave or beam of the wireless signals sent from the antennas having a certain steering angle θ in relation to mechanical boresight, i.e., the normal of the antenna array. A signal to one antenna element is here called an antenna signal.

[0057] When having the same distance between the antenna elements in one direction, e.g., elevation direction in fig.4, such a planar wave with steering angle θ to boresight can be achieved by adding a phase shift ^^ between the antenna signals. As an alternative to adding phase shift ^^, a deliberate delay τ can be added between the antenna signals. The required delay becomes: ^^ =ௗ ^୧^ , where c = speed of light (1) This delay τ is frequency independent and the beam direction becomes the same independent of radio frequency used. If instead applying a phase shift ^^ between the antenna signals the required phase shift becomes ^^ = − ^^ ^^ = −௪ௗ ^୧୬ ఏ^ , where ω=2πf (2) In contrast to applying delay τ, the phase shift ^^ is only correct for one frequency since we compensate for a physical distance between the antenna elements with a phase shift. If the signal bandwidth is wide compared to the operating frequency, we get a beam squint, i.e., different portions of the frequency spectra points in different directions. Such a beam squint is proportional to the steering angle θ and fractional bandwidth of the signal. The beam squint or pointing error Δ ^^, can be calculated from Δ ^^ = arcsin ^^బ^ ^^ ^^ ^^ ^^^^ − ^^^(3) , where ^^^is the steeringfrequency. In other words, the pointing error will be largest for the frequency f of the bandwidth that is further away from the center frequency ^^^.

[0058] For example, and as shown in fig.5a, for beam steering angle θ = 60 degrees, 400 MHz BW and center frequency ^^^= 4GHz, a beam squint of around 4.3 degrees is generated if the phase shifts are designed for the center frequency ^^^. When using analog beamforming, the beam squint can be mitigated by using a true time delay as shown in equation (1). If using frequency domain beamforming, beam squint could easily be compensated by using frequency dependent phase shifts. The phase shifts are then a piecewise linearrepresentation of equation (1). Fig.5b shows for the same values as in fig.5a how the beam squint varies as a function of carrier frequency when using a beam defined for 4 GHz.

[0059] Equation (3) and figs.5a and 5b explain how beam squint typically occurs. Further, beam squint is generally seen as a disadvantage. However, in embodiments of this invention, we like to deliberately produce a squint. Embodiments of the invention provide a different mechanism, i.e., by adding a certain time delay, which can be used to produce a larger, easily controllable beam squint, which in turn enables serving UEs in quite different directions by serving them in different frequency regions. This can be seen as adding a large deliberate beam squint.

[0060] In the following it will be explained how such a beam squint can be generated. In this explanation, the beam squint generated is in the elevation direction, however, it is also applicable in the azimuth direction or any combination of azimuth and elevation direction.

[0061] Figs.6a and 6b show what happens to a transmitted beam in elevation direction when adding a frequency-varying phase shift to antenna signals of two independently phase-controlled antenna elements that are arranged spaced apart in the elevation direction. The antenna signals are here in radio frequency having a center carrier frequency (f0), where ω0=2πf0 and a bandwidth (BW). Fig.6a illustrates a frequency-varying inter-element phase difference between the two antenna signals. In this example, the phase of antenna signal 1, which is the signal to the first antenna element of the two independently controlled antenna elements is kept constant over the BW, whereas the phase of antenna signal 2, which is the signal to the second antenna element of the two independently controlled antenna elements, varies linearly over the BW, with antenna signal 1 and antenna signal 2 having the same phase at the center carrier frequency w0 and the largest difference at the lowest and highest frequencies. In other words, the inter-element phase difference between the two antenna signals varies linearly across the BW.Note that here, the phase varies e.g. linearly with the baseband frequency. Thus the linear variation is different from the linear variation of the equivalent phase shift caused by the time delay of eq. (1), where the fixed time delay causes a phase shift that varies linearly with the actual transmitted frequency and does not cause squint. Fig.6b shows the achieved beam in elevation direction when the two antenna signals are transmitted wirelessly from the two antenna elements, using the frequency-varying phase shift of fig.6a. The area 200 in fig.6b symbolized where in the elevation beam direction that the superposed beam will have its highest gain (within 1 dB). As shown, the highest gain beam direction varies in this case from - 15 degree at the lowest frequency to + 15 degrees at the highest frequency. So, hereby a beam squint is generated over the BW. One can say that the direction of the main lobe of the transmitted / received beam varies from -15 degrees to + 15 degrees elevation over the frequency. A basic idea of embodiments of the invention is to use such a frequency varying phase shift.

[0062] To understand how such a frequency-varying varying phase shift could be generated, a generic block diagram of thought radio transmitter with two ports, each port handling one antenna signal, is shown in fig.7. The purpose with this passage is to show the principle of generating a frequency-varying phase shift, according to embodiments. A transmitter is shown; however, it is equally applicable to a receiver, e.g.in the receive direction signals would flow in the reverse direction, power amplifiers PAs replaced by low noise amplifiers, LNA, and the signal generator replaced by baseband processing, as is known in the art. The radio transmitter comprises a common signal generator 302 that transmits a baseband (BB) signal defined as A*exp(j*ωBBt) into a first and a second port. The first port comprises a BB delay and phase shift unit 304, a BB to RF converter 308, in which the BB signal is mixed with an RF signal at the center carrier frequency f0, possibly from a local oscillator, an RF delay and phase shift unit 312, a power amplifier (PA) 316 and an antenna element or subarray 320. The BB to RF converter 308 may also be called an IQ-modulator. Likewise, the second port comprises a BB delay and phase shift unit 306, a BB to RF converter 310, an RFdelay and phase shift unit 314, a power amplifier (PA) 318 and an antenna element or subarray 322. Usually in practice, the first and second ports will have either a BB delay and phase shift unit or an RF delay and phase shift unit, or a BB delay unit and an RF phase shift unit or an RF delay unit and a BB phase shift unit. They are alternative implementations. In the general case, each antenna branch could have multiple beams and will transmit wideband signals. But for this analysis one frequency at a time is considered. The signals to be transmitted can be delayed and phase shifted both at BB and at RF.

[0063] Fig.7 may depict the two antenna branches of an AAS transmitter. Typically, f0is placed in the center of the transmit band and the BB frequency range spans +-BW / 2 from f0, where BW is the transmitter signal bandwidth. The complex BB signal is upconverted to RF in the BB to RF converter 308, 310. In one embodiment, it is assumed that phase shift is applied to the RF signal, given the notion ^^ோி. An alternative implementation with equal result, and also shown in fig.7 is to shift the phase of the Local Oscillator (LO) signal exp( ^^ ^^^^^) in the up- converter as: exp ( ^^ ^^^^^ + ^^).

[0064] The signals at the two antennas now become: ^^ sin( ^^^( ^^ − ^^ோி^) + ^^^^( ^^ − ^^ோி^− ^^^^^) + ^^ோி^+ ^^^^^) ^^ sin(^^^(^^ − ^^ோிଶ)+ ^^^^(^^ − ^^ோிଶ− ^^^^ଶ)+ ^^ோிଶ+ ^^^^ଶ)(4) From Equations (4) it is observed that to generate a phase shift which is BB frequency dependent, either a delay at RF, ^^ோி, or a delay at BB, ^^^^, can be added. Since ^^ோிalso generates a big frequency independent phase shift, ^^^^is preferably used. Normal beamforming in order to direct the beam can be added on top of this delay by applying BB phase shift, RF phase shift or RF delay.

[0065] In the following, to apply a time delay at baseband is mathematically described. Starting from (4), looking at one equation for the signal to be transmitted from one antenna element (or subarray): ^^ sin( ^^^( ^^ − ^^ோி) + ^^^^( ^^ − ^^ோி− ^^^^) + ^^ோி+ ^^^^)and, for simplicity, looking at the case where no fixed phase shift is added: ^^^^ ^^= 0 , ^^^^ ^^= 0we as argument to the sine function ^^0(^^ − ^^^^ ^^)+ ^^^^ ^^(^^ − ^^^^ ^^− ^^^^ ^^)^ ^^ ^^ ^^ ^^ ^^ ^^ ^to a ^^0+ ^^^^ ^^with a time delay ( ^^^^ ^^+ ( ^^^^ ^^ / ( ^^0+ ^^^^ ^^) ^^^^ ^^) Given that ^^ ≫ ^^^^ ^^, thean delay in the upconverted to frequency. ^^^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^= ( ^^^^ ^^ / ^^0) ^^^^ ^^From (1), for a delay ^^ a ^^ =ௗ ^^^ ఏ^ ^ ^^ ^^ ^^ ^^ =^ఛௗ (valid for^ఛௗ ≤ 1) For a baseband frequency of + / - 200 MHz, ( ^^^^ ^^= 2 ^^ 200 ^^ + 6), no delay at RF ( ^^^^ ^^= 0), and a wanted beam angle of 10 degrees at ^^0+ ^^^^ ^^, and an example subarray spacing of 2 ^^0= 2 ∙ ^^ ∙ 2 ^^ / ^^0:( ^^^^ ^^ / ^^0) ^^^^ ^^=^2 ∙ ^^ ∙ 2 ^^ ^^0^ ∙ ^^ ^^ ^^ 10 / ^^ ^ ^^delay difference of 1.7 ns with an element spacing of 2 ^^0would result in the wanted beam angle of 10 degrees. The above shows how the inter-element time delay / time delay difference to be applied in baseband may be determined in order to achieve the desired squint.

[0066] In the following, to apply a time delay at RF is mathematically described. From the above equations, the equivalent time delay when delaying at RF and baseband is ( ^^^^ ^^+ ( ^^^^ ^^ / ( ^^0+ ^^^^ ^^) ^^^^ ^^), as described in Equation (5) above. The corresponding equivalent phase shift at angular frequency ^^0+ ^^^^ ^^is ( ^^0+ ^^^^ ^^) ( ^^^^ ^^+ ( ^^^^ ^^ / ( ^^0+ ^^^^ ^^) ^^^^ ^^) = ( ^^0+ ^^^^ ^^) ^^^^ ^^+ ( ^^^^ ^^^^^^ ^^) ,term at RF term is caused by the delay at baseband. What is then the effect of a delay at RF? Let ^^^^ ^^= ^^0+ ^^ ∙ 2 ^^ / ^^0, ^^0< 2 ^^ / ^^^, , where n is an integer and ^^ ≥ 1 ( ^^^^ ^^= 0 as delay at RF is sought for) (i.e. ^^^^ ^^is ^^0plus n full periods ^^0of ^^0, thus at ^^0having the same effect on the beam as only ^^0) At ^^0the phase shift caused by delay at RF is ^^0^^^^ ^^= ^^0( ^^0+ ^^ ∙ 2 ^^ ^^0) = ^^0^^0+ ^^ ∙ 2 ^^term second term has no effect on the beam.At ^^0+ ^^^^ ^^the phase shift caused by delay at RF is ( ^^ + ^ 2 ^^ 0 ^^^ ^^) ( ^^0+ ^^ ∙ 2 ^^ / ^^0) = ( ^^0+ ^^^^ ^^) ^^0+ ( ^^0+ ^^^^ ^^) ^^ ∙ ^^0=∙ 2 ^^ + ^^^^ ^^^^ ^^0The first term directs the beam but does not produce any squint as it is proportional to the transmitted / received frequency. The second term has no effect on the beam. The third term, i.e. ^^^^ ^^^^ ^^0, provides a phase shift proportional to ^^^^ ^^and hence produces squint. Selecting a higher value of n produces more squint for a given ^^^^ ^^. If producing squint by delay at baseband rather than at RF, selecting ^^^^ ^^= ^^ ^^^will produce the same squint as selecting ^^^^ ^^= ^^ ^^0for delaying at RF. If a non-integer multiple of ^^0is applied at RF delay, this will have the effect of applying the nearest lower integer factor of ^^0and adding the remainder to ^^0, i.e. for a selected ^^0of zero the beam will nevertheless be directed away from boresight at ^^0. A condition for squint to occur as a result of delaying at RF seems to be that the delay time is greater than the period time at ^^0, i.e. ^^^^ ^^≥ ^^0. As shown above, there are two ways of producing the desired squint, either by delay at baseband, or by delay at RF, where ^^^^ ^^≥ ^^. The above shows how the inter-element time delay / time delay difference to be applied at RF may be determined in order to achieve the desired squint,

[0067] By applying a delay in between adjacent ports, i.e., a time delay difference between antenna signals of adjacent ports, a frequency varying phase difference is generated as depicted in the fig.6a. When combining two such adjacent antenna signals, a combined beam with varying direction as depicted in fig.6aError! Reference source not found. is achieved.

[0068] From Equation (4) it can be derived that when adding a delay ^^ோி^between adjacent ports, a phase shift proportional to ^^^is generated, whereas when adding a delay ^^^^between adjacent ports, a phase shift proportional to ^^^^is generated. Thus, as long as ^^^≫ ^^^^, the required delay to generate afrequency dependent squint is much longer than the delay needed to phase shift, i.e., direct, the full beam.

[0069] Taking the antenna array of fig.4 as an example, one antenna element 185 has an elevation beamwidth of 30 degrees (and an azimuth beamwidth of 120 degrees). If adding antenna signals of two adjacent antenna elements 185, 186,, the beamwidth is halved to 15 degrees. By adding a phase shift in between the two antenna elements using an added delay ^^ between the ports as described above results in a varying beam direction versus frequency. The added delay between the ports may be called an inter-element time delay difference.

[0070] The required inter-element time delay difference depends on the bandwidth of the signal and a distance, aka spacing, between neighboring, aka adjacent, antenna elements. Assuming we want to change the beam direction over the frequency range, i.e. the BW, from -10 degrees to +10 degrees, in order to achieve a desired service area of -15 to + 15 degrees, see fig.6b, and the elevation spacing between two adjacent antenna elements is 2 λ (at ^^^), then when using Equation (2) we get: ∆ ^^ = + −௪ௗ ^= ±4 ^^ sin 10గ^ ^଼^ = ±2.2 ^^ ^^ ^^ ^^ ^^ ^^ ^^ (5)Equation (4): ^^^^^= ௪ಳಳ=ସ.ସଶగସ^^^^ = 1.7 ^^ ^^ (6) This will add somesignal but if the added delay is much smaller than a cyclic prefix (CP) used, any problems occuring due to inter-symbol interference should be avoided. For NR in FR1, Sub-Carrier Spacing (SCS) = 30kHz and CP = 2.4us.

[0071] On top of this delay-based beamforming, the center of the combined beam can be shifted by adding a phase shift in between the two antenna elements, as in normal beamforming. Any such phase / delay combination can be used to form a combined beam with arbitrary center and slope / squint.

[0072] The delay-based beamforming has now combined two antenna signals and formed the elevation service area as in fig.6b. Subsequent beamforming for a plurality of antenna elements can form arbitrarily elevation beams within the elevation range. UEs in different elevation directions could be simultaneously covered using different portions of the frequency range. In other words, UEs in different elevation directions are allocated a sub-band of the frequency range for communication, depending on the UE´s position in the squinted beam / slope of fig. 6b.

[0073] Figs.8a and 8b show an example of an embodiment where port expansion in downlink (DL) and port reduction in uplink (UL) is performed using e.g., time domain digital beamforming (#2 in fig.3), in this example with the factor of N. This time-domain digital beamforming (#2) is performed in addition to the delay-based beamforming discussed above. The port reduction / expansion is performed by adding a phase shift per component carrier. In order to form the four final elevation beams 402, 404, 406, 408 as shown on fig.8b, another elevation port reduction / expansion is needed. This could be done either by continuing using time domain digital beamforming (#2) or by using frequency domain beamforming (#1 in fig.3). If using digital beamforming, i.e. #1 or #2 beamforming, arbitrary beams can be generated within the elevation service area of fig.6b. When doing port reduction in UL, the received antenna signals are combined. When doing port expansion in DL, signals are divided into the antenna signals that are transmitted from the antenna elements.

[0074] An embodiment for achieve the port reduction / port expansion discussed will now be described referring to figs.9-12 where port reduction / expansion is by a factor of 3. In general, with N elements per antenna element group a factor of N is achieved. Two sub-bands and three antenna element groups are shown, but a different number of either can be used in the same manner and by the same principle as shown here. The symbols φ and τ in the figs 9-12 merely signify that phase shift / time delay is performed, they do not indicate that phase shift or time delay necessarily is the same.

[0075] Fig.9 describes port expansion in downlink. The plurality of antenna elements of the network node 130 of fig.1 are divided into a plurality of antenna element groups 420, 422, 424, here exemplified by three such antenna element groups 420, 422, 424, where each group having three antenna elements Y. Each antenna element group 420, 422, 424 comprises unique antenna elements Y of the plurality of antenna elements and each group 420, 422, 424 consists of neighboring of the plurality of antenna elements. Each antenna element group 420, 422, 424 has a port 430, 432, 434 where signals in downlink are input for transmission. Each antenna element group 420, 422, 424 thus has one port signal P1, P2, P3.

[0076] Further, in downlink, a sub-band signal SB1, SB2 on each sub-band 440, 442 is split into one signal for each port, SB1-P1, SB1-P2, SB1-P3,SB2-P1, SB2-P2, SB2-P3. Then phase shifts φ according to an inter-port phase shift difference are applied to each split sub-band signal and the split and phase shifted sub-band signals are fed to the ports 430, 432, 434 of each antenna element group 420, 422, 424. Then at each port 430, 432, 434, the port set of split and phase shifted sub-band signals SB1-P1, SB2-P1, SB1-P2, SB2-P2, SB1-P3, SB2-P3 are combined into the port signals P1, P2, P3. Then, for each antenna element group 420, 422, 424, the port signals P1, P2, P3 are split into the antenna signals and fed towards the antenna elements Y. Before the antenna signals reaches the antenna elements Y, the time delaysτaccording to the determined inter-element time delay difference are applied to the antenna signals of the antenna elements of the respective antenna element group 420, 422, 424.

[0077] Whereas fig.9 shows a schematic diagram of the embodiment for downlink communication, a diagram with further details of a typical implementation is shown in fig.10, in which like numerals as in fig.9 refer to like features. In fig.1, the sub-band signals SB1, SB2 are split, phase shifted and combined into the port signals P1, P2, P3 in digital baseband. The respective port signal P1, P2, P3 is then converted to analog in a digital to analog converter (DAC), upconverted from baseband (BB) to radio frequency (RF) and amplified in a power amplifier (PA) before splitting, delaying (τ)and transmitting through the antenna elements Y ofthe respective antenna element group 420, 422, 424. In the fig.10, the DAC, Upconverter and PA are in one and the same box 450, 452, 454, one box per port 430, 432, 434 in order to simplify the figure.

[0078] Fig.11 describes port reduction in uplink. The plurality of antenna elements of the network node 130 of fig.1 are divided into a plurality of antenna element groups 420, 422, 424, here exemplified by three such antenna element groups 420, 422, 424, where each group has three antenna elements Y. Each antenna element group 420, 422, 424 comprises unique antenna elements Y of the plurality of antenna elements and each group 420, 422, 424 consists of neighboring of the plurality of antenna elements. Each antenna element group 420, 422, 424 has one port 430, 432, 434 where signals in uplink are fed from the antenna element group for further processing. Each antenna element group 420, 422, 424 thus has one port signal P1, P2, P3.

[0079] Further, for each antenna element group 420, 422, 424, the time delays τ according to the determined inter-element time delay difference are applied to the antenna signals of the antenna elements of that antenna element group 420, 422, 424, prior to combining the antenna signals from the antenna elements into the port signal P1, P2, P3. Each of the port signals P1, P2, P3 are then split into one port signal for each sub-band 440, 442, and for each sub-band set of split port signals P1-SB1, P2-SB1, P3-SB1; P1-SB2, P2-SB2, P3-SB2, phase shifts φ according to an inter-port phase shift difference are applied and the split and phase-shifted port signals of one sub-band 440, 442 are combined into a sub-band signal SB1, SB2.

[0080] Whereas fig.11 shows a schematic diagram of the embodiment for uplink communication, a diagram with further details of a typical implementation for uplink is shown in fig.12, in which like numerals as in fig.11 refers to like features. The combined signal from the antenna elements Y of each antenna element group 420, 422, 424 is fed to a low noise amplifier (LNA), down-converted from RF to BB and digitalized in an analog to digital converter (ADC), thereby creating digital baseband port signals P1, P2, P3. The splitting, phase shifting and combining of the port signals into the sub-band signals SB1 and SB2 then takes place in digitalbaseband. In the fig.12, the LNA, Downconverter and ADC are in one and the same box 460, 462, 464, one box per port 430, 432, 434 in order to simplify the figure.

[0081] In the following, another aspect of embodiments is described, starting with a description of a problem solved by some of the embodiments. Assuming analog beamforming is used to combine two adjacent elevation antenna elements, such as the antenna elements 185, 186 in the example of fig.4, some beamforming limitations are introduced, since the analog beamforming can only produce one beam with full antenna gain at a time, i.e., at an OFDM symbol. Such combining is depicted in fig.13. Here two ports are combined by analog beamforming and either Elevation beam #1 or Elevation beam #2 of fig.13 can be produced at a given time instant. Here it is assumed that true-time delay RF beamforming is used, to avoid any not-requested beam squint. Using further digital port combining, arbitrary beams can be formed within either Elevation beam #1 or Elevation beam #2.

[0082] UEs can still be located by using the standardized 3GPP procedure. The network node can send Synchronization Signal Blocks (SSB) in multiple directions and having each UE select the SSB that suits the certain UE best. When a UE has communication needs, the elevation beam can be directed in the correct direction. However if multiple UEs within both Elevation beam #1501 and Elevation beam #2502 want to be served at the same time, they need to be served round-robin. This adds latency and is particularly troublesome in UL, where the capacity is lowered since only a fraction of the UEs can be active at a time and thus aggregated UL power is reduced.

[0083] If instead implementing any of the above embodiments, beam shapes can be produced for example either as in fig.8b or as in fig.13. By using the beam shape of fig.8b and sending different SSBs at different frequencies the elevation direction of the UEs can be located. In general, with the embodiments disclosed herein, by sending different SSBs atdifferent frequencies or sub-bands in a squinted beam, the resources such as sub- bands or beam patterns to be used for communication with the UE may be determined similar to the 3GPP standardized procedure, but where the PRACH resource selected by the UE is instead used to identify the sub-band whose direction in the squinted beam best matches the direction of the UE. The identified sub-band and the squint of the squinted beam then in turn identifies the direction. When doing beam sweeping, a squint may be selected / generated to be large enough to cover the whole of a desired service area (e.g. in elevation; then a single beam pattern may be used in elevation for SSB and may suffice to determine the elevation direction of any UE within the desired service area). Once directions to UEs are known, a different squint may be selected to match the directions of particular UEs for which communication (in particular in uplink) is desired at the moment. The squint may be optimized to cover an optimal set of UEs (e.g. comprising UEs having the largest communication need, comprising a maximum number of UEs, comprising UEs having the highest priority, etc). Now, if multiple UEs in different elevation directions have traffic needs, the beam shape of fig.8b can be used and all UEs with traffic needs within the elevation area of fig.8b (-15 to +15 degrees) can be addressed at the same time. This reduces latency and boosts primarily UL capacity. UL is often power limited, meaning that a UE cannot utilize more than one CC efficiently.

[0084] The embodiments described above are mostly described in the context of beamforming in the elevation direction. However, the embodiments disclosed herein are also applicable to beamforming in azimuth direction or in a combination of elevation and azimuth direction. Further, the embodiments are applicable to downlink communication as well as to uplink communication. Where different features are disclosed in different embodiments, they may be combined with one another. The methods disclosed may be implemented in a network node. Such network node may be controlled by a computer program implementing the methods. The computer program may be stored on a carrier.

[0085] Fig.14, in conjunction with fig.1, discloses a network node 130 configured to operate in a wireless communication network 100, and configured for beamformed communication with a plurality of User Equipment, UE, 140, 145. The network node 130 comprises a plurality of antenna elements, where each antenna element is arranged for handling an antenna signal to be transmitted or received over a frequency band comprising a plurality of sub-bands. The network node 130 comprises a processing circuitry 603 and a memory 604. Said memory contains instructions executable by said processing circuitry, whereby the network node 130 is operative for determining, depending on directions between the network node 130 and each of the plurality of UEs 140, 145, a first direction in which to direct a beam formed by transmitting or receiving the antenna signals by the plurality of antenna elements, and obtaining phase shifts or delays to add to the antenna signals for the beam to get the determined first direction. The network node 130 is further operative for determining an inter-element time delay difference for controlling time delays to add to the antenna signals so that the antenna signals of neighboring of the plurality of antenna elements are mutually delayed with the inter-element time delay difference, the time delays being added in order to squint the beam formed by the transmitted antenna signals or the received antenna signals so that any sub-band of the squinted beam, when the squinted beam is directed in the first direction, reaches the plurality of UEs 140, 145. The network node 130 is further operative for determining, based on information on the directions between the network node 130 and each of the plurality of UEs 140, 145 and the determined first direction, to allocate to each of the plurality of UEs 140, 145, a sub-band of the plurality of sub-bands for communication with the network node 130, wherein the respective sub-band is allocated depending on a position of the direction between the network node and the respective UE 140 in the squinted beam arising from application of the determined inter-element time delay difference to the antenna signals. The network node 130 is further operative for sending an instruction to each of the plurality of UEs 140, 145 to use the allocated sub-band for communication with the network node 130, and transmitting to the plurality of UEs 140, 145 or receiving from the plurality of UEs 140, 145, the antenna signals via the plurality of antennaelements applying time delays to the antenna signals according to the determined inter-element time delay difference and applying the determined phase shifts or delays hereby achieving the squinted beam directed in the first direction.

[0086] According to an embodiment, the network node 130 is operative for the determining of the inter-element time delay difference to be at least one period time of a center frequency of the frequency band greater than any minimum inter- element time delay difference required to direct the beam in the first direction.

[0087] According to another embodiment, the network node 130 is operative for the determining of the inter-element time delay difference based on a distance between neighboring antenna elements of the plurality of antenna elements, bandwidth of the frequency band and a requested transmission angle range or reception angle range.

[0088] According to another embodiment, the network node 130 is operative for time-domain beamforming the antenna signals transmitted to the plurality of UEs or received from the plurality of UEs 140, 145 by individually controllable phase shifters.

[0089] According to another embodiment, the network node 130 is operative for forming the squint of the beam in elevation direction.

[0090] According to yet another embodiment, the network node 130 is operative for determining the inter-element time delay difference so that time delay difference between the two of the plurality of antenna elements that are furthest away from one another is only a fraction of a duration of a cyclic prefix used for the antenna signals.

[0091] According to yet another embodiment, the plurality of antenna elements are divided into a plurality of antenna element groups 420, 422, 424, each antenna element group comprising unique antenna elements Y and each group consisting of neighboring of the plurality of antenna elements. Further, each antenna element group 420, 422, 424 has a port 430, 432, 434 where signals in uplink are output for processing and signals in downlink are input for transmission, each antennaelement group 420, 422, 424 thus having one port signal P1, P2, P3. Further, the network node is operative for, for each antenna element group 420, 422, 424, applying the time delays according to the determined inter-element time delay difference to the antenna signals of the antenna elements of that antenna element group 420, 422, 424, in uplink prior to combining the antenna signals from the antenna elements into the port signal P1, P2, P3, or, in downlink after splitting the port signal P1, P2, P3 to the antenna elements. Further, in uplink, the network node is operative for splitting each of the port signals P1, P2, P3 into one port signal for each sub-band 440, 442, and for each sub-band set of split port signals P1-SB1, P2- SB1, P3-SB1; P1-SB2, P2-SB2, P3-SB2, applying phase shifts according to an inter-port phase shift difference and combining the split and phase-shifted port signals of one sub-band 440, 442 into a sub-band signal SB1, SB2. Further, in downlink, the network node is operative for, for each sub-band 440, 442, splitting the sub-band signal SB1, SB2 into one signal for each port SB1-P1, SB1-P2, SB1-P3 ,SB2-P1, SB2-P2, SB2-P3, applying the phase shifts according to the inter-port phase shift difference to each split sub-band signal and feeding the split and phase shifted sub-band signals to the ports 430, 432, 434 of each antenna element group 420, 422, 424, and combining, at each port 430, 432, 434, the port set of split and phase shifted sub-band signals SB1-P1, SB2-P1, SB1-P2, SB2-P2, SB1-P3, SB2-P3 into the port signals P1, P2, P3.

[0092] According to yet another embodiment, the network node 130 is operative for the applying of time delays to the antenna signals according to the determined inter-element time delay difference when the antenna signals are in baseband frequency.

[0093] According to yet another embodiment, the network node 130 is operative for the applying of time delays to the antenna signals according to the determined inter-element time delay difference when the antenna signals are in radio frequency.

[0094] According to other embodiments, the network node 130 may further comprise a communication unit 602, which may be considered to compriseconventional means for wireless communication with the UEs 140, 145, such as a transceiver for wireless transmission and reception of signals in the communication network. The communication unit 602 may also comprise conventional means for communication with other network nodes of the wireless communication network 100. The instructions executable by said processing circuitry 603 may be arranged as a computer program 605 stored e.g., in said memory 604. The processing circuitry 603 and the memory 604 may be arranged in a sub-arrangement 601. The sub-arrangement 601 may be a micro-processor and adequate software and storage therefore, a Programmable Logic Device, PLD, or other electronic component(s) / processing circuit(s) configured to perform the methods mentioned above. The processing circuitry 603 may comprise one or more programmable processor, application-specific integrated circuits, field programmable gate arrays or combinations of these adapted to execute instructions.

[0095] The computer program 605 may be arranged such that when its instructions are run in the processing circuitry 603, the instructions cause the network node 130 to perform the steps described in any of the described embodiments of the network node 130 and its method. The computer program 605 may be carried by a computer program product connectable to the processing circuitry 603. The computer program product may be the memory 604, or at least arranged in the memory. The computer program product may be called a computer-readable storage medium 606. The memory 604 may be realized as for example a Random-access memory (RAM), Read-Only Memory (ROM) or an Electrical Erasable Programmable ROM (EEPROM). In some embodiments, a carrier may contain the computer program 605. The carrier may be one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electric signal, a radio signal, a microwave signal, or computer readable storage medium. The computer-readable storage medium 606 may be e.g., a CD, DVD or flash memory, from which the program could be downloaded into the memory 604. Alternatively, the computer program 605 may be stored on a server or any other entity to which the network node 130 has access via thecommunication unit 602. The computer program 605 may then be downloaded from the server into the memory 604.

[0096] Although the description above contains a plurality of specificities, these should not be construed as limiting the scope of the concept described herein but as merely providing illustrations of some exemplifying embodiments of the described concept. It will be appreciated that the scope of the presently described concept fully encompasses other embodiments which may become obvious to those skilled in the art, and that the scope of the presently described concept is accordingly not to be limited. Reference to an element in the singular is not intended to mean "one and only one" unless explicitly so stated, but rather "one or more." All structural and functional equivalents to the elements of the above- described embodiments that are known to those of ordinary skill in the art are expressly incorporated herein by reference and are intended to be encompassed hereby. Moreover, it is not necessary for an apparatus or method to address each and every problem sought to be solved by the presently described concept, for it to be encompassed hereby. In the exemplary figures, a broken line generally signifies that the feature within the broken line is optional.

Claims

CLAIMS 1. A method performed by a network node (130) of a wireless communication network (100) for beamformed communication with a plurality of User Equipment, UE, (140, 145), the network node (130) comprising a plurality of antenna elements, each antenna element arranged for handling an antenna signal to be transmitted or received over a frequency band comprising a plurality of sub- bands, the method comprising: determining (202), depending on directions between the network node (130) and each of the plurality of UEs (140, 145), a first direction in which to direct a beam formed by transmitting or receiving the antenna signals by the plurality of antenna elements; obtaining (204) phase shifts or delays to add to the antenna signals for the beam to get the determined (202) first direction; determining (206) an inter-element time delay difference for controlling time delays to add to the antenna signals so that the antenna signals of neighboring of the plurality of antenna elements are mutually delayed with the inter-element time delay difference, the time delays being added in order to squint the beam formed by the transmitted antenna signals or the received antenna signals so that any sub-band of the squinted beam, when the squinted beam is directed in the first direction, reaches the plurality of UEs (140, 145); determining (208), based on information on the directions between the network node (130) and each of the plurality of UEs (140, 145) and the determined first direction, to allocate to each of the plurality of UEs (140, 145), a sub-band of the plurality of sub-bands for communication with the network node (130), wherein the respective sub-band is allocated depending on a position of the direction between the network node and the respective UE (140, 145) in the squinted beam arising from application of the determined inter-element time delay difference to the antenna signals; sending (210) an instruction to each of the plurality of UEs (140, 145) to use the allocated sub-band for communication with the network node (130), and transmitting (212) to the plurality of UEs (140, 145) or receiving from the plurality of UEs (140, 145), the antenna signals via the plurality of antennaelements applying time delays to the antenna signals according to the determined inter-element time delay difference and applying the determined phase shifts or delays hereby achieving the squinted beam directed in the first direction.

2. Method according to claim 1, wherein the inter-element time delay difference is determined (206) to be at least one period time of a center frequency of the frequency band greater than any minimum inter-element time delay difference required to direct the beam in the first direction.

3. Method according to claim 1 or 2, wherein the inter-element time delay difference is determined (206) based on a distance between neighboring antenna elements of the plurality of antenna elements, bandwidth of the frequency band and a requested transmission angle range or reception angle range.

4. Method according to any of the preceding claims, wherein the antenna signals transmitted (212) to the plurality of UEs or received from the plurality of UEs (140, 145) are time-domain beamformed by individually controllable phase shifters.

5. Method according to any of the preceding claims, wherein the squint of the beam is formed in elevation direction.

6. Method according to any of the preceding claims, wherein the inter- element time delay difference is determined (206) so that time delay difference between the two of the plurality of antenna elements that are furthest away from one another is only a fraction of a duration of a cyclic prefix used for the antenna signals.

7. Method according to any of the preceding claims, wherein the plurality of antenna elements are divided into a plurality of antenna element groups (420, 422, 424), each antenna element group comprising unique antenna elements (Y) and each group consisting of neighboring of the plurality of antenna elements, wherein each antenna element group (420, 422, 424) has a port (430, 432, 434) where signals in uplink are output for processing and signals in downlink are input for transmission, each antenna element group (420, 422, 424) thus having one portsignal (P1, P2, P3), wherein, for each antenna element group (420, 422, 424), the time delays according to the determined inter-element time delay difference are applied to the antenna signals of the antenna elements of that antenna element group (420, 422, 424), in uplink prior to combining the antenna signals from the antenna elements into the port signal (P1, P2, P3), or, in downlink after splitting the port signal (P1, P2, P3) to the antenna elements, wherein in uplink, each of the port signals (P1, P2, P3) are split into one port signal for each sub-band (440, 442), and for each sub-band set of split port signals (P1-SB1, P2-SB1, P3-SB1; P1-SB2, P2-SB2, P3-SB2), phase shifts according to an inter-port phase shift difference are applied and the split and phase-shifted port signals of one sub-band (440, 442) are combined into a sub-band signal (SB1, SB2), and / or in downlink, for each sub-band (440, 442), the sub-band signal (SB1, SB2) is split into one signal for each port (SB1-P1, SB1-P2, SB1-P3 ,SB2-P1, SB2-P2, SB2-P3), the phase shifts according to the inter-port phase shift difference is applied to each split sub-band signal and the split and phase shifted sub-band signals are fed to the ports (430, 432, 434) of each antenna element group (420, 422, 424), and at each port (430, 432, 434), the port set of split and phase shifted sub-band signals (SB1-P1, SB2-P1, SB1-P2, SB2-P2, SB1-P3, SB2-P3) are combined into the port signals (P1, P2, P3).

8. Method according to any of the preceding claims, wherein the applying of time delays to the antenna signals according to the determined inter-element time delay difference is performed when the antenna signals are in baseband frequency.

9. Method according to any of claims 1-7, wherein the applying of time delays to the antenna signals according to the determined inter-element time delay difference is performed when the antenna signals are in radio frequency.

10. A network node (130) configured to operate in a wireless communication network (100), and configured for beamformed communication with a plurality of User Equipment, UE, (140, 145), the network node (130) comprising a plurality of antenna elements, each antenna element arranged for handling an antenna signalto be transmitted or received over a frequency band comprising a plurality of sub- bands, the network node (130) comprising a processing circuitry (603) and a memory (604), said memory containing instructions executable by said processing circuitry, whereby the network node (130) is operative for: determining, depending on directions between the network node (130) and each of the plurality of UEs (140, 145), a first direction in which to direct a beam formed by transmitting or receiving the antenna signals by the plurality of antenna elements; obtaining phase shifts or delays to add to the antenna signals for the beam to get the determined first direction; determining an inter-element time delay difference for controlling time delays to add to the antenna signals so that the antenna signals of neighboring of the plurality of antenna elements are mutually delayed with the inter-element time delay difference, the time delays being added in order to squint the beam formed by the transmitted antenna signals or the received antenna signals so that any sub-band of the squinted beam, when the squinted beam is directed in the first direction, reaches the plurality of UEs (140, 145); determining, based on information on the directions between the network node (130) and each of the plurality of UEs (140, 145) and the determined first direction, to allocate to each of the plurality of UEs (140, 145), a sub-band of the plurality of sub-bands for communication with the network node (130), wherein the respective sub-band is allocated depending on a position of the direction between the network node and the respective UE (140) in the squinted beam arising from application of the determined inter-element time delay difference to the antenna signals; sending an instruction to each of the plurality of UEs (140, 145) to use the allocated sub-band for communication with the network node (130), and transmitting to the plurality of UEs (140, 145) or receiving from the plurality of UEs (140, 145), the antenna signals via the plurality of antenna elements applying time delays to the antenna signals according to the determined inter-element time delay difference and applying the determined phase shifts or delays hereby achieving the squinted beam directed in the first direction.

11. Network node (130) according to claim 10, operative for the determining of the inter-element time delay difference to be at least one period time of a center frequency of the frequency band greater than any minimum inter-element time delay difference required to direct the beam in the first direction.

12. Network node (130) according to claim 10 or 11, operative for the determining of the inter-element time delay difference based on a distance between neighboring antenna elements of the plurality of antenna elements, bandwidth of the frequency band and a requested transmission angle range or reception angle range.

13. Network node (130) according to any of claims 10-12, operative for time- domain beamforming the antenna signals transmitted to the plurality of UEs or received from the plurality of UEs (140, 145) by individually controllable phase shifters.

14. Network node (130) according to any of claims 10-13, operative for forming the squint of the beam in elevation direction.

15. Network node (130) according to any of claims 10-14, operative for determining the inter-element time delay difference so that time delay difference between the two of the plurality of antenna elements that are furthest away from one another is only a fraction of a duration of a cyclic prefix used for the antenna signals.

16. Network node (130) according to any of claims 10-15, wherein the plurality of antenna elements are divided into a plurality of antenna element groups (420, 422, 424), each antenna element group comprising unique antenna elements (Y) and each group consisting of neighboring of the plurality of antenna elements, wherein each antenna element group (420, 422, 424) has a port (430, 432, 434) where signals in uplink are output for processing and signals in downlink are input for transmission, each antenna element group (420, 422, 424) thus having one port signal (P1, P2, P3), wherein, the network node is operative for, for each antenna element group (420, 422, 424), applying the time delays accordingto the determined inter-element time delay difference to the antenna signals of the antenna elements of that antenna element group (420, 422, 424), in uplink prior to combining the antenna signals from the antenna elements into the port signal (P1, P2, P3), or, in downlink after splitting the port signal (P1, P2, P3) to the antenna elements, wherein in uplink, the network node is operative for splitting each of the port signals (P1, P2, P3) into one port signal for each sub-band (440, 442), and for each sub-band set of split port signals (P1-SB1, P2-SB1, P3-SB1; P1-SB2, P2-SB2, P3-SB2), applying phase shifts according to an inter-port phase shift difference and combining the split and phase-shifted port signals of one sub-band (440, 442) into a sub-band signal (SB1, SB2), and / or in downlink, the network node is operative for, for each sub-band (440, 442), splitting the sub-band signal (SB1, SB2) into one signal for each port (SB1-P1, SB1-P2, SB1-P3 ,SB2-P1, SB2-P2, SB2-P3), applying the phase shifts according to the inter-port phase shift difference to each split sub-band signal and feeding the split and phase shifted sub-band signals to the ports (430, 432, 434) of each antenna element group (420, 422, 424), and combining, at each port (430, 432, 434), the port set of split and phase shifted sub-band signals (SB1-P1, SB2-P1, SB1- P2, SB2-P2, SB1-P3, SB2-P3) into the port signals (P1, P2, P3).

17. Network node (130) according to any of claims 10-16, operative for the applying of time delays to the antenna signals according to the determined inter- element time delay difference when the antenna signals are in baseband frequency.

18. Network node (130) according to any of claims 10-16, operative for the applying of time delays to the antenna signals according to the determined inter- element time delay difference when the antenna signals are in radio frequency.

19. A computer program (605) comprising instructions, which, when executed by at least one processing circuitry of a network node (130) of a wireless communication network (100), configured for beamformed communication with a plurality of User Equipment, UE, (140, 145), the network node (130) comprising aplurality of antenna elements, each antenna element arranged for handling an antenna signal to be transmitted or received over a frequency band comprising a plurality of sub-bands, causes the network node (130) to perform the following steps: determining, depending on directions between the network node (130) and each of the plurality of UEs (140, 145), a first direction in which to direct a beam formed by transmitting or receiving the antenna signals by the plurality of antenna elements; obtaining phase shifts or delays to add to the antenna signals for the beam to get the determined first direction; determining an inter-element time delay difference for controlling time delays to add to the antenna signals so that the antenna signals of neighboring of the plurality of antenna elements are mutually delayed with the inter-element time delay difference, the time delays being added in order to squint the beam formed by the transmitted antenna signals or the received antenna signals so that any sub-band of the squinted beam, when the squinted beam is directed in the first direction, reaches the plurality of UEs (140, 145); determining, based on information on the directions between the network node (130) and each of the plurality of UEs (140, 145) and the determined first direction, to allocate to each of the plurality of UEs (140, 145), a sub-band of the plurality of sub-bands for communication with the network node (130), wherein the respective sub-band is allocated depending on a position of the direction between the network node and the respective UE (140) in the squinted beam arising from application of the determined inter-element time delay difference to the antenna signals; sending an instruction to each of the plurality of UEs (140, 145) to use the allocated sub-band for communication with the network node (130), and transmitting to the plurality of UEs (140, 145) or receiving from the plurality of UEs (140, 145), the antenna signals via the plurality of antenna elements applying time delays to the antenna signals according to the determined inter-element time delay difference and applying the determined phase shifts or delays hereby achieving the squinted beam directed in the first direction.

20. A carrier containing the computer program (605) according to claim 19, wherein the carrier is one of an electronic signal, an optical signal, a radio signal, an electric signal or a computer readable storage medium.

Images