A method of selecting a downlink beam for downlink communication to a user equipment, as well as a corresponding base station

Abstract

A method of selecting a downlink beam, among multiple candidate beams, by a base station for downlink communication to a User Equipment, UE, in a telecommunication network, said beam selection utilizes two selection methods: a) downlink based beam selection, wherein said beam can be selected, by said base station, based on a received measurement report from said UE; b) uplink based beam selection, wherein said beam is selected, by said base station, based on processing, multiple times, a reference symbol received by said base station from said UE multiple times, each time using one of said multiple candidate beams; wherein said method comprises the steps of performing, by said base station, said uplink based beam selection thereby obtaining a first selection of said beam, performing, by said base station, said downlink based beam selection thereby obtaining a second selection, wherein said step of performing said downlink based beam selection is performed periodically and / or triggered by a triggering condition, selecting, by said base station, a said downlink beam for downlink communication to said UE using either said first selection of said beam or a beam resulting from said second selection.

Description

[0001] Title

[0002] A METHOD OF SELECTING A DOWNLINK BEAM FOR DOWNLINK COMMUNICATION TO A USER EQUIPMENT, AS WELL AS A CORRESPONDING BASE STATION

[0003] Technical field

[0004] The present disclosure generally relates to the field of beam selection and, more specifically, to a method of selecting a downlink beam for downlink communication to a User Equipment.

[0005] Background

[0006] In fifth generation, 5G, wireless networks, maintaining reliable coverage at millimetre wave, mmW, frequencies may be difficult due to high susceptibility to signal blockage and propagation loss. One of the main techniques to overcome this challenge is beamforming, which dynamically adjusts the phase and amplitude of antenna elements to create directed beams aimed at specific user equipment, UEs, improving the link budget. However, the constantly changing channel conditions and UE positions may necessitate continuous measurement and evaluation of various beam directions to maintain an optimal connection with the base station.

[0007] To address this, the 3rd Generation Partnership Project, 3GPP, has defined a set of beam management procedures in its technical documentation, particularly in section 6.1.6 of document 38.802. These procedures may be of importance for improving beam alignment and include three primary phases: Pl for initial access, P2 for beam refinement and tracking at the base station, i.e. gNodeB, and P3 for UE-side beam refinement.

[0008] Beam management in Frequency Range 2, FR2, the mmW band in 5G, may require significant resources. Each procedure involves evaluating the quality of several candidate beams for each UE, which consumes a substantial number of Physical Resource Blocks, PRBs, or Orthogonal Frequency Division Multiplexing, OFDM, symbols. Specifically, the P2 procedure, which is focused on refining and tracking beams, often needs to assess many candidate beams.

[0009] In a traditional FR2 system, during P2 procedures, the gNodeB sends multiple Channel State Information-Reference Signals, CSI-RS, at regular intervals, e.g. every 80 ms, over different candidate beams. These signals are measured by the UE, which then reports the signal quality back to the gNodeB through the Physical Uplink Shared Channel, PUSCH. The gNodeB then identifies the best narrow beam based on the received feedback. While effective, this approach consumes valuable PRB resources that could be used for data transmission, reducing system throughput. The signalling overhead also increases as the number of UEs in the network grows.

[0010] To mitigate this, digital beamforming, DBF, has been suggested as a more resourceefficient approach, though traditional FR2 systems typically use analogue beamforming, ABF, due to the high costs of digital systems, which require large antenna arrays. A proposed solution involves using a digital beamforming receiver that operates over a limited frequency range, which significantly lowers hardware and processing costs, making digital beamforming more feasible in FR2 systems.

[0011] One of the concepts in these reports is an uplink-based beam management approach that reduces reference signal overhead during the P2 procedure. Rather than sending multiple CSI- RS signals over different candidate beams in the downlink, the UE transmits a single reference signal, such as a Sounding Reference Signal, SRS, over a limited frequency range. The gNodeB then applies digital beamforming to the received signal, processing it multiple times to simulate different candidate beams. This allows the gNodeB to determine the optimal beam based on the single SRS transmission, significantly reducing the signaling overhead.

[0012] This approach, however, faces challenges due to the potential for discrepancies between uplink, UL, and downlink, DL, paths, which can lead to a loss of reciprocity. In some cases, the optimal UL and DL beams may not align. For example, regulatory constraints may prevent a device from transmitting through the human body, prompting it to direct its uplink transmission away from the body, resulting in different propagation paths for the UL and DL signals. In such cases, using the same beam for both UL and DL may improve UL quality but lead to a significant degradation in DL quality. This discrepancy makes it difficult to rely solely on either an uplinkbased or downlink -based beam management solution, as both modes have their drawbacks in nonreciprocal scenarios. Moreover, the limited frequency range may lower the accuracy of the uplink beam management.

[0013] Thus, while the proposed uplink-based beam management approach offers a promising reduction in signalling overhead, it must be carefully managed to avoid compromising the quality of either the uplink or downlink, particularly in non-reciprocal situations.

[0014] Summary

[0015] It would be advantageous to achieve a method for selecting a downlink beam, by a base station, for downlink communication to a User Equipment, UE, that is less resource intensive. It would further be advantageous to achieve a corresponding base station. In a first aspect of the present disclosure, there is provided a method of selecting a downlink beam, among multiple candidate beams, by a base station for downlink communication to a User Equipment, UE, in a telecommunication network.

[0016] The beam selection utilizes two selection method.

[0017] The first selection method is directed to downlink based beam selection, wherein the beam can be selected, by said base station, based on a received measurement report from said UE.

[0018] Here, the selection process is for a downlink beam, meaning that the focus is on the communication path from the base station to the UE, i.e. from the gNodeB to the device.

[0019] To be able to select the beam for downlink communication, the base station may rely on a received measurement report from the UE. This report provides information about the quality of a received signal at the UE. Typically, the UE may measure the strength and quality of various beams from the base station and send this information back in the form of a measurement report.

[0020] Based on these reports, the base station can decide which beam to use for the downlink communication to ensure a reliable and strong connection. This process allows the base station to adjust its beamforming, improving the overall network performance, especially in environments with high-frequency signals like millimetre waves.

[0021] The second selection method is directed to uplink based beam selection, wherein the beam is selected, by the base station, based on processing, multiple times, a reference symbol received by said base station from said UE, each time using one of said multiple candidate beams.

[0022] Here, the selection process is also for a downlink beam, thus also for a beam used for communication from the base station to the UE. The difference with the downlink based beam selection is that the uplink based beam selection is performed based on a reference symbol that is received by the base station from the UE. The reference symbol is processed multiple times, each time using one of a multiple candidate beams. After processing, the best beam may be selected.

[0023] Thus, a downlink beam is selected based on a reference symbol that is received on an uplink beam by the base station from the UE. The inventors have found that this uplink based beam selection may also be of assistance when performing downlink beam management, i.e. from the base station to the UE. This is, especially, the case when there is reciprocity.

[0024] Reciprocity may refer to the property that the uplink, UL, and downlink, DL, beams have similar characteristics or quality, meaning that the signal propagation conditions are the same in both directions.

[0025] When reciprocity holds, the same beam used at the base station for downlink transmissions to the UE may be used for receiving uplink communications from the UE to the base station. This similarity arises because the signal experiences the same environment (e.g., the same obstacles, interference, and path loss) in both directions. Loss of reciprocity often occurs when a UE uses different beam directions for uplink transmissions and downlink receptions. This results in distinct UL and DL propagation paths, potentially necessitating different beams at the base station to achieve optimal quality for both UL and DL. This aspect is further elucidated with reference to the figures. finally, the method comprises multiple steps.

[0026] In a first step, the method comprises performing, by said base station, said uplink based beam selection thereby obtaining a first selection of said beam.

[0027] In a second step, the method comprises performing, by said base station, said downlink based beam selection thereby obtaining a second selection, wherein said step of performing said downlink based beam selection is performed periodically and / or triggered by a triggering condition.

[0028] In a third step, the method comprises selecting, by said base station, said downlink beam for downlink communication to said UE using either said first selection of said beam or a beam resulting from said second selection.

[0029] The inventors have realized that the downlink based beam selection is resourceintensive compared to uplink based beam selection. The inventors have also found that the results, i.e. the downlink beams selected, by the uplink based beam selection may be acceptable very frequently. Even though the uplink based beam selection is more tailored to selecting an uplink beam it may thus still very often provide acceptable results for selecting a downlink beam.

[0030] The above described insights have resulted in a method wherein uplink based beam selection is used for selecting the downlink based beam.

[0031] Downlink based beam selection is also performed, for example periodically or triggered by a triggering condition. The downlink based beam selection may be used for evaluating whether the downlink beam selected by the uplink based beam selection is acceptable. The downlink based beam selection may thus be related to a sort of verifying mechanism to verify whether the results from the uplink based beam selection are still acceptable.

[0032] The downlink based beam selection may, for example, be performed periodically. This means that uplink based beam selection may be considered the standard approach for selecting the downlink beam and, every now and then, the results thereof may be verified or evaluated using the downlink based beam selection.

[0033] For example, the downlink based beam selection may be performed periodically and less frequently than the UL based beam selection. Another option is that the downlink based beam selection may be triggered by a triggering condition. This is explained in more detail with reference to the figures and the examples provided below.

[0034] In an example, the method further comprises the step of: evaluating, by said base station, said first selection of said beam using said second selection.

[0035] The second selection may be used for evaluating purposes. This means that the second selection may be used to verify whether the first selection is acceptable or not. In other words, the second selection may be used to verify whether the uplink based beam selection provides results that are acceptable.

[0036] In an example, the step of evaluating said first selection of said beam using said second selection results in said first selection of said beam being acceptable, wherein said step of selecting comprises the step of: selecting, by said base station, said first selection of said beam for downlink communication to said UE.

[0037] If the evaluating step results in the downlink beam selected being acceptable, then the first selection of the beam for downlink communication to the UE may be selected. On top of that, the base station may then continue with the uplink based beam selection method as the default method for selecting a downlink beam in any subsequent instance.

[0038] In a further example, the step of evaluating said first selection of said beam using said second selection results in said first selection of said beam being not acceptable, wherein said method comprises the step of: selecting, by said base station, said beam resulting from said second selection for downlink communication to said UE.

[0039] If the evaluating step results in the downlink beam selecting being not acceptable, then the downlink beam may be selected based on the second selection for downlink communication to the UE.

[0040] It is noted that the second selection may be an actual selected beam. Another option is that the second selection comprises a ranked list of beams with associated quality measures. Yet another option is that the second selection comprises one, or more, beams that are considered to be acceptable. As such, the second selection may not be restricted to one single beam.

[0041] In a further example, the step of performing said downlink based beam selection is performed periodically being any of: performed after fixed time intervals; performed after dynamic time intervals, wherein said time intervals are based on occurrences of at least one of said first selected beam and said second selection.

[0042] Fixed time intervals may entail that the downlink based beam selection is performed at regular, predetermined time intervals. The base station may, for example, periodically evaluate the beam selected by the uplink beam selection using the downlink beam selection. For instance, the base station could trigger a beam selection process every 80 milliseconds, regardless of the network conditions or the current beam performance. This approach is simple and predictable, making it easier to implement.

[0043] Another example is related to dynamic time intervals, which are based on at least one of the occurrence of the first selected beam and the second selection. For example, if the network conditions are stable, performing beam selection too frequently may unnecessarily consume resources and add overhead. Conversely, if the conditions change rapidly, a fixed interval might not allow for timely adjustments to the beam selection. A dynamic time interval may then be beneficial.

[0044] In another example, the triggering condition is any of the following: said first selection of said beam being different to a subsequent first selection of said beam; said first selection of said beam being at least two beams spaced apart to a subsequent first selection of said beam. an evaluation of a previous first selection of said beam using a previous second selection results in said previous first selection of said beam being not acceptable.

[0045] Typically, the uplink beam selection may be used for selecting the downlink beam, i.e. the beam for downlink communication to the UE. Such an uplink beam selection may be performed multiple times in a row, over time. This process may continue until a triggering condition is fulfilled, at which an evaluation process may start. The evaluation process may include performing the downlink beam selection and compare its result with the beam selected using the uplink beam selection.

[0046] The trigger may, for example, be that a beam currently selected using the uplink beam selection differs from a beam that was previously selected. This is an indication that the channel conditions between the base station and the UE, or vice versa, have changed which is an indication that a different beam should be used for downlink communication to the UE.

[0047] Another option is that the first selection of the beam is at least two beams spaced apart to a subsequent first selection of the beam. It may be the case that beams that are, from a spatial point of view, relatively close to one another have very similar properties with respect to the channel to the UE. As such, if subsequent selected beams differ, but are still relative, spatially, close to one another, it may not be necessary to perform an evaluation process. However, if these beams are spatially spaced apart with at least two beams, then that may form an indication that channel conditions have changed too much.

[0048] Yet another example is an evaluation of a previous first selection of the beam using a previous second selection results in the previous first selection of the beam being not acceptable.

[0049] In a further example, the triggering condition is any of a quality measure of said first selection of said beam falls below a predefined threshold; a quality measure difference between said first selection of said beam and a subsequent same first selection of said beam is above a predefined threshold; a power level difference between said first selection of said beam and a subsequent different first selection of said beam is above a predefined threshold a quality measure difference between said multiple candidates corresponding to said first selection of said beam and said multiple candidates corresponding to a subsequent first selection of said beam is above a predefined threshold; a sum of squared power level (or SINR) differences, where each difference is between a candidate out of said multiple candidates corresponding to said first selection of said beam and the same candidate out of said multiple candidates corresponding to a subsequent first selection of said beam, is above a predefined threshold.

[0050] As mentioned above typically, the uplink beam selection may be used for selecting the downlink beam, i.e. the beam for downlink communication to the UE. Such an uplink beam selection may be performed multiple times in a row, over time. This process may continue until a triggering condition is fulfilled, at which an evaluation process may start. The evaluation process may include performing the downlink beam selection and compare its result with the beam selected using the uplink beam selection.

[0051] The trigger may, for example, be related to a quality measure of the first selection of the beam. If the quality of the selected beam, for example the received power strengths using that beam, the signal to noise ratio related to that beam or something similar, fall below a predefined threshold, then that might be an indication that the selected beam is no longer acceptable. An evaluation process may then be initiated to double-check the selected beam.

[0052] Yet another option is related to a quality measure difference between the first selection of the beam and a subsequent same first selection of the beam if above a predefined threshold. In other words, if the relative quality measure for the same beam suddenly differs then that might be an indication that the selected beam is no longer acceptable. A further option is related to a power level difference between the first selection of the beam and a subsequent different first selection of the beam. If there is a sudden drop in receiver power levels, or strengths, then that might also be an indication that an evaluation process is needed.

[0053] In an example, the triggering condition is any of the following: a result from a prediction mechanism for predicting that uplink based results in said first selection of said beam being are not acceptable, said prediction mechanism having input any of received power strengths on any of said candidate beams, UE position, time of day, number of UEs connected to the base station, interference levels between UEs connected to the base station, throughput load and / or energy costs.

[0054] In yet another example, the triggering condition is any of the following: a change in selection of a beam used for uplink communication with said UE; a quality measure of a beam used for uplink communication with said UE falls below a predefined threshold.

[0055] The inventors have found that the beam (and its quality measure) that is selected for the uplink communication, thus communication from the UE to the base station, may be used for determining whether an evaluation process is needed. For example, a beam used for uplink communication, by the base station, changes, then that could also form an indication that there might be the case that a different downlink beam needs to be selected. .

[0056] Another option is that a quality measure related to the beam used for uplink communication with the UE falls below a predefined threshold.

[0057] In an example, the method further comprises any of the steps of: selecting, by said base station, an uplink beam for uplink communication with said UE using a beam resulting from said second selection selecting, by said base station, an uplink beam for uplink communication with said UE using said first selection of said beam.

[0058] The above described example defines that the base station may select an uplink beam for uplink communication with the UE using a beam resulting from the second selection. As mentioned above, the second selection may comprise a list of beams among which a beam may be selected.

[0059] In a further example, said reference symbol, for said uplink based beam selection, is comprised by any of a Sounding Reference Signal, SRS, or a Demodulation Reference Signal.

[0060] The Sounding Reference Signal, SRS, is a signal sent by the UE to the base station primarily for channel quality measurement purposes. It allows the base station to estimate the channel conditions between the base station and the UE, which may be of importance for proper communication. The SRS is typically transmitted by the UE in a time-division manner, often at regular intervals, and is used by the base station to determine the best transmission strategies, such as the best beam to use when communicating with the UE. This helps to enhance overall system performance and efficiency in terms of throughput and coverage.

[0061] On the other hand, the Demodulation Reference Signal, DRS, is used by the UE to demodulate the signals it receives from the base station. The DRS is embedded within the downlink signals transmitted by the base station and helps the UE correctly decode the received data. It provides phase and frequency references, allowing the UE to mitigate distortions such as signal interference and timing errors.

[0062] In a further example, the step of performing, by said base station, said downlink based beam selection results in said second selection being a single beam.

[0063] The second selection may thus be directed to the selection of a specific beam for downlink communication with the UE.

[0064] In yet another example, the step of performing, by said base station, said downlink based beam selection results in a second selection being a ranked list of at least a subset of said candidate beams.

[0065] In a further example, the steps of performing said uplink based beam selection, performing said downlink based beam selection and comparing are performed during a P2 Beam Selection Procedure.

[0066] The P2 beam selection procedure in 5G refers to a method to select a beam from a plurality of candidate beams for downlink communication to the UE. This procedure may be of importance for improving the performance of the downlink communication, especially in environments where multiple antenna beams are used to enhance signal strength, coverage, and data throughput. The P2 beam selection process may occur during an initial connection setup and beam management phases.

[0067] In another example, the received measurement report comprises results of measurements performed by said UE on reference symbols sent by said base station over said multiple candidate beams.

[0068] In a second aspect of the present disclosure, there is provided a computer program product comprising a computer readable medium having instructions stored thereon which, when executed by a base station in a telecommunication network, cause said base station to implement a method in accordance with any of the previous examples.

[0069] It is noted that the advantages as explained with reference to the first aspect, being the method for selecting a beam, are also applicable to the second aspect, being the computer program product. In a third aspect of the present disclosure, there is provided a base station arranged for selecting a downlink beam, among multiple candidate beams, by a base station for downlink communication to a User Equipment, UE, in a telecommunication network, said base station being arranged to perform two selection methods: a) downlink based beam selection, wherein said beam can be selected, by said base station, based on a received measurement report from said UE; b) uplink based beam selection, wherein said beam is selected, by said base station, based on processing, multiple times, a reference symbol received by said base station from said UE, each time using one of said multiple candidate beams.

[0070] The base station comprises process equipment arranged for performing the uplink based beam selection thereby obtaining a first selection of said beam and for performing said downlink based beam selection thereby obtaining a second selection, wherein process equipment is arranged for performing said downlink based beam selection periodically and / or triggered by a triggering condition.

[0071] The base station further comprises select equipment arranged for selecting said downlink beam for downlink communication to said UE using either said first selection of said beam or a beam resulting from said second selection.

[0072] It is noted that the advantages as explained with reference to the first aspect, being the method for selecting a beam, are also applicable to the third aspect, being the base station.

[0073] In an example, the base station further comprises: evaluate equipment arranged for evaluating said first selection of said beam using said second selection.

[0074] In another example, the evaluating results in said first selection of said beam being acceptable, wherein said select equipment is further arranged for: selecting said first selection of said beam for downlink communication to said UE.

[0075] In yet another example, the evaluating results in said first selection of said beam being not acceptable, wherein said select equipment is further arranged for: selecting said beam resulting from said second selection for downlink communication to said UE.

[0076] In a further example, the perform equipment is arranged for performing said downlink based beam selection periodically being any of: performed after fixed time intervals; performed after dynamic time intervals, wherein said dynamic time intervals are based on at least one of said first selected beam and said second selection. In an example, the triggering condition is any of the following: said first selection of said beam being different to a subsequent first selection of said beam; said first selection of said beam being at least two beams spaced apart to a subsequent first selection of said beam. an evaluation of a previous first selection of said beam using a previous second selection results in said previous first selection of said beam being not acceptable.

[0077] In an example, the triggering condition is any of a quality measure of said first selection of said beam falls below a predefined threshold; a quality measure difference between said first selection of said beam and a subsequent same first selection of said beam is above a predefined threshold; a power level difference between said first selection of said beam and a subsequent different first selection of said beam is above a predefined threshold a quality measure difference between said multiple candidates corresponding to said first selection of said beam and said multiple candidates corresponding to a subsequent first selection of said beam is above a predefined threshold; a sum of squared power level differences, where each difference is between a candidate out of said multiple candidates corresponding to said first selection of said beam and the same candidate out of said multiple candidates corresponding to a subsequent first selection of said beam, is above a predefined threshold.

[0078] In an example, the triggering condition is any of the following: a result from a prediction mechanism for predicting that uplink based results in said first selection of said beam being are not acceptable, said prediction mechanism having input any of received power strengths on any of said candidate beams, UE position, time of day, number of UEs connected to the base station, interference levels between UEs connected to the base station, throughput load and / or energy costs.

[0079] In an example, the triggering condition is any of the following: a change in selection of a beam used for uplink communication with said UE; a quality measure of a beam used for uplink communication with said UE falls below a predefined threshold.

[0080] In a further example, the select equipment is further arranged for: selecting an uplink beam for uplink communication with said UE using a beam resulting from said second selection. In yet another example, the selection method is an uplink based beam selection, wherein said beam is selected, by said base station, based on processing a reference symbol received by said base station from said UE multiple times, each using one of said multiple candidate beams wherein said reference symbol is comprised by any of a Sounding Reference Signal, SRS, or a Demodulation Reference Signal.

[0081] In an even further example, the perform equipment is further arranged for: performing said downlink based beam selection results in a second selection of said beam.

[0082] In an example, the perform equipment is further arranged for: performing said downlink based beam selection results in a second selection being a ranked list of at least a subset of said candidate beams.

[0083] In an example, the received measurement report comprises results of measurements performed by said UE on reference symbols sent by said base station over said multiple candidate beams.

[0084] It is noted that in all previous examples, the role of the base station and the User Equipment, UE, may be reversed. The UE may, for example, select an uplink beam for uplink communication with the base station based on reference symbols it may have received in the downlink. The selected uplink beam may be evaluated, every now and then, based on feedback received from the base station with respect to reference symbols that were sent by the UE over multiple candidate beams. This is the same process as described with respect to the base station but then reversed for the UE.

[0085] In the appended figures, similar components and / or features may have the same reference label. Further, various components of the same type may be distinguished by following the reference label by a dash and a second label that distinguishes among the similar components. If only the first reference label is used in the specification, the description is applicable to any one of the similar components having the same first reference label irrespective of the second reference label.

[0086] The above and other aspects of the disclosure will be apparent from and elucidated with reference to the examples described hereinafter.

[0087] Brief description of the drawings

[0088] Figure 1 discloses an example of a beam refinement process in a Service Based Architecture, SBA, telecommunication network;

[0089] Figure 2 discloses an example illustrating different optimal uplink, UL, and downlink, DL, paths; Figures 3a-3g disclose an example of a signalling flow in accordance with the present disclosure;

[0090] Figure 4 discloses a state diagram in accordance with the present disclosure;

[0091] Figures 5a-5g disclose an example of a signalling flow in accordance with the present disclosure;

[0092] Figure 6 discloses an example of a base station in accordance with the present disclosure;

[0093] Figure 7 discloses an example of a method in accordance with the present disclosure.

[0094] Detailed description

[0095] It is noted that in the description of the figures, same reference numerals refer to the same or similar components performing a same or essentially similar function.

[0096] A more detailed description is made with reference to particular examples, some of which are illustrated in the appended drawings, such that the manner in which the features of the present disclosure may be understood in more detail. It is noted that the drawings only illustrate typical examples and are therefore not to be considered to limit the scope of the subject matter of the claims. The drawings are incorporated for facilitating an understanding of the disclosure and are thus not necessarily drawn to scale. Advantages of the subject matter as claimed will become apparent to those skilled in the art upon reading the description in conjunction with the accompanying drawings.

[0097] The ensuing description above and below provides preferred exemplary embodiment(s) only, and is not intended to limit the scope, applicability or configuration of the disclosure. Rather, the ensuing description of the preferred exemplary embodiment(s) will provide those skilled in the art with an enabling description for implementing a preferred exemplary embodiment of the disclosure, it being understood that various changes may be made in the function and arrangement of elements, including combinations of features from different embodiments, without departing from the scope of the disclosure.

[0098] Unless the context clearly requires otherwise, throughout the description and the claims, the words "comprise," "comprising," and the like are to be construed in an inclusive sense, as opposed to an exclusive or exhaustive sense; that is to say, in the sense of "including, but not limited to." As used herein, the terms "connected," "coupled," or any variant thereof means any connection or coupling, either direct or indirect, between two or more elements; the coupling or connection between the elements can be physical, logical, electromagnetic, or a combination thereof. Additionally, the words "herein," "above," "below," and words of similar import, when used in this application, refer to this application as a whole and not to any particular portions of this application. Where the context permits, words in the Detailed Description using the singular or plural number may also include the plural or singular number respectively. The word "or," in reference to a list of two or more items, covers all of the following interpretations of the word: any of the items in the list, all of the items in the list, and any combination of the items in the list.

[0099] Figure 1 discloses an example of a beam refinement process in a Service Based Architecture, SBA, telecommunication network.

[0100] Beam management in Frequency Range 2, FR2, deployments are of importance for maintaining a good communication between the User Equipment, UE, and the base station. However, these procedures are resource-intensive, consuming significant Physical Resource Blocks, PRBs, or Orthogonal Frequency-Division Multiplexing, OFDM, symbols. The primary reason for this is that beam management must be performed periodically for each UE.

[0101] This process involves evaluating the quality of multiple candidate beams to ensure the best beam is selected for communication. The number of candidate beams assessed during the P2 procedures typically exceed the three beams illustrated in the example shown in figure 1, which depicts a simplified three-step beam refinement procedure.

[0102] In traditional FR2 systems, the base station, for example the gNB, determines the beam to be used through periodic "P2 tracking events." For instance, every 80 milliseconds, the gNB transmits multiple Channel State Information-Reference Signals, CSI-RS, consecutively in time domain, each mapped to a different candidate beam. The UE measures the signal quality of each CSI-RS and reports these measurements back to the gNB via a CSI report on the Physical Uplink Shared Channel, PUSCH. Based on these reports, the gNB identifies the best beam for the UE. That is, the gNB selects the beam to be used in downlink communication with the UE. This process is referred to as a Downlink DL-based beam management or selection.

[0103] While effective, the traditional approach comes with some drawbacks. First, transmitting CSI-RS and transmitting CSI reports on PUSCH consumes PRBs that could otherwise be allocated for data transmission, thereby reducing system peak throughput. Second, the signaling overhead increases with the number of UEs, as each UE requires its own set of CSI-RS and PUSCH transmissions. This limits the overall efficiency of the network.

[0104] An alternative approach to reduce this resource consumption involves using digital beamforming, DBF.

[0105] DBF at the gNB involves receiving a reference signal over the limited frequency range, i.e. associated with the SRS or Demodulation Reference Signal, DMRS, and perform DBF on the signal. In other words, the gNB will process the received signal multiple times, where each of the processing outputs represents one of the potential candidate-beams at the gNB side. By comparing the signal quality of the different processing outputs, the optimal gNB beam for the UE can be determined. This means that one single reference signal transmission is sufficient to determine the best P2 beam, which reduces the signaling overhead over the air interface significantly.

[0106] However, DBF is often impractical for FR2 systems due to the high hardware costs associated with mmWave systems, which require large antenna arrays with numerous small antenna elements. Consequently, most FR2 systems rely on analog beamforming, ABF, which, while more cost-effective, is less resource-efficient in beam management.

[0107] There are at least two problems with replacing a downlink based beam selection with an uplink based beam selection.

[0108] If a UE transmits on the uplink, UL, using a different beam direction than it uses for receiving on the downlink, DL, the optimal beam for UL reception at the gNB might differ from the optimal beam for DL transmission at the gNB, as illustrated in Figure 2.

[0109] For example, regulatory constraints might prevent the device from transmitting through the human body. Instead of reducing power, the device could direct the UL transmission in another direction that is not obstructed by the human body. In this case the UE’s transmission is directed towards the same gNB wide beam but the transmission will utilize a different beam direction and a different propagation path compared to the beam direction the UE uses for DL reception.

[0110] In such scenarios, a downlink-based beam selection would use the optimal DL gNB beam for both DL and UL communication. With an UL-based beam selection, the gNB would instead use the optimal UL beam for both UL and DL communication. This could, in principle, improve UL quality; however, it could also result in a significant drop in DL quality.

[0111] Another aspect is the need of downlink transmissions for feedback on the uplink. If the uplink is broken the downlink will suffer, for example less reliable HARQ feedback and Channel State Information, CSI, report supporting DL based beam selection. Consequently, relying on one beam management mode, mode being downlink-based or uplink -based, is problematic in such scenarios of no reciprocity.

[0112] Further, the fact that the SRS bandwidth is rather limited, to reduce the hardware cost of the digital receiver, can lead to inaccuracies in determining the best beam based on SRS decoding. Such inaccuracies can degrade performance compared to DL based beam selection.

[0113] How frequently UL based beam management will give suboptimal results can vary significantly depending on the deployment scenario. Channels with high time dispersion, in other words high level of frequency selectivity, are likely to result in less accurate optimal gNB beam estimations when using narrow-band SRS compared to utilizing wide-band CSI-RS measurements. Even with less time dispersion, the reception of SRS on a limited bandwidth (fewer subcarriers to receive samples on) simply lower the quality in the presence of noise. Moreover, an increased number of users sharing the same SRS resource configurations in time and frequency domains, through different cyclic shifts, can further degrade performance. In summary, there is a trade-off between enhancing the accuracy of SRS-based optimal beam evaluation and minimizing the resource consumption of SRS transmissions used for UL-based beam selection. Allocating more SRS resources for UL-based beam selection will improve accuracy but also increase SRS resource consumption, leading to higher reference signal overhead. This, in turn, diminishes the advantage of UL-based beam selection compared to legacy DL-based beam selection in terms of increasing potential cell peak throughput.

[0114] The present disclosure, to address the previously described technical challenges, is directed to a combination of UL based beam selection and DL based beam selection. This is explained as follows.

[0115] Here below a typical situation is described. It is noted that the present disclosure is also applicable in different types of situations, not limited to the one described here below.

[0116] The base station may first assume that there is reciprocity between the base station and the UE. This means that the same, or similar, channel conditions exist in the downlink as well as in the uplink. This also indicates that the same beam will be used, at the gNB side, for both uplink communication as downlink communication.

[0117] That means that the gNB will perform uplink based beam selection thereby obtaining a first selection of said beam. Again, uplink based beam selection entails that the gNB selects a beam based on a reference symbol that is sent by the UE to the gNB, which reference symbol is processed multiple times using multiple candidate beams. The “best” beam may then be selected, which beam is used for downlink communication. It is noted that this beam may also be used for uplink communication with the UE.

[0118] The above described uplink based beam selection may be performed many times in a row. That means that the gNB may regularly, or event based, perform such an uplink based beam selection to select the beam. Different beams may be selected each time, or the same beam may be selected each time, all dependent on the quality of the channel between the UE and the gNB, i.e. the beams.

[0119] As mentioned above, the UL based beam selection may be not as accurate as desired. Especially in situations wherein no reciprocity is in place between the UE and the gNB results may start to differ. That means that the UL based beam selection may select an inferior beam.

[0120] The inventors have found that it may be beneficial to perform DL based beam selection as well. The DL based beam selection may be performed periodically or may be triggered by a triggering condition. The DL based beam selection may, for example, be viewed as an evaluation mechanism to evaluate whether the selection made by the UL based beam selection is still acceptable. The results of the DL based beam selection may be compared with the selected beam using the UL based beam selection, and, based on this comparison, it may be concluded whether or not UL based beam selection still provides acceptable results.

[0121] If the evaluation is such that UL based beam selection provides for a selected beam that is acceptable, it may be decided to continue with the UL based beam selection for selecting the beam used for downlink communication. If the evaluation is such that the UL based beam selection provides for a selected beam that is not acceptable, it may be decided to abandon the UL based beam selection - at least for a while - and continue with the DL based beam selection.

[0122] In other words, the DL based beam selection may be used, or may be performed, to double-check whether the results of the UL based beam selection are still acceptable.

[0123] The DL based beam selection may be performed periodically, wherein the periodicity of the DL based beam selection is much longer compared to the periodicity of the UL based beam selection. Alternatively, the DL based beam selection may be triggered whenever a change in the (quality of the) selected beam occurs. This may form an indication that there is no longer reciprocity between the gNB and the UE.

[0124] The present disclosure has multiple advantages.

[0125] If the UL and DL paths are the same or similar, and the channel conditions are such that UL based beam selection can be relied upon, UL based beam selection can be used to replace DL based beam selection, or at least to reduce the number of times that the DL based beam selection is to be performed, thereby reducing the reference signal signalling overhead, which will increase the achievable DL cell throughput and which will also help to increase the total capacity for the number of connected UEs that can be served by the system.

[0126] If the UL and DL paths differ from each other, and UL based beam selection can be relied upon, both UL and DL based beam selection can be run in parallel and both paths can be utilized for UL / DL communication thereby improving UL SINR compared to legacy systems, where both UL and DL will always use the optimal DL path only. DL quality will in that case also benefit since it relies on feedback using UL transmissions such as Ack / Nack and CSI-reporting.

[0127] If the channel conditions are such that UL based BM cannot be relied upon, out of coverage for instance, the gNB can fall back to DL based beam selection using, for example, CSI- RS measurements.

[0128] UL based beam selection can be more efficient compared to DL based beam selection from a reference signal resource consumption perspective. However, in certain scenarios, UL based beam selection using SRS over a limited bandwidth may not be as reliable as DL based beam selection with CSI-RS transmission over a large bandwidth. One of the examples of the present disclosure is to mainly use UL based beam selection for periodic P2 tracking, in the SBA based telecommunication network, and to identify situations in which a verification, or evaluation, of the selected beam using DL based beam selection may be needed.

[0129] Here, the gNB will keep UL based beam selection (for example SRS based) based P2 tracking functionality continuously activated, so that the gNB will, with a certain periodicity, receive optimal beam estimations from the SRS digital receiver chain decoding output.

[0130] At the time of the first expected SRS reception, the gNB may trigger one DL based beam selection, i.e. P2 tracking event, to evaluate the selected beam using a CSI-RS signals. If both UL based beam selection and DL based beam selection evaluations result in the same optimal beam, or if the DL based beam selection evaluation results in the conclusion that the UL based beam selection is acceptable, the DL based beam selection may be turned off, for as long as the UL based beam selection may continue to estimate the same beam as the optimal beam. This optimal beam may be used for both UL and DL communication throughout this duration.

[0131] Whenever the optimal beam has changed according to UL based beam selection, the gNB may trigger an additional DL based beam selection to again verify the accuracy of the result.

[0132] The above described approach reduces the resource consumption for beam management procedures significantly, compared to legacy, because in many cases when the gNB receives a P2 tracking report, it will show the same best gNB beam for the UE as the previous report.

[0133] Whenever a mismatch happens between the optimal beams according to UL based and DL based beam selection, the optimal beam determined by DL based beam selection is used for downlink communication to the UE.

[0134] Figures 3a-3g shows such an example signalling flow as described above.

[0135] It is noted that the round circles represent the best beam according to UL based, for example SRS based, beam evaluation and that the squares represent the best beam candidate according to DL based, for example CSI-RS based, beam evaluation.

[0136] It is noted that it may not be necessary to trigger a DL based beam selection every time the optimal traffic beam changes according to UL based beam selection.

[0137] For example, it could be an option to trust the UL based beam selection result whenever: the new best beam has previously been the second strongest beam, and / or the new best beam is a direct neighbour of the previously best beam, and / or SRS power level difference between the new best beam and the previously best beam does not exceed a certain threshold, i.e. UL quality did not get worse by a big margin and it can be concluded that since the UL quality, as measured by SRS decoding, is still good enough, the SRS beam estimation should still be accurate. and / or certain specific threshold rules are met when comparing the Reference Signals Received Power, RSRP, levels between the previously best and the new best beams as well as the RSRP levels of their direct neighbour beam sets, and / or the SRS power level of the best beam candidate does not drop below a minimum RSRP threshold. The minimum tolerable RSRP threshold could - for example - be adjusted based on the last DL based beam selection result. This means it can be assumed that the UL based beam selection gives accurate results, if the UL quality, according to SRS RSRP, is not worse than DL quality.

[0138] This would trust UL based beam selection whenever relatively small changes are detected. An example of a large change could be when a sum of squared power level (or SINR) differences, where each difference is between a candidate out of said multiple candidates corresponding to said first selection of said beam and the same candidate out of said multiple candidates corresponding to a subsequent first selection of said beam, is above a predefined threshold. Then, accordingly, the gNB may trigger a DL based beam selection.

[0139] If the SRS RSRP level on the strongest beams falls below a minimum threshold, it may be decided to turn on DL based BM selection permanently - or at least for a prolonged time period - for that UE until the SRS received signal level on the best beam improves.

[0140] Machine learning algorithms may be used to accurately predict the cases where DL based beam selection should be triggered. The system could be trained on the options provided above, and the machine learning can consider in which situations the DL based beam selection resulted in a different optimal beam compared to the UL based beam selection result. Data input to the machine learning model can include SRS RSRP levels of all the beam candidates in the grid, distances between the relevant beams in the grid, site specific or deployment scenario specific properties, UE type, UE vendor and model information, previous behaviour of this UE, time of day etc.

[0141] Reference is now made to the state diagram shown in figure 4 and the example signalling flow shown in figures 5a-5g.

[0142] Here, an option is described wherein it is acknowledged that the reason for the different beam estimates - between DL based beam selection and UL based beam selection - could either be the SRS based beam estimation result is inaccurate, or UL & DL beam reciprocity is lost, and the UE is sending the UL into a different beam direction than it receives the DL from (no reciprocity).

[0143] Lor the differentiation between these two states a detection mechanism may be needed. Different alternatives are possible. first, the situation is considered wherein a DL beam estimated the same as before, but the UL beam changed significantly. An indication of a loss of the UL / DL reciprocity, could be a sudden significant drop of the SRS RSRP level in the previously best beam in the UL based beam selection, but no drop, or a less significant drop, in the DL based beam selection, while simultaneously receiving a new best UL beam, which is not a direct neighbour of the previous best UL beam but is relatively far away from the previously best beam in the grid.

[0144] In other words, the optimal DL beam stays the same, but the optimal UL beam appears suddenly somewhere else. In that case it may be likely that the UL based beam selection cannot utilize the same optimal path anymore, while the DL still can.

[0145] Second, simultaneously receiving UL data on an alternative UL beam. For data active UEs, it could be an option to receive the UL data on two different beam directions simultaneously and process the received signal twice, i.e. on the beam corresponding to the optimal DL path and on an alternative beam. If the UE experiences a better UL SINR on the alternative beam compared to the beam used for DL transmission, it may be concluded that the optimal UL path is now different from the optimal DL path and that UL / DL reciprocity is lost. The optimal UL gNB beam can then be used for UL reception and the optimal DL gNB beam can then be used for transmission in the DL.

[0146] The second alternative described above of comparing UL data on two different gNB beams for one specific data active UE, can also be used to verify if the detection mechanism worked correctly. UL SINR could be compared on both beam directions to identify, if the SRS best beam result is accurate or not. If SRS best beam is accurate then the detection mechanism concludes there is no reciprocity.

[0147] Given that the present disclosure provides for a method to evaluate and verify which UL gNB beam direction is better for a UE, it would be possible to utilize machine learning algorithms that predict the likelihood that a UE lost UL / DL reciprocity. In that way, the verification mechanism to - i.e. the DL based beam selection - may be used more seldomly in the future.

[0148] Data input parameters to the machine learning algorithm may include: RSRP levels on the different candidate beams for both UL based as well as DL based P2 tracking, UL and DL SINR, UE position (site the UE is connected to and the UEs position in the beam grid) , time of day, number of UEs connected to the system, current level of interference between UEs, current throughput load, current energy cost, etc.

[0149] Figure 6 discloses an example of a base station 601 in accordance with the present disclosure.

[0150] Figure 6 discloses a base station 601 arranged for selecting a downlink beam, among multiple candidate beams, by a base station for downlink communication to a User Equipment, UE, in a telecommunication network, said base station being arranged to perform two selection methods. a) downlink based beam selection, wherein said beam can be selected, by said base station, based on a received measurement report from said UE. b) uplink based beam selection, wherein said beam is selected, by said base station, based on processing, multiple times, a reference symbol received by said base station from said UE, each time using one of said multiple candidate beams;

[0151] The base station comprises process equipment 604 arranged for performing said uplink based beam selection thereby obtaining a first selection of said beam and for performing said downlink based beam selection thereby obtaining a second selection, wherein process equipment is arranged for performing said downlink based beam selection periodically and / or triggered by a triggering condition.

[0152] The base station also comprises select equipment 605 arranged for selecting said downlink beam for downlink communication to said UE using either said first selection of said beam or a beam resulting from said second selection.

[0153] In addition, the base station may comprise receiving equipment 602 for receiving packets and may comprise transmit equipment 603 for transmitting packets. A processor 606 in communication with a memory 607 may be provided as well.

[0154] Figure 7 discloses an example of a method in accordance with the present disclosure.

[0155] Figure 7 discloses a method (701) of selecting a downlink beam, among multiple candidate beams, by a base station for downlink communication to a User Equipment, UE, in a telecommunication network, said beam selection utilizes two selection methods: a) downlink based beam selection, wherein said beam can be selected, by said base station, based on a received measurement report from said UE; b) uplink based beam selection, wherein said beam is selected, by said base station, based on processing, multiple times, a reference symbol received by said base station from said UE, each time using one of said multiple candidate beams;

[0156] Here, the method comprises the steps of: performing (702), by said base station, said uplink based beam selection thereby obtaining a first selection of said beam; performing (703), by said base station, said downlink based beam selection thereby obtaining a second selection, wherein said step of performing said downlink based beam selection is performed periodically and / or triggered by a triggering condition; selecting (704), by said base station, a downlink beam for downlink communication to said UE using either said first selection of said beam or a beam resulting from said second selection.

[0157] It should be noted that the above-mentioned examples illustrate rather than limit the idea, and that those skilled in the art will be able to design many alternative examples without departing from the scope of the appended claims. The word “comprising” does not exclude the presence of elements or steps other than those listed in a claim, “a” or “an” does not exclude a plurality, and a single processor or other unit may fulfil the functions of several units recited in the claims.

[0158] Any reference signs in the claims shall not be construed so as to limit their scope.

Claims

23CLAIMS1. A method (701) of selecting a downlink beam, among multiple candidate beams, by a base station (601) for downlink communication to a User Equipment, UE, in a telecommunication network, said beam selection utilizes two selection methods: a) downlink based beam selection, wherein said beam can be selected, by said base station (601), based on a received measurement report from said UE; b) uplink based beam selection, wherein said beam is selected, by said base station (601), based on processing, multiple times, a reference symbol received by said base station (601) from said UE, each time using one of said multiple candidate beams; wherein said method comprises the steps of: performing (702), by said base station (601), said uplink based beam selection thereby obtaining a first selection of said beam; performing (703), by said base station (601), said downlink based beam selection thereby obtaining a second selection, wherein said step of performing said downlink based beam selection is performed periodically and / or triggered by a triggering condition; selecting (704), by said base station (601), said downlink beam for downlink communication to said UE using either said first selection of said beam or a beam resulting from said second selection.

2. A method in accordance with claim 1, wherein said method further comprises the step of: evaluating, by said base station, said first selection of said beam using said second selection.

3. A method in accordance with claim 2, wherein said step of evaluating said first selection of said beam using said second selection results in said first selection of said beam being acceptable, wherein said step of selecting comprises the step of: selecting, by said base station, said first selection of said beam for downlink communication to said UE.

4. A method in accordance with claim 2 wherein said step of evaluating said first selection of said beam using said second selection results in said first selection of said beam being not acceptable, wherein said method comprises the step of: selecting, by said base station, said beam resulting from said second selection for downlink communication to said UE.

5. A method in accordance with any of the previous claims, wherein said step of performing said downlink based beam selection is performed periodically being any of: performed after fixed time intervals; performed after dynamic time intervals, wherein said time intervals are based on occurrences of at least one of said first selected beam and said second selection.

6. A method in accordance with any of the previous claims, wherein said triggering condition is any of the following: said first selection of said beam being different to a subsequent first selection of said beam; said first selection of said beam being at least two beams spaced apart to a subsequent first selection of said beam. an evaluation of a previous first selection of said beam using a previous second selection results in said previous first selection of said beam being not acceptable.

7. A method in accordance with any of the previous claims, wherein said triggering condition is any of a quality measure of said first selection of said beam falls below a predefined threshold; a quality measure difference between said first selection of said beam and a subsequent same first selection of said beam is above a predefined threshold; a power level difference between said first selection of said beam and a subsequent different first selection of said beam is above a predefined threshold; a quality measure difference between said multiple candidates corresponding to said first selection of said beam and said multiple candidates corresponding to a subsequent first selection of said beam is above a predefined threshold; a sum of squared power level, or SINR, differences, where each difference is between a candidate out of said multiple candidates corresponding to said first selection of said beam and the same candidate out of said multiple candidates corresponding to a subsequent first selection of said beam, is above a predefined threshold.

8. A method in accordance with any of the previous claims, wherein said triggering condition is any of the following: a result from a prediction mechanism for predicting that uplink based results in said first selection of said beam being are not acceptable, said prediction mechanism having input any of received power strengths on any of said candidate beams, UE type, UE vendor and modelinformation, UE position, time of day, number of UEs connected to the base station, interference levels between UEs connected to the base station, throughput load and / or energy costs.

9. A method in accordance with any of the previous claims, wherein said triggering condition is any of the following: a change in selection of a beam used for uplink communication with said UE; a quality measure of a beam used for uplink communication with said UE falls below a predefined threshold.

10. A method in accordance with any of the previous claims, wherein said method further comprises any of the steps of: selecting, by said base station, an uplink beam for uplink communication with said UE using a beam resulting from said second selection selecting, by said base station, an uplink beam for uplink communication with the UE using said first selection of said beam.

11. A method in accordance with any of the previous claims, wherein said reference symbol, for said uplink based beam selection, is comprised by any of a Sounding Reference Signal, SRS, or a Demodulation Reference Signal.

12. A method in accordance with any of the previous claims, wherein said step of performing, by said base station, said downlink based beam selection results in said second selection being a single beam.

13. A method in accordance with any of the previous claims, wherein said step of performing, by said base station, said downlink based beam selection results in a second selection being a ranked list of at least a subset of said candidate beams.

14. A method in accordance with any of the previous claims, wherein said steps of performing said uplink based beam selection, performing said downlink based beam selection and comparing are performed during a P2 Beam Selection Procedure.

15. A method in accordance with any of the previous claims, wherein said received measurement report comprises results of measurements performed by said UE on reference symbols sent by said base station over said multiple candidate beams.2616. Computer program product comprising a computer readable medium having instructions stored thereon which, when executed by a base station in a telecommunication network, cause said base station to implement a method in accordance with any of the claims 1 - 14.

17. A base station (601) arranged for selecting a downlink beam, among multiple candidate beams, by a base station (601) for downlink communication to a User Equipment, UE, in a telecommunication network, said base station being arranged to perform two selection methods a) downlink based beam selection, wherein said beam can be selected, by said base station, based on a received measurement report from said UE; b) uplink based beam selection, wherein said beam is selected, by said base station, based on processing, multiple times, a reference symbol received by said base station from said UE, each time using one of said multiple candidate beams; wherein said base station (601) comprises: process equipment (604) arranged for performing said uplink based beam selection thereby obtaining a first selection of said beam and for performing said downlink based beam selection thereby obtaining a second selection, wherein process equipment is arranged for performing said downlink based beam selection periodically and / or triggered by a triggering condition; select equipment (605) arranged for selecting said downlink beam for downlink communication to said UE using either said first selection of said beam or a beam resulting from said second selection.

18. A base station in accordance with claim 17, wherein said base station further comprises: evaluate equipment arranged for evaluating said first selection of said beam using said second selection.

19. A base station in accordance with claim 18, wherein said evaluating results in said first selection of said beam being acceptable, wherein said select equipment is further arranged for: selecting said first selection of said beam for downlink communication to said UE.

20. A base station in accordance with claim 18, wherein said evaluating results in said first selection of said beam being not acceptable, wherein said select equipment is further arranged for:27 selecting said beam resulting from said second selection for downlink communication to said UE.

21. A base station in accordance with any of the claims 17 - 20, wherein said perform equipment is arranged for performing said downlink based beam selection periodically being any of: performed after fixed time intervals; performed after dynamic time intervals, wherein said dynamic time intervals are based on occurrences of at least one of said first selected beam and said second selection.

22. A base station in accordance with any of the claims 17 - 21, wherein said triggering condition is any of the following: said first selection of said beam being different to a subsequent first selection of said beam; said first selection of said beam being at least two beams spaced apart to a subsequent first selection of said beam. an evaluation of a previous first selection of said beam using a previous second selection results in said previous first selection of said beam being not acceptable.

23. A base station in accordance with any of the claims 17 - 22, wherein said triggering condition is any of: a quality measure of said first selection of said beam falls below a predefined threshold; a quality measure difference between said first selection of said beam and a subsequent same first selection of said beam is above a predefined threshold; a power level difference between said first selection of said beam and a subsequent different first selection of said beam is above a predefined threshold, a quality measure difference between said multiple candidates corresponding to said first selection of said beam and said multiple candidates corresponding to a subsequent first selection of said beam is above a predefined threshold; a sum of squared power level differences, where each difference is between a candidate out of said multiple candidates corresponding to said first selection of said beam and the same candidate out of said multiple candidates corresponding to a subsequent first selection of said beam, is above a predefined threshold.

24. A base station in accordance with any of the claims 17 - 23, wherein said triggering condition is any of the following:28 a result from a prediction mechanism for predicting that uplink based results in said first selection of said beam being are not acceptable, said prediction mechanism having input any of received power strengths on any of said candidate beams, UE position, time of day, number of UEs connected to the base station, interference levels between UEs connected to the base station, throughput load and / or energy costs.

25. A base station in accordance with any of the claims 17 - 24, wherein said triggering condition is any of the following: a change in selection of a beam used for uplink communication with said UE; a quality measure of a beam used for uplink communication with said UE falls below a predefined threshold.

26. A base station in accordance with any of the claims 17 - 25, wherein said select equipment is further arranged for: selecting an uplink beam for uplink communication with said UE using a beam resulting from said second selection.

27. A base station in accordance with any of the claims 17 - 26, wherein said selection method: b) uplink based beam selection, wherein said beam is selected, by said base station, based on processing a reference symbol received by said base station from said UE multiple times, each using one of said multiple candidate beams wherein said reference symbol is comprised by any of a Sounding Reference Signal, SRS, or a Demodulation Reference Signal.

28. A base station in accordance with any of the claims 17 - 27, wherein said perform equipment is further arranged for: performing said downlink based beam selection results in a second selection of said beam.

29. A base station in accordance with any of the claims 17 - 28, wherein said perform equipment is further arranged for: performing said downlink based beam selection results in a second selection being a ranked list of at least a subset of said candidate beams.2930. A base station in accordance with any of the claims 17 - 29, wherein said received measurement report comprises results of measurements performed by said UE on reference symbols sent by said base station over said multiple candidate beams.

Images