Methods and apparatus to support beam management for UE transmission and reception
The closed-loop monitoring system with an NN adapts UE beam patterns to the real-time spatial channel, addressing the suboptimal beamforming issue in mmWave communications by optimizing beamforming gain and coverage.
Patent Information
- Application Number
- PCT/CN2024/101004
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-06-24
- Publication Date
- 2026-01-02
AI Technical Summary
Existing beam management methods for mmWave band communications in 5G NR cellular technologies fail to achieve optimal beamforming gain due to their inability to adapt to the real-time spatial channel environment, especially in the presence of strong NLOS paths, leading to suboptimal communication performance.
A closed-loop monitoring system that utilizes an artificial neural network (NN) to estimate spatial channel parameters and adaptively synthesize optimal UE TX and RX beam patterns based on real-time measurements, enabling better beamforming gain and improved coverage by using a performance monitoring metric to update NN coefficients.
The system achieves improved beamforming gain and UL coverage by dynamically matching beam patterns to the actual spatial channel, reducing transmission power and enhancing communication efficiency.
Smart Images

Figure CN2024101004_02012026_PF_FP_ABST
Abstract
Description
METHODS AND APPARATUS TO SUPPORT BEAM MANAGEMENT FOR UE TRANSMISSION AND RECEPTIONTechnical Field
[0001] The present disclosure relates, in general, to beam management. Aspects relate to beam management for optimising uplink and / or downlink communication between user equipment and a network node.Background
[0002] Millimeter wave (mmWave) band communications have been adopted by 5G NR cellular technologies, and use frequencies from around 24GHz to around 47GHz. Use of such frequencies can significantly improve data throughput for cellular communications. In order to compensate for the gain of an individual radio frequency (RF) antenna being much smaller, when compared with the sub-6GHz band, whilst maintain low power consumption, analog beamforming has been used implemented for mmWave band RF front end implementations in user equipment (UE) . For example, an analog beamforming based transmitter in UE can be implemented by way of a phased antenna array in which the direction of radiation for a transmitted signal is controlled (beamformed) by analog phase shifters associated to the antenna array.Summary
[0003] An objective of the present disclosure is to provide a closed-loop monitoring system for beam management that is better matched to the prevailing spatial channel between user equipment and a network node than pre-defined beam patterns, thereby enabling better beamforming gain to be achieved. Accordingly, an improved UE TX beam, for example, can be run-time computed in a more efficient manner.
[0004] The foregoing and other objectives are achieved by the features of the independent claims.
[0005] Further implementation forms are apparent from the dependent claims, the description and the Figures.
[0006] A first aspect of the present disclosure provides a user equipment, UE, configured to support beam management for UE transmission, the UE comprising a memory, and a processor, operationally coupled to the memory, configured to transmit, to a node, multiple reference signals, each signal of the multiple reference signals transmitted using a different beam pattern selected from a first set of UE beams, generate estimated values for a first set of spatial channel parameters for a channel between the UE and the node using a first set of received power values, wherein the first set of received power values comprises a power value for each one of the multiple signals received at the node, calculate a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams, generate a performance metric on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values, and transmit the performance metric to the node.
[0007] Accordingly, it is possible to run-time synthesize (compute) an optimal UE TX beam pattern, which is better matched to the actual UL spatial channel than pre-defined beam patterns. As such, better UL beamforming gain can be achieved resulting in improved UL coverage and / or reduced UE transmission power.
[0008] A second aspect of the present disclosure provides a user equipment, UE, configured to support beam management for UE reception, the UE comprising a memory, and a processor, operationally coupled to the memory, configured to receive, from a node, multiple reference signals, each signal of the multiple reference signals received by the UE using a different beam pattern selected from a first set of UE beams, measure a first set of received power values, wherein the first set of received power values comprises a received power value for each signal of the multiple signals received at the UE, generate estimated values for a first set of spatial channel parameters for a channel between the UE and the node using the first set of received power values, calculate a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams, and generate a performance metric on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values.
[0009] Similarly to the UE TX case, an optimal UE RX beam pattern can be run-time synthesized (computed) . Such a UE RX beam pattern takes in to account the actual UL spatial channel, and therefore enables an improvement in gain and / or coverage.
[0010] In an implementation of the first or second aspects, the processor can be further configured to generate, using the measures of similarity, a loss value for an artificial neural network, NN, of the UE used to calculate a set of spatial channel parameters, wherein the loss value provides an indication of the degree to which a set of spatial channel parameters calculated using the NN match the first set of spatial channel parameters, wherein the NN is configured to use the loss value to update a set of coefficients of the NN.
[0011] A NN can therefore be run-time adapted to the prevailing real-world communication environment existing between the UE and a node.
[0012] In an example, the processor can be further configured to use the loss value to update a set of coefficients of the NN for one of: a predetermined number of update iterations of the NN, each update iteration using a loss value to modify one or more coefficients of the NN, or until a calculated loss value is below a pre-defined threshold value.
[0013] In an example, corresponding to, e.g., an implementation of the first aspect, the processor can be further configured to calculate a pair of candidate spatial channel parameter sets forming a pair of solutions for uplink channel parameters from the UE to the node, and receive, from the node, on the basis that the performance metric is greater than a predetermined threshold value, a signal configured to trigger a beam selection process at the UE for selection of one of the pair of candidate spatial channel parameter sets.
[0014] In an implementation of the first or second aspects, the processor can be further configured to transmit, to the node, a pair of reference signals, each reference signal of the pair of reference signals transmitted using a different beam pattern, wherein each one of the different beam patterns is configured using a respective one of the pair of candidate spatial channel parameter sets, and receive, from the node, a message indicating one of the pair of solutions for uplink channel parameters from the UE to the node resulting in a highest received power measurement at the node.
[0015] In an example, corresponding to, e.g., an implementation of the first aspect, the processor can be further configured to receive, from the node, the pair of candidate spatial channel parameter sets in a reduced payload message in which a pair of solutions for uplink channel parameters from the UE to the node share the same value for an angle of departure, the payload message comprising a single real-valued angle of departure vector associated with multiple complex-valued gain vectors for the pair of solutions.
[0016] A third aspect of the present disclosure provides an apparatus in a mobile telecommunications network, the apparatus comprising a memory, and a processor, operationally coupled to the memory, configured to transmit, to a UE, a first trigger signal configured to trigger transmission, to the apparatus, of multiple signals, each signal of the multiple signals transmitted by the UE using a different beam pattern from a first set of UE beams, receive, at the apparatus, each signal of the multiple signals and, each time a signal of the multiple signals is received, calculate a received power value of the received signal, generate estimated values for a first set of spatial channel parameters for a channel between the UE and the apparatus using the received power values, calculate a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams, generate a performance metric on the basis of measures of similarity between respective ones of received power values and calculated power values of the second set of received power values, compare the performance metric to a first predetermined threshold performance metric value, and, in the event that the performance metric is greater than the first predetermined threshold performance metric value, generate a final performance metric over all the multiple signals for the first set of UE beams, transmit, to the UE, in the event that the final performance metric is greater than second predetermined threshold value, a second trigger signal configured to trigger a beam selection process at the UE for selection of a set of spatial channel parameters from a pair of candidate spatial channel parameter sets forming a pair of solutions for uplink channel parameters from the UE to the apparatus, receive, at the apparatus, a pair of reference signals, each signal of the pair of reference signals transmitted to the node by the UE using a different beam pattern configured using respective ones of the pair of candidate spatial channel parameters, and transmit, to the UE, a message indicating one of the pair of solutions for uplink channel parameters from the UE to the apparatus resulting in a highest received power measurement at the apparatus.
[0017] In an implementation of the third aspect, the processor can be further configured to transmit, to the UE, the pair of candidate spatial channel parameter sets in a reduced payload message in which the pair of solutions for uplink channel parameters from the UE to the apparatus share the same value for an angle of departure, the payload message comprising a single real-valued angle of departure vector associated with multiple complex-valued gain vectors for the pair of solutions. The processor can be further configured to, in the event that a received power value is less than or equal to the first predetermined threshold power value, transmit an abort message to the UE to trigger termination of transmission of any remaining ones of the multiple signals for the first set of UE beams. The processor can be further configured to, in the event that the final performance metric is less than or equal to the second predetermined threshold value, transmit, to the UE, a message identifying a reference signal transmitted by the UE with the highest received power within a first set of received power values, the identified reference signal representing a fall-back beam pattern for the UE selected from the first set of UE beams. The processor can be further configured to generate, using the measures of similarity, a loss value for an artificial neural network, NN, of the apparatus used to calculate a set of spatial channel parameters, wherein the loss value provides an indication of the degree to which a set of spatial channel parameters calculated using the NN match the first set of spatial channel parameters, wherein the NN is configured to use the loss value to update a set of coefficients of the NN. The processor can be further configured to use the loss value to update a set of coefficients of the NN for one of: a predetermined number of update iterations of the NN, each update iteration using a loss value to modify one or more coefficients of the NN, or until a calculated loss value is below a pre-defined threshold value.
[0018] A fourth aspect of the present disclosure provides a method for beam management in a user equipment, UE, configured to support beam management for UE transmission, the method comprising transmitting, to a node, multiple reference signals, each signal of the multiple reference signals transmitted using respective different beam patterns of a first set of UE beams, generating estimated values for a first set of spatial channel parameters for a channel between the UE and the node using a first set of received power values, wherein the first set of received power values comprises a power value for each one of the multiple signals received at the node, calculating a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams, generating a performance metric on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values, and transmitting the performance metric to the node.
[0019] A fifth aspect of the present disclosure provides a method for beam management in a user equipment, UE, configured to support beam management for UE reception, the method comprising receiving, from a node, multiple reference signals, each signal of the multiple reference signals received by the UE using respective different beam patterns of a first set of UE beams, measuring a first set of received power values, wherein the first set of received power values comprises a received power value for each one of the multiple signals received at the UE, generating estimated values for a first set of spatial channel parameters for a channel between the UE and the node using the first set of received power values, calculating a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams, and generating a performance metric on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values.
[0020] In an implementation of the fourth or fifth aspects, the method can further comprise generating, using the measures of similarity, a loss value for an artificial neural network, NN, of the UE used to calculate a set of spatial channel parameters, wherein the loss value provides an indication of the degree to which a set of spatial channel parameters calculated using the NN match the first set of spatial channel parameters, wherein the NN is configured to use the loss value to update a set of coefficients of the NN.
[0021] In an example, the method can further comprise using the loss value to update a set of coefficients of the NN for one of: a predetermined number of update iterations of the NN, each update iteration using a loss value to modify one or more coefficients of the NN, or until a calculated loss value is below a pre-defined threshold value.
[0022] In an example, corresponding to, e.g., an implementation of the fourth aspect the method can further comprise calculating a pair of candidate spatial channel parameter sets forming a pair of solutions for uplink channel parameters from the UE to the node, and receiving, from the node, on the basis that the performance metric is greater than a predetermined threshold value, a signal configured to trigger a beam selection process at the UE for selection of one of the pair of candidate spatial channel parameter sets. The method can further comprise transmitting, to the node, a pair of reference signals, each reference signal of the pair of reference signals transmitted using a different beam pattern, wherein each one of the different beam patterns is configured using a respective one of the pair of candidate spatial channel parameter sets, and receiving, from the node, a message indicating one of the pair of solutions for uplink channel parameters from the UE to the node resulting in a highest received power measurement at the node. The method can further comprise receive, from the node, the pair of candidate spatial channel parameter sets in a reduced payload message in which a pair of solutions for uplink channel parameters from the UE to the node share the same value for an angle of departure, the payload message comprising a single real-valued angle of departure vector associated with multiple complex-valued gain vectors for the pair of solutions.
[0023] A sixth aspect of the present disclosure provides method for beam management in a node of a mobile telecommunications network, the method comprising transmitting, to a UE, a first trigger signal configured to trigger transmission, to the node, of multiple signals, each signal of the multiple signals transmitted by the UE using respective different beam patterns of a first set of UE beams, receiving, at the node, each signal of the multiple signals and each time a signal of the multiple signals is received, calculating a received power value of the received signal, generating estimated values for a first set of spatial channel parameters for a channel between the UE and the node using the received power values, calculating a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams, generating a performance metric on the basis of measures of similarity between respective ones of received power values and calculated power values of the second set of received power values, comparing the performance metric to a first predetermined threshold performance metric value, and, in the event that the performance metric is greater than the first predetermined threshold performance metric value, generating a final performance metric over all the multiple signals for the first set of UE beams, transmitting, to the UE, in the event that the final performance metric is greater than second predetermined threshold value, a second trigger signal configured to trigger a beam verification process at the UE for selection of a set of spatial channel parameters from a pair of candidate spatial channel parameters forming a pair of solutions for uplink channel parameters from the UE to the node, receiving, at the node, a pair of reference signals, each signal of the pair of reference signals transmitted to the node by the UE using different beam patterns configured using respective ones of the pair of candidate spatial channel parameters, and transmitting, to the UE, a message indicating one of the pair of solutions for uplink channel parameters from the UE to the node resulting in a highest received power measurement at the apparatus.
[0024] In an implementation of the sixth aspect, the method can further comprise transmitting, to the UE, the pair of candidate spatial channel parameter sets in a reduced payload message in which the pair of solutions for uplink channel parameters from the UE to the apparatus share the same value for an angle of departure, the payload message comprising a single real-valued angle of departure vector associated with multiple complex-valued gain vectors for the pair of solutions. The method can further comprise, in the event that a received power value is less than or equal to the first predetermined threshold power value, transmitting an abort message to the UE to trigger termination of transmission of any remaining ones of the multiple signals for the first set of UE beams. The method can further comprise, in the event that the final performance metric is less than or equal to the second predetermined threshold value, transmitting, to the UE, a message identifying a reference signal transmitted by the UE with the highest received power within a first set of received power values, the identified reference signal representing a fall-back beam pattern for the UE selected from the first set of UE beams. The method can further comprise generating, using the measures of similarity, a loss value for an artificial neural network, NN, of the apparatus used to calculate a set of spatial channel parameters, wherein the loss value provides an indication of the degree to which a set of spatial channel parameters calculated using the NN match the first set of spatial channel parameters, wherein the NN is configured to use the loss value to update a set of coefficients of the NN. The method can further comprise using the loss value, updating a set of coefficients of the NN for one of: a predetermined number of update iterations of the NN, each update iteration using a loss value to modify one or more coefficients of the NN, or until a calculated loss value is below a pre-defined threshold value.
[0025] These and other aspects of the invention will be apparent from the embodiment (s) described below.Brief Description of the Drawings
[0026] In order that the present disclosure may be more readily understood, embodiments will now be described, by way of example, with reference to the accompanying drawings, in which:
[0027] Figure 1 is a schematic representation of beam management performance monitoring metric generation, according to an example;
[0028] Figure 2 is a schematic representation of beam management performance monitoring metric generation, according to an example;
[0029] Figure 3 is a schematic representation for a messaging sequence between a UE and a BS, according to an example;
[0030] Figure 4 is a schematic representation for a messaging sequence between a UE and a BS, according to an example; Figure 5 is a flowchart of a method for beam management in a UE configured to support beam management for UE transmission, according to an example;
[0031] Figure 6 is a flowchart of a method for beam management in a UE configured to support beam management for UE reception, according to an example; and
[0032] Figure 7 is a schematic representation of a machine according to an example.Detailed Description
[0033] Example embodiments are described below in sufficient detail to enable those of ordinary skill in the art to embody and implement the systems and processes herein described. It is important to understand that embodiments can be provided in many alternate forms and should not be construed as limited to the examples set forth herein.
[0034] Accordingly, while embodiments can be modified in various ways and take on various alternative forms, specific embodiments thereof are shown in the drawings and described in detail below as examples. There is no intent to limit to the particular forms disclosed. On the contrary, all modifications, equivalents, and alternatives falling within the scope of the appended claims should be included. Elements of the example embodiments are consistently denoted by the same reference numerals throughout the drawings and detailed description where appropriate.
[0035] The terminology used herein to describe embodiments is not intended to limit the scope. The articles “a, ” “an, ” and “the” are singular in that they have a single referent, however the use of the singular form in the present document should not preclude the presence of more than one referent. In other words, elements referred to in the singular can number one or more, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises, ” “comprising, ” “includes, ” and / or “including, ” when used herein, specify the presence of stated features, items, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, items, steps, operations, elements, components, and / or groups thereof. The term “and / or” is only an association relationship for describing associated objects and represents that three relationships may exist such that A and / or B may indicate that A exists alone, A and B exist at the same time, or B exists alone. The character “ / ” generally represents that the associated objects are in an “or” relationship.
[0036] Unless otherwise defined, all terms (including technical and scientific terms) used herein are to be interpreted as is customary in the art. It will be further understood that terms in common usage should also be interpreted as is customary in the relevant art and not in an idealized or overly formal sense unless expressly so defined herein.
[0037] The following contains specific information related to implementations of the present disclosure. The drawings and their accompanying detailed disclosure are merely directed to implementations. However, the present disclosure is not limited to these implementations. Other variations and implementations of the present disclosure will be obvious to those skilled in the art.
[0038] The phrases “in one implementation, ” or “in some implementations, ” may each refer to one or more of the same or different implementations. The term “coupled” is defined as connected whether directly or indirectly through intervening components and is not necessarily limited to physical connections. The expression “at least one of A, B and C” or “at least one of the following: A, B and C” means “only A, or only B, or only C, or any combination of A, B and C” .
[0039] The terms “system” and “network” may be used interchangeably.
[0040] For the purposes of explanation and non-limitation, specific details such as functional entities, techniques, protocols, and standards are set forth for providing an understanding of the present disclosure. In other examples, detailed disclosure of well-known methods, technologies, systems, and architectures are omitted so as not to obscure the present disclosure with unnecessary details.
[0041] Persons skilled in the art will immediately recognize that any network function (s) or algorithm (s) disclosed may be implemented by hardware, software or a combination of software and hardware. Disclosed functions may correspond to modules which may be software, hardware, firmware, or any combination thereof.
[0042] A software implementation may include machine-and / or computer-readable and / or executable instructions stored on a machine-and / or computer-readable medium such as memory or other types of storage devices. One or more microprocessors or general-purpose computers with communication processing capability may be programmed with corresponding executable instructions and perform the disclosed network function (s) or algorithm (s) .
[0043] The microprocessors or general-purpose computers may include Applications Specific Integrated Circuitry (ASIC) , programmable logic arrays, and / or using one or more Digital Signal Processor (DSPs) . Although some of the disclosed implementations are oriented to software installed and executing on computer hardware, alternative implementations implemented as firmware or as hardware or as a combination of hardware and software are well within the scope of the present disclosure. The computer readable medium includes but is not limited to Random Access Memory (RAM) , Read Only Memory (ROM) , Erasable Programmable Read-Only Memory (EPROM) , Electrically Erasable Programmable Read-Only Memory (EEPROM) , flash memory, Compact Disc Read-Only Memory (CD-ROM) , magnetic cassettes, magnetic tape, magnetic disk storage, or any other equivalent medium capable of storing computer-readable instructions.
[0044] The following acronyms and / or abbreviations may be used herein, as follows:
[0045] Full Name Acronym / Abbreviation
[0046] 3rd Generation Partnership Project 3GPP
[0047] Long Term Evolution LTE
[0048] New Radio access technology for the Fifth Generation 5G NR
[0049] mobile network
[0050] Beam Management BM
[0051] Sounding Reference Signal SRS
[0052] Orthogonal Frequency-Division Multiplexing OFDM
[0053] Cyclic Prefix CP
[0054] Radio Frequency RF
[0055] Base Station BS
[0056] User Equipment UE
[0057] Power Amplifier PA
[0058] Low Noise Amplifier LNA
[0059] Customer Premises Equipment CPE
[0060] Line Of Sight LOS
[0061] Non Line Of Sight NLOS
[0062] Uplink UL
[0063] Downlink DL
[0064] Radio Resource Control RRC
[0065] Downlink Control Information DCI
[0066] Medium Access Control Control Element MAC CE
[0067] Physical Uplink Shared Channel PUSCH
[0068] Physical Uplink Control Channel PUCCH
[0069] Physical Downlink Shared Channel PDSCH
[0070] Physical Downlink Control Channel PDCCH
[0071] Reference Signal Received Power RSRP
[0072] Neural Network NN
[0073] Artificial Intelligence AI
[0074] Transmitter TX
[0075] Receiver RX
[0076] Channel State Information CSI
[0077] Eigen Value Decomposition EVD
[0078] Angle of Departure AoD
[0079] Uniform Linear Array ULA
[0080] Uniform Rectangular Array URA
[0081] Mean Square Error MSE
[0082] In general, a UE (user equipment) refers to a terminal device with cellular communication capability, such as a smart phone device, a customer premises equipment device, a cellular communication capable vehicle, etc.
[0083] For a ULA, the spatial frequency corresponding to a UL channel path l, is defined as:
[0084] wherein, Δ: antenna spacing, λ: wavelength, θl angle of departure.
[0085] For a URA, the spatial frequencies corresponding to a UL channel path l, are defined as:
[0086] For the horizontal spatial frequency:
[0087] For the vertical spatial frequency:
[0088] wherein, Δh: horizontal antenna spacing, Δv: vectical antenna spacing, λ: wavelength, θv, l: angle of departure in vertical direction, θh, l: angle of departure in horizontal direction.
[0089] To enable efficient analog beamforming, a UE can identify a phase shifter setting that will steer a TX signal towards the direction of a base station node / receiver. Herein, a TX phase shifter setting is called a UE TX beam. A phase shifter setting for a base station that maximizes gain in relation to an RX signal from the UE is correspondingly referred to as a BS RX beam.
[0090] In 3GPP, the procedure to identify the best UE TX beam and the best BS RX beam, aiming to optimize the uplink (UL) communication from the UE to the base station is called UL beam management (BM) . Similarly, the procedure to identify the best BS TX beam and the UE RX beam, aiming to optimize downlink (DL) communication is called DL BM.
[0091] AI based BM has been introduced in the 3GPP DL BM framework. Two sets of pre-defined beams are defined as part of this framework:
[0092] ● SetB: a set of beams corresponding to measurements (used to learn the features of the current wireless spatial channel) ;
[0093] ● SetA: a set of candidate communications beams targeted for prediction (prediction means to select the best candidate beams for communication) .
[0094] The basic concept of the current AI based BM is that the RSRP measurements, which are determined using one or more beams from setB, can be input to a NN. By assuming that the NN has been pre-trained to learn an association between RSRP values under beams of setB and corresponding best communication beam indexes from beams of setA, the NN can (at run-time) predict the indexes of the top-K (K>=1) best suitable communication beams from setA based on the run-time RSRP measurements under setB beams.
[0095] In the context of mmWave TX beamforming, a first device (transmission device, or transmitter) can transmit a set of reference signals (for example sounding reference signals (SRS) ) to a second device (reception device, or receiver) . The transmitter can be equipped with an analog phased array with N antennas and can transmit M reference signals at M different time instances, each associated with a different TX beam. The set of those TX beams is denoted as the setB, which are used to capture of the features of the spatial channel.
[0096] Meanwhile, the reception device can measure the received reference signal powers (RSRPs) of the received reference signals which are transmitted under the setB beams by the transmission device. The receiving device can have a single antenna or an analog phased array. When the receiver is equipped with a phased array, it can fix its RX beam during the setB TX beam transmissions. Ignoring the measurement noise, the measured RSRP vector corresponds to the following formula: RSRP = diag {WRhhWH} ∈CM×1 (1)
[0097] Where N is the number of antennas within the TX phased array, M is the number of TX beams in setB, Rhh∈CN×N is the spatial covariance matrix of the wireless channel, and W∈CM×N is the codebook matrix corresponding to the setB beams, wherein each row of W corresponds to one TX beam in setB. An example of a typical setB codebook is the DFT codebook.
[0098] If Rhh can be estimated from the measured RSRPs, then instead of selecting a beam index from a pre-defined beam set (setA) , an optimal TX communication beam, which is matched to the actual spatial channel, can be directly computed based on Rhh in the run-time, such that the beamforming gain can be maximized. One example of how to compute the optimal communication beam based on Rhh is to first apply the eigen value decomposition (EVD) of Rhh , and then select the eigen vector which corresponds to the maximal eigen value: The phase vector of the conjugated version of the selected eigen vector, is the optimal phase shifter setting (optimal TX beam) , which maximizes the beamforming gain for the transmission.
[0099] The projection from Rhh to the RSRPs is trivial and linear. However, the inverse operation (from the measured RSRPs to recover the Rhh) belongs to the class of phase retrieval problems and is neither linear nor trivial. Furthermore, the solution for the recovery is not unique. Specifically, due to the quadratic form of (1) , there can exist more than 2 different Rhh matrices as the solutions, both of which fulfill the same formula (1) .
[0100] In the context of UL beam management for, e.g., 5G NR mmWave band communications, a BS and a UE need to determine the optimal UE TX communication beam which can optimize the beamforming gain for UL communication. Existing methods, which are based on pre-defined beam patterns, cannot achieve the best beamforming gain because they are not able to be fully adapted to the run-time spatial channel environment, especially when there are strong NLOS paths.
[0101] According to an example, there is provided a closed-loop performance monitoring method for RSRP based TX beam online beam synthesis that is based on measured RSRPs under setB beams (with those RSRPs being denoted herein as “input RSRPs” ) . Instead of predicting the Top-K communication beams from setA, a UE can use the input RSRPs to estimate UL spatial channel parameters. An RSRP replay operation can then be made, which reconstructs the input RSRPs by the projecting the estimated spatial channel parameters with the setB codebook. A measure of the similarity between the reconstructed input RSRPs and the actual (measured) input RSRPs can be computed to provide a performance monitoring metric. Upon having achieved good spatial channel estimates, an optimal UE TX beam pattern (s) can be computed.
[0102] Figure 1 is a schematic representation of a BM performance monitoring metric generation, according to an example. In the example of figure 1, input RSRPs 101 are provided to an estimator 103. The input RSRPs 101 can be determined by a network node that has received a set of signals transmitted from a UE using setB beams, for example, with the network node having calculated the input RSRPs 101 and returned the calculated values to the UE, which can then use estimator 103 to calculate estimated spatial channel parameters 107 on the basis of the input RSRPs 101 received from the network node (BS) . The estimated spatial channel parameters 107 are used by an RSRP replay module 109, along with the set of setB beam patterns 111 to calculate a set of reconstructed input RSRPs 113. The reconstructed input RSRPs 113 can be used, along with the measured input RSRPs 101, to generate a performance monitoring metric 115, providing a measure of the similarity between the reconstructed input RSRPs 113 and the actually measured input RSRPs 101, using a similarity computation module 105. The performance monitoring metric 115 can be reported from the UE to the BS so that UL BM procedures can be adapted by conditional early termination of UL SRS transmissions.
[0103] Accordingly, with reference to figure 1, measured input RSRPs 101, which are associated with UL reference signals 150 (such as SRSs) and which are transmitted from a UE 100 to a BS 110 using setB beams, are provided to estimator 103 as input. The estimator 103 generates the estimated spatial channel parameters 107 as output. If the antenna array of the UE 100 is, for example, a uniform linear array (ULA) with N antennas, then the estimated spatial channel vector 107, which is constructed by the estimated spatial channel parameters, can be calculated according to:
[0104] where, L is a pre-defined number of effective paths, gl is the complex-valued gain for the path l, μl is the real-valued spatial frequency corresponding to the angle of departure (AoD) for the channel path l, and a (μl) is the steering vector corresponding to the path l, defined as Accordingly, for ULA, the estimated spatial channel parameters (output of the estimator / NN) are the spatial frequency μl and the path gain gl (1≤l≤L) .
[0105] If the antenna array is a uniform rectangular array (URA) with N= Nv×Nh antennas, then the estimated spatial channel vector, which is constructed by the estimated spatial channel parameters, can be calculated according to:
[0106] where, L is a pre-defined number of effective paths, gl is the complex-valued gain for the path l, μv, l and μh, l are the real-valued spatial frequencies of the channel path l, corresponding to the vertical AoD and the horizontal AoD respectively, and are the vertical and horizontal steering vectors respectively, and is the Kronecker product operator. Accordingly, for URA, the estimated spatial channel parameters (outputs of the estimator / NN) are the spatial frequency pair (μv, l, μh, l) and the path gain gl (1≤l≤L) .
[0107] For both ULA and URA, the estimator could also directly estimate the spatial covariance matrix
[0108] Having estimated the spatial channel parameters 107, an RSRP replay operation 109 can be performed in which a channel vector is first re-constructed using (2) or (3) , and then the spatial channel covariance matrix is re-constructed according to the following:
[0109] where (. ) H is the transpose conjugate operator.
[0110] With the input RSRPs can further be reconstructed according to:
[0111] where W∈C M×N is the codebook matrix corresponding to setB beams.
[0112] Note that the RSRP vector in (1) is the actual RSRPs 101 from the physical measurements, which are related to the true spatial channel covariance matrix Rhh, which is projected by the setB codebook matrix W. The RSRP vector forms the physical measurements and the input to the spatial channel estimator 103. The vector in (5) is the reconstructed input RSRPs 113 based on RSRP replay module 109. It is related to the estimated spatial channel covariance matrix which is projected by the same the codebook matrix W.
[0113] If is correctly estimated by the estimator 103, then the reconstructed input vector in (5) will be very similar to the actual input measured RSRP vector 101 in (1) . Hence, the similarity between RSRP (101) and (113) can be computed at 105 as the performance monitoring metric 115 and can be used to monitor the estimation performance of the estimator 103. The performance monitoring metric 115 could be, e.g., the mean square error (MSE) between RSRP (101) and (113) . Alternatively, the performance monitoring metric 115 could be, e.g., the cosine-similarity between RSRP (101) and (113) .
[0114] Figure 2 is a schematic representation of a BM performance monitoring metric generation, according to an example. In the example of figure 2, the estimator 103 of figure 1 is replaced by a neural network (NN) 201. As such, the closed-loop performance monitoring metric computation as described with reference to figure 1 can be used to provide a loss value 203 to NN 201 to enable the coefficients of the NN 210 to be updated / fitted in an unsupervised manner. As such, NN 201 can be trained online in the field without the need for, e.g., ground-truth labels. Since the loss computation is differentiable, classic NN fitting methods such as gradient descent (GD) and back-propagation (BP) can be reused.
[0115] That is, a proposed performance monitoring metric can be used to compute a loss value 203, which can be used to fit the coefficients of the NN 201. When the performance monitoring metric shows a high similarity between the input RSRP vector 101 and the replayed RSRP vector (113) , it means that the estimated spatial channel parameters 107 have good accuracy, hence a smaller loss value 203 could be generated. Conversely, when the metric shows low similarity between the input RSRP vector 101 and the replayed RSRP vector (113) , it means the that estimated spatial channel parameters 107 have low accuracy, hence a higher loss value 203 could be generated.
[0116] Since the computation of the similarity between the input RSRPs 101 and the replayed RSRPs (113) is differentiable, the loss value 203 is also differentiable. Hence, e.g., classic gradient descent (GD) based methods can be reused to fit the coefficients of the NN 201 by backpropagation, in an iterative manner. Furthermore, by considering that the fitting of the NN coefficients is done in an unsupervised manner, online fitting is also possible, meaning that, according to an example, the NN coefficients can be updated at run-time in the field based on the run-time measured RSRP vector (101) . Such online NN fitting could be terminated when, e.g., the loss value 203 is below a pre-defined threshold value, or when a maximal number of fitting iterations is reached. In an example, when the online fitting procedure is finished, the latest outputs of the spatial channel parameters from the NN 201 can be used to compute an optimal TX beam which is matched to the run-time spatial channel.
[0117] As noted above, if the spatial channel covariance matrix Rhh can be estimated from RSRPs, one example of how to compute the optimal communication TX beam based on Rhh is to first apply the eigen value decomposition (EVD) of Rhh , and then select the eigen vector which corresponds to the maximal eigen value: The phase vector of the conjugated version of the selected eigen vector, is the optimal phase shifter setting (optimal TX beam) , which maximizes the beamforming gain for the transmission.
[0118] As noted above, due to the quadratic form of formula (1) , there could exist more than one solution of the problem. That is, multiple can exist as solutions, which can all be projected to the same RSRP measurement vector. Hence, the estimator 103 and the NN 201 can generate more than one set of estimated spatial channel parameters 107, each corresponding to one possible solution for Within those solutions there is one true solution for which can be transformed to the optimal TX beam. The other solution for will result in a sub-optimal TX beam. In order to determine the solution for resulting in the optimal TX beam, according to an example, a beam verification process can be performed as described in more detail below.
[0119] Figure 3 is a schematic representation for a messaging sequence between a UE and a BS, according to an example. In the example of figure 3, the messaging sequence relates to an adaptive UL BM procedure for UE TX beam synthesis. In an example, the performance monitoring metric, e.g., based on RSRP replay as described above, can be reported from the UE 301 to the BS 303. In the example of figure 3, unsupervised AI processing used to estimate UL spatial channel parameters based on RSRPs can be implemented at the UE 301.
[0120] As noted above, due to the quadratic form of (1) more than one estimate which is a solution to (1) can exist. Accordingly, if the performance monitoring metric shows a low measure of similarity between the actually measured SRS RSRP vector and the replayed SRS RSRP vector, the estimated spatial channel parameters have high errors. On the other hand, if the performance monitoring metric shows a high measure of similarity between the input RSRP vector and the replayed RSRP vector, it does not mean that the estimated spatial channel parameter vector is well matched to the actual true spatial channel. This is because it could be another false, i.e., sub-optimal, solution of (1) that does not correspond to the true spatial channel that exists between the UE 301 and the BS 301. Hence, an additional beam verification process can be used to resolve such solution ambiguity.
[0121] With reference to figure 3, at (300) , a UE capability message is sent from the UE 301 to the BS 303. The UE capability message comprises information relating to whether the UE 301 supports online learning for RSRP based spatial UL channel estimation. It can also comprise information representing of a minimum time interval (denoted as ‘Tproc’ in figure 3) in between when the UE 301 has received SRS RSRP measurement information (e.g., SRS RSRP measurement information) from the BS 301 to when the UE 301 is asked to report a calculated performance monitoring metric back to the BS 303.
[0122] At (301) , the BS 303 configures the association information of two SRS resource sets to the UE 301, so that the UE 301 can link the two sets -one for the feature collection and the other for beam verification. Note that messages at (300) and (301) are, in an example, semi-static messages.
[0123] At (302) , the BS 303 sends a first trigger message to the UE 301 to trigger the transmission of the first set of SRSs. At (303) , the UE 301 transmits the triggered set of SRSs, wherein each SRS is transmitted using a distinct TX beam pattern from setB. That is, UE 301 transmits, to the BS 303, multiple reference signals, each signal of the multiple reference signals transmitted using a different beam pattern selected from a first set of UE beams.
[0124] The BS 303 measures the RSRPs of the received SRSs. In an example, BS 303 measures the RSRPs of the received SRSs as the SRSs are received by the BS 303.
[0125] At (304) , the BS 303 sends a message to the UE 301 to indicate the measured SRS RSRPs. Generally speaking, (302) to (304) can be characterized as a feature collection stage during which the BS 303 receives signals from the UE 301 that are sent using respective different transmission beam settings (from setB) .
[0126] At (305) , the UE 301 estimates the spatial UL channel parameters, based on the received SRS RSRPs that were calculated by the BS 303 and transmitted to the UE 301 at (304) . The UE 301 also generates the performance monitoring metric based on, e.g., RSRP reply as described above. The UE 301 can also compute candidate TX communication beams based on the estimated spatial channel parameters. However, due to the ambiguity of the problem there can be multiple TX communication beams, which need to be further selected in a beam verification process.
[0127] Accordingly, at (306) the UE 301 can send a message to the BS 303 to indicate the performance monitoring metric that has been calculated. Note that, as a scheduling master, BS 303 should ensure that the time interval between the SRS-RSRP indication at (304) and the performance monitoring metric indication at (306) is not smaller than Tproc, so that the UE 301 processing time at (305) can be ensured.
[0128] At (307) , upon having received the UE 301 indicated performance monitoring metric, BS 303 compares it with a pre-defined threshold value. If the metric is higher than the pre-defined threshold value it indicates high confidence in the estimation. However, as mentioned, due to the quadratic form of formula (1) , there exist multiple solutions of the equation. Hence the UE 301 estimator can generate more than one set of spatial channel parameters, each corresponding to one of the solutions. Even in the case that the performance monitoring metric shows a high degree of similarity between the measured SRS RSRP vector and the replied SRS RSRP vector, the estimated spatial channel parameters could still be sub-optimal (e.g., providing a false-alarmed solution) comparing with a ground-true solution. Hence, a beam verification process is used in order to select the true solution. This is done art (308) , where the BS 303 sends a second trigger message to the UE 301 which triggers the UE 301 to transmit the second set of SRSs. As such, each of the SRS is transmitted by the UE 301 using one candidate communication beam which is matched to one possible solution (asolution is a set of estimated spatial channel parameters which correspond to one possible ) . The BS 303 meanwhile measures the SRS RSRPs of the received SRSs. If an SRS is transmitted by a candidate communication beam which is matched to the true spatial channel parameters, it shall result in a high SRS RSRP measurement in the BS 303 side. Otherwise, if an SRS is transmitted by a candidate communication beam which is matched to the false-alarmed spatial channel parameters, it shall result in a low SRS RSRP measurement the BS 303 side.
[0129] At (309) , based on the RSRP levels the BS 303 can select the best received SRS and indicate its index to the UE 301. The UE 301 can then determine the best TX communication beam.
[0130] Referring back (307) , if the performance monitoring metric is lower or equal than the threshold it indicates low confidence in the estimation. In this case beam verification is no longer meaningful as the estimation fails completely. Hence the procedure directly jumps to (309) . Although the 2nd set of SRSs are not measured, the BS 303 can still select one of the best SRS index based on the SRS RSRP measurements obtained in (303) , as a fallback method, such that the UE 301 can still select the best beam corresponding to set, as the communication beam.
[0131] Accordingly, the performance monitoring metric, which is reported from UE 301 to the BS 301 at (306) , can be the similarity metric between the measured SRS RSRP vector and the replayed SRS RSRP vector as both correspond to setB beams. It can also be other performance monitoring metrics. For example, it can comprise a single 1-bit piece information to inform the BS 303 whether the RSRP based spatial channel estimation was successful (e.g., equal to 1) or a failure (e.g., equal to 0) .
[0132] According to an example, for UL BM, the AI structure outlined with respect to figure 2 can also be implemented at or for a BS. In this case, instead of indicating estimated UE TX beam patterns to the UE, by exploring the specific structure of the solution space for RSRP based spatial channel estimation, an optimized indicating structure can be implemented in which a BS can indicate 2 effective complexed-valued channel gain vectors, shared by 1 AoD vector, to the UE. This approach reduces the payload for messaging between the BS and the UE. The UE can then use this information to compute the two candidate TX communication beams. One of the two will correspond to the optimal UE TX communication beam while the other corresponds to a solution that does not provide an optimal UE TX communication beam. A beam verification process can then be used to select a final beam.
[0133] In an adaptive UL BM procedure for UE TX beam synthesis, the unsupervised AI processing, which is used to estimate the UL spatial channel parameters based on the RSRPs, can be located at the BS. As noted, in this case the performance monitoring metric is computed together with the UL spatial channel parameters at the BS.
[0134] Again, as noted, due to the quadratic form of formula (1) and the multi-solutions property of the problem, a beam verification process is still needed to rule out sub-optimal or false alarm solutions. According to an example, when the generation of the solutions (estimation of the multiple spatial channel parameter sets based on the RSRP measurements) is done in the BS side, the generated solutions have to be indicated to the UE. Upon having received those, the UE can compute the corresponding UE TX candidate communication beams. The UE can further transmit the SRSs using the computed UE TX candidate communication beams in the beam verification process.
[0135] According to an example, to reduce the payload of the indication message it is noted that the multiple solutions for RSRP based spatial channel parameters share a same AoD vector but only differ the complexed channel gains.
[0136] For one example, when the UE TX antenna array is a ULA with N antennas, the multi-solution structure for the RSRP based spatial channel parameters is as follows:
[0137] Solution a:
[0138] Solution b:
[0139] where, L is a pre-defined number of effective paths, is the complex-valued gain for the effective path l, corresponding to the 1st solution ha, is the complex-valued gain for the effective path l, corresponding to the 2nd solution hb, μl is the real-valued spatial frequency corresponding to the angle of departure (AoD) for the channel path l, a (μl) is the steering vector corresponding to the path l, defined as For each of the effective paths l, a (μl) is the same for ha and hb.
[0140] In another example, when the UE TX antenna array is a URA with N= Nv×Nh antennas, the multi-solution structure for the spatial UL channel parameters is as follows:
[0141] Solution a:
[0142] Solution b:
[0143] where, L is a pre-defined number of effective paths, is the complex-valued gain for the path l, corresponding to the solution ha, is the complex-valued gain for the path l, corresponding to the solution hb, μv, l and μh, l are the real-valued spatial frequencies of the channel path l, corresponding vector AoD and the horizontal AoD, respectively, and are the vertical and horizontal steering vectors respectively, and is the Kronecker product operator. For each of the effective paths l, is the same for ha and hb .
[0144] According to an example, exploring such structures in order to save payload bits, the indication message could contain a single real-valued AoD vector associated with multiple complex-valued gain vectors. Configuration examples for ULA and URA cases are:
[0145] Figure 4 is a schematic representation for a messaging sequence between a UE and a BS, according to an example.
[0146] At (400) , a UE (301) antenna information message is sent from the UE 301 to the BS 303. In an example, the message comprises information about UE TX antenna arrays. For example, in case that UE 301 comprises a ULA, then information comprises the number of antennas (N) within the array. In case that UE 301 comprises a URA, the information comprises a number of horizontal antennas (Nh) and a number of vertical antennas (Nv) . This is used by the BS 303 to compute spatial channel parameters based on SRS RSRP measurements and also to generate an indication message for the multi-solutions as described with reference to the configuration examples for the ULA and URA cases shown above.
[0147] At (401) , the BS 303 configures the association information of two SRS resource sets to the UE 301, so that UE 301 can link the two sets (one for feature collection and the other for the beam verification process) . Note that the messages at (400) and (401) can be, in an example, semi-static messages.
[0148] At (402) , the BS 303 transmits a first trigger message to the UE 301 to trigger the transmission of the first set of SRSs.
[0149] At (403) , the UE 301 transmits the triggered set of SRSs, wherein each SRS is transmitted using a district TX beam pattern from the setB. The BS 303 measures the RSRPs of the received SRSs at the same time as each of the SRSs is received from the UE 301. Since the RSRP based spatial channel parameter estimation is done at the BS 303 side, the BS 303 can update the computation of the performance monitoring metric every time a new SRS RSRP is measured (thereby representing a streaming mode) , and compare it with a pre-defined threshold value (Th1) . When the metric is smaller than Th1 the estimation already fails, so in this case the BS 303 can terminate the transmission of the current triggered SRS set by sending an abort message to the UE 301. Otherwise, (i.e., the metric is larger than or the same as Th1) the BS 303 can keep on updating the performance monitoring metric until all SRS RSRPs corresponding to the 1st SRS set are measured.
[0150] At (404) , the BS 303 compares the finally updated performance monitoring metric with a second pre-defined threshold (Th2) . If the metric is higher than Th2 it means the estimations have high confidence. However, as mentioned, due to the quadratic form of formula (1) , there exist multiple solutions of the equation. Hence the estimator 103 or NN 201 can generate more than one set of spatial channel parameters, each corresponding to one of the possible solutions. Note that, even when the performance monitoring metric shows a high degree of similarity between the measured SRS RSRP vector and the replied SRS RSRP vector, the estimated spatial channel parameters could still be a false-alarm / sub-optimal solution compared with the ground-true solution. Hence, in an example, a further beam verification process can be implemented to select the true solution. In such a procedure, beam verification can be started by indicating the multi-solutions for the spatial channel parameters from the BS 303 to the UE 301, at (405) . The indicated message structure for the multi-solutions can be as described with reference to the configuration examples for the ULA and URA cases shown above, in which an AoD vector is shared among the two solutions to reduce message payload.
[0151] At (406) , upon having received the indicated spatial channel parameters corresponding to multi-solutions, the UE 301 computes the communication TX beam candidates, each matched to a different solution (asolution is a set of estimated spatial channel parameters which correspond to one possible ) .
[0152] At (407) , a second trigger message is sent from the BS 303 to the UE 301 to trigger the second set of SRSs for beam verification. Accordingly, the UE 301 transmits the second set of SRSs by using the computed TX communication beam candidates obtained at (406) . Meanwhile, the BS 303 measures the SRS RSRPs.
[0153] At (408) , the BS 303 indicates the SRS ID with the best RSPR measurement to the UE 301. Based on the ID the UE 301 can determine the best optimal UE TX communication beam.
[0154] Referring to (404) , if the performance monitoring metric is lower or equal than the Th2, it means the estimation has low confidence. In this case the beam verification is no longer meaningful as the estimation fails. Hence, the procedure directly jumps to (408) . Although the 2nd set of SRSs are not measured, the BS 303 can still select one of the best SRS measurement based on the SRS RSRPs obtained at (403) as a fallback such that UE 301 can still select the best beam corresponding to setB as the TX communication beam.
[0155] Figure 5 is a flowchart of a method for beam management in a user equipment, UE, configured to support beam management for UE transmission, according to an example.
[0156] In block 501 UE 301 transmits, to a node (or BS) 303, multiple reference signals, each signal of the multiple reference signals transmitted using respective different beam patterns of a first set of UE beams.
[0157] In block 503, node 303 generates estimated values for a first set of spatial channel parameters for a channel between the UE 301 and the node 303 using a first set of received power values. In an example, the first set of received power values comprises a power value for each one of the multiple signals received at the node.
[0158] In block 505, a second set of received power values is calculated using the estimated values for the first set of spatial channel parameters and the first set of UE beams.
[0159] In block 507, a performance metric is generated on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values.
[0160] In block 509, the performance metric is transmitted to the node 303.
[0161] Figure 6 is a flowchart of a method for beam management in a user equipment, UE, configured to support beam management for UE reception, according to an example.
[0162] In block 601 multiple reference signals are received at a UE 301, from a node 303, each signal of the multiple reference signals received by the UE 301 using respective different beam patterns of a first set of UE beams. That is, in the example of figure 6, different profiles for a set of antennae of the UE can be used for reception of signals.
[0163] In block 603, a first set of received power values is measured, wherein the first set of received power values comprises a received power value for each one of the multiple signals received at the UE.
[0164] In block 605, estimated values for a first set of spatial channel parameters for a channel between the UE and the node are generated using the first set of received power values.
[0165] In block 607, a second set of received power values is calculated using the estimated values for the first set of spatial channel parameters and the first set of UE beams.
[0166] In block 609, a performance metric is generated on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values.
[0167] Examples in the present disclosure can be provided as methods, systems or machine-readable instructions, such as any combination of software, hardware, firmware or the like. Such machine-readable instructions may be included on a computer readable storage medium (including but not limited to disc storage, CD-ROM, optical storage, etc. ) having computer readable program codes therein or thereon.
[0168] The present disclosure is described with reference to flow charts and / or block diagrams of the method, devices and systems according to examples of the present disclosure. Although the flow diagrams described above show a specific order of execution, the order of execution may differ from that which is depicted. Blocks described in relation to one flow chart may be combined with those of another flow chart. In some examples, some blocks of the flow diagrams may not be necessary and / or additional blocks may be added. It shall be understood that each flow and / or block in the flow charts and / or block diagrams, as well as combinations of the flows and / or diagrams in the flow charts and / or block diagrams can be realized by machine readable instructions.
[0169] The machine-readable instructions may, for example, be executed by a machine such as a general-purpose computer, a platform comprising user equipment such as a smart device, e.g., a smart phone, a special purpose computer, an embedded processor or processors of other programmable data processing devices to realize the functions described in the description and diagrams. In particular, a processor or processing apparatus may execute the machine-readable instructions. Thus, modules of apparatus (for example, a module implementing an estimator 103, module 105 and / or 109, NN 201 and so on) may be implemented by a processor executing machine readable instructions stored in a memory, or a processor operating in accordance with instructions embedded in logic circuitry. The term 'processor' is to be interpreted broadly to include a CPU, processing unit, ASIC, logic unit, or programmable gate set etc. The methods and modules may all be performed by a single processor or divided amongst several processors.
[0170] Such machine-readable instructions may also be stored in a computer readable storage that can guide the computer or other programmable data processing devices to operate in a specific mode. For example, the instructions may be provided on a non-transitory computer readable storage medium encoded with instructions, executable by a processor.
[0171] Figure 7 is a schematic representation of a machine according to an example. The machine 700 can be, e.g., a system or apparatus, user equipment, node / BS, or part thereof. The machine 700 comprises a processor 703, and a memory 705 to store instructions 702, executable by the processor 703. The machine can comprise a storage 709 that can be used to store data 711 representing, e.g., measured and / or calculated RSRPs, estimated channel parameters, beam pattern information, reconstructed input RSRPs, loss values, NN weights, performance monitoring metrics and so on as described above with reference to figures 1 to 6 for example.
[0172] The instructions, executable by the processor 703, can cause the machine 700, in an example, to transmit, to a node 303, multiple reference signals, each signal of the multiple reference signals transmitted using a different beam pattern selected from a first set of UE beams, generate estimated values for a first set of spatial channel parameters for a channel between the UE 301 and the node 303 using a first set of received power values, wherein the first set of received power values comprises a power value for each one of the multiple signals received at the node 303, calculate a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams, generate a performance metric on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values, and transmit the performance metric to the node 303.
[0173] The instructions, executable by the processor 703, can cause the machine 700, in an example, to receive, from a node 303, multiple reference signals, each signal of the multiple reference signals received by the UE 301 using a different beam pattern selected from a first set of UE beams, measure a first set of received power values, wherein the first set of received power values comprises a received power value for each signal of the multiple signals received at the UE 301, generate estimated values for a first set of spatial channel parameters for a channel between the UE 301 and the node 303 using the first set of received power values, calculate a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams, and generate a performance metric on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values.
[0174] Accordingly, the machine 700 can implement a method for supporting beam management for UE transmission and / or reception.
[0175] Such machine-readable instructions may also be loaded onto a computer or other programmable data processing devices, so that the computer or other programmable data processing devices perform a series of operations to produce computer-implemented processing, thus the instructions executed on the computer, or other programmable devices provide an operation for realizing functions specified by flow (s) in the flow charts and / or block (s) in the block diagrams.
[0176] Further, the teachings herein may be implemented in the form of a computer or software product, such as a non-transitory machine-readable storage medium, the computer software or product being stored in a storage medium and comprising a plurality of instructions, e.g., machine readable instructions, for making a computer device implement the methods recited in the examples of the present disclosure.
[0177] In some examples, some methods can be performed in a cloud-computing or network-based environment. Cloud-computing environments may provide various services and applications via the Internet. These cloud-based services (e.g., software as a service, platform as a service, infrastructure as a service, etc. ) may be accessible through a web browser or other remote interface of the user equipment for example. Various functions described herein may be provided through a remote desktop environment or any other cloud-based computing environment.
[0178] While various embodiments have been described and / or illustrated herein in the context of fully functional computing systems, one or more of these exemplary embodiments may be distributed as a program product in a variety of forms, regardless of the particular type of computer-readable-storage media used to actually carry out the distribution. The embodiments disclosed herein may also be implemented using software modules that perform certain tasks. These software modules may include script, batch, or other executable files that may be stored on a computer-readable storage medium or in a computing system. In some embodiments, these software modules may configure a computing system to perform one or more of the exemplary embodiments disclosed herein. In addition, one or more of the modules described herein may transform data, physical devices, and / or representations of physical devices from one form to another.
[0179] According to examples, a closed-loop performance monitoring metric for RSRP based spatial channel parameters estimation using RSRP replay is provided. In an implementation, a NN structure is provided, which estimates spatial channel parameters based on RSRP measurements, wherein a training loss, which is used to fit NN coefficients, is computed based on RSRP reply. In an example, a UE capability indication method is provided which indicates a Tproc time interval, in between having received SRS-RSRPs from a BS, until reporting an AI performance monitoring metric to a BS. In an example, an adaptive UL BM procedure, which consists of a feature collection step and a beam verification step or process is provided, wherein the beam verification step can be earlier terminated based on the performance monitoring metric. As such, the 1st step and the 2nd step correspond to two different sets of SRSs, and association information between the two SRS sets is configured from the BS to the UE. In an example, a configuration method for indicating the multi-solutions for RSRP based spatial channel estimation is provided, wherein multiple solutions share a same real-valued AoD vector but with different complexed-valued path gains vectors.
[0180] The preceding description has been provided to enable others skilled in the art to best utilize various aspects of the exemplary embodiments disclosed herein. This exemplary description is not intended to be exhaustive or to be limited to any precise form disclosed. Many modifications and variations are possible without departing from the spirit and scope of the instant disclosure. The embodiments disclosed herein should be considered in all respects illustrative and not restrictive. Reference should be made to the appended claims and their equivalents in determining the scope of the instant disclosure.
Claims
1.A user equipment, UE, configured to support beam management for UE transmission, the UE comprising:a memory; anda processor, operationally coupled to the memory, configured to:transmit, to a node, multiple reference signals, each signal of the multiple reference signals transmitted using a different beam pattern selected from a first set of UE beams;generate estimated values for a first set of spatial channel parameters for a channel between the UE and the node using a first set of received power values, wherein the first set of received power values comprises a power value for each one of the multiple signals received at the node;calculate a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams;generate a performance metric on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values; andtransmit the performance metric to the node.2.A user equipment, UE, configured to support beam management for UE reception, the UE comprising:a memory; anda processor, operationally coupled to the memory, configured to:receive, from a node, multiple reference signals, each signal of the multiple reference signals received by the UE using a different beam pattern selected from a first set of UE beams;measure a first set of received power values, wherein the first set of received power values comprises a received power value for each signal of the multiple signals received at the UE;generate estimated values for a first set of spatial channel parameters for a channel between the UE and the node using the first set of received power values;calculate a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams; andgenerate a performance metric on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values.3.The UE of claim 1 or 2, wherein the processor is further configured to:generate, using the measures of similarity, a loss value for an artificial neural network, NN, of the UE used to calculate a set of spatial channel parameters, wherein the loss value provides an indication of the degree to which a set of spatial channel parameters calculated using the NN match the first set of spatial channel parameters, wherein the NN is configured to use the loss value to update a set of coefficients of the NN.4.The UE of claim 3, wherein the processor is further configured to:use the loss value to update a set of coefficients of the NN for one of: a predetermined number of update iterations of the NN, each update iteration using a loss value to modify one or more coefficients of the NN, or until a calculated loss value is below a pre-defined threshold value.5.The UE of any preceding claim when dependent on claim 1, wherein the processor is further configured to:calculate a pair of candidate spatial channel parameter sets forming a pair of solutions for uplink channel parameters from the UE to the node; andreceive, from the node, on the basis that the performance metric is greater than a predetermined threshold value, a signal configured to trigger a beam selection process at the UE for selection of one of the pair of candidate spatial channel parameter sets.6.The UE of claim 5, wherein the processor is further configured to:transmit, to the node, a pair of reference signals, each reference signal of the pair of reference signals transmitted using a different beam pattern, wherein each one of the different beam patterns is configured using a respective one of the pair of candidate spatial channel parameter sets; andreceive, from the node, a message indicating one of the pair of solutions for uplink channel parameters from the UE to the node resulting in a highest received power measurement at the node.7.The UE of any preceding claim when dependent on claim 1, wherein the processor is further configured to:receive, from the node, the pair of candidate spatial channel parameter sets in a reduced payload message in which a pair of solutions for uplink channel parameters from the UE to the node share the same value for an angle of departure, the payload message comprising a single real-valued angle of departure vector associated with multiple complex-valued gain vectors for the pair of solutions.8.Apparatus in mobile telecommunications network, the apparatus comprising a memory, and a processor, operationally coupled to the memory, configured to:transmit, to a UE, a first trigger signal configured to trigger transmission, to the apparatus, of multiple signals, each signal of the multiple signals transmitted by the UE using a different beam pattern from a first set of UE beams;receive, at the apparatus, each signal of the multiple signals and, each time a signal of the multiple signals is received, calculate a received power value of the received signal;generate estimated values for a first set of spatial channel parameters for a channel between the UE and the apparatus using the received power values;calculate a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams;generate a performance metric on the basis of measures of similarity between respective ones of received power values and calculated power values of the second set of received power values;compare the performance metric to a first predetermined threshold performance metric value, and, in the event that the performance metric is greater than the first predetermined threshold performance metric value, generate a final performance metric over all the multiple signals for the first set of UE beams;transmit, to the UE, in the event that the final performance metric is greater than second predetermined threshold value, a second trigger signal configured to trigger a beam selection process at the UE for selection of a set of spatial channel parameters from a pair of candidate spatial channel parameter sets forming a pair of solutions for uplink channel parameters from the UE to the apparatus;receive, at the apparatus, a pair of reference signals, each signal of the pair of reference signals transmitted to the node by the UE using a different beam pattern configured using respective ones of the pair of candidate spatial channel parameters; andtransmit, to the UE, a message indicating one of the pair of solutions for uplink channel parameters from the UE to the apparatus resulting in a highest received power measurement at the apparatus.9.The apparatus of claim 8, wherein the processor is further configured to:transmit, to the UE, the pair of candidate spatial channel parameter sets in a reduced payload message in which the pair of solutions for uplink channel parameters from the UE to the apparatus share the same value for an angle of departure, the payload message comprising a single real-valued angle of departure vector associated with multiple complex-valued gain vectors for the pair of solutions.10.The apparatus of claim 8 or 9, wherein the processor is further configured to:in the event that a received power value is less than or equal to the first predetermined threshold power value, transmit an abort message to the UE to trigger termination of transmission of any remaining ones of the multiple signals for the first set of UE beams.11.The apparatus of claim 8 or 9, wherein the processor is further configured to:in the event that the final performance metric is less than or equal to the second predetermined threshold value, transmit, to the UE, a message identifying a reference signal transmitted by the UE with the highest received power within a first set of received power values, the identified reference signal representing a fall-back beam pattern for the UE selected from the first set of UE beams.12.The apparatus of any of claims 8 to 11, wherein the processor is further configured to:generate, using the measures of similarity, a loss value for an artificial neural network, NN, of the apparatus used to calculate a set of spatial channel parameters, wherein the loss value provides an indication of the degree to which a set of spatial channel parameters calculated using the NN match the first set of spatial channel parameters, wherein the NN is configured to use the loss value to update a set of coefficients of the NN.13.The apparatus of claim 12, wherein the processor is further configured to:use the loss value to update a set of coefficients of the NN for one of: a predetermined number of update iterations of the NN, each update iteration using a loss value to modify one or more coefficients of the NN, or until a calculated loss value is below a pre-defined threshold value.14.A method for beam management in a user equipment, UE, configured to support beam management for UE transmission, the method comprising:transmitting, to a node, multiple reference signals, each signal of the multiple reference signals transmitted using respective different beam patterns of a first set of UE beams;generating estimated values for a first set of spatial channel parameters for a channel between the UE and the node using a first set of received power values, wherein the first set of received power values comprises a power value for each one of the multiple signals received at the node;calculating a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams;generating a performance metric on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values; andtransmitting the performance metric to the node.15.A method for beam management in a user equipment, UE, configured to support beam management for UE reception, the method comprising:receiving, from a node, multiple reference signals, each signal of the multiple reference signals received by the UE using respective different beam patterns of a first set of UE beams;measuring a first set of received power values, wherein the first set of received power values comprises a received power value for each one of the multiple signals received at the UE;generating estimated values for a first set of spatial channel parameters for a channel between the UE and the node using the first set of received power values;calculating a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams; andgenerating a performance metric on the basis of measures of similarity between respective ones of received power values of the first set of received power values and calculated power values of the second set of received power values.16.The method of claim 14 or 15, further comprising:generating, using the measures of similarity, a loss value for an artificial neural network, NN, of the UE used to calculate a set of spatial channel parameters, wherein the loss value provides an indication of the degree to which a set of spatial channel parameters calculated using the NN match the first set of spatial channel parameters, wherein the NN is configured to use the loss value to update a set of coefficients of the NN.17.The method of claim 16, further comprising:using the loss value to update a set of coefficients of the NN for one of: a predetermined number of update iterations of the NN, each update iteration using a loss value to modify one or more coefficients of the NN, or until a calculated loss value is below a pre-defined threshold value.18.The method of any of claims 14 to 17 when dependent on claim 14, further comprising:calculating a pair of candidate spatial channel parameter sets forming a pair of solutions for uplink channel parameters from the UE to the node; andreceiving, from the node, on the basis that the performance metric is greater than a predetermined threshold value, a signal configured to trigger a beam selection process at the UE for selection of one of the pair of candidate spatial channel parameter sets.19.The method of claim 18, further comprising:transmitting, to the node, a pair of reference signals, each reference signal of the pair of reference signals transmitted using a different beam pattern, wherein each one of the different beam patterns is configured using a respective one of the pair of candidate spatial channel parameter sets; andreceiving, from the node, a message indicating one of the pair of solutions for uplink channel parameters from the UE to the node resulting in a highest received power measurement at the node.20.The method of any of claims 14 to 19 when dependent on claim 14, further comprising:receive, from the node, the pair of candidate spatial channel parameter sets in a reduced payload message in which a pair of solutions for uplink channel parameters from the UE to the node share the same value for an angle of departure, the payload message comprising a single real-valued angle of departure vector associated with multiple complex-valued gain vectors for the pair of solutions.21.A method for beam management in a node of a mobile telecommunications network, the method comprising:transmitting, to a UE, a first trigger signal configured to trigger transmission, to the node, of multiple signals, each signal of the multiple signals transmitted by the UE using respective different beam patterns of a first set of UE beams;receiving, at the node, each signal of the multiple signals and each time a signal of the multiple signals is received, calculating a received power value of the received signal;generating estimated values for a first set of spatial channel parameters for a channel between the UE and the node using the received power values;calculating a second set of received power values using the estimated values for the first set of spatial channel parameters and the first set of UE beams;generating a performance metric on the basis of measures of similarity between respective ones of received power values and calculated power values of the second set of received power values;comparing the performance metric to a first predetermined threshold performance metric value, and, in the event that the performance metric is greater than the first predetermined threshold performance metric value, generating a final performance metric over all the multiple signals for the first set of UE beams;transmitting, to the UE, in the event that the final performance metric is greater than second predetermined threshold value, a second trigger signal configured to trigger a beam verification process at the UE for selection of a set of spatial channel parameters from a pair of candidate spatial channel parameters forming a pair of solutions for uplink channel parameters from the UE to the node;receiving, at the node, a pair of reference signals, each signal of the pair of reference signals transmitted to the node by the UE using different beam patterns configured using respective ones of the pair of candidate spatial channel parameters; andtransmitting, to the UE, a message indicating one of the pair of solutions for uplink channel parameters from the UE to the node resulting in a highest received power measurement at the apparatus.22.The method of claim 21, further comprising:transmitting, to the UE, the pair of candidate spatial channel parameter sets in a reduced payload message in which the pair of solutions for uplink channel parameters from the UE to the apparatus share the same value for an angle of departure, the payload message comprising a single real-valued angle of departure vector associated with multiple complex-valued gain vectors for the pair of solutions.23.The method of claim 21 or 22, further comprising:in the event that a received power value is less than or equal to the first predetermined threshold power value, transmitting an abort message to the UE to trigger termination of transmission of any remaining ones of the multiple signals for the first set of UE beams.24.The method of claim 21 or 22, further comprising:in the event that the final performance metric is less than or equal to the second predetermined threshold value, transmitting, to the UE, a message identifying a reference signal transmitted by the UE with the highest received power within a first set of received power values, the identified reference signal representing a fall-back beam pattern for the UE selected from the first set of UE beams.25.The method of any of claims 21 to 24, further comprising:generating, using the measures of similarity, a loss value for an artificial neural network, NN, of the apparatus used to calculate a set of spatial channel parameters, wherein the loss value provides an indication of the degree to which a set of spatial channel parameters calculated using the NN match the first set of spatial channel parameters, wherein the NN is configured to use the loss value to update a set of coefficients of the NN.26.The method of claim 25, further comprising:using the loss value, updating a set of coefficients of the NN for one of: a predetermined number of update iterations of the NN, each update iteration using a loss value to modify one or more coefficients of the NN, or until a calculated loss value is below a pre-defined threshold value.
Citation Information
Patent Citations
Beam mangement systems and methods
CN110710133A
Reporting channel statistics for beam management
CN115362637A
Set of rules for triggering coordinated beamforming
US20220376852A1