Methods and Devices for Determination of Beamforming Information

TECHNICAL FIELD

Embodiments presented herein relate to a method, an RE, a computer program, and a computer program product for determining beamforming weights or decoding user data for terminal devices. Embodiments presented herein further relate to a method, an REC, a computer program, and a computer program product for determining beamforming weights or decoding user data for terminal devices.

BACKGROUND

In communications systems, there may be a challenge to obtain good performance and capacity for a given communications protocol, its parameters and the physical environment in which the communications system is deployed.

For example, the introduction of digital beamforming antenna systems in access nodes, such as radio base stations, etc., could allow multiple simultaneous narrow beams to be used to provide network access to, and thus server, multiple simultaneous served terminal devices, such as user equipment (UE), etc. However, the current split in the access nodes between a radio equipment controller (REC) and a radio equipment (RE) as interconnected by the Common Public Radio Interface (CPRI) may no longer be feasible as passing the data for each individual radio chain over the CPRI interface could drive prohibitively high data rates.

In more detail, the bit rate of the current CPRI interface scales directly to the number of independent radio chains in the RE. When having e.g., a 200 MHz carrier bandwidth and 128 physical antenna elements in the beamforming antenna system, a bit rate of 530 Gbps would be needed for the CPRI interface with currently used sample coding. A further potential drawback with CPRI is the extra latency from uplink (UL; from terminal device to access node) sampling to the time the data can be used in downlink (DL; from access node to terminal device), as any information needs to loop in the REC.

One way to address the above-mentioned issues is to collapse the CPRI based architecture by removing the CPRI interface and putting the functionality of the REC in the RE. This approach has at least two drawbacks. Firstly, due to faster technological development of the REC compared to the RE, the technical lifetime of the REC is assumed to be shorter than that of the RE. Replacing the RE is more costly than replacing the REC. From this aspect it could thus be beneficial to keep the functionality of the RE as simple as possible. Secondly, the REC could be configured to make decisions spanning over multiple REs in order to make coordinated multi-sector decisions, e.g. when some REs represent coverage regions of the access node within the coverage regions of other REs (e.g. a so-called micro cell within a so-called macro cell). A collapsed architecture loses this overarching coordination possibility.

Hence, there is a need for an improved communication between the REC and the RE.

SUMMARY

An object of embodiments herein is to provide efficient communication between the REC and the RE.

According to a first aspect there is presented a method for obtaining information for determining beamforming weights for terminal devices. The method is performed by an RE of an access node. The RE has an interface to an REC of the access node. The method comprises obtaining beamforming information per direction from the REC over the interface. The method comprises transforming the beamforming information per direction to beamforming information per antenna, the beamforming information per antenna representing the beamforming weights. The method comprises applying the beamforming weights.

According to a second aspect there is presented an RE of an access node for obtaining information for determining beamforming weights for terminal devices. The RE has an interface to an REC of the access node and comprises processing circuitry. The processing circuitry is configured to cause the RE to obtain beamforming information per direction from the REC over the interface. The processing circuitry is configured to cause the RE to transform the beamforming information per direction to beamforming information per antenna, the beamforming information per antenna representing the beamforming weights. The processing circuitry is configured to cause the RE to apply the beamforming weights.

According to a third aspect there is presented an RE of an access node for obtaining information for determining beamforming weights for terminal devices. The RE has an interface to an REC of the access node and comprises processing circuitry and a computer program product. The computer program stores instructions that, when executed by the processing circuitry, causes the RE to perform operations, or steps. The operations, or steps, cause the RE to obtain beamforming information per direction from the REC over the interface. The operations, or steps, cause the RE to transform the beamforming information per direction to beamforming information per antenna, the beamforming information per antenna representing the beamforming weights. The operations, or steps, cause the RE to apply the beamforming weights.

According to a fourth aspect there is presented an RE of an access node for obtaining information for determining beamforming weights for terminal devices. The RE has an interface to an REC of the access node. The RE comprises an obtain module configured to obtain beamforming information per direction from the REC over the interface. The RE comprises a transform module configured to transform the beamforming information per direction to beamforming information per antenna, the beamforming information per antenna representing the beamforming weights. The RE comprises an apply module configured to apply the beamforming weights.

According to a fifth aspect there is presented a computer program for obtaining information for determining beamforming weights for terminal devices. The computer program comprises computer program code which, when run on processing circuitry of an RE, causes the RE to perform a method according to the first aspect.

According to a sixth aspect there is presented a method for providing information for determining beamforming weights for terminal devices. The method is performed by an REC of an access node. The REC has an interface to an RE of the access node. The method comprises obtaining beamforming information per antenna, the beamforming information per antenna representing the beamforming weights. The method comprises transforming the beamforming information per antenna to beamforming information per direction. The method comprises providing the beamforming information per direction to the RE over the interface.

According to a seventh aspect there is presented an REC of an access node for providing information for determining beamforming weights for terminal devices. The REC has an interface to an RE of the access node and comprises processing circuitry. The processing circuitry is configured to cause the REC to obtain beamforming information per antenna, the beamforming information per antenna representing the beamforming weights. The processing circuitry is configured to cause the REC to transform the beamforming information per antenna to beamforming information per direction. The processing circuitry is configured to cause the REC to provide the beamforming information per direction to the RE over the interface.

According to an eighth aspect there is presented an REC of an access node for providing information for determining beamforming weights for terminal devices. The REC has an interface to an RE of the access node and comprises processing circuitry and a computer program product. The computer program stores instructions that, when executed by the processing circuitry, causes the REC to perform operations, or steps. The operations, or steps, cause the REC to obtain beamforming information per antenna, the beamforming information per antenna representing the beamforming weights. The operations, or steps, cause the REC to transform the beamforming information per antenna to beamforming information per direction. The operations, or steps, cause the REC to provide the beamforming information per direction to the RE over the interface.

According to a ninth aspect there is presented an REC of an access node for providing information for determining beamforming weights for terminal devices. The REC has an interface to an RE of the access node. The REC comprises an obtain module configured to obtain beamforming information per antenna, the beamforming information per antenna representing the beamforming weights. The REC comprises a transform module configured to transform the beamforming information per antenna to beamforming information per direction. The REC comprises a provide module configured to provide the beamforming information per direction to the RE over the interface.

According to a tenth aspect there is presented a computer program for providing information for determining beamforming weights for terminal devices. The computer program comprises computer program code which, when run on processing circuitry of an REC, causes the REC to perform a method according to the sixth aspect.

According to an eleventh aspect there is presented a method for determining beamforming weights for terminal devices. The method is performed by an RE of an access node. The RE has an interface to an REC of the access node. The method comprises obtaining beamforming information per antenna from at least one of the terminal devices, the beamforming information per antenna representing the beamforming weights. The method comprises transforming the beamforming information per antenna to beamforming information per direction. The method comprises providing the beamforming information per direction to the REC over the interface.

According to a twelfth aspect there is presented an RE of an access node for determining beamforming weights for terminal devices. The RE has an interface to an REC of the access node and comprises processing circuitry. The processing circuitry is configured to cause the RE to obtain beamforming information per antenna from at least one of the terminal devices, the beamforming information per antenna representing the beamforming weights. The processing circuitry is configured to cause the RE to transform the beamforming information per antenna to beamforming information per direction. The processing circuitry is configured to cause the RE to provide the beamforming information per direction to the REC over the interface.

According to a thirteenth aspect there is presented an RE of an access node for determining beamforming weights for terminal devices. The RE has an interface to an REC of the access node and comprises processing circuitry and a computer program product. The computer program stores instructions that, when executed by the processing circuitry, causes the RE to perform operations, or steps. The operations, or steps, cause the RE to obtain beamforming information per antenna from at least one of the terminal devices, the beamforming information per antenna representing the beamforming weights. The operations, or steps, cause the RE to transform the beamforming information per antenna to beamforming information per direction. The operations, or steps, cause the RE to provide the beamforming information per direction to the REC over the interface.

According to a fourteenth aspect there is presented an RE of an access node for determining beamforming weights for terminal devices. The RE has an interface to an REC of the access node. The RE comprises an obtain module configured to beamforming information per antenna from at least one of the terminal devices, the beamforming information per antenna representing the beamforming weights. The RE comprises a transform module configured to transform the beamforming information per antenna to beamforming information per direction. The RE comprises a provide module configured to provide the beamforming information per direction to the REC over the interface.

According to a fifteenth aspect there is presented a computer program for determining beamforming weights for terminal devices. The computer program comprises computer program code which, when run on processing circuitry of an RE, causes the RE to perform a method according to the eleventh aspect.

According to a sixteenth aspect there is presented a method for obtaining information for determining beamforming weights for terminal devices. The method is performed by an REC of an access node. The REC has an interface to an RE of the access node. The method comprises obtaining beamforming information per direction from the RE over the interface. The method comprises determining the beamforming weights per direction based on the beamforming information per direction. The method comprises providing the beamforming weights per direction to the RE over the interface.

According to a seventeenth aspect there is presented an REC of an access node for providing information for obtaining information for determining beamforming weights for terminal devices. The REC has an interface to an RE of the access node and comprises processing circuitry. The processing circuitry is configured to cause the REC to obtain beamforming information per direction from the RE over the interface. The processing circuitry is configured to cause the REC to determine the beamforming weights per direction based on the beamforming information per direction. The processing circuitry is configured to cause the REC to provide the beamforming weights per direction to the RE over the interface.

According to a eighteenth aspect there is presented an REC of an access node for obtaining information for determining beamforming weights for terminal devices. The REC has an interface to an RE of the access node and comprises processing circuitry and a computer program product. The computer program stores instructions that, when executed by the processing circuitry, causes the REC to perform operations, or steps. The operations, or steps, cause the REC to obtain beamforming information per direction from the RE over the interface. The operations, or steps, cause the REC to determine the beamforming weights per direction based on the beamforming information per direction. The operations, or steps, cause the REC to provide the beamforming weights per direction to the RE over the interface.

According to an nineteenth aspect there is presented an REC of an access node for obtaining information for determining beamforming weights for terminal devices. The REC has an interface to an RE of the access node. The REC comprises an obtain module configured to obtain beamforming information per direction from the RE over the interface. The REC comprises a determine module configured to determine the beamforming weights per direction based on the beamforming information per direction. The REC comprises a provide module configured to provide the beamforming weights per direction to the RE over the interface.

According to a twentieth aspect there is presented a computer program for obtaining information for determining beamforming weights for terminal devices. The computer program comprises computer program code which, when run on processing circuitry of an REC, causes the REC to perform a method according to the sixteenth aspect.

According to a twenty first aspect there is presented a computer program product comprises a computer program according to at least one of the fifth aspect, the tenth aspect, the fifteenth aspect, and the twentieth aspect and a computer readable storage medium on which the computer program is stored. The computer readable storage medium could be a non-transitory computer readable storage medium.

Advantageously these methods, these REs, these RECs, and these computer programs allows for efficient communications between the RE and the REC when configuring resources for terminal devices.

Advantageously these methods, these REs, these RECs, and these computer programs allows for large-scale digital beamforming in the access node without significantly upgrading the data rate of the interface between the RE and the REC.

Other objectives, features and advantages of the enclosed embodiments will be apparent from the following detailed disclosure, from the attached dependent claims as well as from the drawings.

Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to “a/an/the element, apparatus, component, means, step, etc.” are to be interpreted openly as referring to at least one instance of the element, apparatus, component, means, step, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated.

BRIEF DESCRIPTION OF THE DRAWINGS

The inventive concept is now described, by way of example, with reference to the accompanying drawings, in which:

FIGS. 1, 2, 3, 4, 5, and 6 are schematic diagrams illustrating an access node according to embodiments;

FIG. 7 is a schematic illustration of a time-frequency resource grid according to embodiments;

FIG. 8 is a schematic illustration comparing energy in antenna space and direction space;

FIG. 9 is a schematic illustration of a compression subsystem according to an embodiment;

FIGS. 10, 11, 12, and 13 are flowcharts of methods according to embodiments;

FIG. 14 is a schematic diagram showing functional units of an RE according to an embodiment;

FIG. 15 is a schematic diagram showing functional modules of an RE according to an embodiment;

FIG. 16 is a schematic diagram showing functional units of an REC according to an embodiment;

FIG. 17 is a schematic diagram showing functional modules of an REC according to an embodiment; and

FIG. 18 shows one example of a computer program product comprising computer readable means according to an embodiment.

DETAILED DESCRIPTION

The inventive concept will now be described more fully hereinafter with reference to the accompanying drawings, in which certain embodiments of the inventive concept are shown. This inventive concept may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided by way of example so that this disclosure will be thorough and complete, and will fully convey the scope of the inventive concept to those skilled in the art. Like numbers refer to like elements throughout the description. Any step or feature illustrated by dashed lines should be regarded as optional.

FIG. 1 is a schematic diagram illustrating an access node 100 where embodiments presented herein can be applied. The access node could be a radio base station such as a radio access network node, base transceiver station, node B, evolved node B, or access point. As disclosed above, the access node comprises at least one Radio Equipment Controller (REC) 300 and at least one Radio Equipment (RE) 200. In the illustrative example of FIG. 1 the access node comprises one REC and two REs, where the REC has one interface 700 to each of the REs. The REs are configured to perform DL transmissions to, and UL receptions from, terminal devices boo in beams 500 by using appropriate beamforming weights at the antennas 400. The beamforming weights define at least the pointing direction and the width of the beams. How to determine the beamforming weights will be disclosed below.

As defined herein the REC does not send in-phase/quadrature (I/Q) samples per physical radio branch to the RE, but rather multiple-input multiple-output (MIMO) streams, i.e., I/Q samples per layer. According to the current CPRI specification, the REC can directly address the antennas in the RE, but in the herein disclosed access node that is configured to perform beamforming, the RE performs the functionality of mapping a MIMO stream to a set of physical antenna elements in order to generate a wanted beam form. Further, in order to enable efficient simultaneous multi user beamforming, the Fast Fourier Transform (FFT) and Inverse Fast Fourier Transform (IFFT) functions are performed in the RE. In addition, the execution of the beamforming data plane functionality is added to the RE. Further, the interface 700 between REC and RE could be a packet-based interface, and hence no longer a streaming interface, sending the (frequency domain) samples to the RE symbol by symbol. This allows for quick and flexible allocation of resources on the interface to different terminal devices. The REC is configured to maintain knowledge about the terminal devices, and schedules the air interface between the access node and the terminal devices. The RE is configured to act on commands received from the REC.

As an illustrative example, consider a communications system having an air interface with a system bandwidth of 400 MHz and that provides support for 4 MIMO streams and utilizes access nodes with 64 antennas for beamforming. Using CPRI interfaces between the REC and the RE exposing all 64 antennas for the REC would require approximately 54 CPRI interfaces of 10 Gbps, since a CPRI interface carries about 480 MHz. Further, an interface using virtual antenna ports would require 4 MIMO streams of 400 MHz, and would require about 4 CPRI interfaces of 10 Gbps, since one 10 Gbps CPRI interface still carries data for about 480 MHz. By also moving the modulation DL to the RE, the 4 MIMO streams of 400 MHz would require 7 Gbps (assuming 256QAM and 20 LTE 20 MHz carriers), or one 10 Gbps CPRI interface. A higher bitrate of the CPRI interface is required in the UL if the whole system bandwidth is used, as demodulation is still performed in the REC.

Consider the REC and RE illustrated in FIG. 2 where the REC and RE are interconnected via an interface denoted RUI, for Radio Unit Interface. In the DL direction the REC sends unmodulated bits for each terminal device. The RE modulates the data, places it on the correct subcarrier (thus performing resource mapping), applies beamforming weights (individual per radio branch) defining e.g. width and/or direction of the beams and finally convert it to time domain and transmits it to the terminal devices. FIG. 2 shows 1-8 DL MIMO layers on 128 radio branches. In the UL direction the RE samples the signals for each individual radio branch, converts it to frequency domain, applies beamforming weights, combines the signals from the different radio branches and sends a selection of the modulated combined signal to the REC for further demodulation. FIG. 2 shows 128 radio branches and 1-16 receive beams. The receive beams weakly relates to UL MIMO branches as the more MIMO branches the more receive beams are needed. Typically, more receive beams than MIMO branches are required. FIG. 2 also shows beamforming as performed before de-mapping in the RE. The two stages are interlinked as the beamforming is performed individually per terminal device. FIG. 2 further illustrates User Plane Control (UPC) with its air interface scheduling and link adaptation placed in the REC.

FIG. 3 shows an embodiment of an REC and an RE similar to those in FIG. 2. In FIG. 3 the RE is configured to decode reference symbols such as Sounding Reference symbols (SRS) in the UL, to store information of how the reference symbols are best received (e.g., what beamforming weights maximizes the SNR of the reference symbols) and to use this stored information when performing beamforming in the DL and UL. In FIG. 3 the RE could thus be regarded as autonomous in this respect. To accomplish this, the RU needs to maintain a storage of information identifying the best beam shapes for each terminal device, and be configured to apply that information when the corresponding terminal device is scheduled. This minimizes the communication needed over the interface between the REC and the RE with respect to beamforming, but also hides the channel for the UPC.

The scheduling and the LA are based on the dimension reduced information while further dimensions are used for the determination of beamforming weights. This is equally applicable for UL and DL. The radio channel should typically be dispersive to be able to gain from adding more dimensions to the beam calculation.

FIG. 4 shows an embodiment of an REC and an RE similar to those in FIG. 2. In FIG. 4 the beamforming control is performed primarily by the UPC in the REC. The RE still receives the reference symbols and determines information of how the reference symbols are best received. This information is then sent to the REC. The REC uses this information to make an optimal decision on MIMO streams, link adaptation and coordinated scheduling with other terminal devices, such as MU-MIMO or nulling.

To make the communication and functionality performed in the UPC generic and not heavy dependent on the actual implementation of the RE (e.g., in terms of number of branches, antenna layout, etc.) or operating mode (e.g., power save, faulty branches, etc.), the communications between the REC and the RE regarding beamforming properties is expressed in beam direction space rather than antenna element space. That is, instead of presenting a beam as a set of weights of physical antenna elements the beam is presented as a combination of a set of predetermined beams. This also allows for a more compressed format for this communication, thus saving bit rate on the interface between the REC and the RE. For instance, a linear combination of 3 predetermined beams could be expressed as 3 times 24 bits (an 8-bit beam number+an 8-bit amplitude+an 8-bit phase) rather than 128 times 16 bits (an 8-bit amplitude+an 8-bit phase for each of the 128 physical antenna elements). The transformation from the physical antenna element space to the beam direction space is performed by a Dimensions Reduction entity, and the inverse transformation is performed by a Beam Weight Transformation entity.

The operations as performed by the REC and RE in FIG. 4 could be preferred when the load of the access node is comparatively high and UPC needs to determine the channel state for the best decision on link adaptation.

The operations as performed by the REC and RE in FIG. 4 could be preferred when MU-MIMO or nulling is used, as the UPC then needs channel state information for the LA of the combination, and to determine which users are suitable for simultaneous scheduling, so called MU-MIMO scheduling. Note that nulling may also be applied between terminal devices served by different REs, and is therefore impossible to perform within one autonomous RE (as in FIG. 3).

FIG. 5 illustrates an embodiment of an REC and an RE combining the functionality of the REC and RE in the embodiments of FIG. 3 and FIG. 4. In comparison to FIG. 3 and FIG. 4, the embodiment of FIG. 5 comprises a channel state memory, an UL/DL BF coefficient calculation entity (where BF is short for beamforming), an UL/DL quality calculation entity, an UL/DL SNR calculation entity (where SNR is short for signal to noise ratio), and a CSI feedback function (where CSI is short for Channel State Information).

The channel state memory is configured to store reference symbols since the reference symbols are not sent continuously. The transmission rate of the reference symbols is controlled by the access node, and different terminal devices could have different transmission rates of the reference symbols to allow for the access node to follow channel state changes. Although the channel state memory is illustrated as storing data expressed in the beam space in both the RE and the REC, the RE could instead have the channel state memory storing data expressed in the physical antenna element space.

The UL/DL BF coefficient calculation entity is configured to determine beamforming weights based on the channel state memory and possible other constraints (such as nulling and MU-MIMO) in the RE.

The UL RX weights calculation entity (where RX is short for reception) is configured to determine beamforming weights for the uplink. The UL beamforming weight determination can be part of maximum-ratio combining (MRC) in the UL demodulator/equalizer. The UPC will order more beams than layers, and then the demodulation will combine these to improve SNR or suppress interferers. In the REC, the determination of beamforming weights could be performed in conjunction with the link adaptation whereas in the case of beamforming weights determined by the RE, the link adaptation is done independently.

The UL/DL quality calculation entity is configured to determine a quality estimate in respect of each terminal device subject to MU-MIMO scheduling. This quality estimate should reflect the spatial separation between wireless terminals as well as the quality achieved when co-scheduling wireless terminals on the same time/frequency resource. The quality estimate is based on the information in the channel state memory.

The UL/DL SNR calculation entity is configured to determine the beamforming weights for each terminal device, and to provide the link adaptation function with estimates of the resulting SNR for each of the terminal devices being scheduled, including the mutual effect of co-scheduled terminal devices, so called MU-MIMO scheduling. The UE feedback (CSI) entity is configured to extract information about the channel provided by the terminal devices (in the UL data plane). Especially, in the case of FDD, where the reciprocity of the DL and UL of the channel to the terminal device is not perfect, it can be beneficial for the terminal device to send measurements (e.g., CSI) on the DL signal back to the access node. The CSI reports are extracted by the REC and used in the channel state memory and thus being considered in the determination of the beamforming weights.

In configurations where the bitrate of the interface between the REC and the RE is low (such as below a threshold), the determination of beamforming weights is executed by the RE (as in FIG. 3), otherwise the determination is executed by the REC (as in FIG. 4). That is, in case of limited available capacity of the interface between the REC and the RE, the REC (such as in the UPC) determines that the determination of beamforming weights is to be executed by the RE. Cell performance of the access node could be maximized by determining the beamforming weights in the REC for all co-scheduled terminal devices (within the same time and frequency domain) in order to accomplish best link adaptation and orthogonality. This requires that the RE sends extra information to the REC to allow for such determination, and that the REC sends the determined beamforming weights to the RE.

Still further, although the REC sends the determined beamforming weights to the RE, the RE can determine beamforming weights in parallel and thus combine these internally determined beamforming weights with the beamforming weights received from the REC. For access nodes where the interface between the REC and the RE is constrained but not minimized, the dual loops (as defined by the embodiment in FIG. 5) allow for the access node to select some terminal devices which can be handled locally in the RE, thereby making bitrate of the interface between the REC and the RE available to handle terminal devices which are eligible to MU-MIMO scheduling, e.g. terminal devices using a streaming service.

Having both loops also allows for efficient beamforming coefficient compression. If a terminal device is note exposed to constraints from the UPC, the beamforming coefficient calculation in the REC will be the same as performed in the RE, and the REC can send a minimal message requesting the RE to use its locally calculated values. In case of heavy dimensioning reduction, this may even improve the beamforming gain as the RE has uncompressed channel estimation data. Correspondingly, if constraints have changed any beamforming weights, it can be evaluated if the constraints have less bits than the new beamforming weights, and can be transferred instead.

Further aspects applicable to both the embodiments of the RE and the REC will now be disclosed with reference to the embodiment of REC and RE illustrated in FIG. 6. The description of those entities already having been described with reference to any of FIGS. 2-5 is omitted for brevity.

A reference symbol extraction entity is configured to extract the reference symbols from the Resource Elements that are provided by the UL OFDM FFT from all antenna ports.

The spatial Discrete Fourier Transform (DFT) entity and the channel estimation entity are configured to collectively provide a quality value for the fixed beam directions. The quality value is typically based on a filtering of the reference symbols per involved transmission antennas at the terminal device or MIMO layer within the resource block or for a filtering of a further processed channel estimate per involved transmission antennas at the terminal device or MIMO layer and extracted reference symbols. The beam direction space is provided by processing the reference symbols from all antennas through the spatial DFT entity.

A Dimension Reduction entity is configured to reduce dimension of data being inputted to the Dimension Reduction entity. For beamforming weight determination performed via the REC, the dimension of which fixed beam directions to use for the beamforming is reduced in order to limit the processing load when calculating the weights and the interface rate from the RE to the REC. The dimension reduction is based on the quality values from the Spatial DFT entity and the channel estimation entity. For beamforming weight determination performed internally in the RE, more dimensions can be stored, and thus providing better SNR in case of SU-MIMO transmission.

A Channel state memory is provided in the REC when beamforming weight determination is performed via the REC. For the terminal devices that are scheduled to be measured, the reference symbol based channel estimates can be sent to the REC and stored in the REC Channel State Memory. These stored channel estimates can then be used for link adaption as well as determination of beamforming weights. The content of the Channel state memory can be used when the data channel is active and is updated for every new measurement of reference symbols. In addition to this the Channel state memory can also store a covariance matrix for all beam directions that have been measured. Those values can be calculated in the REC. If no MU-MIMO pairing shall be done the beam direction related information does not need to be stored in the Channel state memory, which will lower the demand on the interface between the REC and the RE.

A Channel state memory is provided in the RE when beamforming weight determination is performed internally in the RE. For the reference symbols of the terminal devices that are scheduled to be measured, the reference symbol based Channel estimates are stored in the RE Channel State Memory. These stored channel estimates can be used for determining beamforming weights. The content of the Channel state memory can be used when the data channel is active and could be updated for every new measurement of reference symbols. If no MU-MIMO pairing is done, if no beam direction related information is stored in the REC based channel state memory, and if the covariance matrix will be used, the same covariance matrix as described in the channel state memory in the REC can instead be stored in the channel state memory in the RE.

A Source Select entity is configured to select and/or combine the beamforming weights that either originate from the REC or locally from the RE. Even in the case where the channel estimate is sent to the REC, a local copy of the channel estimate can be stored in the RE. In case the REC will not send beamforming weights, the RE will have to use beamforming weights as determined internally. This can be due to the REC being satisfied with the beamforming weights determined internally or that the beamforming weights are not received properly by the RE (e.g., due to a lost message). Further, as will be further disclosed below the local stored channel estimates can be used to calculate a default setting of the beamforming weights to which the REC sends differential information, see the compression chapter below. The RE can signal to the REC if it has stored a local copy of the channel estimate. This allows the REC to know if such compression is possible. The RE can run out of local memory, thus such a signalling is recommended (but not mandated). Also, the REC can explicitly order the RE to store a local copy. Also in case no complete channel estimate is sent to the REC, a reduced channel estimate could be transmitted from the RE to the REC to aid the link adaptation and rank selection.

FIG. 7 gives an example of a time-frequency resource grid 1700 applicable to embodiments disclosed herein. The time-frequency resource grid illustrates a possible allocation of resources to terminal devices in terms of a scheduled entity comprising symbols, where each symbol corresponds to a Resource Element. Each Resource Element in time corresponds to one orthogonal frequency-division multiplexing (OFDM) symbol and in frequency corresponds to a subcarrier. One subframe comprises 7 symbols; one subcarrier group comprises 24 subcarriers, one subcarrier chunk comprises 10 subcarrier groups; and one carrier comprises at least one subcarrier chunk.

According to some aspects there are provided mechanisms for efficient dimensions reduction for expressing beamforming weights and information related to determining the beamforming weights.

Reference is now made to FIG. 10 illustrating methods for determining beamforming weights for terminal devices as performed by the RE of the access node according to embodiments.

As will be disclosed below, the REC in step S206 provides beamforming information per direction to the RE 200 over the interface 700. Hence, the RE 200 is configured to perform step S102:

S102: The RE 200 obtains beamforming information per direction from the REC 300 over the interface 700.

The beamforming information is to be applied not per direction but per antenna. The RE 200 is therefore configured to perform step S106:

S106: The RE 200 transforms the beamforming information per direction to beamforming information per antenna. The beamforming information per antenna represents the beamforming weights.

The beamforming weights are then applied. Thus, the RE 200 is configured to perform step S108:

S108: The RE 200 applies the beamforming weights.

Further aspects of determining beamforming weights for terminal devices as performed by the RE will now be disclosed.

In general terms, the beamforming weights are to be applied at antennas 400 of the RE 200.

In some aspects the number of beamforming weights in the direction space is smaller than the number of beamforming weights in the antenna space. In this aspect the transformation thus results in an expansion of dimensionality of the of beamforming information. That is, according to an embodiment the beamforming information per antenna is associated with a first dimensionality, and the beamforming information per direction is associated with a second dimensionality being smaller than the first dimensionality.

In some aspects the beamforming information is transmitted between the RE 200 and the REC 300 as linear combinations of predetermined beams. Particularly, according to an embodiment the beamforming information per direction is obtained as a linear combination of predetermined beamforming weights per direction.

There could be different ways to perform the transformation. In some aspects the transformation is achieved based on using a DFT. Particularly, according to an embodiment the beamforming information is transformed using a DFT.

As will be further disclosed below, indices i₀, i₁, . . . could be used that appoint which of the fixed beam directions (as defined by the predetermined beams) to use. Particularly, according to an embodiment the RE 200 is configured to perform (optional) step S104:

S104: The RE 200 obtains information appointing fixed beam directions from the REC 300 over the interface 100. The beamforming information per direction is then transformed to the beamforming information per antenna only for the appointed beam directions.

Reference is now made to FIG. 11 illustrating methods for determining beamforming weights for terminal devices as performed by the REC of the access node according to embodiments.

In general terms, beamforming information is applied per antenna and beamforming information is therefore often determined per antenna. Therefore the REC 300 is configured to perform step S202:

S202: The REC 300 obtains beamforming information per antenna. The beamforming information per antenna represents the beamforming weights.

Beamforming information per direction could enable more efficient transmission of the beamforming information to the RE 200 than when the beamforming information is per antenna. Therefore, the REC 300 is configured to perform step S204:

S204: The REC 300 transforms the beamforming information per antenna to beamforming information per direction.

The beamforming information is to be applied by the RE 200 and is therefore provided to the RE 200. That is, the REC 300 is configured to perform step S206:

S206: The REC 300 provides the beamforming information per direction to the RE 200 over the interface 100.

Further aspects of determining beamforming weights for terminal devices as performed by the REC will now be disclosed.

The beamforming information per antenna could be obtained from at least one of the terminal devices boo.

As disclosed above, in some aspects the number of beamforming weights in the direction space is smaller than the number of beamforming weights in the antenna space. That is, the embodiment wherein the beamforming information per antenna is associated with a first dimensionality, and wherein the beamforming information per direction is associated with a second dimensionality being smaller than the first dimensionality is also applicable for the REC 300.

In some aspects only high quality beamforming information per direction is provided. Particularly, according to an embodiment only the beamforming information per direction having a signal quality measure above a threshold is provided to the RE 200 over the interface 700 in step S206. The beamforming information per direction that does not fulfil the requirement defined by the threshold is thus not transmitted over the interface 100, thus enabling a reduction in bitrate in comparison to transmitting all the beamforming information per direction or all the beamforming information per antenna.

As disclosed above, in some aspects the transformation is achieved based on using a DFT. That is, the embodiment wherein the beamforming information is transformed using a DFT is also applicable for the REC 300.

As disclosed above, in some aspects indices i₀, i₁, . . . could be used that appoint which of the fixed beam directions to use. That is, according to an embodiment the REC 300 is configured to perform (optional) step S208:

S208: The REC 300 provides information to the RE 200 over the interface 700 appointing which fixed beam directions to use when transforming the beamforming information per direction to beamforming information per antenna.

Reference is now made to FIG. 12 illustrating methods determining beamforming weights for terminal devices as performed by the RE of the access node according to embodiments.

S302: The RE 200 obtains beamforming information per antenna from at least one of the terminal devices boo. The beamforming information per antenna represents the beamforming weights.

Beamforming information per direction could enable more efficient transmission of the beamforming information to the REC 300 than when the beamforming information is per antenna. Therefore, the RE 200 is configured to perform step 3204:

S304: The RE 200 transforms the beamforming information per antenna to beamforming information per direction.

The beamforming information is then provided to the REC 300. That is, the RE 200 is configured to perform step S306:

S306: The RE 200 provides the beamforming information per direction to the REC 300 over the interface 700.

Further aspects of determining beamforming weights for terminal devices as performed by the RE will now be disclosed.

As disclosed above, in some aspects only high quality beamforming information per direction is provided. That is, the embodiment wherein only the beamforming information per direction having a signal quality measure above a threshold is provided to the REC 300 over the interface 700 is also applicable for the RE 200.

As will be further disclosed below, in some aspects the REC 300 provide beamforming weights per direction to the RE 200 over the interface 700.

Hence, according to an embodiment the RE 200 is configured to perform (optional) step S308:

S308: The RE 200 obtains updated beamforming weights per direction from the REC 300 over the interface 700.

The beamforming weights per direction are transformed to beamforming weights per antenna. That is, according to this embodiment the RE 200 is configured to perform (optional) step S310:

S310: The RE 200 obtains transforms the updated beamforming weights per direction to updated beamforming weights per antenna.

The beamforming weights per antenna can then be used by the RE 200. Particularly, according to this embodiment the RE 200 is configured to perform (optional) step S312:

S312: The RE 200 applies the updated beamforming weights per antenna.

The updated beamforming weights could be applied at antennas 400 of the RE 200.

Reference is now made to FIG. 13 illustrating a method for determining beamforming weights for terminal devices as performed by the REC of the access node according to embodiments.

As disclosed above, the RE 200 in step S306 provides beamforming information per direction to the REC 300 over the interface 700. Therefore, the REC 300 is configured to perform step S402:

S402: The REC 300 obtains beamforming information per direction from the RE 200 over the interface 700.

Beamforming weight to be applied at the RE 200 are then determined. Particularly, the REC 300 is configured to perform step S406:

S406: The REC 300 determines the beamforming weights per direction based on the beamforming information per direction.

The beamforming weights are then provided to the RE 200. Thus, the REC 300 is configured to perform step S408:

S408: The REC 300 provides the beamforming weights per direction to the RE 200 over the interface 700.

As disclosed above, in some aspects only high quality beamforming information per direction is used. That is, according to an embodiment only the beamforming information per direction having a signal quality measure above the threshold is used when determining the beamforming weights per direction.

In some aspects the beamforming weights are determined based on the rank of the radio channel, either for multiple users or a single user (where each user is defined by a respective terminal device 600). Therefore, according to an embodiment the REC 300 is configured to perform (optional) step S404:

S404: The REC 300 obtains channel rank information from the RE 200 over the interface 700. The beamforming weights per direction are then determined based on the channel rank information.

Particular aspects of efficient dimensions reduction of the beamforming weights when communicating information of the beamforming weights, such as beamforming information, between the REC 300 and the RE 200 will now be disclosed in more detail.

The energy or Signal-to-Noise and Interference Ratio (SNIR) of the signals is approximately the same for all the individual antennas. However, since the received UL signals (or transmitted DL signals) are associated with pointing directions due to the beamforming weights, the energy or SNIR will be unevenly distributed between different pointing directions. This is illustrated in FIG. 8. At 1800a is illustrated energy per antenna and at 1800b is illustrated energy per direction. The energy or SNIR will be high in pointing directions from the access node to the terminal devices and low in other directions. This motivates beamforming information per direction to be communicated between the RE 200 and the REC 300 in accordance with embodiments disclosed above.

If the majority of the energy is located in a subset of all possible pointing directions, and signals in other directions contains mostly noise, it is possible to reduce the dimensionality of the signals to process by only selecting the directions with significant energy or SNIR. This motivates the above embodiments wherein the beamforming information per antenna is associated with a first dimensionality, and wherein the beamforming information per direction is associated with a second dimensionality being smaller than the first dimensionality.

In general terms, beam information, such as information relating to the beamforming weights, can be transmitted between the RE and the REC as linear combinations of predetermined beams. Hence, in accordance with what has been disclosed above, the beamforming information per direction could be represented by a linear combination of predetermined beamforming weights per direction

The UL OFDM FFT in FIG. 6 provides the output of Resource Elements for all antennas with N_aantenna elements used for beamforming. The Resource Elements include either data or pre-known Reference Symbols.

By extracting the same Reference Symbol from all N_aantennas elements and perform a transformation, e.g., a DFT, on the thus N_aReference Symbols, a transformation to a beam direction space from the antenna space is done. Each value in the N_along output from the transform provides information about the quality of the signal in one of the N_afixed beam directions.

All beamforming weight calculation can thereby be made in the beam direction space to minimize complexity. The beamforming weights can be determined based on the rank of the radio channel either for multiple users or a single user and be determined to limit the influence of UL interference as well as spreading of interference in the DL. This could be accomplished by combining enough of the discrete beam directions properly weighted to capture the majority of the energy that will result in a small throughput loss on network level.

The beamforming weights in the direction space (of size N_d) for beam b are given by:

w
_d
^b
,d=i
₀
,i
₁
, . . . ,i
_N
_d
₋₁

When applying the weights to either the UL or DL digital beamforming, they need to be transformed to the antenna space to handle the Resource Elements with data that also is in the antenna space.

The beamforming weights in the antenna space (of size N_a) for beam b are given by:

w
_a
^b
,a=0,1, . . . ,N_a−1

If the transformation matrix used for performing the inverse transform (i.e. from the direction space to the antenna space) is denoted F_ad, then:

w
_a
^b=Σ_dF_adw_d^b

The number of bits for per physical resource block (PRB) group and beam is N_aB_weight, whilst the number of bits for w_d^bper PRB group and beam is N_dB_weight.

As disclosed above, in order to reduce the load on the interface 700 between the REC 300 and the RE 200 the beamforming weights can be transferred from the REC to the RE in the direction space. The number of beamforming weights N_din the direction space is smaller than the number of beamforming weights N_a, in the antenna space. That is, the beamforming information per antenna is associated with a first dimensionality, and the beamforming information per direction is associated with a second dimensionality being smaller than the first dimensionality.

In addition to the beamforming weights in the direction space the indices i₀, i₁, . . . appointing which of the fixed beam directions to use are also transferred to the beamforming weight transformation block. The number of bits to transmit for the indices per PRB group and beam is ≤N_dB_ind.

The beam weight transformation block also include a hardware abstraction transformation function that hide the radio physical implementation and is capable to handle different configurations of number of antenna elements in the beamforming array antenna and other implementation aspects and states. In more detail, since most of the signal processing can be performed in the directional space, it will be possible to hide some hardware properties from the signal processing software. For example, this enables identical signal processing software to be used for different number of antennas but with the same number of used dimensions in the directional space.

Particular aspects of efficient compression of beamforming weights will now be disclosed in more detail.

Compression of the beamforming weights could be achieved by applying different encoding techniques to reduce the channel estimate in the UL direction and the beamforming weights in the DL direction. For example, the encoding techniques could be based on reducing the resolution of information in a transform domain, and/or sending differential information.

FIG. 9 shows a compression subsystem 1900 according to an embodiment where the interface 700 separates the REC 300 and the RE 200. The compression subsystem comprises a quantizer (“Quant.”) placed in the REC and respective compression memories (“Compression DB” where DB is short for database) in the REC and RE. The compression subsystem could be controlled by the UPC in the REC. The use of beam direction space as disclosed above is optionally part of the compression subsystem.

An embodiment of the quantizer of the compression subsystem will now be disclosed. In general terms, the quantizer can apply reduction of word widths, have variable length lists of coefficients and/or apply other compression techniques. The beamforming weights are applied per OFDM symbol. A first reduction of the data is to use the same beamforming weights for a set of consecutive OFDM symbols (e.g., a set of consecutive subcarriers) in the time domain, typically for a complete transmission. A second reduction is to use the same beamforming weights for a set of consecutive OFDM symbols in the frequency domain. The REC could then inform which beamforming weights to use for each N subcarriers, and the RE could then perform interpolation for determining the beamforming weights of the subcarriers between these N subcarriers.

In some aspects the value of N is dependent on the channel coherence. Thus, the channel estimate can be processed in order for the REC to determine a value of N that is small enough to give sufficient beamforming performance. Each value of N will give a certain SNR performance, so the choice of N can be determined jointly with other related parameters in the link adaptation, such as output power and modulation and coding scheme (MCS).

An embodiment of the compression memories of the compression subsystem will now be disclosed. In order to reduce the load of the interface between the REC and the RE is for to the REC is to send differential information of the beamforming weights rather than a complete set of (quantized) beamforming weights to the RE. Two sources of base information on which the differential information can be based are previous transmissions of beamforming weights between the REC and the RE and beamforming weights determined internally in the RE. Previous transmissions of beamforming weights between the REC and the RE can either be from the last transmission (i.e., recent-most used beamforming weights), or as a continuous incremental refinement of beamforming weights. For continuous incremental refinement of beamforming weights, beamforming weights are first transmitted with a coarse resolution and then more refined, e.g. first using a large value N and then with incrementally smaller values of N. In terms of beamforming weights determined internally in the RE, the RE could be configured to determine the beamforming weights from its locally stored channel estimate. In the case that the REC is satisfied (e.g., the RE's assumption of single-user transmission is correct; one example where the REC is not satisfied is when it has received additional information (e.g. about interference) that it wants to use to create better beamforming weights) with the beamforming weights determined by the RE (e.g. as optimized for single-user operation), the REC can indicate this to the RE, thus requiring less signaling on the interface between the REC and the RE than if beamforming weight information is transmitted. This allows the REC to only send beamforming weights for terminal devices which beamforming weights have by the REC been modified (e.g. as optimized for multi-user operation).

An advantage of using the herein disclosed compression memories is that a communications system with stationary terminal devices will have very little signaling relating to the beamforming weights over the interface between the REC and the RE since such stationary terminal devices could use the same beamforming weights for two or more consecutive transmissions.

There are different ways of providing the differential information. Examples relating thereto will be provided next. A first example is to provide additive differential information whereby differential beamforming weights are added to the stored beamforming weights. A second example is to provide interpolating differential information whereby complementary beamforming weights, e.g. in frequency domain, are sent in order to improve the resolution (of beamforming weights having been sent with lower value of N) of the beamforming weights. A second example is to provide multiplicative differential information whereby sending a beam form to multiply the beamforming weights with in direction space. Assuming that the beamforming weights are represented as beamforming weight vectors, then each element of the beamforming weight vector corresponds to one direction. Setting one of those elements to zero (by multiply with zero) will create a beam pattern with a null in that direction. This can be very efficient to express a nulling, i.e. setting a restriction on a transmission.

The transmission of information over the interface between the REC and the RE in the UL generally is based on the same features as for the DL, but with information relating to channel estimates being transferred (instead of information relating to beamforming weights).

The (S)RS channel estimation function is configured to set a suitable value of N, and a suitable quatization. The UPC can, ahead of a measurement, state the maximum amount to data to send, not to overload the interface between the REC and the RE. One way is for the UPC in the REC to provide the RE with a value of N and where the RE then executes.

Particular aspects of efficient selection of UL beams will now be disclosed in more detail.

It can here be assumed that the UL beams are selected based on the channel state information. Selecting as a few beams as possible could reduce the bit rate on the interface between the REC and the RE. At the same time, more beams allow for more energy to be received and better suppression of interferers.

The more terminal devices that are scheduled together and the more angular spread of the beams needed to reach the terminal devices, the more beams need to be selected in order to reach good demodulation performance. Also the types of beams used (such as beam widths and energy per beam) will affect the number of beams needed. Beams with beamforming weights determined according to SVD (so-called SVD-beams) will require fewer selected beams than beams with beamforming weights determined according to a DFT (so-called DFT-beams) but could require more computational resources to compute. Hence there is a tradeoff between reduction of bitrate on the interface between the REC and the RE and processing complexity of the beam selection. DTF-beams imply that a fixed set of beams that evenly cover the whole spatial view seen from the RE is used. The DTF-beams can be spread out one-dimensionally (either spread out vertically or horizontally) or they can be spread out two-dimensionally (spread out both vertically and horizontally). DFT is one example of creating fixed beams. As understood by the person skilled in the art other methods for creating fixed beams, optionally including uneven spread of the beams, are equally applicable. SVD-beams imply that the set of beams are dynamically decided depending on the received signals from the desired terminal devices and optionally also dependent on received signals from the interfering terminal devices. As understood by the person skilled in the art other methods for creating dynamic beams are equally applicable.

The selection of beams can be based on channel estimates of reference symbol signals, or on channel estimates of demodulation reference signals (DMRS) embedded in an LTE physical uplink shared channel (PUSCH).

The selection of beams can be performed for each resource block, or for groups of resource blocks, or even with one common selection for the whole carrier bandwidth. The selection of beams can be performed with one single, common, selection for the whole carrier bandwidth, since the beams generally correspond to physical directions and even though the phase of the radio propagation is frequency dependent, the physical directions are not.

If multiple terminal devices are scheduled in the same resource blocks, and the RE is connected to one single REC, the selections can be made as the union of the beams needed for each terminal device. If, for example, three beams are selected for a first terminal device and three beams are selected for a second terminal device and one of the beams is common for the first terminal device and the second terminal device, then only five beams need to be transferred from the RE to the REC. If, on the other hand, the RE is connected to two RECs, and the first terminal device is handled by one REC and the second terminal device is handled by another REC, then three beams need to be sent to each REC.

According to some aspects the maximum number of beams that can be selected is controlled by the REC. The information about how many beams that can be selected are then provided by the REC to the RE over the interface. This information can be expressed in different ways. According to a first example the information specifies the maximum number of beams selected per terminal device. According to a second example the information specifies the maximum number of beams selected in total. According to a third example the information specifies a combined restriction of both the maximum number of beams selected per terminal device and the maximum number of beams selected in total. Given these limitations, the RE will decide how many beams that should be selected per terminal device. This can be done by the RE comparing the signal quality for the terminal devices in each of the beams, and select beams with the given limitations. In additions to the limitations given by the REC, the RE could also need to make the selection such that each terminal gets at least M≥1 number of beams selected.

FIG. 14 schematically illustrates, in terms of a number of functional units, the components of an RE according to an embodiment. Processing circuitry 210 is provided using any combination of one or more of a suitable central processing unit (CPU), multiprocessor, microcontroller, digital signal processor (DSP), etc., capable of executing software instructions stored in a computer program product 1810a (as in FIG. 18), e.g. in the form of a storage medium 230. The processing circuitry 210 may further be provided as at least one application specific integrated circuit (ASIC), or field programmable gate array (FPGA).

Particularly, the processing circuitry 210 is configured to cause the RE to perform a set of operations, or steps, S102-S108, S302-S312, as disclosed above. For example, the storage medium 230 may store the set of operations, and the processing circuitry 210 may be configured to retrieve the set of operations from the storage medium 230 to cause the RE to perform the set of operations. The set of operations may be provided as a set of executable instructions. Thus the processing circuitry 210 is thereby arranged to execute methods as herein disclosed.

The storage medium 230 may also comprise persistent storage, which, for example, can be any single one or combination of magnetic memory, optical memory, solid state memory or even remotely mounted memory.

The RE may further comprise a communications interface 220 for communications at least with the REC and the terminal devices. As such the communications interface 220 may comprise one or more transmitters and receivers, comprising analogue and digital components.

The processing circuitry 210 controls the general operation of the RE e.g. by sending data and control signals to the communications interface 220 and the storage medium 230, by receiving data and reports from the communications interface 220, and by retrieving data and instructions from the storage medium 230. Other components, as well as the related functionality, of the RE are omitted in order not to obscure the concepts presented herein.

FIG. 15 schematically illustrates, in terms of a number of functional modules, the components of an RE 200 according to an embodiment. According to some aspects the RE 200 comprises an obtain module 210a configured to perform step S102, an obtain module 210b configured to perform step S104, a (optional) transform module 210C configured to perform step S106, and an apply module 210d configured to perform step S108. According to some aspects the RE 200 comprises an obtain module 210e configured to perform step S302, a transform module 210 of configured to perform step S304, a provide module 210g configured to perform step S306, an (optional) obtain module 210h configured to perform step S308, a (optional) transform module 210i configured to perform step S310, and an (optional) apply module 210j configured to perform step S312.

In general terms, each functional module 210a-210j may be implemented in hardware or in software. Preferably, one or more or all functional modules 210a-210j may be implemented by the processing circuitry 210, possibly in cooperation with functional units 220 and/or 230. The processing circuitry 210 may thus be arranged to from the storage medium 230 fetch instructions as provided by a functional module 210a-210j and to execute these instructions, thereby performing any steps of the RE as disclosed herein.

FIG. 16 schematically illustrates, in terms of a number of functional units, the components of an REC according to an embodiment. Processing circuitry 310 is provided using any combination of one or more of a suitable central processing unit (CPU), multiprocessor, microcontroller, digital signal processor (DSP), etc., capable of executing software instructions stored in a computer program product 1810b (as in FIG. 18), e.g. in the form of a storage medium 330. The processing circuitry 310 may further be provided as at least one application specific integrated circuit (ASIC), or field programmable gate array (FPGA).

Particularly, the processing circuitry 310 is configured to cause the REC to perform a set of operations, or steps, S202-S208, S402-S408, as disclosed above. For example, the storage medium 330 may store the set of operations, and the processing circuitry 310 may be configured to retrieve the set of operations from the storage medium 330 to cause the REC to perform the set of operations. The set of operations may be provided as a set of executable instructions. Thus the processing circuitry 310 is thereby arranged to execute methods as herein disclosed.

The storage medium 330 may also comprise persistent storage, which, for example, can be any single one or combination of magnetic memory, optical memory, solid state memory or even remotely mounted memory.

The REC may further comprise a communications interface 320 for communications at least with the RE. As such the communications interface 320 may comprise one or more transmitters and receivers, comprising analogue and digital components.

The processing circuitry 310 controls the general operation of the REC e.g. by sending data and control signals to the communications interface 320 and the storage medium 330, by receiving data and reports from the communications interface 320, and by retrieving data and instructions from the storage medium 330. Other components, as well as the related functionality, of the REC are omitted in order not to obscure the concepts presented herein.

FIG. 17 schematically illustrates, in terms of a number of functional modules, the components of an REC according to an embodiment. According to some aspects the REC 300 comprises an obtain module 310a configured to perform step S302, a transform module 310b configured to perform step S304, a provide module 210C configured to perform step S306, and a (optional) provide module 210d configured to perform step S308. According to some aspects the REC 300 comprises an obtain module 310e configured to perform step S402, an (option) obtain module 210f configured to perform step S404, a determine module 310g configured to perform step S406, and a provide module 310h configured to perform step S408.

In general terms, each functional module 310a-310h may be implemented in hardware or in software. Preferably, one or more or all functional modules 310a-310h may be implemented by the processing circuitry 310, possibly in cooperation with functional units 320 and/or 330. The processing circuitry 310 may thus be arranged to from the storage medium 330 fetch instructions as provided by a functional module 310a-310h and to execute these instructions, thereby performing any steps of the REC as disclosed herein.

The RE and REC may be provided as standalone devices or as a part of at least one further device. For example, as disclosed above the RE and REC may be provided in an access node. Alternatively, functionality of the RE and the REC may be distributed between at least two devices, or nodes.

Thus, a first portion of the instructions performed by the RE or REC may be executed in a first device, and a second portion of the of the instructions performed by the RE or REC may be executed in a second device; the herein disclosed embodiments are not limited to any particular number of devices on which the instructions performed by the RE or REC may be executed.

Hence, the methods according to the herein disclosed embodiments are suitable to be performed by an RE or REC residing in a cloud computational environment. Therefore, although a single processing circuitry 210, 310 is illustrated in FIGS. 14 and 16 the processing circuitry 210, 310 may be distributed among a plurality of devices, or nodes. The same applies to the functional modules 210a-210j, 310a-310h of FIGS. 15 and 17 and the computer programs 1820a, 1820b, 1820c, 1820d of FIG. 18 (see below).

FIG. 18 shows one example of a computer program product 1810a, 1810b, 1810c, 1810d comprising computer readable means 1830. On this computer readable means 1830, a computer program 1820a, 1820c can be stored, which computer program 1820a, 1820c can cause the processing circuitry 210 and thereto operatively coupled entities and devices, such as the communications interface 220 and the storage medium 230, to execute methods according to embodiments described herein. The computer program 1820a, 1820c and/or computer program product 1810a, 1820c may thus provide means for performing any steps of the RE as herein disclosed. On this computer readable means 1830, a computer program 1820b, 1820d can be stored, which computer program 1820b, 1820d can cause the processing circuitry 310 and thereto operatively coupled entities and devices, such as the communications interface 320 and the storage medium 330, to execute methods according to embodiments described herein. The computer program 1820b, 1820d and/or computer program product 1810b, 1810d may thus provide means for performing any steps of the REC as herein disclosed.

In the example of FIG. 18, the computer program product 1810a, 1810b, 1810c, 1810d is illustrated as an optical disc, such as a CD (compact disc) or a DVD (digital versatile disc) or a Blu-Ray disc. The computer program product 1810a, 1810b, 1810c, 1810d could also be embodied as a memory, such as a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), or an electrically erasable programmable read-only memory (EEPROM) and more particularly as a non-volatile storage medium of a device in an external memory such as a USB (Universal Serial Bus) memory or a Flash memory, such as a compact Flash memory. Thus, while the computer program 1820a, 1820b, 1820c, 1820d is here schematically shown as a track on the depicted optical disk, the computer program 1820a, 1820b, 1820c, 1820d can be stored in any way which is suitable for the computer program product 1810a, 1810b, 1810c, 1810d.

The inventive concept has mainly been described above with reference to a few embodiments. However, as is readily appreciated by a person skilled in the art, other embodiments than the ones disclosed above are equally possible within the scope of the inventive concept, as defined by the appended claims. For example, although the embodiments mainly have been described in a time division duplex (TDD) context, at least some of the embodiments are also applicable for frequency division duplex (FDD). One difference in FDD compared to TDD is that the UL measurements cannot for sure be used for DL, due to the different frequencies UL and DL. In FDD the terminal device sends information about what beam direction the terminal device deems is best (from a raster of predetermined beams that the access node is repeatedly transmitting, i.e. a “code book”). In this case, the REC may need to provide beamforming weights to the RE (thus defining the external information).

Methods and Devices for Determination of Beamforming Information

Information

Publication Number

Date Filed

Date Published

Inventors

CPC

International Classifications

Abstract

Description

Claims

PCT Information

Provisional Applications (1)