Distributed AAS beam refinement

Abstract

At least partial beam-finding is moved from a central processor into each of a plurality of Digital Front End (DFE) circuits, by performing FFT operations in the DFEs. This significantly reduces the processor / DFE interface bitrate. Additionally, aspects are presented where the DFEs are connected to the central processor serially, to support better scaling and a more distributed system than is possible with parallel connections. Still further, calculation of at least part of the covariance matrix is placed in the DFEs, to derive dominating beam directions, further reducing the amount of information required to be transmitted to the central processor for beam-finding. Decisions on which beamforming to use are preserved to the central processor, providing flexibility and enabling an optimum joint decision. This also enables the use of a joint coherent beam across the full antenna array.

Description

[0001] DISTRIBUTED AAS BEAM REFINEMENT

[0002] TECHNICAL FIELD

[0003] The present disclosure relates generally to wireless communications, and in particular to performing beam-finding in Digital Front End (DFE) circuits of an Advanced Antenna System (AAS) by performing domain transforms on narrowband signals in the DFEs.

[0004] BACKGROUND

[0005] Wireless communication networks are ubiquitous in many parts of the world. These networks continue to grow in capacity and sophistication. To accommodate more users, different types of devices, and different use cases, the technical standards governing the operation of wireless communication networks continue to evolve. The fourth generation (4G) of network standards has been deployed, the fifth generation (5G) is in development and early deployment, and the sixth generation (6G) is in design. With each generation, technological advances improve the capacity, spectral efficiency, and achieved bitrate of the wireless communication system. As spectrum in lower frequencies fills up, higher frequency spectrum is taken into use.

[0006] 5G added a second frequency range, FR2. This provided significant new available spectrum in the range 24.25-52.6 GHz (FR1 spans 410 - 7125 MHz). At these high frequencies, wavelengths are small. Correspondingly, antenna elements are small, and each captures / radiates less energy. Accordingly, more antenna elements are required to cover the same area with the same link performance and the same inter site distance. AAS implementations contemplate hundreds, or even thousands, of antenna elements. This increases the number of data streams to be processed, increasing the system computational load. Another problem with RF carriers at these high frequencies is that they suffer higher path loss, and hence have limited range, compared to conventional wireless telecom operating frequencies. Beamforming is one technique featured in 5G and 6G, to improve both coverage and capacity.

[0007] Beamforming refers to the use of antennas having increased and controllable directionality, whereby an RF transmission (or reception sensitivity) is narrow, and is “aimed” in a specific direction. This is enabled by transmitting or receiving signals with controlled relative phase and gain in the antenna elements (or subarrays of antenna elements). The relative phases of, e.g., transmit signals sent to each antenna element are controlled to create constructive or destructive interference, thus amplifying the signal in some directions, and attenuating it in others, and hence controlling the direction in which the beam is transmitted. Similar phase manipulation of signals from antenna elements (or subarrays) in a receive antenna can also result in beamforming the sensitivity of an antenna array in receiving signals. FIG. 1 shows how a successively larger phase shift at each of adjacent antenna elements results in a directionally steered RF beam. Another advantage to beamforming is that, particularly with large antenna arrays, multiple orthogonal beams can be formed and aimed in different directions, thus simultaneously addressing multiple wireless devices, also known as User Equipment (UE). For example, both a direct beam (line of sight) and a reflected beam may be targeted to a UE. Additionally, frequency-selective beamforming may be implemented, wherein subcarriers are assigned different weights, thus pointing beams in different directions as a function of frequency.

[0008] Because beamforming combines the outputs of multiple antenna elements (or subarrays), it increases beam gain, concentrating greater RF signal energy towards a receiver. This mitigates the inherent path loss of higher frequency carrier signals, and restores the rated equivalent isotropic radiated power (EIRP) rating of base stations operating in FR1 and FR2 to usable levels.

[0009] Beamforming may be implemented in several ways. Some basic methods are briefly reviewed.

[0010] One popular, low complexity option is analog beamforming. Here, the signals to / from the antennas are beamformed by phase and amplitude shifters in the RF domain, close to the antenna. The rest of the signal chain is common to all or a portion of the antenna elements. In this scenario, all of the data is converted into a time domain stream early, before being sent to the radio ASICs and antennas. Since one set of beam weights is applied during the duration of one Orthogonal Frequency Division Multiplexing (OFDM) symbol, the beam is spatially fixed for all data. Although it may have peaks in multiple directions, the data stream will be transmitted through one beam pattern, which obviously limits the possibility to simultaneously transmit data to multiple users. This technique is thus problematic when it would be advantageous to direct different data streams in different directions, by frequency selective scheduling. Additionally, it may be difficult for the base station to find the UEs for initial access. Present FR2 AAS systems use beam sweeping, or wider initial beams, to address this deficiency. However, this adds cost in terms of coverage, latency, and / or capacity.

[0011] The other extreme is full digital frequency selective beamforming where each antenna element and each subcarrier could have individual beam forming weights and support the possibility to simultaneously transmit data to multiple users as well as frequency selective beamforming. However, this requires Inverse Fast Fourier Transform (IFFT) processing per antenna element, which adds complexity. It also results in very high interface bitrates, which becomes particularly problematic when the number of antennas approaches 1000, and with channel bandwidths exceeding 1 GHz.

[0012] The problem of high interface bitrates for digital beamforming with a large number of antennas may be mitigated by implementation of distributed beamforming. This retains the Frequency Division Multiplexing (FDM) and Spatial Division Multiplexing (SDM) advantages. However, there is no access to individual antenna elements independently from the central processor; hence beam sweeping is required in the uplink (UL) for UE directional finding.

[0013] FIG. 2 depicts one architecture for implementing a variety of beamforming operations, shown roughly in actual layout relationship. RFICs are located under the antenna array, providing short interconnects between antenna elements (or subarrays of antenna elements) and RF signal processing paths. DFE circuits are placed around the antenna array and RFICs. The DFEs contain digital signal processing circuitry. A central processor, such as a baseband processor, performs beamforming calculations.

[0014] FIG. 3 depicts these architectural components in a data flow view, in which the DFEs are interposed between the RFICs / antenna elements and the central processor. Considering the data flow for the receiver case - i.e., antennas to processor - the information rate is reduced in several steps.

[0015] Several antenna elements can be combined to form a subarray. The RFICs contain analog time domain beamforming (ATDBF) circuits. The DFEs include circuitry to perform digital time domain beamforming (DTDBF). DTDBF circuits process the antenna ports provided by the ATDBF, and can provide a number of independent beams to support single or multiple users within the analog beam created by the ATDBF. Like the ATDBF, the DTDBF is applied to the OFDM time domain symbols. DTDBF can apply different beam weights for different portions of the spectrum. Normally, not per sub-carrier, but for groups of sub-carriers. The group of subcarriers is first filtered out from rest of the signal using digital filters in time domain, and then the beam weight is applied on the sub-set of the spectrum. Accordingly, DTDBF can support multiple UEs, on multiple beams. The DTDBF is calculated from uplink signals, such as the UL Sounding Reference Signal (SRS). Finally, digital frequency domain beamforming (DFDBF) is applied in the central processor, allowing frequency selective beamforming, within the beams created by the ATDBF and the DTDBF. This allows for Single-User (SU) and Multi-User (MU) Multiple Input Multiple Output (MIMO), as well as spatial multiplexity to a single UE where the user can be multiplexed, both spatially and in the frequency domain.

[0016] One problem is how to locate and track UEs. The port expansion in the DTDBF is normally dominating; in many cases, ATDBF and DFDBF are very limited or even omitted. The ability to detect all UEs simultaneously would require access to all antenna ports entering DTDBF, and forwarding this information to the central unit for joint processing; however, this is not feasible due to interface data rate limitations.

[0017] One approach to mitigate these problems with distributed digital BF is a parallel narrowband receiver, which extracts a small frequency portion from each antenna element and forwards that for digital processing. This architecture is depicted in FIG. 4. The received narrowband signals are sufficient for estimation of the main directions of the received signal. Since directions are more stable than the complex channel, a second wideband receiver can then, in a second step, use the directions to perform wideband beamformed reception in the directions obtained from the narrowband receiver. For those directions, wideband reference signals from the UE can be used to determine the DFDBF. When all the beamformers are configured (ATDBF, DTDBF and DFDBF), the full bandwidth may be utilized for communications with a reduced complexity and interface load. This allows a wideband reception that retains a high signal to noise ratio (SNR), while reducing the number of data streams that needs to be interfaced for further combining - down to one for each direction. This approach is the subject of US Published Patent Application 2023 / 0170973. This approach provides a good solution for efficient reception across the full spatial domain within a narrow bandwidth. Bitrates on the interface to the central processor are reduced significantly, enhancing beamforming performance and enabling new use cases.

[0018] However, in some use cases, the resulting interface rate using a time domain interface between the DFEs and the central processor is still high, and causes high latency or requires over-dimensioning of the interface. This is particularly troublesome if the DFEs are connected in series. Series connection of DFEs is attractive for several reasons. For FR1, it is the limited range of Serializer / Deserializer (SerDes) interfaces combined with large antenna arrays. For FR2, it is the limited number of interfaces available in the central processor, in combination with a wide RF bandwidth. Distributed MIMO applications envision many spatially separated nodes, such as radio stripes and various 6G architectures. FIG. 5 shows the architectural difference between a 4G / 5G central MIMO base station, and a 6G radio stripe, in which a large number of distributed MIMO antenna arrays are arranged serially.

[0019] The Background section of this document is provided to place aspects of the present disclosure in technological and operational context, to assist those of skill in the art in understanding their scope and utility. Approaches described in the Background section could be pursued, but are not necessarily approaches that have been previously conceived or pursued. Unless explicitly identified as such, no statement herein is admitted to be prior art merely by its inclusion in the Background section.

[0020] SUMMARY

[0021] The following presents a simplified summary of the disclosure in order to provide a basic understanding to those of skill in the art. This summary is not an extensive overview of the disclosure and is not intended to identify key / critical elements of aspects of the disclosure or to delineate the scope of the disclosure. The sole purpose of this summary is to present some concepts disclosed herein in a simplified form as a prelude to the more detailed description that is presented later.

[0022] According to aspects of the present disclosure described and claimed herein, at least partial beam-finding is moved from a central processor into each of a plurality of DFEs, by performing FFT operations in the DFEs. This significantly reduces the processor / DFE interface bitrate, for several reasons. First, the FFT allows for suppression of unwanted adjacent “blocker” signals, such as wireless communication signals from other networks or operators, which can have greater energy (e.g., >20dB) than desired signals when a mobile device is at the edge of a cell’s coverage area. Suppressing blocker signals early is important, as it enables compressing the desired information based on the SNR for the desired signal, rather than a sum of the desired signal and a blocker signal. Second, the early FFT allows for removal of the cyclic prefix (CP) and selection of the desired part of the frequency range ( / .e., removing any oversampling of the desired signal). Additionally, aspects are presented where the DFEs are connected to the central processor serially, to support better scaling and a more distributed system than is possible with parallel connections. Still further, calculation of at least part of the covariance matrix is placed in the DFEs, to derive dominating beam directions. This further reduces the amount of information required to be transmitted to the central processor when the NBR is used for beam-finding. Aspects presented herein preserve decisions on which beamforming to use to the central processor, providing flexibility and enabling an optimum joint decision. This also enables the use of a joint coherent beam across the full antenna array. The covariance matrix can be split and combined for efficient combination of the antenna streams from different DFEs, requiring only a single FFT per joint beam.

[0023] One aspect relates to a DFE circuit of an AAS comprising a plurality of DFEs connected in series. The DFE is configured to be part of, or connected to, one or more Radio Frequency Integrated Circuits (RFIC), which are connected to a number M of antenna elements or subarrays of antenna elements. The DFE includes a plurality of narrowband receivers (NBR). Each NBR comprises circuitry configured to receive a first signal from a corresponding antenna element or subarray of antenna elements, and process the first signal to produce a corresponding narrowband signal. The DFE also includes a corresponding plurality of domain transformation circuits. Each domain transformation circuit comprises circuitry configured to receive the corresponding narrowband signal and transform the corresponding narrowband signal between domains. The DFE further includes a selection circuit. The selection circuit is configured to select a subset of Resource Elements (RE) from the plurality of domain transformed signals. The DFE further includes circuitry configured to calculate a local partial covariance matrix (R) from the subset of REs, and interface circuitry configured to operatively connect to one or more other DFEs.

[0024] Another aspect relates to a method of performing partial beamforming in a DFE circuit of an AAS. The DFE is configured to be part of, or connected to, a RFIC, which is connected to a number M of antenna elements or subarrays of antenna elements. A first signal is received from a corresponding antenna element or subarray of antenna elements in each of a plurality of narrowband receivers and the received first signals are processed to generate narrowband signals. The narrowband signals are transformed between domains in a corresponding plurality of domain transformation circuits. A subset of REs is selected from the plurality of domain transformed signals. A local partial covariance matrix is calculated from the subset of REs. The DFE is connected to one or more other DFEs.

[0025] Yet another aspect relates to a wireless communication device operative in a wireless communication network. The wireless communication device includes processing circuitry configured to control wireless communications between the wireless device and a base station, and communication circuitry operatively connected to the processing circuitry. The communication circuitry includes a plurality of DFE circuits of an AAS. The plurality of DFEs are connected in series. Each DFE is configured to be part of, or connected to, one or more RFICs, which is connected to a number M of antenna elements or subarrays of antenna elements. Each DFE includes a plurality of narrowband receivers, each comprising circuitry configured to receive a first signal from a corresponding antenna element or subarray of antenna elements and process the first signal to produce a narrowband signal; a corresponding plurality of domain transformation circuits, each comprising circuitry configured to receive a corresponding narrowband signal and transform the narrowband signal between domains; a selection circuit configured to select a subset of Resource Elements from the plurality of domain transformed signals; circuitry configured to calculate a local covariance matrix (R) from the subset of Resource Elements; and interface circuitry configured to operatively connect to one or more other DFEs.

[0026] BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The present disclosure will now be described more fully hereinafter with reference to the accompanying drawings, in which aspects of the disclosure are shown. However, this disclosure should not be construed as limited to the aspects set forth herein. Rather, these aspects are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. Like numbers refer to like elements throughout.

[0028] FIG. 1 is a diagram showing the relationship between beam steering and phase shift.

[0029] FIG. 2 is hardware block diagram of RFICs under an antenna array, with DFEs and a central processor.

[0030] FIG. 3 is a data flow view of the hardware blocks of FIG. 2.

[0031] FIG. 4 is an AAS receiver block diagram showing parallel connections of DFEs, with NBRs in each DFE.

[0032] FIG. 5A is an illustration of a central MIMO base station.

[0033] FIG. 5B is an illustration of a 6G radio stripe.

[0034] FIG. 6 is an AAS receiver block diagram showing serial connections of DFEs, with NBRs and a multiplexer in each DFE.

[0035] FIG. 7 is an AAS receiver block diagram showing serial connections of DFEs, with UL

[0036] TD precoding in each DFE. FIG. 8 is an AAS transmitter block diagram showing serial connections of DFEs, with DL TD precoding in each DFE.

[0037] FIG. 9 is an AAS receiver block diagram showing serial connections of DFEs, with beam-finding and coherent summing of R in each DFE.

[0038] FIG. 10 is an AAS receiver block diagram showing serial connections of DFEs, with multiplexing of beam indexes.

[0039] FIG. 11 is a graph of signal power vs. number of beams for LOS and NLOS transmission.

[0040] FIG. 12 is a flow diagram of a method of performing partial beamforming in a DFE circuit of an AAS receiver.

[0041] FIG. 13 is a hardware block diagram of a wireless device in a wireless communication network.

[0042] FIG. 14 is a functional block diagram of a wireless device in a wireless communication network.

[0043] FIG. 15 is a hardware block diagram of a base station in a wireless communication network.

[0044] FIG. 16 is a functional block diagram of a base station in a wireless communication network.

[0045] DETAILED DESCRIPTION

[0046] For simplicity and illustrative purposes, the present disclosure is described by referring mainly to an exemplary aspect thereof. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. However, it will be readily apparent to one of ordinary skill in the art that the present disclosure may be practiced without limitation to these specific details. In this description, well known methods and structures have not been described in detail so as not to unnecessarily obscure the present disclosure.

[0047] Aspects of the present disclosure are generally described herein with respect to an AAS in a base station (e.g., gNB), performing beamforming on UL signals from UEs. Those of skill in the art will readily recognize the same inventive concepts are equally applicable to DL beamformed signals transmitted from the base station, as well as an AAS on a UE.

[0048] Referring again to FIG. 4, in this aspect the central processor has access to all resource elements (RE) - both UL reference symbols as well as data symbols - and can process them jointly.

[0049] It is well known to process signals in the time domain and frequency domain. Additional signal processing domains are useful to consider. When RF signals are received at the antenna, the power at each antenna element is weak. This may be considered an “element domain” view. Similarly, analyzing the signals over frequency yields a “subcarrier domain” view, in which per- subcarrier power is similar, and also weak. Little insight into beamforming is obtained in the element or subcarrier domains, which may be combined to form the “subcarrier antenna” domain.

[0050] A full digital frequency selective system where data received on each antenna element is transformed into frequency domain by its IFFT resulting in data in "subcarrier-element" domain. Here the known reference signal resource elements (REs) are selected. Applying, e.g., a DFT over element dimension transforms the “subcarrier-element” reference signal data to “subcarrier-beam” domain, which concentrates the power the significant beam direction. Further, applying, e.g., a DFT over the subcarriers transforms the "subcarrier-beam" domain data to "delay-beam" domain, which concentrates the reference signal power to significant beam "direction". This "delay-beam" domain is for most propagation channels a good way to concentrate the signal power and separate it from noise, since white noise is not affected by DFT processing and therefore not concentrated. Hence, in "delay-beam" domain the signals are often easier to distinguish from noise, and signal “de-noising” or noise suppression / cancellation can be performed. This is typically done by having an average noise power estimate (from, e.g., empty REs), this average noise power estimate is then used as a threshold separating the signal from noise. Data above the average noise power estimate are categorized as signal and below are categorized as noise and removed. The de-noised signals can then be transformed into whatever domain is suitable for further processing. This is the basis of de-noising and noise suppression.

[0051] As shown in FIG. 4, the antenna array is divided into N segments, each comprising M antenna elements (or subarrays). Each DFE includes NBRs (e.g., M NBRs) for beamforming detection and processing. UL reference symbols are received on some subcarriers, and are denoised by transforming the RS from subcarrier-antenna domain to delay-beam domain by FFT processing in the central processor. From these, the noise power is estimated and somewhat removed / supressed..

[0052] After de-noising, a covariance matrix R, representing relative phase differences between antenna elements (or subarrays), is calculated, and from that the time domain (TD) precoders, which hold weight vectors. In this aspect, full freedom on how to use R is possible. For example, it could be used with the highest possible spatial resolution RFULL where all elements are involved in the beam finding. Alternatively, a reduced RPARTIAL with wider beams and lower spatial resolution could be considered. Each DFE can then be configured with TD decoders by using either the complex vector beam directly, or by using an index into to a stored table of complex beamforming decoders.

[0053] FIG. 6 depicts another aspect, in which the DFEs are connected serially. Each DFE receives UL narrow band data channels (e.g., PRACH, PUSCH, PUCCH) via the NBR. These data signals are selected. Each DFE (except the first one) multiplexes the data signals received from its neighboring DFE with its own data signals before forwarding the combined signals to the next connected DFE in line (or central processor for the last one). The central processor unit then performs the conventional channel estimations and decoding to reveal the actual information received.

[0054] FIG. 7 depicts an aspect featuring UL TD precoding. The TD precoders, which contain weight vectors, are configured and activated in each DFE either by sending each DFE its part of the complex TD vector, or sending it a beam index from which the DFE checks its Look-Up Table (LUT) to find the corresponding prestored TD vector from DFE memory. The wideband time domain signals received on all physical antennas are TD-precoded by each DFE. Since each TD-precoded vector is a sub-set of the total TD-precoder vector defined by RFULL, the partial results from each DFE are coherently summed together by the sum block. This coherent summation sustains a low data rate over the interfaces. In the central processor unit, the received time domain signal is transformed to the frequency domain by an FFT. Each of B beams requires its own FFT operation. Finally, a FD precoder is applied, as well as equalizing, to extract the soft symbols for further processing.

[0055] FIG. 8 shows the inverse operation: TD precoding for DL transmissions from the base station. In the central processor, the Tx symbols are FD precoded, and then transformed into time domain for each beam by the I FFT block. The same information is sent to all DFEs; each DFE extracts its portion of the time domain signal in a split block. The wideband time domain signal is then TD-precoded by each DFE before being transmitted on the physical antennas. Since each TD-precoded vector is a sub-set of the total TD-precoded vector defined by RFULL, the partial result from each DFE is coherently summed together, creating a narrow beam when transmitted on the channel. Those of skill in the art will readily recognize that all aspects of the present disclosure have corresponding downlink / transmitter implementations, the realization of which is within the skill of those of ordinary skill in the art, given the teachings of the present disclosure.

[0056] In the data flow diagram of FIG. 3, the DFEs are connected to the central processor in parallel. In this configuration, the number of connections to the central processor is proportional to the number of DFEs. To allow better scaling and a more distributed system (such as for D- MIMO or large arrays), series connected DFEs are preferred. This is possible in the baseline solution, but it means that the NBR data load in the interface scales with number of DFEs (see FIG. 4). In the following discussion of aspects of the present disclosure, the DFEs are connected in series, as the benefits of the aspects are more apparent in this use case.

[0057] FIG. 9 depicts DFEs of an AAS connected serially. In this aspect, the DFTs perform beam-finding based on UL reference symbols. This approach features reduced complexity, reduced angular resolution, and reduced possibility to de-noise in the spatial domain, but with enhanced SNR, due to summation of the partial covariance matrices Rs.

[0058] In this configuration, each DFE connects (via RFIC) to M antenna elements (or subarrays). Accordingly, each DFE includes a plurality of narrowband receivers (NBR), M domain transformation circuits, a selection circuit, optionally (as indicated by dashed lines) a denoising circuit, a local partial covariance matrix calculating circuit, and coherent summing circuit. Each NBR comprises circuitry configured to receive a signal from a corresponding antenna element (or subarray of antenna elements) and process the signal to produce a narrowband signal. Each domain transformation circuit comprises circuitry configured to receive a corresponding narrowband signal and transform the narrowband signal between domains. For example, FIG. 9 depicts FFT circuits, which transform signals from a time domain to a frequency domain representation. The selection circuit is configured to select a subset of Resource Elements (reference symbols or data symbols) from the M domain transformed signals. The optional de-noising circuit is configured to de-noise the selected REs. The local partial covariance matrix calculating circuit is configured to calculate a local partial covariance matrix RPARTIAL from the subset of Resource Elements.

[0059] A coherent summing circuit coherently adds the local partial covariance matrix R to a partial covariance matrix received from an upstream DFE (except for the first one), and outputs the sum to a downstream circuit (the next DFE or, for the last one, the central processor). The spatially reduced covariance matrix RRED, output by the last DFE in the chain, cannot provide the same angular resolution as RFULL, but it has a better SNR and it is calculated with lower complexity compared to RFULL. The central processor then operates with RRED, from which the TD precoders are calculated. These can take the form of (i) full complex TD vector, or (ii) as an index into a TD beam value look-up table (LUT). The TD precoders in each DFE are then configured and activated, either by sending each DFE its part of the complex TD vector, or as a beam index from which the DFE checks its LUT to find the corresponding prestored TD vector from DFE memory. This summation of covariance matrices R requires the arrays connected to each DFE to be symmetric. Furthermore, the arrays connected to each DFE should be somewhat centralized, to experience the same radio propagation environment in terms of direction and relative amplitude of the incoming multipath components. More distributed DFE arrays will violate this constraint.

[0060] Distributing the calculation of the covariance matrix RRED to multiple DFEs is a compromise between accuracy (precision and resolution) and bit rate in the processor / DFE interface, as well as better scalability. The number of antenna elements (or subarrays) controls resolution, which refers to the ability to distinguish between multiple beams. Accuracy refers to precision in beamforming direction.

[0061] FIG. 10 depicts an aspect in which series-connected DFEs perform beam-finding based on UL reference symbols per DFE. This does not achieve full spatial resolution, hence there is a reduced possibility to de-noise in the spatial domain. In this configuration, each DFE finds its preferred beam directions in terms of, e.g., beam angular direction, beam power (denoted as beam idx in the FIG. 10), and sends this information (or some other information describing its spatial view) to the central processor. The processor concludes on the beam direction(s) and schedules traffic accordingly over the DFEs. When the time domain precoders are activated in each DFE, additional reference signals could be requested from the UE, and received on the formed TD beams. Based on these additional reference signals, a consecutive beamformer in the central processor could be formed to coherently transmit over all DFEs’ associated antennas toward a target. Only information of the B strongest beams is required at the central processor, resulting in a very low interface bitrate. This is suitable for distributed arrays, without constraints on the antenna element distributions or on DFE array colocation (as required for the configuration of FIG. 8). Each DFE monitors its own antenna array, and can do its spatial processing of the incoming reference signals and estimate some general parameters, such as multi-path angle of arrival and power. This supports quasi co-located or distributed DFE arrays.

[0062] In FIGs. 7 and 8, a combination of TD and FD precoding is shown. The FD precoding is in the central processor, and the TD precoders are part of the channel. The resulting interface data rate is then proportional to the number of beams used. The number of beams required for each UE depends on the propagation properties, and particularly if a UE is in Line Of Sight (LOS) or not (NLOS). FIG. 11 graphs the relative beam strengths in LOS and in deep NLOS conditions. If UEs are in LOS, only one beam is sufficient, and no FD precoding is required. The total number of beams is limited by the interface in between the central processor and the DFEs.

[0063] FIG. 12 shows the steps in a method 100 of performing partial beamforming in a DFE circuit of an AAS. The DFE is configured to be part of or connected to a RFIC which is connected to a number M of antenna elements or subarrays of antenna elements. A first signal is received from a corresponding antenna element or subarray of antenna elements in each of a plurality of narrowband receivers and the received reference signals are processed to generate narrowband signals (block 102). The narrowband reference signals are transformed between domains in M corresponding domain transformation circuits (block 104). A subset of Resource Elements is selected from the M domain transformed signals (block 106). The subset of Resource Elements is used to calculate a local partial covariance matrix (block 108). The DFE connects to one or more other DFEs (block 110).

[0064] Apparatuses described herein may perform the method 100 herein and any other processing by implementing any functional means, modules, units, or circuitry. In one embodiment, for example, the apparatuses comprise respective circuits or circuitry configured to perform the steps shown in the method figures. The circuits or circuitry in this regard may comprise circuits dedicated to performing certain functional processing and / or one or more microprocessors in conjunction with memory. For instance, the circuitry may include one or more microprocessor or microcontrollers, as well as other digital hardware, which may include digital signal processors (DSPs), special-purpose digital logic, and the like. The processing circuitry may be configured to execute program code stored in memory, which may include one or several types of memory such as read-only memory (ROM), random-access memory, cache memory, flash memory devices, optical storage devices, etc. Program code stored in memory may include program instructions for executing one or more telecommunications and / or data communications protocols as well as instructions for carrying out one or more of the techniques described herein, in several embodiments. In embodiments that employ memory, the memory stores program code that, when executed by the one or more processors, carries out the techniques described herein.

[0065] Figure 13 for example illustrates a hardware block diagram of a wireless device 10 as implemented in accordance with one or more embodiments. As shown, the wireless device 10 includes processing circuitry 14 and communication circuitry 18. The communication circuitry 18 (e.g., radio circuitry) is configured to transmit and / or receive information to and / or from one or more other nodes, e.g., via any communication technology. The communication circuitry 18 includes one or more DFEs 19 configured to perform beam-finding, as described herein. Such communication may occur via one or more antennas 20 that are either internal or external to the wireless device 10, as indicated by dashed lines. The processing circuitry 14 is configured to perform processing described above, such as by executing instructions stored in memory 16. The processing circuitry 14 in this regard may implement certain functional means, units, or modules.

[0066] Figure 14 illustrates a functional block diagram of a wireless device 30 in a wireless network according to still other embodiments. As shown, the wireless device 30 implements various functional means, units, or modules, e.g., via the processing circuitry 14 in Figure 13 and / or via software code. These functional means, units, or modules, e.g., for implementing the method 100 herein, include for instance: narrowband signal receiving / processing unit 32, domain transforming unit 34, RE selecting unit 36, local partial covariance matrix calculating unit 38, and interfacing unit 40. Narrowband signal receiving / processing unit 32 is configured to receive a first signal from a corresponding antenna element or subarray of antenna elements in each of a plurality of narrowband receivers and process the first signal to generate narrowband signals. Domain transforming unit 34 is configured to transform the narrowband reference signals between domains in M corresponding domain transformation circuits. RE selecting unit 36 is configured to select a subset of Resource Elements from the M domain transformed signals. Local partial covariance matrix calculating unit 38 is configured to calculate a local partial covariance matrix. Interfacing unit 40 is configured to connect to one or more other DFEs.

[0067] Figure 15 illustrates a hardware block diagram of a network node 50 as implemented in accordance with one or more embodiments. As shown, the network node 50 includes processing circuitry 52 and communication circuitry 56. The communication circuitry 56 is configured to transmit and / or receive information to and / or from one or more other nodes, e.g., via any communication technology. The communication circuitry 56 includes one or more

[0068] DFEs 57 configured to perform beam-finding, as described herein. The network node 50 may function as a base station (e.g., eNB, gNB, etc.), and may wirelessly communicate with a plurality of wireless devices 10, 30 via one or more antennas 58. As indicated by the broken line, the antennas 58 may be located remotely from the network node 50, such as on a tower or building. The processing circuitry 52 is configured to perform processing described above, such as by executing instructions stored in memory 54. Although represented as being within the network node 50, those of skill in the art understand that some or all of the processing circuitry 54 may be implemented as virtualized servers in a data center, e.g., in the so-called cloud. The processing circuitry 54 in this regard may implement certain functional means, units, or modules.

[0069] Figure 16 illustrates a functional block diagram of a network node 60 in a wireless network according to still other embodiments. As shown, the network node 60 implements various functional means, units, or modules, e.g., via the processing circuitry 54 in Figure 15 and / or via software code. These functional means, units, or modules, e.g., for implementing the method 100 herein, include for instance: narrowband signal receiving / processing unit 62, domain transforming unit 64, RE selecting unit 66, local partial covariance matrix calculating unit 68, and interfacing unit 70. Narrowband signal receiving / processing unit 62 is configured to receive a first signal from a corresponding antenna element or subarray of antenna elements in each of a plurality of narrowband receivers and process the received reference signals to generate narrowband signals. Domain transforming unit 64 is configured to transform the narrowband reference signals between domains in M corresponding domain transformation circuits. RE selecting unit 66 is configured to select a subset of Resource Elements from the M domain transformed signals. Local partial covariance matrix calculating unit 68 is configured to calculate a local partial covariance matrix. Interfacing unit 70 is configured to connect to one or more other DFEs.

[0070] Those skilled in the art will also appreciate that embodiments herein further include corresponding computer programs.

[0071] A computer program comprises instructions which, when executed on at least one processor of an apparatus, cause the apparatus to carry out any of the respective processing described above. A computer program in this regard may comprise one or more code modules corresponding to the means or units described above.

[0072] Embodiments further include a carrier containing such a computer program. This carrier may comprise one of an electronic signal, optical signal, radio signal, or computer readable storage medium.

[0073] In this regard, embodiments herein also include a computer program product stored on a non-transitory computer readable (storage or recording) medium and comprising instructions that, when executed by a processor of an apparatus, cause the apparatus to perform as described above. Embodiments further include a computer program product comprising program code portions for performing the steps of any of the embodiments herein when the computer program product is executed by a computing device. This computer program product may be stored on a computer readable recording medium. Aspects of the present disclosure present numerous advantages over the prior art. By performing at least partial beam-finding and beam-refinement in DFEs, the interface data rate between the DFEs and a central processor is greatly reduced. The system also provides for lower latency in beam management. As AAS for 5G advanced and 6G system are expected to be extremely large and feature distributed MIMO, reducing the interface load is critical. Particular advantages flow when the DFEs are connected in series.

[0074] The present invention may, of course, be carried out in other ways than those specifically set forth herein without departing from essential characteristics of the invention. The present embodiments are to be considered in all respects as illustrative and not restrictive, and all changes coming within the meaning and equivalency range of the appended claims are intended to be embraced therein.

Claims

CLAIMSClaims:

1. A Digital Front End, DFE, circuit of an Advanced Antenna System, AAS, comprising a plurality of DFEs connected in series, the DFE configured to be part of or connected to one or more Radio Frequency Integrated Circuits, RFIC, which are connected to a number M of antenna elements or subarrays of antenna elements, the DFE comprising: a plurality of narrowband receivers, each comprising circuitry configured to receive a first signal from a corresponding antenna element or subarray of antenna elements and process the first signal to produce a corresponding narrowband signal; a corresponding plurality of domain transformation circuits, each comprising circuitry configured to receive the corresponding narrowband signal and transform the corresponding narrowband signal between domains; a selection circuit configured to select a subset of Resource Elements from the plurality of domain transformed signals; circuitry configured to calculate a local partial covariance matrix (R) from the subset of Resource Elements; and interface circuitry configured to operatively connect to one or more other DFEs.

2. The DFE of claim 1 further comprising a de-noising circuit.

3. The DFE of claim 1 or 2 wherein the received signals are reference signals.

4. The DFE of any of claims 1-3 wherein each domain transformation circuit is configured to transform the corresponding narrowband signal from a time domain representation to a frequency domain representation.

5. The DFE of claim 4 wherein each domain transformation circuit is configured to perform a Fast Fourier Transform operation on the corresponding narrowband signal.

6. The DFE of any of claims 1-5 further characterized by a coherent summing circuit configured to coherently add the local partial covariance matrix to a partial covariance matrix received from an upstream DFE, and to output the sum to a downstream circuit.

7. The DFE of claim 6 wherein the downstream circuit is a downstream DFE.

8. The DFE of claim 6 wherein the downstream circuit is a central processor.

9. The DFE of claim 8 further characterized by: a plurality of wideband receivers, each comprising circuitry configured to receive a second signal from a corresponding antenna element or subarray of antenna elements and process the signal; and a time domain, TD, precoder circuit configured to apply TD precoding weights to the received second signals.

10. The DFE of claim 9 wherein the TD precoder circuit receives the TD precoding weights from the central processor.

11. The DFE of claim 9 wherein the TD precoder circuit retrieves the TD precoding weights from a preconfigured lookup table.

12. The DFE of claim 1 further characterized by: a beam identification circuit configured to detect one or more beam spatial angles or beam powers from the local partial covariance matrix; and a multiplexing circuit configured to combine the local beam spatial angles and beam powers with beam spatial angles or beam powers received from an upstream DFE and to output the multiplexed beam spatial angles or beam powers to a downstream circuit.

13. The DFE of claim 12 wherein the downstream circuit is a downstream DFE.

14. The DFE of claim 12 wherein the downstream circuit is a central processor.

15. A method (100) of performing partial beamforming in a Digital Front End, DFE, circuit of an Advanced Antenna System, AAS, the DFE configured to be part of or connected to a Radio Frequency Integrated Circuit, RFIC, which is connected to a number M of antenna elements or subarrays of antenna elements, the DFE characterized by: receiving (102) a first signal from a corresponding antenna element or subarray of antenna elements in each of a plurality of narrowband receivers and processing the received first signals to generate narrowband signals; transforming (104) the narrowband signals between domains in a corresponding plurality of domain transformation circuits; selecting (106) a subset of Resource Elements from the plurality of domain transformed signals; and calculating (108) a local partial covariance matrix from the subset of Resource Elements; and connecting (110) to one or more other DFEs.

16. The method (100) of claim 15 further comprising de-noising the subset of Resource Elements.

17. The method (100) of claim 15 or 16 wherein the received signals are reference signals.

18. The method (100) of any of claims 15-17 wherein transforming (106) the narrowband signals between domains comprises transforming the narrowband signals from a time domain representation to a frequency domain representation.

19. The method (100) of claim 18 wherein transforming the narrowband signals from a time domain representation to a frequency domain representation comprises performing Fast Fourier Transform operations on a narrowband signals.

20. The method (100) of any of claims 15-19 further characterized by in response to receiving a partial covariance matrix from an upstream DFE, coherently adding the local partial covariance matrix to the received partial covariance matrix; and outputting the sum partial covariance matrix to a downstream circuit.

21. The method (100) of claim 20 wherein the downstream circuit is a downstream DFE.

22. The method (100) of claim 20 wherein the downstream circuit is a central processor.

23. The method (100) of claim 20 further characterized by: receiving a second signal from a corresponding antenna element or subarray of antenna elements in each of a plurality of wideband receivers and processing the received signals; and applying time domain, TD, precoding weights to the processed signals.

24. The method (100) of claim 23 further characterized by receiving the TD precoding weights from the central processor.

25. The method (100) of claim 23 further characterized by: receiving a TD precoding index from the central processor; and retrieving the TD precoding weights from a preconfigured lookup table using the index.

26. The method (100) of claim 15 further characterized by: detecting one or more beam spatial angles or beam powers from the local partial covariance matrix; and in response to receiving a partial covariance matrix from an upstream DFE, combining the beam spatial angles or beam powers with beam spatial angles or beam powers received from an upstream DFE; and outputting the combined partial covariance matrix to a downstream circuit.

27. The method (100) of claim 26 wherein the downstream circuit is a downstream DFE.

28. The method (100) of claim 26 wherein the downstream circuit is a central processor.

29. A wireless communication device (10, 30, 50, 60) operative in a wireless communication network, comprising: processing circuitry (14, 52) configured to control wireless communications between the wireless device (10, 30, 50, 60) and other nodes in the wireless communication network; and communication circuitry (18, 56) operatively connected to the processing circuitry (14, 52), the communication circuitry (18, 56) including a plurality of Digital Front End, DFE, circuits of an Advanced Antenna System, AAS, the plurality of DFEs connected in series, wherein each DFE is configured to be part of or connected to one or more Radio Frequency Integrated Circuits, RFIC, which is connected to a number M of antenna elements or subarrays of antenna elements, each DFE characterized by: a plurality of narrowband receivers, each comprising circuitry configured to receive a first signal from a corresponding antenna element or subarray of antenna elements and process the first signal to produce a narrowband signal;a corresponding plurality of domain transformation circuits, each comprising circuitry configured to receive a corresponding narrowband signal and transform the narrowband signal between domains; a selection circuit configured to select a subset of Resource Elements from the plurality of domain transformed signals; circuitry configured to calculate a local covariance matrix (R) from the subset ofResource Elements; and interface circuitry configured to operatively connect to one or more other DFEs.

30. The device (10, 30) of claim 29 wherein the wireless communication device (10, 30) is aUser Equipment (10, 30).

31. The device (50, 60) of claim 29 wherein the wireless communication device (50, 60) is a base station (50, 60).

Images