System and method for supporting RIS beamforming in a wireless network

Abstract

The present disclosure provides a system and method for supporting beamforming in a wireless network with zero signaling overhead in operation. The system includes a Reconfigurable Intelligent Surface (RIS) controller associated with an RIS panel that autonomously enables communication between an access point and one or more user equipments (UEs) in a wireless network. The RIS controller is configured to detect a target UE present in the vicinity of the RIS panel based on one or more signals received from the target UE, localize the target UE to identify the relative position of the target UE with respect to one or more UEs, and select an optimal reflection coefficient matrix (RCM) associated with the RIS panel to enable beamforming for the target UE.

Description

Technical Field

[0001] Reservation of Rights Part of the disclosure of this patent document includes, without limitation, subject matter that is the subject of intellectual property rights such as copyrights, designs, trademarks, integrated circuit (IC) layout designs, and / or trade dress protection belonging to Jio Platforms Limited (JPL) or its partners (hereinafter referred to as the owner). The owner does not object to the facsimile reproduction by any person of the patent document or patent disclosure as it appears in the patent file or record at the Patent and Trademark Office, but reserves all rights otherwise. All rights in such intellectual property are fully reserved by the owner.

[0002] Field of the Disclosure Embodiments of the present disclosure generally relate to beamforming in wireless communication networks. In particular, the present disclosure relates to autonomous beamforming and tracking by a Reconfigurable Intelligent Surface (RIS) in a wireless communication network.

Background Art

[0003] Background of the Disclosure The following description of related art is intended to provide background information related to the field of the present disclosure. This section may include specific aspects of technologies that may be related to various features of the present disclosure. However, it should be understood that this section is used only to enhance the reader's understanding of the present disclosure and is not used as an admission of prior art.

[0004] The current 5th generation (5G) wireless communication technology being developed in the 3rd Generation Partnership Project (3GPP) means providing higher multi-gigabit per second (Gbps) peak data speeds, ultra-low latency, improved reliability, large-scale network capacity, increased availability, and a more uniform user experience to more users. Higher performance and efficiency enhance new user experiences and connect new industries. Although some goals have been achieved, there are still some problems that need to be solved, especially when it comes to enabling support for industrial sectors, architectures for supporting civilian networks, and flexible network deployment.

[0005] Regarding the above, a 6th generation (6G) network architecture capable of addressing the issue of network flexibility has been proposed. The proposed 6G network should be able to realize newly emerging technologies such as artificial intelligence, terahertz communication, optical wireless technology, free space optical network, 3D networking, quantum communication, unmanned aerial vehicles, self-free communication, integration of wireless information and energy transfer, integration of sensing and communication, integration of access backhaul network, dynamic network slicing, holographic beamforming, and big data analysis.

[0006] One such new technology proposed for use in 5G and beyond 5G networks is the Reconfigurable Intelligent Surface (RIS). The RIS corresponds to a smart reflecting surface equipped with many small reconfigurable metamaterial elements, also called "unit cells", which enable the control of the propagation environment by the coherent scattering of electromagnetic waves. These intelligent surfaces are reconfigurable and have reflection, refraction, and absorption characteristics that can adapt to the wireless channel environment. The RIS enables the control of wireless signals between transmitters and receivers in a dynamic and target-oriented manner, thereby changing the wireless environment for services that provide enhancements to various network Key Performance Indicators (KPIs) such as capacity, coverage, energy efficiency, positioning, and security.

[0007] The RIS can be constructed in a way that enables the control of an intelligent and programmable wireless environment, performs passive reflection, passive absorption, and passive scattering, and can change the physical environment into an intelligent and interactive one. The RIS can change the electromagnetic characteristics of the elements and generate phase shifts independently of the incident signals without using radio frequency (RF) signal processing. Also, the RIS technology has many technical features that exceed the current mainstream technologies. Compared with the massive multiple-input multiple-output (MIMO) system, the RIS-assisted wireless network significantly improves the system performance by smartly optimizing signal propagation.

[0008] Currently available RIS systems provide passive reflecting surfaces that rely on access points for reflection beamforming. This dependency causes delays when high traffic or a large number of users need to be served.

[0009] Therefore, there is a need in the art to provide an RIS system that can overcome the drawbacks of existing prior arts. SUMMARY OF THE INVENTION

[0010] Problems of the present disclosure Some of the problems of the present disclosure satisfied by at least one example herein are listed below.

[0011] A problem of the present disclosure is to provide autonomous reflective beamforming in a reconfigurable intelligent surface (RIS).

[0012] A problem of the present disclosure is to autonomously identify and continuously track the positions of one or more user equipment (UE) within the coverage area of the RIS.

[0013] Another problem of the present disclosure is to identify the UE and direct the beam from the access point to the UE identified based on the optimal reflection coefficient matrix (RCM).

[0014] Yet another problem of the present disclosure is to provide a communication technique for agnostic autonomous beamforming in the RIS.

[0015] Yet another problem of the present disclosure is to train a neural network to obtain an RCM codebook based on different UE positions.

[0016] Yet another problem of the present disclosure is to utilize multiple variants of a deep neural network to trigger autonomous beamforming in the RIS.

[0017] Yet another problem of the present disclosure is to provide a joint sensing communication system.

[0018] Yet another problem of the present disclosure is to provide joint sensing communication that utilizes an IRS for autonomous beam management and tracking.

[0019] Summary This section is provided to introduce specific problems and aspects of the present disclosure in a simplified form that will be further described later in the detailed description.

[0020] In one aspect, the present disclosure relates to a system that enables autonomous beamforming in a wireless network. The system has a RIS controller associated with a RIS panel that enables communication between an access point and one or more user equipment (UE) in the wireless network. The RIS controller detects a target UE present in the vicinity of the RIS panel based on one or more signals received from the target UE, locates the target UE to identify the relative position of the target UE with respect to the one or more UEs, and is configured to select a first optimal reflection coefficient matrix (RCM) associated with the RIS panel to enable beamforming for the target UE.

[0021] In some embodiments, the selected first optimal RCM may enable optimal reflection of the beam from the access point to the target UE.

[0022] In some embodiments, the RIS panel may have one or more reflection elements and one or more sensing elements.

[0023] In some embodiments, the one or more sensing elements may assist the RIS controller in detecting the presence of the target UE and detecting the movement associated with the target UE.

[0024] In some embodiments, the RIS controller may be configured to receive one or more uplink (UL) transmissions associated with the target UE from the one or more sensing elements and estimate the angle of arrival (AoA) associated with the target UE based on the received one or more UL transmissions.

[0025] In some embodiments, the RIS controller may be configured to select a second optimal RCM based on the detected movement associated with the target UE.

[0026] In some embodiments, the RIS controller may be configured to select first and second optimal RCMs from an RCM look-up table obtained based on training a neural network for different RCMs associated with different UE positions.

[0027] In some embodiments, the RIS controller may be configured to form a reflection beam based on the selected first and second optimal RCMs to direct one or more signals from an access point to a target UE.

[0028] In some embodiments, the RIS controller may be configured to group one or more reflection elements and one or more sensing elements in an array to form a plurality of non-uniform sub-arrays, and create an operation schedule for the plurality of non-uniform sub-arrays for serving one or more UEs in a wireless network.

[0029] In another aspect, the present disclosure relates to a method enabling autonomous beamforming in a wireless network having a RIS controller associated with a RIS panel that enables communication between an access point and one or more user equipment (UE). The method includes the RIS controller detecting a target UE present in the vicinity of the RIS panel based on one or more signals received from the target UE, the RIS controller determining the location of the target UE to identify the relative position of the target UE with respect to one or more UEs, and the RIS controller selecting a first optimal reflection coefficient matrix (RCM) associated with the RIS panel to enable beamforming for the target UE.

[0030] In some embodiments, the method may include the RIS controller detecting at least one of the presence of the target UE and the movement associated with the target UE via one or more sensing elements.

[0031] In some embodiments, the method may include the RIS controller receiving one or more UL transmissions related to the target UE from one or more sensing elements, and the RIS controller estimating the AoA related to the target UE based on the received one or more UL transmissions.

[0032] In some embodiments, the method may include the RIS controller selecting a second optimal RCM based on the detected movement related to the target UE.

[0033] In some embodiments, the method may include the RIS controller grouping one or more reflection elements and one or more sensing elements in the array to form a plurality of non-uniform sub-arrays, and the RIS controller creating an operation schedule for the plurality of non-uniform sub-arrays for serving one or more UEs in the wireless network.

[0034] In some embodiments, the method may include the RIS controller selecting a first and a second optimal RCM from an RCM look-up table obtained based on training a neural network for different RCMs related to different UE positions.

[0035] In some embodiments, the method may include the RIS controller forming a reflection beam based on at least one of the selected first and second optimal RCMs to direct one or more signals from the access point to the target UE.

[0036] In another aspect, the present disclosure relates to a UE having one or more processors and a memory operatively coupled to the one or more processors, the memory including processor-executable instructions that, when executed, cause the one or more processors to transmit one or more UL signals to a RIS controller to provide the position of the UE, and receive signals from an access point via one or more reflection beams formed by the RIS controller based on a selected optimal reflection coefficient matrix (RCM).

[0037] In other aspects, the present disclosure stores, in a non-transitory computer-readable medium, one or more instructions that, when executed by a processor, cause the processor to detect a target UE present in the vicinity of the RIS panel based on one or more signals received from the target UE, to localize the target UE to determine the relative position of the target UE with respect to one or more UEs present in a wireless communication network, and to select an optimal reflection coefficient matrix (RCM) associated with the RIS panel to enable beamforming for the target UE.

Brief Description of the Drawings

[0038] The accompanying drawings, which are incorporated herein and constitute a part of this disclosure, illustrate exemplary embodiments of the disclosed methods and systems with like reference numerals referring to the same parts throughout the different drawings. The components in the drawings are not necessarily to scale, and instead emphasis is placed on clearly illustrating the principles of the present disclosure. Some of the drawings may use block diagrams to show components and may not represent the internal circuitry of each component. Those skilled in the art will understand that the disclosure of such drawings includes the disclosure of the electrical components, electronic components, or circuits that are typically used to implement such components.

[0039]

Figure 1A

Figure 1B

Figure 1C

Figure 1D

Figure 2

Figure 3

Figure 4

Figure 5A

Figure 5B

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10A

Figure 10B

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

[0040] The above will become clearer from the following more detailed description of the present disclosure.

DETAILED DESCRIPTION OF THE INVENTION

[0041] In the following description, for the purpose of explanation, various specific details are set forth in order to provide a thorough understanding of embodiments of the present disclosure. However, it will be apparent that embodiments of the present disclosure may be practiced without these specific details. Some of the features described below may be used independently of each other or in any combination with other features. Each individual feature may not address all of the above-described problems, and may only address some of the above-described problems. Some of the above-described problems may not be fully addressed by any of the features described herein.

[0042] The following description provides only exemplary embodiments and is not intended to limit the scope, applicability, or configuration of the present disclosure. Rather, the following description of the exemplary embodiments provides those skilled in the art with an explanation that enables the practice of the exemplary embodiments. Various changes may be made to the functions and configurations of the elements without departing from the spirit and scope of the present disclosure below.

[0043] Specific details are provided in the following description in order to provide a thorough understanding of the embodiments. However, it will be understood by those skilled in the art that the embodiments may be practiced without these specific details. For example, circuits, systems, networks, processes, and other components may be shown as components in block diagram form so as not to obscure the embodiments with unnecessary detail. In other instances, well-known circuits, processes, algorithms, structures, and techniques may be shown without unnecessary detail in order to avoid obscuring the embodiments.

[0044] Note also that each individual embodiment may be described as a process shown as a flowchart, a flow diagram, a data flow diagram, a structure diagram, or a block diagram. A flowchart may describe the process as a sequential process, but many of the processes may be executed in parallel or simultaneously. Further, the order of the processes may be rearranged. A process ends when the process is completed, but may have additional steps not included in the figure. A process may correspond to a method, a function, a procedure, a subroutine, a subprogram, etc. When a process corresponds to a function, its termination may correspond to the return of the function to the calling function or the main function.

[0045] The terms “exemplary” and / or “illustrative” are used herein to mean serving as a specific example, instance, or illustration. To avoid misunderstanding, the subject matter disclosed herein is not limited by such specific examples. Further, any aspect or design described herein as “exemplary” and / or “illustrative” should not necessarily be construed as more preferable or advantageous than other aspects or designs, and is not intended to exclude equivalent exemplary structures and techniques known to those skilled in the art. Further, when the terms “comprising,” “having,” and other similar terms are used in any of the detailed description or claims, such terms are intended to be inclusive in the same manner as the term “comprising” as an open transitional word without excluding additional or other elements.

[0046] References throughout this specification to “one embodiment,” “an embodiment,” “an instance,” or “an example” mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Thus, the appearances of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment. Further, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0047] The terms used herein are for the purpose of describing particular embodiments only and are not intended to limit the present disclosure. As used herein, the singular forms "a", "an" and "the" may be intended to include the plural forms as well, unless the context clearly indicates otherwise. The terms "comprises" and / or "comprising", when used herein, specify the presence of the described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0048] Throughout the disclosure, certain items and phrases are used and have the following meanings with respect to the following disclosure.

[0049] The term "RIS" may refer to a reconfigurable intelligent surface, an intelligent reflecting surface (IRS), or a smart reflecting surface.

[0050] The term "autonomous" may refer to the stand-alone mode of operation of the RIS.

[0051] The term "autonomous beamforming" may refer to beamforming in the RIS in stand-alone mode without assistance from an access point.

[0052] Various embodiments throughout the present disclosure are described in more detail with reference to FIGS. 1-19.

[0053] FIGS. 1A-1D show various exemplary use case scenarios (102, 104, 105, 108 respectively) related to the implementation of a reconfigurable intelligent surface (RIS) in a wireless communication network according to some embodiments of the present disclosure.

[0054] RIS is implemented in various scenarios for its many associated advantages. One of the main advantages is that RIS elements are completely passive and thus have low power consumption, are environmentally friendly, and are sustainable and green. Furthermore, it does not include high-cost components such as analog-to-digital converters / digital-to-analog converters (ADC / DAC) and power amplifiers that enable large-area deployment. Moreover, the electromagnetic wave can be reconfigured at any point on its continuous surface and thus forms any shape to fit various application scenarios and support higher spatial resolution.

[0055] Furthermore, RIS intelligently controls the propagation environment, improves transmission reliability, and achieves higher spectral efficiency. RIS is applicable to the following typical scenarios: (i) overcoming non-line-of-sight (NLOS) limitations and addressing coverage hole problems in an environmentally friendly way, (ii) benefiting cell-edge users, mitigating multi-cell co-channel interference, expanding coverage, and enabling dynamic mobile user tracking, (iii) reducing electromagnetic pollution and solving the multipath problem, (iv) positioning, sensing, holographic communication, and virtual reality.

[0056] Based on the above-mentioned advantages, RIS may be deployed in one or more scenarios as shown in FIGS. 1A-1D.

[0057] Referring to FIG. 1A, an exemplary implementation scenario for RIS includes a first scenario (102) showing RIS-assisted communication between one or more UEs and an unmanned aerial vehicle (UAV). In FIG. 1A, RIS is deployed on the ground and is, for example, placed in a building to assist communication between the UAV and the UE. In some embodiments, it may be attached to the UAV to utilize smart passive reflection from above.

[0058] Figure 1B shows a second scenario (104) of RIS-assisted millimeter-wave (mmWave) communication. Generally, mmWave communication has a very short transmission range, and the use of RIS in such communication extends the transmission range. In Figure 1B, a RIS panel is shown that enhances communication between an access point and a user device.

[0059] Figure 1C shows a third scenario (106) of the use of RIS in the simultaneous wireless information and power transfer (SWIPT) operation mode. Figure 1C shows far-field power transfer using RIS. The RIS-assisted SWIPT system enables wireless transmission from an access point (AP) to a multi-antenna receiver including an information receiver (IR) and an energy receiver (ER).

[0060] Figure 1D shows a fourth scenario (108) of RIS-assisted device-to-device (D2D) communication. Figure 1D shows D2D communication using RIS. RIS improves the received signal power at remote D2D users, thereby improving the quality of service (QoS) in D2D communication.

[0061] Figure 2 shows an exemplary channel impulse response (200) of a RIS-assisted wireless communication network according to some embodiments of the present disclosure.

[0062] Figure 2 shows the end-to-end impulse response associated with the RIS system. Generally, for a conventional channel model, the impulse response is given with respect to a n,pb (t), b n,pb (t), h n,pb (t), and for a channel model associated with the RIS system, it includes the RIS control variables θ1, ···, θ n,pb in v N and represents the phase shifts provided by the RIS.

[0063] Figure 3 shows an exemplary system (300) of autonomous beamforming according to some embodiments of the present disclosure.

[0064] FIG. 3 shows a RIS panel (310), an access point (320), and a RIS controller (330). The RIS panel (310) includes a front plane (302), a back plane (304), and a control circuit board (306) connected to the RIS controller (330) via a physical channel (312). Further, the RIS controller (330) is connected to the access point (320) via a virtual control channel (308). The access point (320) may include any node capable of providing communication services to a user equipment (UE) (not shown in FIG. 3), such as, without limitation, a wireless fidelity (Wi-Fi) access point, a base station, an evolved Node B (eNodeB), a fifth generation Node (gNodeB), etc. Further, the virtual control channel (308) between the RIS controller (330) and the access point (320) may include an over-the-air (OTA) channel such as, without limitation, an integrated access and backhaul (IAB) or a radio frequency (RF) channel.

[0065] In a communication network using a RIS panel (310), the RIS panel (310) can reflect uplink (UL) communication from a UE to an access point (320), while on the other hand, reflect downlink (DL) communication from the access point (320) to the UE. To enable signal reflection, the position associated with the RIS panel (310) can be controlled. For example, the virtual tilt associated with the RIS panel (310) is controlled to achieve maximum reflection of the signal. In existing systems, the virtual tilt (beam reflection angle) of the RIS panel (310) is controlled by the access point (320). There are two ways to control the tilt of the RIS panel by the access point (320). In the first way, the access point (320) collects a set of signal-to-interference-plus-noise ratio (SINR) profiles via extended user measurements at the access point (320), and then generates an RF signature profile for the area in the vicinity of the RIS. In the second way, a scheduler at the access point (320) calculates the virtual tilt (or the associated beam reflection angle) required to achieve the SINR target. Therefore, in existing systems, the RIS panel (310) may detect the UL signal from the UE and send that information to the access point (320) such that the access point (320) calculates tilt control information over a given time period and conveys it to the RIS panel (310) via the virtual control channel (308). The tilt control information includes parameters related to the reflection characteristics of the RIS panel, such as the reflection coefficient matrix (RCM). In existing systems, the RCM is calculated by the access point (320), and the optimal RCM is sent to the RIS controller (330) to control the tilt of the RIS panel (310). In other words, the tilt of the RIS panel (310) depends on the decision from the access point (320).

[0066] According to the present disclosure, the optimal RCM is calculated in the RIS controller (330) to enable autonomous beamforming in the RIS panel (310) in the stand-alone mode, i.e., without relying on the access point (320). Further, in order to calculate the optimal RCM, the exact position of the UE may be known so that the RIS controller (330) can provide the optimal tilt of the RIS panel (310). According to some embodiments, the RIS controller (330) can perform RIS beamforming based on UE localization. For example, in some embodiments, without limitation, the RIS controller (330) may detect the relative position of the active terminal under its coverage and select the first optimal beam RCM to autonomously optimize the received SINR (or equivalent signal quality parameter) in UL and DL. In an embodiment, the relative position of the UE may be detected by the RIS controller (330) based on estimating the angle of arrival (AoA) of the UL signal from the UE in the RIS panel (310). In real time, the active UE may continue to move under the RIS coverage area, which leads to the need to change the AoA and the optimal RCM. In some embodiments, the RIS controller (330) may autonomously update the optimal beam RCM, i.e., select a second optimal RCM based on detecting the movement related to the active terminal. In an embodiment, the RIS controller (330) includes an RCM codebook, and the RCM codebook includes the RCM at a specific UE position and is obtained by training a neural network at different positions of the UE. For example, the RIS controller (330) may select the first and second optimal RCMs from the RCM codebook.

[0067] FIG. 4 shows beamforming (400) in the RIS according to some embodiments of the present disclosure.

[0068] FIG. 4 shows a beam (402) formed in the RIS panel (310). Accordingly, the RIS system (300) may be configured to generate a reflected beam to direct a DL signal from the access point (320) towards the UE or to direct a UL signal from the UE towards the access point (320).

[0069] FIG. 5A shows an exemplary RIS metasurface (500-A) having reflecting elements and sensing elements according to some embodiments of the present disclosure.

[0070] In FIG. 5A, a RIS metasurface (500-A) having reflecting elements (502) and sensing elements (504) arranged in L rows and C columns is shown. The sensing elements (504) can detect the presence of active UEs in the vicinity of the RIS panel (310). For example, an active UE may be present in the line of sight (LoS) of the RIS panel or within a distance where reflections from the RIS panel can be received without loss. In some embodiments, the sensing elements (504) detect signals from active UEs based on the AoA of UL pilot signals from the UEs. The information from the sensing elements (504) can be utilized by the RIS controller (330) to determine the relative position of the UEs, i.e., to perform localization or positioning, and thereby to select an appropriate RCM for beamforming.

[0071] FIG. 5B shows an exemplary distribution (500-B) of sensing elements on a RIS metasurface according to some embodiments of the present disclosure.

[0072] In FIG. 5B, the distribution of sensing elements (504) on the RIS metasurface is shown. The sensing elements (504) are distributed as row A and column B. When combined, the activated sensing elements (504) can form an RF sensing subarray as shown in FIG. 8. This active subarray formation can provide the RIS controller (330) of FIG. 3 with the ability to estimate the UL arrival angle. In some embodiments, the sensor array may be one-dimensional (all sensors in a single line), or may measure the UL AoA along the horizontal axis. In some other embodiments, the sensor array may be two-dimensional and may measure the UL arrival direction or AoA along both the horizontal and vertical axes.

[0073] In embodiments, sensing of active terminals in the RIS coverage area is achieved by utilizing periodically shifted reference symbols or pilots. These pilots or reference symbols are used in the uplink to assist channel sounding by the access point (320) of FIG. 3. These pilot symbols may correspond, without limitation, to a family of pseudo-random sequences having very good correlation properties, such as zero-forcing (ZF) sequences, Gold sequences, etc. The RIS controller 330 of FIG. 3 may utilize a mechanism to correlate the shifted reference symbol or pilot symbol with the signal received from the active UE in the RIS coverage area.

[0074] FIG. 6 shows a possible pilot arrangement (600) in the UL signal from a user equipment according to some embodiments of the present disclosure.

[0075] FIG. 6 shows an exemplary arrangement (600) of UL pilot elements or resource elements (602). For example, without limitation, the arrangement (600) can include sounding reference signal (SRS) elements on a given 5G NR (New Radio) system resource block (RB). In some embodiments, when the UE transmits a pilot shown as s(t) ∈ C that has a single number of elements "m" to obtain a certain power level, the received reference signal for the element "m" can be given as follows. [Number] However, s(t - τ l ) is the version of s(t) that is delayed by the delay of τ l . Further, n m (t) is the additive white Gaussian noise (AWGN) at the receiver.

[0076] In some embodiments, the UE transmits a resource unit s(t), and the exact configuration of the resource unit is known at the RIS controller (330). Since the cross-correlation between different periodically shifted sequences from the same root sequence is zero, the RIS controller (330) can estimate the channel h(t) by correlating y(t) and s(t). [Number]

[0077] In some embodiments, to detect the presence of an active UE, only a subset (even a single element) of the RIS detection elements (504) in FIG. 5B needs to be activated. This makes the proposed mechanism very efficient.

[0078] In some embodiments, the RIS controller (330) of FIG. 3 can detect active UEs within its coverage area without knowing the exact cyclic shift applied to the pilot sequence used by the active UEs. This can be achieved by correlating the concatenated pilot reference signal and the received signal in a moving window fashion. In such a case, the RIS controller (330) of FIG. 3 can search for all possible shifts of the pilot sequence and compare it with a threshold. If the correlation of the concatenated pilot sequence exceeds the threshold T at index i th , the RIS controller (330) can declare that an active UE is under the RIS coverage area.

[0079] FIG. 7 shows an exemplary real-world setting (700) for the training and deployment of a system for autonomous beamforming according to some embodiments of the present disclosure.

[0080] FIG. 7 shows a training setup (700) comprising a RIS panel (710), an access point (720), a RIS controller (730), and one or more test UEs (740-1, 740-2, 740-3, ···, 740-n). Further, the RIS panel (710) is connected to the RIS controller (730) by a physical channel (704). The access point (720) is connected to the RIS controller (730) by a virtual control channel (702). In some embodiments, the training setup (700) enables training a neural network associated with the RIS controller (730) to record a set of RCMs associated with different UE positions to form an RCM codebook. The training includes three phases. The first phase includes detecting the presence of active UEs within the RIS coverage area based on one or more pilot sequences received from the active UEs, the second phase includes localizing the detected UEs based on estimating the angle of arrival of uplink transmissions from the active UEs, and the third phase includes selecting an appropriate RCM from the codebook for reflection beamforming by the RIS.

[0081] According to some embodiments, for training the neural network, test UEs (740) are placed at different positions (740-1···740-n) in the vicinity of the RIS coverage area, and signals from the test UEs (740) at each position are transmitted to the access point (720) to determine the RCM. Each of the obtained various RCMs is stored in the RCM codebook and can be used by the RIS controller (730) during beamforming operations.

[0082] In an exemplary embodiment, to train a neural network, the sounding reference signal (SRS) of a 5G NR system is considered, and the length of the SRS sequence depends on the number of time-frequency physical resource blocks (PRBs) used in transmission. A PRB is the smallest unit of a resource block that can be allocated to a user equipment (UE). Each PRB has six SRS resource elements (REs). An RE corresponds to one time-frequency point. Thus, the received signal includes the REs of each PRB used. The number of PRBs depends on the configuration of the test UE (740). Further, the cyclic shift related to the SRS can be varied from 1 to 8 to generate up to eight different SRSs that are orthogonal to each other. The access point (720) can set the SRS for up to eight UEs within the same subframe and frequency resource. However, since different cyclic shifts are used, the cyclic shift multiplexed signals need to have the same bandwidth to maintain orthogonality. The number of SRSs transmitted and the cyclic prefix allocated to the UE can be variable, and it should be understood that the steps applied to all possible combinations of the cyclic shift numbers used in the radio technology can be variable.

[0083] In the embodiment, the positioning settings and mechanisms for detecting the UL arrival direction are shown with reference to FIG. 8.

[0084] FIG. 8 shows an exemplary representation (800) of the angle of arrival (AoA) related to the UE UL signal at the sensing element on the RIS metasurface according to some embodiments of the present disclosure.

[0085] The location-specific mechanism helps to identify the relative position of the detected active UE (802). Peaks in the observed channel estimates may be extracted and compiled in a detection matrix of A×B (=M) elements, and the angle of arrival of the signal from the UE (802) to the RIS can be used for location identification. For location identification, all RIS detection elements (804-1 to 804-M) are activated, and the array may be rotated in one direction at a time to measure the output power level. The rotation is performed by weighting each array response and then linearly combining them. The output of one sample is

Number

[0086] The weight vector w is equal to the scanning vector a(θ B ), and the assumed angle θ B is scanned over the angular region. The steering vector of the detection array element is defined as follows for the scanning angle θ B .

Number

Number

[0087] The RIS receiver may have a set of scanning vectors in the form of matrix A corresponding to the possible range from the angle of arrival θ1 to θ p of the UL signal.

[0088] For each assumed angle, the output power is

Number

[0089] For the assumed angle θ BWhen it is the same as the actual angle of the signal, P(w) has a peak in the spectrum.

[0090] For practical calculations, the weight vector is

Equation

[0091] When the UE is detected and located in the RIS coverage area, the beam training phase is performed. RIS beam training is the phase where, although the UE position is known, it is necessary to determine which beamforming matrix or codebook or RCM to use so that the beam from the access point is properly reflected towards the located UE.

[0092] FIG. 9 shows an implementation of an exemplary time-division duplex (TDD) scheme (900) between a UE and a RIS according to some embodiments of the present disclosure. In FIG. 9, TDD communication between an access point (904) and a UE (902) is shown. Generally, TDD communication utilizes different time slots for uplink and downlink transmissions. The TDD duplex scheme brings several important advantages and flexibilities to the communication system. One advantage is channel reciprocity. Channel reciprocity means that the channel characteristics are the same for the uplink and the downlink.

[0093] Therefore, by estimating the channel in the uplink direction, assuming that the channel does not change in the estimation interval, the downlink direction is also estimated. As a result, reciprocity leads to better transmission parameter optimization for resource allocation.

[0094] FIG. 10A shows an implementation of an exemplary frequency-division duplex (FDD) scheme (1000-A) between a UE and an access point according to some embodiments of the present disclosure.

[0095] In Figure 10A, FDD communication between a UE (1002) and an access point (1004) is shown. FDD multiplexing operates UL and DL in separate frequency spectra, and thus, channel reciprocity (such as applicable to TDD UL-DL) is no longer effective. Therefore, different reflection coefficient matrices are required to reflect the UL and DL links of the same UE. According to some embodiments, to enable RCM allocation in FDD communication, a given RIS metasurface is virtually divided into two sections, as will be described later with reference to Figure 10B.

[0096] Figure 10B shows an exemplary distribution (1000-B) of sensing elements on a RIS metasurface utilized by an FDD scheme according to some embodiments of the present disclosure.

[0097] Figure 10B shows a virtually divided RIS metasurface including a first portion (1006) including N / 2 sensing elements and a second portion (1008) including N / 2 sensing elements. FDD utilizes different frequencies for UL and DL, and thus, two different virtual regions or portions (1006, 1008) provide FDD. In an embodiment, the first portion (1006) including N / 2 elements may be utilized for UL communication from the UE (1002) to the access point (1004), and the second portion (1008) including N / 2 elements may be utilized for DL communication from the access point (1004) to the UE (1002).

[0098] Figure 11 shows an exemplary system configuration (1100) for autonomous beamforming according to some embodiments of the present disclosure.

[0099] FIG. 11 shows two main stages, namely, a training stage related to autonomous beamforming in the RIS controller (1130) and an RCM selection stage. The training stage includes stage 1 (1102), which is the second stage for detecting the UE position, and stage 2 (1104), which is the second stage for beam switching through the interaction between the RIS controller (1130) and the access point (1120). When training the neural network associated with the RIS controller (1130), an RCM codebook is generated to enable autonomous beamforming in the RIS panel (1110). The selection stage (1106) selects the RCM from the RCM codebook during the actual implementation.

[0100] Referring to FIG. 11, stage 1 (1102) forms part of the training during the initial RIS deployment. The training stage may include the following steps. 1. The test UE (1140) transmits a constant pilot signal on the uplink during the training period. 2. The RIS controller (1130) detects the user position with respect to the angle of arrival A(φ1) of the signal at the RIS (not in user coordinates [X,Y]) and stores it locally. 3. Then, the RIS controller (1130) starts the test codeword selection for the sequence of beam reflection coefficients in the UL. 4. For each iteration of the UL beam codeword selection, the RIS controller (1130) determines the SINR of each test UE (1140) observed at the access point (1120). 5. At the end of the search for possible codewords in the RIS codebook, the RIS controller (1130) marks the UE position for the beam codeword corresponding to the highest UL SINR observed at the access point (1120) and creates one entry in the codebook lookup table. The training process is repeated by placing the test UE at various positions in the given RIS coverage area. The second stage (1104) corresponds to a live network where the RIS controller (1130) assists the UE (1140) for communication with the access point (1120). The reflection coefficient selected by the RIS controller (1130) is directly mapped to the relative UE position, and the UE position is obtained based on the direction of arrival of the UL signal from the UE detected by the detection element in the RIS panel (1110).

[0101] Any change in the UE position may be detected by the detection element in the RIS panel (1110) as a change in the direction of arrival (DoA) from the UE (1140).

[0102] In some embodiments, when the UE (1140) and the access point (1120) utilize TDD communication, the RIS controller (1130) may estimate the accurate UE position (AoA) with the assistance of the UL transmission detection mechanism (the detection element in the RIS panel (1110)). Then, the RIS controller (1130) may perform an optimal lookup of the beamforming codeword from the training codebook table, i.e., the RCM codebook obtained from stage 1 (1102), using this UE position identification information. The RIS controller (1130) may perform beamforming reflection of the signal of the access point (1120) for the target UE (1140) using the selected beamforming codeword described above. When the UE (1140) moves, the RIS detection element updates the UE position for the RIS controller (1130). The RIS controller (1130) switches the beamforming codeword based on the search for the updated UE position from the training codebook table.

[0103] In some embodiments, beamforming in the RIS panel (1110) is based on the angle of arrival of the UE, where the angle of arrival continues to change as the UE moves, making the beamforming process dynamic. That is, the term "dynamic beamforming" may refer to beamforming in the RIS controller (1130) based on the position of the UE and the number of sub - panels in the RIS panel (1110).

[0104] In some embodiments, the RIS controller (1130) may utilize the disclosed mechanism for the RIS to arrive at the exact UE position and perform RIS fingerprinting, detection, and tracking of the UE in the RIS coverage area that assists the access point (1120) with this information.

[0105] In some other embodiments, when the UE (1140) and the access point (1120) communicate using an FDD system, the RIS controller (1130) may include an additional training phase where the access point (1120) transmits in the DL and the test UE (1140) performs channel estimation for various DL reflection codewords at a given UE position. Since there is no direct interface between the test UE (1140) and the RIS controller (1130), the SINR report for each UL beamforming codeword is performed via UE - access point feedback and then from the access point (1120) to the RIS controller (1130).

[0106] In some embodiments, a deep neural network may be utilized to select the optimal RCM in the RIS controller (1130). The proposed deep neural network for the selection of the exact RCM is a multi - layer perceptron (MLP) network, as will be described later with reference to FIG. 12. The function of the DNN may be part of the RIS controller (1130) or on the cloud that is easily accessible to the RIS controller (1130) using a backhaul mechanism.

[0107] It may be noted that the MLP is suitable for classification problems where the output of the network is discrete or categorical and the input data is labeled. The design of the DNN depends on the corresponding problem to be solved. The first step is to select the correct network type, the number of hidden layers, and the number of nodes in each layer. Further, the activation function and the connections between the nodes can be defined. These variables are called hyperparameters that determine the structure of the network. Once the hyperparameters are determined, the network model needs to be trained and tested. Training means that the weights and biases of the activation function are adjusted to receive accurate estimates. Before the network can be trained, the weights and biases are initialized. This is required for the first iteration of training. The training data includes the correct target, i.e., the desired value for the response related to the input. These targets can be compared against the estimates of the output given by the network using a specific metric. One common metric is the loss function (also called the cost function). The loss function indicates how well the estimate compares to the target. Thus, the smaller the output of the loss function, the better the model for the problem. Training can assist in minimizing the value of the loss function through multiple iterations. In each iteration, an example from the training data is input into the input layer of the network. After that, the weights and biases are adjusted so that the value of the loss function decreases. There are multiple different methods for detecting the optimal weights and biases that minimize the loss function.

[0108] Figure 12 shows a DNN architecture (1200) implemented for training RIS beamforming according to some embodiments of the present disclosure.

[0109] In FIG. 12, a DNN including an input layer (1202), an output layer (1206), and at least two hidden layers (1204) is shown. Generally, the number of hidden layers (1204) is selected to balance between accuracy and computational complexity. In some embodiments, the input layer (1202) includes M nodes related to the size of the RIS sensor array (input data). The hidden layer (1204) includes a plurality of nodes more than L. Here, L is the number of RCMs supported by a given RIS panel (e.g., 1110), and since each channel estimation corresponds to a selection among multi-class settings, the output layer (1206) includes L nodes. When there are more layers in a neural network, the increase in accuracy can be significant. Also, the network can become more complex due to the increase in arithmetic operations. On the other hand, when there are fewer layers, the accuracy can be significantly reduced. In some embodiments, to start from the training process of the DNN, the weight matrix is initialized using some initialization strategies and updated in each epoch according to the update formula as given below to reach the most accurate result.

Number

[0110] The following Table 1 and Table 2 show various parameters used in the DNN model according to some embodiments of the present disclosure. Table 1 provides the DNN parameters in the training stage, and Table 2 provides the hyperparameters related to the DNN.

Number

[0111] The following assumptions can be used to train the DNN. 1. The RIS controller (1130) may support a set of M-beam RCMs and the form factor of a given RIS panel (1110) depending on physical characteristics. Here, this set of RCMs forms a pre-programmed beamforming codebook of a given RIS system. 2. The preliminary stage of RIS training has already been executed with the assistance of a test UE to construct a beamforming codebook by the RIS controller (1130).

[0112] Therefore, the use of DNN can assist the RIS controller (1130) to select the optimal beamforming codeword in a stand-alone mode without the assistance from the access point (1120). This minimizes the delay, complexity, and efficiency of the overall process of reflective beamforming. The use of DNN for RCM selection provides one or more advantages including avoiding calculating the optimal RCM in all UE interactions, reducing synchronization with the access point (1120) for all UE reflections, saving the RIS controller from explicitly identifying UE resource blocks (RBs), autonomously improving the RIS function in terms of the proposed delay and channel quality in the wireless coverage area, and designing the DNN such that the selection / calculation of the reflection coefficient for non-tested DoA values is also optimally executed.

[0113] FIG. 13 shows an exemplary flowchart (1300) related to the initialization of the RIS training stage according to some embodiments of the present disclosure. In FIG. 13, one or more steps related to initializing a training stage or process for autonomous beamforming in the RIS controller (1330) are shown.

[0114] In step 1302, the RIS controller (1330) may perform discovery and registration to the access point (1320). Further, in step 1304, the test UE (1340) starts synchronization with the access point (1320). Further, when the test UE (1340) and the access point (1320) are synchronized, in step 1306, the test UE (1340) may send a RIS training request to the access point (1320). The RIS training request includes a RIS identifier (ID). Further, in step 1308, the AP (1320) may send a training start message to the RIS controller (1330). The training start message may include, without limitation, pilot information, timing synchronization, and uplink information. Upon receiving the training start message, in step 1312, the RIS controller (1330) may initialize the weights associated with the neural network (NN) (1350). In step 1314, the RIS controller (1330) may activate the RIS sensing elements in the RIS array (1310). The activation message may include pilot information, timing synchronization, and uplink information. Upon receiving the activation message, in step 1316, the RIS array (1310) may send a sensor ready signal to the RIS controller (1330). Further, in step 1318, the NN (1350) may send an NN ready message to the RIS controller (1330). Upon receiving the sensor ready signal and the NN ready message, in step 1322, the RIS controller (1330) may send a RIS ready message to the access point (1320), where the access point (1320) forwards a message indicating the start of training to the test UE (1340). Table 3 below specifies one or more parameters used for the initialization of the training phase.

Number

[0115] After initialization, the training may be performed based on the type of communication used, such as TDD or FDD, as described in detail below with reference to FIGS. 14 and 15.

[0116] FIG. 14 shows an exemplary flowchart (1400) related to RIS training for autonomous beamforming in a TDD system according to some embodiments of the present disclosure.

[0117] In FIG. 14, steps for training a RIS NN (1450) in a TDD system are shown. In step 1404, the RIS controller (1430) may register with the access point (1420). By registering the RIS controller (1430), in step 1406, the training process starts a loop to obtain an RCM codebook based on different positions of the test UE (1440). In step 1408, the test UE (1440) transmits a reference signal from a first position (position#1) within the RIS coverage area to the access point (1420), and the signal may be further transmitted to the RIS controller (1430). In step 1412, the RIS controller (1430) may activate a detection element (1402) in the RIS panel (1410) to detect the test UE (1440). In step 1414, the detection element (1420) may determine whether a UE pilot is detected. If the UE pilot is not detected, in step 1458, the detection element (1402) may send a UE detection failure message to the RIS controller (1430). In step 1462, the RIS controller (1430) sends a training next message to the access point (1420). In step 1462, the access point (1420) may further forward a RIS training message with a RIS ID to the test UE (1440).

[0118] On the other hand, in step 1416, the detection element (1402) may further notify the RIS controller (1430) whether the test UE (1440) has been detected. In step 1418, when detecting the test UE (1440), the RIS controller (1430) may activate channel estimation. In step 1422, the RIS detection element (1402) may perform channel estimation on the strongest path of the UE signal detected at each sensor on the RIS sensor array (1410) and send the estimation result to the RIS controller (1430). The channel estimation is

Number

[0119] Referring to FIG. 14, upon completion of the RCM iteration, at step 1438, the RIS controller (1430) may train the RIS NN (1450) by taking as input M normalized channel estimates and providing a selected set of reflection coefficients as the labeled result for the iteration. In some embodiments, the access point (1420) may update the test UE (1440) for which the training iteration has now ended, and the test UE (1440) may move to different positions within the RIS coverage area. Further, the test UE (1440) may repeat the above steps to determine the next RCM for the new position.

[0120] When the training is executed, at step 1442, the RIS NN (1450) may transmit training execution information together with the associated cost function. The RIS NN logic continues to calculate the cost function as the training progresses. At step 1444, if the NN cost function reaches a predetermined threshold and the cost function approaches the predetermined threshold, the RIS controller (1430) may update the access point (1420) with a training completion message (1446). At step 1448, the access point (1420) notifies the test UE (1440) of the training completion status with the RIS ID. On the other hand, if the NN cost function does not approach the predetermined threshold, at step 1454, the RIS controller (1430) transmits a training next message to the access point (1420), and at step 1456, the access point (1420) forwards the message to the test UE (1440) together with the RID ID.

[0121] Various parameters regarding the training of the RIS NN (1450) are shown in Table 4 below.

Number

[0122] FIG. 15 shows an exemplary flowchart (1500) related to RIS training for autonomous beamforming in an FDD system according to some embodiments of the present disclosure.

[0123] In FIG. 15, steps for training UL RIS NN (1550) and DL RIS NN (1560) in an FDD system are shown. The training of UL RIS NN (1550) and DL RIS NN (1560) relates to a similar set of steps for training RIS NN (1450) as described above with reference to FIG. 14. The difference between a TDD system and an FDD system is that the FDD system operates on UL and DL at different frequencies, and thus, channel reciprocity is disabled. Therefore, different reflection coefficient matrices are required to reflect the UL and DL links of the same UE. Therefore, to enable the training of the FDD system, the RIS reflection array (1510) is virtually divided into two sections, one section providing the uplink and the other section providing the downlink, each having N / 2 reflection elements and M / 2 sensing elements. All the training steps of the TDD system described above with reference to FIG. 14 are applicable to the training of the FDD system. As will be understood by those skilled in the art, for simplicity, the description is not repeated here. Further, in the FDD system, the second training stage can be realized by the access point (1520) transmitting on the DL and the test UE (1540) performing channel estimation for various DL reflection codewords at a given UE position.

[0124] Referring to FIG. 15, UL RIS NN (1550) may include an RCM codebook related to reflection beamforming on the RIS surface (1510) for uplink communication (from UE (1540) to access point (1520)), and DL RIS NN (1560) may include an RCM codebook related to reflection beamforming on the RIS surface (1510) for downlink communication (from access point (1520) to UE (1540)).

[0125] FIG. 16 shows an exemplary flowchart (1600) related to autonomous beamforming in a RIS in an actual implementation according to some embodiments of the present disclosure.

[0126] FIG. 16 shows a signal flow related to an actual deployment scenario of an autonomous beamforming system. The actual deployment scenario includes one or more of the following considerations. 1. The reflection coefficient selected by the RIS controller is directly mapped to the relative UE position. 2. In an embodiment, the DoA is an indicator of the relative UE position. 3. A change in the UE position can be detected by the RIS sensor (1602) array as a change in the DoA from the UE.

[0127] In some embodiments, to enable a reliable estimation of the angle of arrival (DoA = [AoA Horizontal, AoA Vertical]) of the UL signal in the RIS array, the size of the steering matrix (M) used for the estimation of the direction of arrival should be long enough to provide good correlation characteristics. The steering matrix is

Number

Number

[0128] In some embodiments, when the RIS controller (1630) is trained, registered with the AP (1620), and provided in a ground deployment, the RIS controller (1630) continuously activates the RIS sensor array (1602) periodically to detect active UEs (1640) in its vicinity. When the RIS sensor array (1602) detects an active UE (1640) in its vicinity, the RIS controller (1630) is updated to start the RCM selection process. The RIS controller (1630) activates the complete sensor array (1602) to calculate accurate normalized UL channel estimates in the spatial domain. M normalized channel estimation results are provided as inputs to the pre-trained DNN (1650). The DNN (1650) generates an output for selecting an appropriate RCM for a given UE (1640).

[0129] Referring to FIG. 16, the actual deployment may include, at step 1604, completing the registration of the RIS controller (1630) to the access point (1620) and completing the training of the RIS NN (1650). At step 1606, the RIS controller (1630) may activate the RIS sensor array (1602) to detect active UEs (1640) in the RIS coverage area. At step 1608, the RIS sensor array (1602) may send a message detected by the UE along with a pilot carrier to the RIS controller (1630). At step 1612, the RIS controller (1630) may further activate channel estimation based on the cyclic shift of the pilot carrier. At step 1614, the RIS sensor array (1610) may further send the channel estimation result to the RIS controller (1630) according to the angle A of the direction of arrival (DoA). At step 1616, the RIS controller (1630) may further send the channel estimation normalized as an input to the RIS NN (1650). At step 1618, the RIS NN (1650) may send the selected RCM along with the confidence level. The RIS controller (1630) may check whether the confidence level is greater than a threshold. At step 1622, if the confidence level is greater than the threshold, the RIS controller (1630) may activate the RCM index of the RIS reflection array (1610). At step 1624, the reflection array (1610) may send a confirmation message to the RIS controller (1630) that the RCM has been activated. The RIS reflection array (1610) may utilize the scrubbed RCM to relay the beam from the active UE (1640) to the access point (1620).

[0130] Table 5 shows various algorithm parameters used for dynamic UE tracking and beam selection.

Number

[0131] Figure 17 shows an exemplary autonomous beamforming system (1700) in a RIS for supporting multiple UEs according to some embodiments of the present disclosure.

[0132] In Figure 17, a single RIS array 1710 is shown that supports multiple reflection beamformings (1704-1, ···, 1704-4) for multiple UEs (1740-1, ···, 1740-4). In some embodiments, the RIS metasurface may be virtually divided into non-uniform subarrays (A1-A14). Further, the reflection beamforming may be initiated in one or more non-uniform subarrays by a RIS controller to provide for each UE. For example, UE1 (1740-1) may be provided by reflection beamforming (1704-1) from the first subarray A1. Similarly, other UEs may be provided by reflection beamforming from other subarrays. According to some embodiments, time / frequency scheduling may be realized among the non-uniform subarrays such that one subarray forms a reflection beam at one particular time / frequency.

[0133] Figure 18 shows an exemplary flowchart of a method (1800) for enabling autonomous beamforming in a wireless network according to some embodiments of the present disclosure. In Figure 18, a method (1800) for enabling autonomous beamforming in a RIS controller is described. The method (1800) may include, at step 1802, detecting a target UE present in the vicinity of the RIS panel. The target UE may be within the line-of-sight range of the RIS panel or alternatively within the signal reception range. Further, the method (1800) may include, at step 1804, localizing the target UE to identify the relative position of the target UE with respect to one or more UEs. The method (1800) may further include, at step 1806, selecting a first optimal RCM associated with the RIS panel to enable beamforming for the target UE.

[0134] Those skilled in the art will understand that these are merely specific examples and will in no way limit the scope of the present disclosure.

[0135] FIG. 19 shows an exemplary computer system (1900) in which embodiments of the present disclosure may be utilized. As shown in FIG. 19, the computer system (1900) may include an external storage device (1910), a bus (1920), a main memory (1930), a read-only memory (1940), a mass storage device (1950), a communication port (1960), and a processor (1970). Those skilled in the art will understand that the computer system (1900) may include two or more processors and communication ports. The processor (1970) may include various modules related to embodiments of the present disclosure. The communication port (1960) may be any of an RS-232 port for use in a modem-based dial-up connection, a 10 / 100 Ethernet® port, a gigabit or 10 gigabit port using copper wire or fiber, a serial port, a parallel port, or any other existing or future port. The communication port (1960) may be selected according to the network, such as a local area network (LAN), a wide area network (WAN), or any network to which the computer system (1900) is connected. The main memory (1930) may be a random access memory (RAM) or any other dynamic storage device commonly known in the art. The read-only memory (1940) may be any static storage device including, without limitation, a programmable read-only memory (PROM) chip for storing static information such as startup or basic input / output system (BIOS) instructions for the processor (1970). The mass storage device (1950) may be any current or future mass storage means that may be used to store information and / or instructions.

[0136] The bus (1920) is communicatively coupled to the processor (1970) and other memory, storage, and communication blocks. The bus (1920) can be, for example, a PCI (Peripheral Component Interconnect) / PCI-X (PCI Extended) bus, SCSI (Small Computer System Interface), USB (Universal Serial Bus), etc. for connecting to other buses such as a front-side bus (FSB) that connects the processor (1970) to a computer system (1900) along with expansion cards, drives, and other subsystems.

[0137] Optionally, an operator and management interface, such as a display, keyboard, and cursor control device, can also be coupled to the bus (1920) to support direct operator interaction with the computer system (1900). Other operator and management interfaces can be provided via a network connection connected through the communication port (1960). The exemplary computer system (1900) described above is in no way intended to limit the scope of the present disclosure.

[0138] Although much emphasis has been placed on the preferred embodiments here, it will be understood that many embodiments can be made and many changes can be made to the preferred embodiments without departing from the principles of the present disclosure. These and other changes in the preferred embodiments of the present disclosure will be apparent to those skilled in the art from the present disclosure herein, whereby it should be clearly understood that the above description should be implemented merely as an exemplification of the present disclosure and not as a limitation.

[0139] Advantages of the Present Disclosure The present disclosure provides autonomous reflective beamforming in a RIS (Reconfigurable Intelligent Surface) that reduces the latency associated with existing reflective beamforming techniques.

[0140] The present disclosure provides a reflection code matrix (RCM) in RIS based on a deep neural network (DNN).

[0141] The present disclosure provides reduced computations associated with calculating an optimal reflection coefficient matrix for all user equipment (UE) interactions.

[0142] The present disclosure provides an improvement in dynamic channel quality in the RIS coverage area.

[0143] The present disclosure provides an advanced communication system.

[0144] The present disclosure enhances the user experience.

[0145] The present disclosure solves one or more network-related problems such as call drops and signal strength.

[0146] The present disclosure provides a joint sensing and communication system.

[0147] The present disclosure provides an advanced joint sensing and communication system that utilizes an IRS for autonomous beam management and tracking.

Claims

1. A system (1100) enabling autonomous beamforming in a wireless network, wherein the system (1100) has an RIS (Reconfigurable Intelligent Surface) controller (1130) associated with an RIS panel (1110) enabling communication between an access point (1120) and one or more user equipment (UE) (1140) in the wireless network, wherein the RIS controller (1130) detects a target UE present in the vicinity of the RIS panel (1110) based on one or more signals received from the target UE (1140-2), locates the target UE (1140-2) to identify the relative position of the UE (1140-2) with respect to the one or more UEs (1140), selects a first optimal reflection coefficient matrix (RCM) associated with the RIS panel (1110) to enable beamforming for the target UE (1140-2), and is configured as such, the system (1100).

2. The system (1100) according to claim 1, wherein the selected first optimal RCM enables optimal reflection of the beam from the access point (1120) for the target UE (1140-2).

3. The system (1100) according to claim 1, wherein the RIS panel (1110) has one or more reflection elements (502) and one or more sensing elements (504).

4. The one or more sensing elements (504) detect the presence of the target UE (1140-2), and detect the movement associated with the target UE (1140-2), and assist the RIS controller (1130) as such, the system (1100) according to claim 3.

5. The RIS controller (1130) receives one or more uplink (UL) transmissions related to the target UE (1140-2) from the one or more sensing elements (504), and estimates the angle of arrival (AoA) related to the target UE (1140-2) based on the received one or more UL transmissions, and is configured as such, the system (1100) according to claim 3.

6. The RIS controller (1130) selects a second optimal RCM based on the detected movement related to the target UE (1140-2). The system (1100) according to claim 4, configured as such.

7. The RIS controller (1130) is configured to select the first and second optimal RCMs from an RCM look-up table obtained based on training a neural network for different RCMs associated with different UE positions, for the system (1100) according to claim 6.

8. The RIS controller (1130) forms a reflection beam based on the selected first and second optimal RCMs to direct one or more signals from the access point (1120) to the target UE (1140-2). The system (1100) according to claim 6, configured as such.

9. The RIS controller (1130) groups the one or more reflection elements (502) and the one or more sensing elements (504) in the array to form a plurality of non-uniform sub-arrays. creates an operation schedule for the plurality of non-uniform sub-arrays for use in the wireless network for the one or more UEs (1140). The system (1100) according to claim 3, configured as such.

10. A method (1800) enabling autonomous beamforming in a wireless network having a RIS (Reconfigurable Intelligent Surface) controller (1130) associated with a RIS panel (1110) enabling communication between an access point (1120) and one or more user equipment (UEs) (1140), wherein the method in the wireless network the RIS controller (1130) detects a target UE present in the vicinity of the RIS panel (1110) based on one or more signals received from the target UE (1140-2) (1802); the RIS controller (1130) locates the target UE (1140-2) to identify the relative position of the UE (1140-2) with respect to the one or more UEs (1140) (1804); the RIS controller (1130) selects a first optimal reflection coefficient matrix (RCM) associated with the RIS panel (1110) to enable beamforming for the target UE (1140-2) (1806). The method (1800) comprising the above.

11. The method (1800) according to claim 10, wherein the selected first optimal RCM enables optimal reflection of a beam from the access point (1120) to the target UE (1140-2).

12. The method (1800) according to claim 10, wherein the RIS panel (1110) has an array of one or more reflection elements (502) and one or more sensing elements (504).

13. The method (1800) according to claim 12, wherein the RIS controller (1130) detects at least one of the presence of the target UE (1140-2) and the movement associated with the target UE (1140-2) via the one or more sensing elements (504).

14. The RIS controller (1130) receives one or more uplink (UL) transmissions related to the target UE (1140-2) from the one or more sensing elements (504); The RIS controller (1130) estimates the angle of arrival (AoA) related to the target UE (1140-2) based on the received one or more UL transmissions; The method (1800) according to claim 12, comprising the above steps.

15. The method (1800) according to claim 13, wherein the RIS controller (1130) selects a second optimal RCM based on the detected movement associated with the target UE (1140-2).

16. The RIS controller (1130) groups the one or more reflection elements (502) and the one or more sensing elements (504) in the array to form a plurality of non-uniform sub-arrays; The RIS controller (1130) creates an operation schedule for the plurality of non-uniform sub-arrays for serving the one or more UEs (1140) in the wireless network; The method (1800) according to claim 12, comprising the above steps.

17. The method (1800) according to claim 15, wherein the RIS controller (1130) selects the first and second optimal RCMs from an RCM look-up table obtained based on training a neural network for different RCMs related to different UE positions.

18. The method (1800) of claim 15, wherein the RIS controller (1130) forms a reflected beam based on at least one of the selected first and second optimal RCMs to direct one or more signals from the access point (1120) to the target UE (1140-2).

19. A user equipment (UE), one or more processors, a memory operatively coupled to the one or more processors, and having, the memory, when executed, transmits one or more uplink (UL) signals to a RIS (Reconfigurable Intelligent Surface) controller (1130) to provide the location of the UE, receives signals from an access point (1120) via one or more reflected beams formed by the RIS controller (1130) based on a selected optimal reflection coefficient matrix (RCM), A user equipment (UE) having processor-executable instructions for causing the one or more processors to perform the above.

20. When executed by a processor, detecting a target UE (1140-2) present in the vicinity of the RIS (Reconfigurable Intelligent Surface) panel (1110) based on one or more signals received from the target UE (1140-2), locating the target UE (1140-2) to determine the relative position of the target UE (1140-2) with respect to one or more UEs (1140) present in the wireless communication network, selecting an optimal reflection coefficient matrix (RCM) associated with the RIS panel (1110) to enable beamforming for the target UE (1140-2), A non-transitory computer-readable medium storing one or more instructions for causing the processor to perform the above.

Images