Hybrid based visibility region identification method for low-power devices in XL-MIMO systems

Abstract

The invention is related to a novel visibility region (visibility region) identification methodology for Ambient IoT devices in extremely large-scale multiple-input multiple-output (XL-MIMO) systems, addressing one of the key challenges in ambient IoT: massive connectivity.

Description

[0001] DESCRIPTION

[0002] HYBRID BASED VISIBILITY REGION IDENTIFICATION METHOD FOR LOW- POWER DEVICES IN XL-MIMO SYSTEMS

[0003] Technical Field

[0004] The invention is related to a novel visibility region (visibility region) identification methodology for Ambient loT devices in extremely large-scale multiple-input multiple-output (XL-MIMO) systems, addressing one of the key challenges in ambient loT: massive connectivity.

[0005] Prior Art

[0006] MIMO is a technology used in wireless communication systems that employs multiple antennas at both the transmitter and receiver ends. This configuration enables the transmission of multiple data streams simultaneously, improving data rates, spectral efficiency, and reliability by exploiting multipath propagation in the environment.

[0007] Ambient loT devices, which rely on backscatter communication, share limited communication resources for data transmission. As the number of deployed loT devices grows, the likelihood of resource contention and collision at the receiver increases, leading to data loss and inefficient communication. The sporadic transmission from numerous devices results in interference, making it difficult to maintain reliable communication. Traditional methods in MIMO systems are insufficient in handling the scale and density of Ambient loT deployments.

[0008] The visibility region is the area around an antenna array where the signal characteristics (such as power and phase) experienced by each antenna element differ significantly due to the physical environment. This concept is crucial in MIMO systems as it affects the performance of the communication link and the potential for interference among different users or devices.

[0009] Visibility region detection in XL-MIMO systems was typically approached through powerbased methods, where the goal was to detect the visibility region of a given user based on either a direct Line-of-Sight (LoS) link or signals received through regular scatterers. These methods were predominantly used in traditional MIMO systems for characterizing spatial regions where users could be located and leveraging the signal's power decay for localization purposes. Several techniques have been proposed in the literature to identify visibility regions. These are energy detection, location-based approaches, beam-sweeping techniques and cluster-based detection methods.

[0010] In the work by Zhang et al. [1], energy detection is employed during the uplink (UL) training phase, where XL array nodes estimate visibility regions by analyzing the energy levels of received signals.

[0011] Liu et al. [2] present a location-based method for visibility region recognition, which involves a training phase where the network conducts channel measurements with pilot users to determine the visibility region distribution across the XL-MIMO environment and identifies the locations of these nodes.

[0012] Christophe et al. [3] propose a beam-sweeping method that utilizes both power and angular information to detect visibility regions effectively.

[0013] Kihero et al. [4] introduce a technique that identifies users' visibility regions while also utilizing information regarding clusters associated with each visibility region, combining insights from both power and beamspace domains.

[0014] Despite the advancements in visibility region (VR) detection techniques for XL-MIMO systems, several significant disadvantages limit their effectiveness, particularly in the context of Ambient loT devices.

[0015] Inability to Handle Low-Power Signals: Most traditional power-based methods rely on detecting the strength of received signals to identify visibility regions. However, these methods are ill-suited for Ambient loT devices that primarily communicate via backscatter techniques, which generate significantly weaker signals. The low power of backscattered signals makes it challenging to distinguish them from noise or regular scatterers, leading to potential misidentification of visibility regions.

[0016] Sensitivity to Direct Path Interference: Techniques that utilize energy detection or locationbased approaches often struggle with the dominance of direct LoS paths from beacons. In environments where Ambient loT devices coexist with strong direct signals, the higher power of these signals can overshadow backscattered communications. This interference complicates the visibility region identification process, resulting in unreliable estimates and potentially erroneous spatial configurations for loT devices. Limited Scalability: Many existing methods, such as beam-sweeping approaches, may not scale well in dense environments with numerous loT devices. As the number of devices increases, the likelihood of signal overlap and collisions also rises. Power-based detection methods can quickly become inefficient, leading to increased latency and reduced overall system performance due to the inability to manage multiple simultaneous transmissions effectively.

[0017] As a result, all of the problem mentioned above has made it necessary to provide a novelty in the related field.

[0018] Brief Description and Objects of the Invention

[0019] The main object of the present invention is to establish a novel visibility region identification methodology that enhances operational efficiency in XL-MIMO systems. This methodology allows for optimized resource allocation and reduced interference among numerous concurrent loT devices, improving throughput and minimizing latency in relay systems. In small cell networks, it facilitates simultaneous backhaul and access operations, crucial for accommodating the varying bandwidth requirements of Ambient loT applications.

[0020] Furthermore, the invention enables real-time spectrum sensing and dynamic access in cognitive radio networks, ensuring reliable connectivity under fluctuating conditions. Additionally, it supports Integrated Access and Backhaul technologies in 5G and beyond, leading to more efficient resource utilization, which is vital for the growing density of Ambient loT devices, thereby paving the way for robust and efficient wireless communication infrastructures.

[0021] The present invention provides a methodology for identifying visibility regions (VRs) for ambient loT devices within XL-MIMO systems. This methodology addresses critical challenges associated with conventional power-based visibility region detection methods, which struggle to differentiate between the signals from low-power backscattered communications of ambient loT devices and those from normal scatterers, particularly in the presence of strong direct Line-of-Sight (LoS) links from beacons.

[0022] The invention contributes to solving the scientific problem of massive connectivity in ambient loT environments, where the effective utilization of resources for data transmission is paramount. Specifically, it addresses the difficulties in distinguishing the visibility regions of ambient loT devices (BDs) from the visibility regions associated with normal scatterers. Existing power-based methods are inadequate, as they fail to provide reliable identification of visibility regions in low-power contexts, leading to data collisions and loss. The present invention proposes method steps that accurately differentiate the visibility regions of various ambient loT devices from those of normal scatterers which is critical for minimizing signal collisions and enhancing data transmission efficiency and that recognize multiple visibility regions associated with a single beacon in a complex environment where both ambient loT devices and scatterers coexist.

[0023] To achieve these goals, the present invention proposes an approach for antenna activation within the XL-MIMO array. This approach is hybrid of hierarchical search-based antenna activation and window-based antenna activation. The hierarchical search-based antenna activation which systematically activates antenna elements in a hierarchical manner, enabling a structured search for visibility regions that maximizes the likelihood of accurately distinguishing between the signals from different BDs and normal scatterers. The windowbased antenna activation which leverages a dynamic window of activated antennas that adapts based on the signal environment, allowing for more responsive and efficient identification of visibility regions as conditions change over time.

[0024] By focusing on the unique characteristics of ambient loT communications, the methodology provides a more reliable means of identifying visibility regions, thus facilitating improved resource allocation and reduced data collisions. Furthermore, this method enhances the capability of XL-MIMO systems to exploit visibility regions as additional radio resources, leading to more efficient network operations in dense environments. Lastly, it enables efficient scaling with the increasing number of ambient loT devices, addressing the challenges of connectivity and resource management in next-generation wireless networks.

[0025] Briefly, the advantages of the proposed approach are as follows:

[0026] • Enhanced visibility region Identification: The invention provides a robust methodology for identifying visibility regions (visibility regions) of ambient loT devices within XL-MIMO systems, overcoming the limitations of traditional powerbased detection methods. By effectively differentiating between the visibility regions of backscatter communication from ambient loT devices and normal scatterers, the invention reduces data collisions and improves overall communication efficiency.

[0027] • Adaptive Antenna Activation: The dual approach of hierarchical search-based and sliding window-based antenna activation allows for dynamic and adaptive responses to changing signal environments. This flexibility enhances the system's capability to accurately identify multiple visibility regions, even in densely populated environments with numerous loT devices.

[0028] • Scalability for Dense Deployments: The proposed solutions are designed to scale effectively with increasing numbers of ambient loT devices. This is particularly important in modern urban environments, where the density of devices can complicate visibility region detection. The invention ensures that XL-MIMO systems can maintain performance and reliability as the network grows.

[0029] • Improved Resource Utilization: By leveraging identified visibility regions as additional radio resources, the invention optimizes resource allocation within the network. This leads to more efficient communication and enhanced throughput, ultimately contributing to better service quality in ambient loT scenarios.

[0030] • Robustness to Interference: The methodology is designed to mitigate the impact of strong direct signals from beacons, which often overshadow backscattered signals in conventional systems. This robustness enables more accurate visibility region identification, even in challenging signal conditions.

[0031] • Support for Integrated Access and Backhaul: The invention aligns with the principles of Integrated Access and Backhaul (IAB) in 5G and beyond, facilitating simultaneous access and backhaul operations. This integration improves the overall efficiency of wireless networks and supports the evolving needs of ambient loT applications.

[0032] Description of the Figures of the Invention

[0033] The figures and related descriptions necessary for the subject matter of the invention to be understood better are given below.

[0034] Figure la. A schematic view of XL-MIMO system.

[0035] Figure lb. A schematic view of XL-MIMO system with visibility regions.

[0036] Figure 2. Illustration of the flow diagram of basic operations in accordance with certain aspects of the present disclosure.

[0037] Figure 3a. A schematic view of proposed hierarchical approach design activating the first part of XL array.

[0038] Figure 3b. A schematic view of hierarchical approach design activating the second part of XL array. Figure 4 A schematic view of sliding window approach in accordance with certain aspects of the present disclosure.

[0039] Figure 5. A schematic view of the hybrid search approach in accordance with certain aspects of the present disclosure

[0040] Reference Numbers

[0041] The parts and components are given in the figures are referenced for the subject matter of the invention to be understood better.

[0042] 110. Transmitter Node

[0043] 120. XL-MIMO Receiver

[0044] 130. Ambient loT devices

[0045] 140. Visibility Region

[0046] 140a. Part of the Visibility Region

[0047] 150. Environmental Scatterers

[0048] 160. Signal

[0049] 170. Backscatter Modulated Signal

[0050] 210. Active antenna element

[0051] 210a. Sliding window

[0052] 220. Inactive antenna element

[0053] 310. Antenna element

[0054] 320. Sliding section

[0055] Detailed Description of the Invention

[0056] The invention is related to a method to identify the visibility regions (140) of backscatter devices (130), such as backscatter modulation node, ambient loT device, passive loT device, loT device in an XL-MIMO system, leveraging those visibility regions (140) as additional radio resources for optimized communication.

[0057] Referring to Figures la and lb; The extremely large-scale multiple-input multiple-output system (XL-MIMO) comprises at least one transmitter node (110) that transmit a signal (160), such as a pilot signal to a medium where the backscatter devices (130) is positioned. The backscatter device (130) is a device that communicates by reflecting (or backscattering) signals sent by a beacon rather than generating its own signal.

[0058] The medium may also comprise environmental scatters (150), too. Furthermore, the XL-MIMO also comprises an XL-MIMO receiver which defines the visibility region (140). The visibility region (140) includes part of visibility regions which defines specific visibility regions of the backscatter devices (130) or environmental scatters (150).

[0059] Referring to Figure 2; the present invention proposes a method to identify visibility regions of the backscatter devices (130). According to the method, the transmitter node (110) transmits to a medium where the backscatter devices (130) and the XL-MIMO receiver (120) is positioned. The backscatter devices (130) perform backscatter modulation to create backscatter modulated signal (170) which received by XL-MIMO receiver (120). The XL-MIMO receiver (120) identifies the backscatter modulated signals (170) by using parts of antenna elements which are provided by dividing of the XL-MIMO receiver (120). Each part separately identifies visibility regions (140) of one backscatter device (130) and that process is performed for each backscatter device (130).

[0060] Identification can be carried out by hybrid-based antenna selection approach.

[0061] For each method, the transmitter node (110) that transmits pilot signals (160) to XL-MIMO receiver (120), and multiple visibility regions in the environment (some visibility regions originate from backscatter device (130) and others from environmental scatters (150). Each backscatter device (130) is assumed to have one unique visibility region that needs to be identified over the XL-MIMO receiver (120).

[0062] Firstly, visibility region of environmental scatters (150) is identified. The transmitter node (110) transmits pilot signal (160) while backscatter devices (130) are inactive. Only visibility regions from environmental scatters (150) can be received at XL-MIMO receiver (120). The XL-MIMO receiver (120) performs any of the conventional visibility region identification methods in the literature [l]-[4] to identify the visibility region of the environmental scatters (150). This identification also can be used for hierarchical search-based antenna activation, sliding windowbased antenna activation, hybrid-based antenna selection approach which will be explained below. For identifying visibility region of backscatter devices (130), the transmitter node (110) transmits pilot signal (160) while the backscatter devices (130) are active. Visibility regions (140) from the environmental scatters (150) and the backscatter devices (130) are received at the XL-MIMO receiver (120). The XL-MIMO receiver (120) eliminates the effect of visibility regions of the environmental scatters (150) according to previous step.

[0063] Referring to Figure 3a and 3b; in the hierarchical search-based antenna activation, large antenna array on the XL-MIMO receiver (120) is divided into at least two sub-arrays which are active antenna (210) and inactive antenna (220). According to process steps the sub arrays can be active antenna (210) and inactive antenna (220). Following this step, the sub-arrays are sequentially or simultaneously activated, depending on the number of available RF chains.

[0064] When a sub-array in an XL-MIMO system is activated, it means that the antennas within this sub-array are connected to a single RF chain. An RF chain is responsible for receiving and processing signals from the active antenna elements that are connected to it. The choice of sequential or simultaneous activation methods depends on the number of available RF chains.

[0065] Sequential activation is used in the case that there are fewer RF chains than sub-arrays. In this approach, only one sub-array is connected to an RF chain at a time, and sub-arrays are activated one after another. This approach reduces the hardware cost but may result in slower data processing.

[0066] Simultaneous activation is used in the case that here are enough RF chains to match the number of sub-arrays, all sub-arrays can be connected to separate RF chains and activated at the same time. This allows faster data processing, but requires more RF chains and, thus, more complex and costly hardware.

[0067] The transmitter node (110) transmits pilot signals (160), and the backscatter device (130) modulate their sequences onto the pilot signals. The visibility region (140) of the backscatter device (130) in each sub-array by detecting the backscatter modulated signal (170). The subarrays which no the backscatter device (130) is detected are deactivated.

[0068] This process is repeated until the number of the backscatter device (130) matches the available orthogonal sequences provided in the system (from the XL-MIMO receiver (120)) or until no devices remain in the current sub-array. After that, the obtained info related visibility regions (140) can be used for resource assignment. For that, non-orthogonal resources to the backscatter device (130) are assigned with unique visibility regions, and orthogonal resources to devices with overlapping visibility regions. This resource assignment also can be used for sliding window-based antenna activation, hybridbased antenna selection approach which will be explained below.

[0069] Hierarchical search-based antenna selection approach provides efficient for large arrays, scalable, reduces search space, and simplifies detection in sparse environments.

[0070] Referring Figure 4; in the window-based antenna activation, The transmitter node (110) transmits pilot signals (160), and the backscatter device (130)modulate their sequences onto the pilot signals. After that, size of the sliding window (210a) is defined.

[0071] Define the sliding window size considering but not limited to the following factors:

[0072] 1. visibility region size of transmitter node (110).

[0073] 2. Minimum visibility region size of the backscatter device (130).

[0074] 3. Number of available RF chains in XL-MIMO receiver (120).

[0075] 4. Number of antenna elements in the XL-MIMO receiver (120).

[0076] One method to define the window size in the sliding window (210a) approach is with the use of beamspace representation of the XL MIMO channel. A beamspace representation is a representation of the signal in the spatial domain, focusing on the angular information of the signal paths, often used to enhance spatial resolution in MIMO systems. Similar to the approach discussed in [4], XL-MIMO receiver (120) estimates the channel and transforms it into a beamspace representation, which is used to identify the beams, paths, or principal components corresponding to each backscatter device (130). The beam widths in beamspace representation correlate with the size of the visibility region. The visibility region size can be calculated using the relationship: o 1.3 x hjB — _

[0077] A 3 dB beamwidth Q3dBfor a given beam / path / principal component and the size NyRof the visibility region associated with it, is the wavelength, and d is the antenna element spacing. Select the window size based on the largest visibility region in the beamspace to ensure all backscatter devices (130) are detected. A smaller window (210a) could result in multiple detections of the same backscatter devices (130) or missed detections.

[0078] After decision of size of the window (210a), the window (210a) slides across the XL array and identify the visibility regions of each backscatter device (130).

[0079] Sliding window-based antenna selection approach provides continuous coverage, adjustable window size, efficient in dense deployments, and provides high localized resolution.

[0080] Referring to Figure 5; in the hybrid-based antenna activation, the transmitter node (110) transmits pilot signals (160), and the backscatter device (130) modulate their sequences onto the pilot signals. The multiple RF chains are utilized. At least three RF chains are used, each assigned to distinct groups of antenna elements for efficient parallel processing. Two of the RF chains control two sliding windows placed serially across the array (vertically aligned), while the third RF chain is used for a parallel window positioned at the intersection of the two vertical windows. The two vertical windows corresponding to the two RF chains are positioned such that they each cover a portion of the antenna array. The third, parallel window is placed at the connection points of the two vertical windows to capture the overlapping signals between them, as shown in Figure 5. This configuration ensures that signals received from multiple BDs or scatterers located at the window boundaries are effectively captured, reducing the likelihood of missed detections.

[0081] Size of windows are determined in same or similar way with window-based method.

[0082] The two vertical windows slide across the antenna array, maintaining an overlap, preferably a 75% overlap between successive positions. This overlap is chosen to ensure continuous coverage of the array, reducing the chance of missing any visibility regions. The third, parallel window moves concurrently

[0083] Once the larger windows have scanned the entire array, the system refines the detection by leveraging the beamspace representation of the received signals. The beamspace domain provides information on the angular spread of the signals, which corresponds to the visibility region sizes of the backscatter device (130). Identifying the smallest visibility regions from the beamspace provides perform high-resolution detection within the larger windows. This step improves the accuracy of visibility region identification for closely positioned the backscatter devices (130) or those with smaller spatial footprints.

[0084] Hybrid-Based Antenna Selection Approach combines the strengths of hierarchical and sliding window techniques. It also enhances detection accuracy through multi-window scanning and beamspace refinement, while maintaining computational efficiency by leveraging parallel RF chains and overlapping window structures.

[0085] For clarity of the invention, the multiple terms are explained below:

[0086] • Array Non-Stationarity: Array non-stationarity describes the phenomenon where the characteristics of a multiple antenna array, such as channel gains and phase shifts, vary significantly due to changes in the environment, such as movement of users or objects, or variations in the propagation medium. This leads to each antenna element experiencing different signal conditions, complicating signal processing and communication strategies.

[0087] • Non-overlapped visibility regions: Non-overlapped visibility regions are segments of the visibility region for multiple users or devices that do not share any common antenna elements. In this scenario, users can communicate without experiencing selfinterference, leading to optimal signal quality and performance.

[0088] • Partially Overlapped Visibility Region: Partially overlapped visibility regions occur when two or more users share some but not all antenna elements within their visibility regions. This overlapping can lead to some level of interference, which must be managed to maintain communication quality, but is less problematic than fully overlapping regions.

[0089] • Fully Overlapped Visibility Region: Fully overlapped visibility regions exist when two or more users use the same antenna elements within their visibility regions. This scenario results in significant interference and often necessitates additional strategies for self-interference mitigation, as the users may experience substantial degradation in signal quality. REFERENCES

[0090] [1] J. Zhang, J. Zhang, Y. Han, J. Wang and S. Jin, "Average Spectral Efficiency for TDD- based non- Stationary XL-MIMO with visibility region Estimation," 2022 14th International Conference on Wireless Communications and Signal Processing (WCSP), Nanjing, China, 2022, pp. 973-977, doi: 10.1109 / WCSP55476.2022.10039284.

[0091] [2] D. Liu et al., "Location-Based Visible Region Recognition in Extra-Large Massive MIMO Systems," in IEEE Transactions on Vehicular Technology, vol. 72, no. 6, pp. 8186-8191, June 2023.

[0092] [3] C. M. Christophe, AB Kihero, H. Arslan, “A method to enable visibility regions (visibility regions) detection and identification, and a visibility region-aware user scheduling technique in

[0093] Extremely Large Aperture Arrays (ELAA) based Wireless Networks” Submitted to Turkish Patent, Application number: 2023 / 015732

[0094] [4] A. B. Kihero, L. Afeef, and H. Arslan, “A method for identifying visibility regions in ELAA- based wireless networks,” Jan. 15, 2024, Turk Patent (EPATS), App. No: 2024 / 001049.

Claims

CLAIMS1. A computer implemented method for identifying of visibility region of backscatter devices (130) in extremely large-scale multiple-input multiple-output systems having at least one a transmitter node (110), an XL-MIMO receiver (120) and backscatter device (130), characterized by; a) transmitting a signal (160) into the medium while the backscatter devices (130) are inactive and identifying visibility region of environmental scatterers (150) to remove effect visibility region of environmental scatterers (150) for identification of visibility region of the backscatter devices (130); b) Transmitting a signal (160) into the medium while the backscatter devices (130) are active; c) using at least two RF chains of the XL-MIMO receiver (120) which are assigned distinct groups of antenna elements and controls two sliding vertical windows placed serially across the array and at least one another RF chain of the XL-MIMO receiver (120) which controls a parallel window that is positioned at the intersection of the two sliding vertical windows, d) sliding the two sliding vertical windows along the antenna element, maintaining an overlap between successive positions and sliding the parallel window concurrently with the sliding vertical windows, e) determining angular spread of the signals (160), which corresponds to the visibility region sizes of the backscatter device (130).

2. Method according to claim 1, characterized by size of the sliding windows is defined by visibility region of the transmitter node (110), minimum size of the visibility region of backscatter device (130) and available RF chains in XL-MIMO receiver (120) and number of antennas XL-MIMO receiver (120).

3. Method according to claim 1, characterized by estimating channels and transforming the channel into a beamspace representation, and defining the size of the sliding window (210a) by the formula of hjBo— _ 1.3 xVRd x tidBwherein lxdBis the angular width of the main lobe of the radiation pattern where the signal strength is within x dB of its peak value and x is a number and the NyRis size of the sliding window (210a), is the wavelength, and d is the antenna element spacing.

4. Method according to claim 1, characterized by identifying the smallest visibility region to perform high-resolution detection within the larger windows.

5. Method according to any of preceding claims, characterized by assigning non- orthogonal resources to backscatter devices (130) with unique visibility region, and orthogonal resources to backscatter devices (130) devices with overlapping visibility region.

6. A data processing device comprising means for carrying out the steps of the method of any of preceding claims.

7. A computer program comprising instructions which, when the program is executed by the data processing device of claim 6, cause the data processing device to carry out the method of any of claim 1 to 5.

8. A computer-readable medium comprising instructions which, when executed by the data processing device of claim 6, cause the data processing device of claim 7 to carry out the method of any of claim 1 to 5.

Images