Method performed by a node device in a communication system and node device

Abstract

Embodiments of the present disclosure provide a method performed by a node device in a communication system, comprising: obtaining a noise interference correlation matrix corresponding to each of a plurality of resource blocks (RBs) on a DMRS respectively; obtaining a first feature of each of the RBs based on the noise interference correlation matrix corresponding to the RB, wherein the first feature of each of the RBs is related to an interference signal strength corresponding to the RB; segmenting the plurality of RBs based on the first feature of each of the RBs, wherein a difference between interference signal distributions corresponding to each of the RBs belonging to a same RB segment and / or a difference between interference signal strengths is less than a first preset threshold; obtaining a whitening scheme corresponding to each of the RB segments by using a second AI / ML model based on a result of the segmenting and the noise interference correlation matrix corresponding to each of the RBs; and performing whitening processing on corresponding each of the RB segments on a data symbol based on the obtained whitening scheme.

Description

Technical Field

[0001] This disclosure relates to the field of communications, and more specifically, to methods and node devices performed by node devices in a communication system. Background Technology

[0002] Given the successive generations of wireless communication development, these technologies have primarily been developed for human-oriented services such as voice calls, multimedia services, and data services. With the commercialization of 5th-generation (5G) communication systems, the number of connected devices is expected to grow exponentially. These will increasingly connect to communication networks. Examples of the Internet of Things (IoT) can include vehicles, robots, drones, home appliances, displays, smart sensors connected to various infrastructures, construction machinery, and factory equipment. Mobile devices are expected to evolve in various forms, such as augmented reality glasses, virtual reality headsets, and holographic devices. Efforts are underway to develop improved 6G communication systems to provide a wide range of services by connecting hundreds of billions of devices and things in the sixth-generation (6G) era.

[0003] The 6G communication system, expected to be commercially available around 2030, will significantly improve upon existing 5G communication systems in all aspects. Its peak speed will reach at least 50 Gbit / s, user experience speed will reach at least 300 Mbit / s, air interface latency will be less than 1 ms, and air interface reliability will reach 10... -5 In addition to the basic communication indicators mentioned above, 6G communication systems will also have sensing capabilities, AI-related capabilities, and better security, interoperability, and sustainability.

[0004] To achieve the aforementioned performance indicators for 6G communication systems, more advanced air interface and network technologies are needed. Currently, the evolution of extreme multiple input multiple output (MIMO) is being considered, including the use of very large-scale antenna arrays, the development and evolution of distributed antenna systems, and the design of MIMO air interface algorithms assisted by artificial intelligence (AI). This technology can achieve higher spectral efficiency, greater coverage, and more precise positioning and sensing capabilities. Furthermore, technologies that contribute to improving high-frequency coverage, including metamaterial-based lenses and antennas, novel antenna architectures, and reconfigurable intelligence surfaces (RIS), also require further evolution and development.

[0005] To meet the new functionalities added to 6G communication systems, it is necessary to develop new technologies in areas such as network energy saving, air interface security, and network security, while also studying the feasibility of converged technologies such as integrated communication and sensing.

[0006] In addition, to improve spectrum efficiency and overall network performance, the following technologies have been developed for 6G communication systems: full-duplex technology to enable uplink and downlink transmissions to use the same frequency resources simultaneously; network technologies that utilize satellites, high-altitude platform stations (HAPS), etc., in a comprehensive manner; improved network architecture to support mobile base stations, etc., and to enable network operation optimization and automation; dynamic spectrum sharing technology based on spectrum usage prediction and conflict avoidance; the use of artificial intelligence (AI) in wireless communication to improve overall network operation by utilizing AI from the design phase of 6G development and internalizing end-to-end AI support functions; and next-generation distributed computing technologies that overcome the computing power limitations of user equipment (UE) by leveraging ultra-high-performance communication and computing resources (such as mobile edge computing (MEC), cloud, etc.) achievable on the network. Furthermore, efforts are continuing to enhance connectivity between devices, optimize networks, promote the software-defined networking of network entities, and increase the openness of wireless communications by designing new protocols to be used in 6G communication systems, developing mechanisms for achieving hardware-based secure environments and secure data use, and developing technologies for maintaining privacy.

[0007] The research and development of 6G communication systems, encompassing hyper-connectivity for both person-to-machine (P2M) and machine-to-machine (M2M) interactions, is expected to deliver the next wave of hyper-connected experiences. Specifically, services such as truly immersive extended reality (XR), high-fidelity mobile holograms, and digital replicas are anticipated to be provided through 6G communication systems. Furthermore, services such as remote surgery for enhanced security and reliability, industrial automation, and emergency response will be available via 6G communication systems, enabling the technology to be applied across a wide range of sectors including industry, healthcare, automotive, and home appliances. Summary of the Invention

[0008] According to one aspect of this disclosure, a method performed by a node device in a communication system is provided, the method comprising: obtaining a noise-interference correlation matrix corresponding to each of a plurality of resource blocks (RBs) on a demodulation reference signal (DMRS), wherein the diagonal elements of the matrix correspond to the autocorrelation values ​​of noise and interference signals received on each antenna included in the node device, and the off-diagonal elements of the matrix correspond to the cross-correlation values ​​of noise and interference signals received on different antennas included in the node device; obtaining a first feature of each RB based on the noise-interference correlation matrix corresponding to each RB, wherein the first feature of each RB is correlated with the interference signal strength corresponding to each RB; segmenting the plurality of RBs based on the first feature of each RB, wherein the difference between the distribution of interference signals and / or the difference between the interference signal strengths corresponding to each RB belonging to the same RB segment is less than a first preset threshold; and obtaining a whitening scheme corresponding to each RB segment by using a second artificial intelligence / machine learning (AI / ML) model based on the segmentation result and the noise-interference correlation matrix corresponding to each RB; and performing whitening processing on the corresponding RB segments on data symbols based on the obtained whitening scheme.

[0009] In conjunction with any of the above embodiments, according to the method performed by a node device in a communication system provided by this disclosure, the method includes: determining a demodulation reference signal DMRS for calculating a noise interference correlation matrix for data symbols based on the received signal strength indication RSSI of the received signal; and determining the noise interference correlation matrix corresponding to each RB on each data symbol based on the determined DMRS.

[0010] In conjunction with any of the above embodiments, according to the method performed by a node device in a communication system provided by this disclosure, obtaining a first feature of each RB based on the noise interference correlation matrix corresponding to each RB includes: taking a portion or all of the diagonal elements and / or off-diagonal elements of the noise interference correlation matrix corresponding to each RB as the first feature corresponding to each RB.

[0011] In conjunction with any of the above embodiments, according to the method performed by a node device in a communication system provided by this disclosure, the step of segmenting the plurality of RBs based on the first feature of each RB includes: segmenting the plurality of RBs based on the first feature corresponding to each RB by using a first AI / ML model.

[0012] In conjunction with any of the above embodiments, according to the method performed by a node device in a communication system provided by this disclosure, the step of segmenting the plurality of RBs based on the first feature of each RB includes: comparing the value associated with the first feature of each RB with the value associated with the first feature of the preceding RB of the RB; and determining, based on the comparison result, whether each RB is a boundary of an RB segment.

[0013] In conjunction with any of the above embodiments, according to the method executed by a node device in a communication system provided by this disclosure, the method for obtaining a whitening scheme corresponding to each RB segment based on the segmentation results and the noise interference correlation matrix corresponding to each RB segment by using a second artificial intelligence / machine learning (AI / ML) model includes: obtaining a second feature corresponding to each RB segment based on the segmentation results and the noise interference correlation matrix corresponding to each RB segment, wherein the second feature is related to the relative intensity of the noise and interference signal corresponding to each RB segment; and obtaining a whitening scheme corresponding to each RB segment by using the second AI / ML model based on the second feature corresponding to each RB segment.

[0014] In conjunction with any of the above embodiments, according to the method performed by a node device in a communication system provided by this disclosure, the step of obtaining the second feature corresponding to each RB segment based on the segmentation result and the noise interference correlation matrix corresponding to each RB includes: averaging the first part of the elements of the noise interference correlation matrix corresponding to each RB, and using each obtained average value as the first average value of each RB, wherein the first average value of each RB is related to the interference plus noise intensity of each RB; averaging the second part of the elements of the noise interference correlation matrix corresponding to each RB, and using each obtained average value as the second average value of each RB, wherein the second average value of each RB is related to the interference intensity of each RB; and obtaining the second feature corresponding to each RB segment based on the first average value and the second average value of each RB included in each RB segment.

[0015] In conjunction with any of the above embodiments, according to the method performed by a node device in a communication system provided by this disclosure, the first part of the elements includes a portion or all of the diagonal elements in the noise interference correlation matrix, and / or the second part of the elements includes a portion or all of the off-diagonal elements in the noise interference correlation matrix.

[0016] In conjunction with any of the above embodiments, according to the method performed by a node device in a communication system provided by this disclosure, the second feature corresponding to the RB segment includes at least one of the average, maximum, or minimum value of the ratio of the first average value and the second average value of each RB included in the RB segment.

[0017] In conjunction with any of the above embodiments, according to the method performed by a node device in a communication system provided by this disclosure, the first AI / ML model and / or the second AI / ML model includes at least one of the following: multilayer perceptron (MLP), convolutional neural network (CNN), recurrent neural network (RNN), and transformer.

[0018] In conjunction with any of the above embodiments, according to the method performed by a node device in a communication system provided by this disclosure, the whitening scheme includes at least one of the following: whitening based on RB normalization (IWNRB); whitening based on signal bandwidth normalization (IWNBW); and whitening based on RB (IWRB).

[0019] In conjunction with any of the above embodiments, the method performed by a node device in a communication system according to the present disclosure includes a user equipment (UE) and / or a base station.

[0020] According to another aspect of this disclosure, a node device in a wireless communication system is provided, the node device comprising: a transceiver configured to transmit and / or receive signals; and a controller configured to control the transceiver to perform the methods described above performed by the node device in the wireless communication system.

[0021] According to another aspect of this disclosure, a non-transitory computer-readable recording medium is provided, on which a program is stored for execution by a computer as described above. Attached Figure Description

[0022] To gain a more complete understanding of this disclosure and its advantages, reference is now made to the following description in conjunction with the accompanying drawings, wherein like reference numerals denote like parts:

[0023] Figure 1 An example wireless network according to an embodiment of the present disclosure is shown;

[0024] Figure 2 An example base station according to an embodiment of the present disclosure is shown;

[0025] Figure 3 An example user equipment according to an embodiment of the present disclosure is shown;

[0026] Figure 4A schematic diagram of uplink (UL) co-channel interference (CCI) according to an embodiment of the present disclosure is shown;

[0027] Figure 5 A schematic diagram of co-channel interference (CCI) between downlink (DL) and uplink (UL) according to an embodiment of the present disclosure is shown;

[0028] Figure 6 A schematic diagram of the frequency domain distribution of interference according to an embodiment of the present disclosure is shown;

[0029] Figure 7 A flowchart is shown of a method performed by a device in a communication system according to an embodiment of the present disclosure;

[0030] Figure 8 A flowchart of an AI-based adaptive interference cancellation receiver method according to an embodiment of the present disclosure is shown.

[0031] Figure 9 A schematic diagram of the channel frequency domain characteristic distribution according to an embodiment of the present disclosure is shown;

[0032] Figure 10 A schematic diagram of interference distribution in the time domain of a channel according to an embodiment of the present disclosure is shown;

[0033] Figure 11 A schematic diagram of a first AI / ML model according to an embodiment of the present disclosure is shown;

[0034] Figure 12 A schematic diagram of labels in a first AI / ML model according to an embodiment of the present disclosure is shown;

[0035] Figure 13 A schematic diagram of a second AI / ML model according to an embodiment of the present disclosure is shown;

[0036] Figure 14 A block diagram of a node device in a communication system according to an embodiment of the present disclosure is shown. Detailed Implementation

[0037] Before proceeding with the following detailed description, it may be advantageous to define certain words and phrases used throughout the patent literature. The term “connection” and its derivatives refer to any direct or indirect communication between two or more elements, regardless of whether those elements are physically in contact with each other. The terms “transmit,” “receive,” and “transmit,” and their derivatives encompass both direct and indirect communication. The terms “comprise” and “include,” and their derivatives mean inclusion without limitation. The term “or” is concurrent, meaning both and / or. The phrase “associated with,” and its derivatives mean including, being included in, interconnected with, containing, being contained within, connected to or connected with, coupled to or coupled with, able to communicate with, cooperate with, intertwine, juxtapose, proximate, bound to or bound with, having, possessing attributes, having a relationship with, or having a relationship with, etc. The term “controller” means any device, system, or part thereof that controls at least one operation. Such a controller may be implemented in hardware or in a combination of hardware and software and / or firmware. The functionality associated with any particular controller, whether local or remote, can be centralized or distributed. The phrase "at least one" when used to list items means that different combinations of one or more of the listed items can be used, and it is possible that only one item in the list is needed. For example, "at least one of A, B, and C" includes any of the following combinations: A, B, C; A and B; A and C; B and C; and only A, only B, and only C. Similarly, the term "set" means one or more. Therefore, a set of items can be a single item or a set of two or more items.

[0038] Furthermore, the various functions described below can be implemented or supported by one or more computer programs, each function being formed by computer-readable program code and embodied in a computer-readable medium. The terms "application" and "program" refer to one or more computer programs, software components, instruction sets, procedures, functions, objects, classes, instances, associated data, or portions thereof suitable for implementation in appropriate computer-readable program code. The phrase "computer-readable program code" includes any type of computer code, including source code, object code, and executable code. The phrase "computer-readable medium" includes any type of medium accessible by a computer, such as read-only memory (ROM), random access memory (RAM), hard disk drive, compact disc (CD), digital video disc (DVD), or any other type of storage. "Non-transitory" computer-readable media excludes wired, wireless, optical, or other communication links that transmit transient electrical or other signals. Non-transitory computer-readable media includes media in which data can be permanently stored and media such as rewritable optical discs or erasable memory devices in which data can be stored and later rewritten.

[0039] Definitions for certain other words and phrases are provided throughout this patent document. Those skilled in the art will understand that, in many, if not the most, instances, such definitions apply to both prior and future uses of the words and phrases defined in this way.

[0040] The figures and various embodiments included herein, used to illustrate the principles of this disclosure, are merely illustrative and should not be construed in any way as limiting the scope of this disclosure. Furthermore, those skilled in the art will understand that the principles of this disclosure can be implemented in any suitably arranged wireless communication system.

[0041] The following Figures 1 to 3 Various embodiments of this disclosure implemented in wireless communication systems are described. Figures 1 to 3 The description does not imply any physical or architectural limitations on the ways in which different embodiments can be implemented. Different embodiments of this disclosure can be implemented in any suitably arranged communication system.

[0042] Figure 1 An example wireless network according to an embodiment of this disclosure is shown. Figure 1 The embodiments of the wireless network shown are for illustrative purposes only. Other embodiments of the wireless network 100 may be used without departing from the scope of this disclosure.

[0043] like Figure 1As shown, the wireless network includes a base station (next generation nodeB, gNB or gNodeB) 101, gNB 102, and gNB 103. gNB 101 communicates with gNB 102 and gNB 103. gNB 101 also communicates with at least one network 130 such as the Internet, Internet Protocol (IP) networks, or other data networks.

[0044] gNB 102 provides wireless broadband access to network 130 to multiple first user equipments (UEs) within coverage area 120 of gNB 102. The multiple first UEs include UE 111, which may be located in a small business (SB); UE 112, which may be located in an enterprise (E); UE 113, which may be located in a WiFi hotspot (HS); UE 114, which may be located in a first residence (R1); UE 115, which may be located in a second residence (R2); and UE 116, which may be a mobile device (M) such as a cellular phone, wireless laptop, or wireless personal digital assistant (PDA). gNB 103 provides wireless broadband access to network 130 to multiple second UEs within coverage area 125 of gNB 103. The multiple second UEs include UE 115 and UE 116, and subscriber stations (SS, such as UEs) 117, 118, and 119. In some embodiments, one or more of gNBs 101103 may communicate with each other and UEs 111 116 using existing wireless communication technologies, and one or more of UEs 111 119 may communicate directly with each other (e.g., UEs 117 119) using other existing or proposed wireless communication technologies.

[0045] Depending on the network type, the term "base station" or "BS" can refer to any component (or set of components) configured to provide wireless access to a network, such as a transmit point (TP), transmit-receive point (TRP), enhanced (or "evolved") base station (eNodeB or eNB), 5G base station (gNB), macro cell, femtocell, wireless fidelity (WiFi) access point (AP), or other wireless-capable devices. A base station can provide wireless access according to one or more wireless communication protocols, such as 3GPP 5G new radio (NR), Long Term Evolution (LTE), LTE-Advanced (LTE A), high-speed packet access (HSPA), WiFi 802.11a / b / g / n / ac, etc. For convenience, various names for base station type devices and functions may be used interchangeably in this patent document to refer to network infrastructure components that provide wireless access to remote terminals. Furthermore, depending on the network type, the term "User Equipment" (UE) can refer to any component such as a mobile station (MS), user station (SS), remote terminal, wireless terminal, receiving point, or user device. For convenience, various names for user equipment type devices and functions may be used interchangeably in this patent document to refer to remote wireless devices that wirelessly access the BS regardless of whether the UE is a mobile device (such as a mobile phone or smartphone) or is generally considered a fixed device (such as a desktop computer or vending machine).

[0046] The dashed lines indicate the approximate extent of coverage areas 120 and 125, which are shown as roughly circular for illustrative and explanatory purposes only. It should be clearly understood that coverage areas such as 120 and 125 associated with the gNB can have other shapes, including irregular shapes, depending on the configuration of the gNB and variations in the wireless environment associated with natural and man-made obstacles.

[0047] As described in more detail below, one or more of UEs 111 and 119 include circuitry, programming, or a combination thereof. In some embodiments, one or more of gNBs 101 and 103 include circuitry, programming, or a combination thereof.

[0048] although Figure 1 An example of a wireless network is shown, but more can be found on... Figure 1Various modifications can be made. For example, wireless network 100 can include any number of gNBs and any number of UEs in any suitable arrangement. Furthermore, gNB 101 can communicate directly with any number of UEs and provide those UEs with wireless broadband access to network 130. Similarly, each gNB 102 or 103 can communicate directly with network 130 and provide UEs with direct wireless broadband access to network 130. Additionally, gNBs 101, 102, and / or 103 can provide access to other or additional external networks, such as external telephone networks or other types of data networks.

[0049] Figure 2 An example base station according to an embodiment of the present disclosure is shown. Figure 2 The embodiment of gNB 102 shown is for illustrative purposes only, and Figure 1 gNBs 101 and 103 can have the same or similar configurations. However, gNBs come in a variety of configurations, and Figure 2 This disclosure is not intended to limit the scope to any particular implementation of gNB.

[0050] like Figure 2 As shown, gNB 102 includes multiple antennas 200a 200n, multiple radio frequency (RF) transceivers 201a 201n, transmit (TX) processing circuitry 203, and receive (RX) processing circuitry 204. gNB 102 also includes a controller / processor 205, a memory 206, and a backhaul or network interface 207.

[0051] RF transceivers 201a and 201n receive incoming RF signals, such as signals transmitted by the UE in network 100, from antennas 200a and 200n. RF transceivers 201a and 201n down-convert the incoming RF signals to generate intermediate frequency (IF) or baseband signals. The IF or baseband signal is sent to RX processing circuitry 204, which generates a processed baseband signal by filtering, decoding, and / or digitizing the baseband or IF signal. RX processing circuitry 204 sends the processed baseband signal to controller / processor 205 for further processing.

[0052] TX processing circuit 203 receives analog or digital data (such as voice data, web data, email, or interactive video game data) from controller / processor 205. TX processing circuit 203 encodes, multiplexes, and / or digitizes the outgoing baseband data to generate a processed baseband or IF signal. RF transceivers 201a and 201n receive the processed baseband or IF signal from TX processing circuit 203 and up-convert the baseband or IF signal into an RF signal transmitted via antennas 201a and 201n.

[0053] The controller / processor 205 may include one or more processors or other processing devices that control the overall operation of the gNB 102. For example, the controller / processor 205 may control the reception of forward channel signals and the transmission of reverse channel signals by the RF transceivers 201a 201n, the RX processing circuit 204, and the TX processing circuit 203, according to known principles. The controller / processor 205 may also support additional functions, such as more advanced wireless communication capabilities.

[0054] For example, the controller / processor 205 can support beamforming or directional routing operations, where outgoing signals from multiple antennas 200a 200n are weighted differently to effectively redirect the outgoing signals in the desired direction. Any of a variety of other functions can be supported in the gNB 102 via the controller / processor 205.

[0055] The controller / processor 205 is also capable of executing programs and other processes located in the memory 206, such as the operating system (OS). The controller / processor 205 can move data into or out of the memory 206 as needed by the executing process.

[0056] The controller / processor 205 is also connected to a backhaul or network interface 207. The backhaul or network interface 207 allows the gNB 102 to communicate with other devices or systems via a backhaul connection or over a network. Interface 207 can support communication via any suitable wired or wireless connection(s). For example, when the gNB 102 is implemented as part of a cellular communication system (such as a cellular communication system supporting 5G, LTE, or LTE A), interface 207 can allow the gNB 102 to communicate with other gNBs via a wired or wireless backhaul connection. When the gNB 102 is implemented as an access point, interface 207 can allow the gNB 102 to communicate via a wired or wireless local area network or via a wired or wireless connection to a larger network (such as the Internet). Interface 207 includes any suitable structure that supports communication via wired or wireless connections such as Ethernet or RF transceivers.

[0057] Memory 206 is connected to controller / processor 205. A portion of memory 206 may include random access memory (RAM), and another portion of memory 206 may include flash memory or other read-only memory (ROM).

[0058] although Figure 2 An example of gNB 102 is shown, but it is possible to see more. Figure 2 Various changes can be made. For example, gNB 102 can include any number of Figure 2 Each component is shown in the diagram. As a specific example, an access point may include multiple interfaces 207, and the controller / processor 205 may support routing functionality to route data between different network addresses. As another specific example, although shown as a single instance of TX processing circuitry 203 and a single instance of RX processing circuitry 204, gNB102 may include multiple instances of each (such as one per RF transceiver). For example, Figure 2 The various components can be combined, further subdivided, or omitted, and additional components can be added as needed.

[0059] Figure 3 An example user equipment according to an embodiment of the present disclosure is shown. Figure 3 The embodiment of UE 116 shown is for illustrative purposes only, and Figure 1 UEs 111, 115, 117, and 119 can have the same or similar configurations. However, UEs appear in multiple configurations, and Figure 3 This disclosure is not intended to limit the scope to any particular implementation of the UE.

[0060] like Figure 3 As shown, UE 116 includes an antenna 301, a radio frequency (RF) transceiver 302, a TX processing circuit 303, a microphone 304, and a receive (RX) processing circuit 305. UE 116 also includes a speaker 306, a controller or processor 307, an input / output (I / O) interface (IF) 308, an input device 309, a touchscreen display 310, and memory 311. Memory 311 includes an OS 312 and one or more applications 313.

[0061] RF transceiver 302 receives incoming RF signals transmitted by gNB of network 100 from antenna 301. RF transceiver 302 down-converts the incoming RF signals to generate IF or baseband signals. The IF or baseband signals are sent to RX processing circuitry 305, which generates processed baseband signals by filtering, decoding, and / or digitizing the baseband or IF signals. RX processing circuitry 305 sends the processed baseband signals to speaker 306 (e.g., for voice data) or processor 307 for further processing (e.g., for web browsing data).

[0062] TX processing circuit 303 receives analog or digital voice data from microphone 304 or other outgoing baseband data (such as web data, email, or interactive video game data) from processor 307. TX processing circuit 303 encodes, multiplexes, and / or digitizes the outgoing baseband data to generate a processed baseband or IF signal. RF transceiver 302 receives the processed baseband or IF signal from TX processing circuit 303 and up-converts the baseband or IF signal into an RF signal transmitted via antenna 301.

[0063] Processor 307 may include one or more processors or other processing devices and executes OS 312 stored in memory 311 to control the overall operation of UE 116. For example, processor 307 may control the reception of forward channel signals and the transmission of reverse channel signals by RF transceiver 302, RX processing circuitry 305, and TX processing circuitry 303 according to known principles. In some embodiments, processor 307 includes at least one microprocessor or microcontroller.

[0064] Processor 307 is also capable of executing other processes and programs located in memory 311, such as processes for CSI reporting on the uplink channel. Processor 307 can move data into or out of memory 311 as needed for executing processes. In some embodiments, processor 307 is configured to execute application 313 based on OS 312 or in response to signals received from gNB or operator. Processor 307 is also coupled to I / O interface 308, which provides UE 116 with the ability to connect to other devices such as laptops and laptops. I / O interface 308 is the communication path between these accessories and processor 307.

[0065] The processor 307 is also connected to the touchscreen display 310. The user of the UE 116 can use the touchscreen display 310 to input data into the UE 116. The touchscreen display 310 can be a liquid crystal display, a light-emitting diode display, or other display capable of rendering text and / or at least limited graphics such as those from a website.

[0066] Memory 311 is connected to processor 307. A portion of memory 311 may include RAM, and another portion of memory 311 may include flash memory or other ROM.

[0067] although Figure 3 An example of UE 116 is shown, but it is possible to modify it. Figure 3 Make various changes. For example, Figure 3 The various components can be combined, further subdivided, or omitted, and additional components can be added as needed. As a specific example, processor 307 can be divided into multiple processors, such as one or more central processing units (CPUs) and one or more graphics processing units (GPUs). Moreover, although... Figure 3 The UE 116 is shown configured as a mobile phone or smartphone, but the UE can be configured to operate as other types of mobile or fixed devices.

[0068] Exemplary embodiments of this disclosure are further described below with reference to the accompanying drawings.

[0069] The text and accompanying drawings are provided by way of example only to aid in understanding this disclosure. They should not be construed as limiting the scope of this disclosure in any way. Although certain embodiments and examples have been provided, it will be apparent to those skilled in the art, based on the content disclosed herein, that changes may be made to the illustrated embodiments and examples without departing from the scope of this disclosure.

[0070] Inter-cell interference severely impacts channel spectral efficiency and network coverage performance. To suppress inter-cell interference and improve channel capacity, both 4G and 5G systems use MMSE-IRC (Minimum mean square error-interference rejection combining) receivers to handle spatial interference. 3GPP Rel-19 specifies NR (New Radio) FR1 (Frequency Range) as the standard.

[0071] 1) The MMSE-IRC requirement was determined to suppress inter-cell interference from uplink (UL). Please refer to [reference needed]. Figure 4 , Figure 4 A schematic diagram of uplink (UL) to uplink (UL) co-channel interference (CCI) according to an embodiment of the present disclosure is shown. Figure 4 As shown, the uplink signal transmitted by UE2 will cause co-channel interference between uplink (UL) and uplink (UL) signals transmitted by UE1, which is the target signal that BS1 needs to detect. Furthermore, in real-world network scenarios, the interference of downlink (DL) signals from the same serving cell or neighboring cells located in different subbands on uplink signals is becoming increasingly significant. Please refer to [reference needed]. Figure 5 , Figure 5 A schematic diagram of co-channel interference (CCI) between downlink (DL) and uplink (UL) according to an embodiment of the present disclosure is shown. Figure 5 As shown, the downlink signal transmitted by the aggressor gNB will cause co-channel interference (DL) to uplink (UL) to the uplink signal transmitted by the target UE that the victim gNB needs to detect. Because NR systems offer greater configuration flexibility and the interference scenarios are more complex, many network vendors and equipment manufacturers have proposed introducing the MMSE-IRC assessment requirement for Co-Channel Interference (CCI) between uplink and downlink in 3GPP Rel-20 to address this issue.

[0072] To implement the MMSE-IRC algorithm, existing receivers estimate the noise-interference correlation matrix based on the demodulation reference signal (DMRS) in units of resource blocks (RBs). The following section introduces current receiver algorithms for suppressing interference.

[0073] Current receiver algorithms for suppressing interference treat interference as colored noise and whiten it in the spatial domain (Interference whitening, IW). The specific steps are as follows.

[0074] Step 1: Estimate the noise-interference correlation matrix for each RB on the DMRS. Taking uplink transmission as an example, assuming there is one interfering UE (but it is understandable that there may be interference signals from multiple sources in the network environment), the estimated interference and noise v can be calculated using the following formula:

[0075]

[0076] Where y is the signal received by the receiver, and h1 is the channel response of the target UE obtained after channel estimation of the signal. and The signals transmitted by the target UE and the interfering UE are respectively, h2 represents the channel response of the interfering UE, and n represents the received noise.

[0077] Furthermore, the noise interference correlation matrix is ​​estimated in units of RB using the following formula.

[0078]

[0079] Where b represents the b-th RB, b = 0, 1, ..., B. B represents the number of RBs within the signal bandwidth. Each RB contains 12 resource elements (REs).

[0080] Step 2: Calculate the noise interference correlation matrix over the entire bandwidth. Depending on the interference scenario, there are three whitening schemes.

[0081] Option 1: Whitening with normalization over RB (IWNRB). This option is used in scenarios where interference is dominant and the correlation between interference between different antennas is low. Here, `diag()` represents the operation of taking the diagonal elements of the matrix.

[0082]

[0083] Option 2: Whitening with normalization over signal bandwidth (IWNBW). This option is used in noise-dominant scenarios.

[0084]

[0085] Option 3: IW over RB (IWRB) whitening. This option is used in scenarios where interference is dominant and the correlation between interference between different antennas is high.

[0086]

[0087] Step 3: Select one of the three whitening schemes mentioned above to suppress interference. If the noise is high, IWNBW is selected to obtain gain through average noise reduction across the entire bandwidth. If the interference is high, IWNRB or IWRB is selected for whitening to suppress interference based on the threshold set according to the signal-to-noise ratio (SNR) estimated during the channel estimation stage.

[0088] Step 4: Based on the noise interference correlation matrix calculated in the above steps Using the selected whitening scheme, interference whitening and MMSE detection are performed on the received data to obtain the recovered estimated signal.

[0089] Due to the complexity of the interference distribution, current methods cannot achieve optimal detection performance and are highly complex.

[0090] For example, when estimating the noise interference correlation matrix over the entire signal bandwidth, the three IW schemes mentioned above offer only two options: averaging on a per-RB basis (i.e., using the noise interference correlation matrix within each RB as the noise interference correlation matrix for the entire bandwidth) and averaging across the entire bandwidth. In noise-dominant scenarios, averaging across the entire bandwidth is chosen to obtain the noise reduction gain. In interference-dominant scenarios, since the distribution of interference cannot be determined, the noise interference correlation matrix is ​​calculated separately for each RB.

[0091] However, in real-world communication scenarios, interference types are diverse. For example, interference can be uplink-to-uplink co-channel interference, downlink-to-downlink co-channel interference, or co-channel interference between uplink and downlink. Due to flexible Time Division Duplexing (TDD) configurations, downlink-to-uplink interference or uplink-to-downlink interference can also occur between different cells. Furthermore, to improve uplink coverage, 3GPP Rel-18 introduced sub-band full-duplex technology, allowing different sub-bands to be configured with different transmission schemes, and interference can also exist between sub-bands. In addition, the characteristics of interference, including its distribution and intensity, are random. First, the frequency domain distribution of interference is random, potentially occupying only a portion of the signal bandwidth. On the other hand, because NR systems employ flexible scheduling methods in the time domain—at the slot level or mini-slot level—supporting time-domain scheduling of different lengths, the time-frequency distribution of interference can also be random, considering its time-domain characteristics. Furthermore, the NR system supports different UE types. Considering the different transmit power levels and distribution of different UE types, the intensity of interference is also random. Please refer to... Figure 6 , Figure 6 A schematic diagram of the frequency domain distribution of interference according to an embodiment of the present disclosure is shown. For example... Figure 6 As shown, the interference signal may be distributed across different RBs, and the interference intensity on different RBs is random. Therefore, averaging on each RB or averaging across the entire bandwidth may not be the optimal detection method.

[0092] Furthermore, the receiver cannot predict the intensity of interference relative to noise. Current methods select the whitening scheme based on predetermined thresholds determined by empirical values ​​(e.g., step 3 above). However, the threshold setting is not precise, and the estimation error for SNR is also large when interference or noise is high. In short, using thresholds to select whitening schemes is not precise enough, and the whitening schemes selected by thresholds are limited to IWNRB and IWRB. Selecting an inappropriate whitening scheme will reduce detection performance.

[0093] It is understood that the above-mentioned technical problems are merely illustrative and are not intended to limit the technical problems that this disclosure can solve. Any technical problem that can be solved by the methods provided in this disclosure is within the scope of protection of this disclosure.

[0094] To address one or more of the aforementioned problems, a novel receiver algorithm for suppressing interference is needed to improve network throughput performance under various interference scenarios.

[0095] Various embodiments of this disclosure provide a method executed by a node device in a communication system, comprising: obtaining a noise-interference correlation matrix corresponding to each of a plurality of resource blocks (RBs), wherein the diagonal elements of the matrix correspond to the autocorrelation values ​​of noise and interference signals received on each antenna included in the node device, and the off-diagonal elements of the matrix correspond to the cross-correlation values ​​of noise and interference signals received on different antennas included in the node device; obtaining a first feature of each RB based on the noise-interference correlation matrix corresponding to each RB, wherein the first feature of each RB is related to the interference signal strength corresponding to each RB; segmenting the plurality of RBs based on the first feature of each RB, wherein the difference between the distribution of interference signals and / or the difference between the interference signal strengths corresponding to each RB belonging to the same RB segment is less than a first preset threshold; and obtaining a whitening scheme corresponding to each RB segment by using a second artificial intelligence / machine learning (AI / ML) model based on the segmentation result and the noise-interference correlation matrix corresponding to each RB based on the noise-interference correlation matrix; and performing whitening processing on the corresponding RB segments on data symbols based on the obtained whitening scheme.

[0096] In the methods provided in various embodiments of this disclosure, RBs with similar interference signal distributions and / or interference signal intensities are divided into the same RB segment based on a first feature of each RB (wherein the first feature of each RB is related to the interference signal strength corresponding to each RB). For the same RB segment, a whitening scheme most suitable for that RB segment is obtained through a second artificial intelligence / machine learning (AI / ML) model. Whitening is then performed using the determined whitening schemes for each RB segment, thereby obtaining the optimal whitening scheme under different interference scenarios and thus achieving optimal interference cancellation performance, thereby improving the network throughput performance.

[0097] Please refer to Figure 7 , Figure 7 A schematic diagram of a method performed by a node device in a communication system according to an embodiment of the present disclosure is shown. The method may be performed by a user equipment (UE) and / or a base station, and may include one or more of the following steps.

[0098] Step S410: Obtain the noise-interference correlation matrix corresponding to each RB among the multiple resource blocks RB on the demodulated reference signal DMRS. The diagonal elements of the matrix correspond to the autocorrelation values ​​of the noise and interference signals received on each antenna included in the node device, and the off-diagonal elements correspond to the cross-correlation values ​​of the noise and interference signals received on different antennas included in the node device.

[0099] In one manner, the DMRS can be one or more.

[0100] Step S420: Based on the noise interference correlation matrix corresponding to each RB, obtain the first feature of each RB. The first feature of each RB may be related to the interference signal strength corresponding to each RB.

[0101] As one implementation method, some or all of the diagonal and / or off-diagonal elements of the noise interference correlation matrix corresponding to each RB can be used as the first feature corresponding to each RB.

[0102] Step S430: Based on the first feature of each RB, the plurality of RBs are segmented. The difference in interference signal distribution and / or the difference in interference signal intensity between RBs belonging to the same RB segment can be less than a first preset threshold (that is, the interference signal distribution and / or the interference signal intensity of RBs belonging to the same RB segment are approximately the same). The first preset threshold can be preset as needed.

[0103] As one implementation method, based on the first feature corresponding to each RB, the multiple RBs can be segmented by using a first AI / ML model.

[0104] In one implementation, the first AI / ML model may include at least one of the following: multilayer perceptron (MLP), convolutional neural network (CNN), recurrent neural network (RNN), and transformer.

[0105] In one implementation, the value associated with the first feature of each RB (e.g., the average value of each RB) can be compared with the value associated with the first feature of the previous RB of the RB; and based on the comparison results, it can be determined whether each RB is the boundary of an RB segment.

[0106] Step S440: Based on the segmentation results and the noise interference correlation matrix corresponding to each RB, a whitening scheme corresponding to each RB segment is obtained by using a second artificial intelligence / machine learning (AI / ML) model.

[0107] In one implementation, the second AI / ML model may include at least one of the following: multilayer perceptron (MLP), convolutional neural network (CNN), recurrent neural network (RNN), and transformer.

[0108] As one implementation method, a second feature corresponding to each RB segment can be obtained based on the segmentation results and the noise interference correlation matrix corresponding to each RB, wherein the second feature is related to the relative intensity of the noise and interference signal corresponding to each RB segment; further, based on the second feature corresponding to each RB segment, a whitening scheme corresponding to each RB segment can be obtained by using the second AI / ML model.

[0109] As one implementation, the first part of the elements of the noise interference correlation matrix corresponding to each RB (e.g., it may include a portion or all of the diagonal elements in the noise interference correlation matrix) can be averaged, and the resulting averages can be used as the first averages of each RB, wherein the first averages of each RB can be related to the interference plus noise intensity of each RB; and the second part of the elements of the noise interference correlation matrix corresponding to each RB (e.g., it may include a portion or all of the off-diagonal elements in the noise interference correlation matrix) can be averaged, and the resulting averages can be used as the second averages of each RB, wherein the second averages of each RB can be related to the interference intensity of each RB; furthermore, based on the first and second averages of each RB segment, a second feature corresponding to each RB segment can be obtained (e.g., the second feature corresponding to the RB segment may include at least one of the average, maximum, or minimum values ​​of the ratio of the first and second averages of each RB segment).

[0110] As one implementation, the whitening scheme may include at least one of the following: whitening based on RB normalization (IWNRB); whitening based on signal bandwidth normalization (IWNBW); and whitening based on RB (IWRB).

[0111] As one embodiment, the method may further include the steps of: determining a demodulation reference signal DMRS for calculating the noise interference correlation matrix for data symbols based on the received signal strength indication RSSI of the received signal; and determining the noise interference correlation matrix corresponding to each RB on each data symbol based on the determined DMRS.

[0112] Step S450: Based on the obtained whitening scheme, perform whitening processing on the corresponding RB segments on the data symbols.

[0113] In the methods provided in various embodiments of this disclosure, based on the first feature of each RB (wherein the first feature of each RB is related to the interference signal strength corresponding to each RB), and based on a first AI / ML model or a non-AI method, RBs with similar interference signal distribution and / or interference signal strength are divided into the same RB segment. Furthermore, for the same RB segment, a whitening scheme most suitable for that RB segment is obtained through a second AI / ML model. Whitening is then performed using the determined whitening schemes for each RB segment, thereby obtaining the optimal whitening scheme under different interference scenarios and thus achieving optimal interference cancellation performance, thereby improving the network throughput performance.

[0114] The following is combined with Figure 8The overall architecture of this disclosure is described. Figure 8 A flowchart of an AI-based adaptive interference cancellation receiver method according to an embodiment of this disclosure is shown.

[0115] like Figure 8 As shown, this disclosure may consist of two main parts, one being AI-based time-frequency domain interference segmentation detection, and the other being AI-based interference cancellation algorithm selection.

[0116] AI-based time-frequency domain interference segmentation detection mainly consists of the following steps:

[0117] Step 0: By detecting the Received Signal Strength Indication (RSSI) of the received signal in the time domain, determine the DMRS used to calculate the noise interference correlation matrix (e.g., interference covariance matrix) for each data symbol. This includes two cases: the noise interference correlation matrix calculated for each symbol using only one neighboring DMRS, and the matrix obtained by averaging the noise interference correlation matrices calculated for each symbol using all DMRS. If it is the first case, divide the received signal in the time domain into multiple segments, the number of which is the same as the number of DMRS symbols. Perform the following steps for the noise interference correlation matrix of each DMRS. If it is the second case, divide the received signal in the time domain into one segment, and perform the following steps for the matrix obtained by averaging the noise interference correlation matrices of all DMRS.

[0118] Step 1: Calculate the noise-interference correlation matrix for each RB on the DMRS to extract the frequency domain channel features. The frequency domain channel features are as follows: Figure 9 As shown, for subbands without interference, the channel response is relatively stable. For subbands with interference, the channel response fluctuates significantly, especially near the RB where the interference distribution changes.

[0119] Step 2: Based on the extracted frequency domain features, perform frequency domain interference segmentation detection based on AI or non-AI to obtain segmentation results. The interference distribution within each segment is approximate.

[0120] Based on the above segmented detection results, for each data symbol, the optimal algorithm is selected from the candidate algorithms based on the AI / ML model within each segment, and finally whitening is performed separately within each segment.

[0121] The principles and solutions of the present invention will now be described in conjunction with several exemplary embodiments.

[0122] First, the AI-based time-frequency domain interference segmentation detection process proposed in this invention will be introduced, which includes the following multiple stages, modes, or operations:

[0123] Step 0: Time-domain segmentation detection:

[0124] Based on the signals received in the time domain (which include data symbols and DMRS), the base station calculates the Received Signal Strength Indicator (RSI) for each signal. Based on the RSSI of each signal, a demodulation reference signal (DMRS) is determined for calculating the noise interference correlation matrix (NOC) of data symbols. This further determines the NOC of the DMRS and the corresponding NOC of the data symbols for subsequent steps, used for interference whitening detection of data symbols. For example, if the difference between the RSSI of all data symbols and the RSSI of the DMRS is less than a predetermined threshold X, then the average of the NOC of all DMRS can be used as the NOC of all data symbols for interference whitening detection, and the received data is divided into segments in the time domain. If the difference between the RSSI of any data symbol and the RSSI of any DMRS is greater than the predetermined threshold X, then each data symbol uses only the NOC of the DMRS closest to it in the time domain as its NOC for interference whitening detection, and the received data is divided into multiple segments in the time domain, the number of which is the same as the number of DMRS. The receiving entity can also be a UE.

[0125] Please refer to Figure 10 , Figure 10 A schematic diagram of interference distribution in the time domain of a channel according to an embodiment of the present disclosure is shown. Figure 10 The document includes three different scenarios: assuming each signal frame contains two DMRS symbols, in scenario 1, the interference is distributed across the entire frame; in scenario 2, the signal is not scheduled based on time slots, and the interference signal is only distributed on the 4th and 5th symbols of time slot 0; in scenario 3, the interference is distributed on the first 7 symbols. The following details the implementation examples for each scenario.

[0126] Scenario 1: In the time domain, interference is distributed throughout the entire frame. The RSSI of 14 symbols, including the data Symbol and DMRS, is calculated. In this scenario, for each data symbol, if the difference between its RSSI and the RSSI of both DMRS is less than a predetermined threshold X, then the data symbol can use the average result of the noise-interference correlation matrices on the two DMRS (the specific calculation method for the noise-interference correlation matrices on the DMRS (in this paper, the noise-interference correlation matrix of the DMRS can be interchanged with the noise-interference correlation matrix of each RB (in the spatial domain) on the DMRS) as the noise-interference correlation matrix of that data symbol to perform interference whitening detection. That is, the received data is divided into segments in the time domain.

[0127] Scenario 2: In the time domain, interference is only distributed on the 4th and 5th data symbols. The RSSI of the 14 symbols, including the data symbols and DMRS, is calculated. In this scenario, for the 4th and 5th data symbols, if the difference between their RSSI and the RSSI of any DMRS is greater than a predetermined threshold X, then for all data symbols, the noise-interference correlation matrix on the nearest DMRS is used as the noise-interference correlation matrix for that data symbol to perform interference whitening detection. That is, the received data is divided into two segments in the time domain, with the segment boundary being the middle of two adjacent DMRSs.

[0128] Scenario 3: In the time domain, the interference is only distributed in the first slot, i.e., the first 7 data symbols. Calculate the RSSI of the 14 symbols including the data symbols and DMRS. In this scenario, for each of the first 7 data symbols, if the difference between the RSSI of any data symbol and the RSSI of any DMRS is greater than a predetermined threshold X, the processing method is the same as in Scenario 2, that is, the received data is divided into two segments in the time domain.

[0129] Step 1: Obtain the spatial noise interference correlation matrix for each RB on the DMRS. There can be multiple DMRSs; when there are multiple DMRSs, calculate the spatial noise interference correlation matrix for each RB on each DMRS separately. The following description uses a single DMRS as an example:

[0130] Estimating the noise interference correlation matrix for each RB, the noise interference correlation matrix for the b-th RB is as follows:

[0131]

[0132] The noise interference correlation matrix has a dimension of Nr*Nr, where Nr represents the number of receiving antennas at the receiver. The diagonal elements of the matrix correspond to the autocorrelation values ​​of the noise and interference signals received on each antenna included in the node device, and can represent the sum of the power of the interference signal and the power of the noise. The off-diagonal elements of the matrix correspond to the cross-correlation values ​​of noise and interference signals received on the different antennas included in the node device, and can represent the power of the interference signal and the correlation of the interference signal between different antennas.

[0133] for The element in the i-th row and j-th column of the matrix represents the correlation results of the noise and interference signals between antenna i and antenna j of the b-th RB. It can be calculated using the following formula:

[0134]

[0135] Where y represents the DMRS signal received by the base station in the frequency domain, with dimensions Nr*K, where K represents the number of resources in the frequency domain, and Nr represents the number of receiving antennas at the receiver. H represents the DMRS channel response of the target UE obtained after channel estimation, with dimensions Nt*Nr*K, where Nt represents the number of transmitting antennas at the terminal, and x represents the DMRS signal transmitted by the target UE, with dimensions K*1. Where i, j = 0, ..., Nr-1, k = 0, ..., K-1, nt = 0, ..., Nt-1, Nr represents the number of receiving antennas at the receiver, K represents the number of resources in the frequency domain, and Nt represents the number of transmitting antennas at the terminal. Furthermore, each RB includes 12 frequency domain resources.

[0136] For example, when the system subcarrier spacing is 15kHz and the bandwidth is 10MHz, the maximum number of RBs is 52, and b = 1, ..., 52. For instance, if the base station is configured with 8 receiving antennas and the terminal is configured with 1 transmitting antenna, then... It is an 8x8 square matrix.

[0137] Taking two DMRS as an example, based on the time-domain segmentation results in step 1, determine whether to average the spatial noise interference correlation matrices of the two DMRS. For example, if the segmentation result in step 1 is one segment, then for each RB, the spatial noise interference correlation matrices of the two DMRS are averaged, and subsequent steps are performed on the averaged noise interference correlation matrix; if the segmentation result in step 1 is two segments, then for each RB, the spatial noise interference correlation matrices of the two DMRS are not averaged, and subsequent steps are performed on the two noise interference correlation matrices separately.

[0138] Step 2: Feature Extraction

[0139] Based on the noise interference correlation matrix of each RB on the obtained DMRS Extract the interference signal features (also called the first feature, or other names) of each RB. If there are multiple DMRSs, and the segmentation result of step 1 is the number of segments equal to the number of DMRSs, then for each RB on each DMRS, the noise-interference correlation matrix is... Extract the interference signal features for each RB. If there are multiple DMRSs and the segmentation result in step 1 is a single segment, then the noise-interference correlation matrix for each RB on each DMRS is calculated. The interference signal features of each RB are extracted from the matrix after averaging the matrix along the dimension of the number of DMRS.

[0140] The interference signal features can be diagonal elements, second-diagonal elements, or elements of the entire noise interference correlation matrix for each RB. The interference signal features can include at least Nr noise interference correlation matrices. The diagonal elements, reflecting the interference intensity at each RB, are used to determine the subband segmentation of the interference in the frequency domain. For multiple frequency subbands obtained, the multiple RBs within each subband have similar interference distribution characteristics. For example, when extracting the 52 noise-interference correlation matrices estimated from the full bandwidth... When extracting diagonal elements, the extracted feature dimension is 52*8, assuming the base station is configured with 8 receiving antennas.

[0141] Step 3: Frequency domain segmentation detection:

[0142] The method for determining the subband segmentation of interference in the frequency domain can be based on AI / ML or on non-AI / ML methods.

[0143] If based on a non-AI / ML approach, for example, the first feature extracted in step 2 is averaged within that RB to obtain a value (e.g., average) related to the first feature. Taking three RBs as an example, assuming they are RB1, RB2, and RB3, the first feature of RB1 is averaged within RB1 to obtain an average value (average 1) related to RB1; the first feature of RB2 is averaged within RB2 to obtain an average value (average 2) related to RB2; and the first feature of RB3 is averaged within RB3 to obtain an average value (average 3) related to RB3. If the difference between average 2 and average 1 is less than a predefined value X, then RB2 is not a segment boundary; if the difference between average 3 and average 2 is greater than the predefined value X, then RB3 is a segment boundary.

[0144] If based on AI / ML, the features of each RB extracted in step 2 are used as input to the model. The output of the AI / ML model (also called the first AI / ML model, or other names) is used to identify the sub-band distribution of interference in the frequency domain, thereby obtaining multiple frequency domain sub-bands. As one implementation, if a DMRS is included, the features of each RB on that DMRS (assuming each DMRS corresponds to 52 RBs) can be used as input to the AI / ML model, and the output of the AI / ML model is the segmentation result for these 52 RBs. As another implementation, if multiple (e.g., N) DMRSs are included, and the segmentation result in step 1 is N, the features of each RB on each DMRS (assuming each DMRS corresponds to 52 RBs) can be used as input to the AI / ML model, outputting multiple segmentation results. Each segmentation result corresponds to the segmentation result of an RB on one DMRS; that is, each segmentation result corresponds to the segmentation result of 52 RBs. As another implementation, if there are multiple (e.g., N) DMRSs and the segmentation result of step 1 is 1, the features of each RB (assuming each DMRS corresponds to 52 RBs) on the matrix after averaging the noise interference correlation matrices of the N DMRSs can be used as the input of the AI / ML model, and the output of the AI / ML model is the segmentation result of these 52 RBs.

[0145] There are no restrictions on AI / ML models. They can be multilayer perceptrons (MLP), convolutional neural networks (CNN), recurrent neural networks (RNN), long short-term memory networks (LSTM), gated recurrent units (GRU), or even Transformers, etc.

[0146] For AI / ML-based approaches, when the selected model is a CNN, the internal structure of the AI / ML model is as follows: Figure 11 As shown.

[0147] Input to the AI / ML model: For example, the first feature of each RB extracted in step 2, which can be a 52*8 dimension matrix.

[0148] The parameters involved in AI / ML models include: The AI / ML model's label, obtained through preprocessing of the training data (where the intensity and distribution of interference are known), indicates whether the current boundary element (RB) is a segment boundary. If the RB is a segment boundary, the label is 1; if the RB is a non-segment boundary, the label is 0. Figure 12 As shown, labels are set according to the distribution and intensity of interference. If the power of multiple (more than one) interferences is significantly greater than the power of a single interference, the last RB with the large interference power can be used as the segment boundary.

[0149] The structure and processing flow of the AI / ML model: The input data is convolved through two convolutional layers to obtain a 52*1 dimensional output. The CNN convolution output is converted into a probability representation using a sigmoid function. If the probability is greater than or equal to a certain threshold, the output is set to 1; otherwise, it is set to 0. Based on the 0 / 1 output of the 52*1 dimensional matrix, the final segmentation result is obtained, that is, segmentation is performed at position 1.

[0150] Step 4:

[0151] Based on the frequency domain sub-band of each segment, interference whitening processing is performed according to the configured interference scheme or the selected interference scheme. The selected interference scheme can be an AI / ML-based scheme (see the next embodiment for details).

[0152] This embodiment achieves interference segmentation in the time-frequency domain based on the distribution and intensity of the interference. The time-frequency domain signal is divided into signal segments with similar interference distribution characteristics, and each segment is subsequently whitened, improving interference cancellation performance while achieving noise reduction gain.

[0153] Next, the adaptive selection scheme of the AI-based interference cancellation algorithm proposed in this invention will be introduced, the process of which includes the following multiple stages, modes or operations:

[0154] After detecting time-frequency domain interference using AI, assuming multiple RBs are divided into N RB segments, the number of RBs included in the i-th RB segment is denoted as RB. i The following steps are performed within each RB segment.

[0155] Step 1: Data preprocessing:

[0156] The noise interference correlation matrix for each RB The first set of elements (e.g., all or part of the diagonal elements) are averaged along the antenna dimension as shown in the following formula, to obtain... The average value of each RB obtained It can reflect the interference plus noise intensity N+I of each RB, where N represents the statistical noise power and I represents the statistical interference power. Rx represents the number of receiving antennas at the receiver.

[0157]

[0158] The noise interference correlation matrix for each RB The modulus of the second part of the elements (e.g., all or part of the second diagonal elements, where the second diagonal elements are complex numbers) is averaged as shown in the following formula to obtain... The dimension of the average operation can be N, for example, N equals Rx-1, which can reflect the interference intensity I of each RB.

[0159]

[0160] Step 2: Feature Extraction

[0161] and The ratio can be approximately reflected in the noise-to-power ratio (N+I) / I within the b-th RB, or in the average dominant interference ratio (DIP = I / (N+I)). Based on (N+I) / I or DIP, the effective features (also called the second feature, or other names) of the b-th RB are extracted. For example, the maximum, minimum, and average values ​​of (N+1) / I within the RB segment can be extracted as the second feature.

[0162]

[0163] For example, the maximum, minimum, and average values ​​of DIP within the RB segment can be extracted as a second feature.

[0164]

[0165] Step 3: The second feature extracted based on each RB segment can be used as input to the subsequent AI / ML model (also known as the second AI / ML model, or other names). The optimal interference whitening process is selected based on the output of the second AI / ML model.

[0166] There are no restrictions on AI / ML models; they can be MLPs, CNNs, or Transformers, etc.

[0167] As one implementation method, when the selected model is an MLP, the internal structure of the second AI / ML model is as follows: Figure 13 As shown.

[0168] The input to the second AI / ML model: g1 and g2 are used as the network inputs, with a dimension of 52*1, where g1 represents and The average value of the ratio along the RB dimension, g2 represents and The minimum value of the ratio in the RB dimension is calculated using the following formula:

[0169]

[0170] As another way of saying this, g1 represents and The average value of the ratio along the RB dimension, g2 represents and The maximum value of the ratio in the RB dimension is calculated using the following formula:

[0171]

[0172] The parameters involved in the second AI / ML model include: the label of the second AI / ML model, which can be obtained by preprocessing the data used for training (for the training data, the optimal whitening scheme for obtaining the best detection result is known). The label dimension of the second AI / ML model is 3*1, with a value of 0 or 1, used to indicate whether the detection result of the three whitening schemes corresponding to the current input is correct. If the detection result of the whitening scheme corresponding to the current input (one of the three schemes IWNRB, IWNBW, and IWRB) is correct, the corresponding label is set to 1; the labels of the other two schemes with incorrect detection results are set to 0.

[0173] The structure and processing flow of the second AI / ML model: g1 and g2 are used as inputs to the second AI / ML model. After passing through an intermediate layer containing 16 nodes, the output is obtained, with a 3*1 dimension. The output Softmax is converted into a probability representation, setting the dimension with the highest probability to 1 and the other two dimensions to 0. Finally, the whitening scheme corresponding to the dimension with the highest probability is selected as the preferred whitening scheme.

[0174] In this embodiment, effective features are selected as input to the AI / ML model for each RB segment, and a preferred whitening scheme is chosen based on the AI / ML model to obtain optimal interference cancellation performance, thereby improving throughput performance. Furthermore, this embodiment, while achieving optimal interference cancellation performance, allows for the selection of a less complex whitening scheme based on the AI / ML model, further reducing the overall complexity of interference whitening.

[0175] Figure 14 A block diagram of a node device in a communication system according to an embodiment of the present disclosure is shown.

[0176] refer to Figure 14The node device 500 includes a transceiver 501 and a controller 502. The transceiver 501 is configured to transmit signals to and / or receive signals from the outside. The controller 502 is configured to perform the methods described above by the user device. The node device 500 may be implemented in hardware, software, or a combination of hardware and software to enable it to perform the methods described herein performed by a node device (e.g., a UE or a base station).

[0177] Those skilled in the art will understand that the illustrative embodiments described above are not intended to be limiting. It should be understood that any two or more of the embodiments disclosed herein can be combined in any combination. Furthermore, other embodiments may be utilized and other changes may be made without departing from the spirit and scope of the subject matter presented herein. It will be readily understood that aspects of the invention disclosed herein, as generally described herein and illustrated in the accompanying drawings, can be arranged, substituted, combined, separated, and designed in a variety of different configurations, all of which are contemplated herein.

[0178] Those skilled in the art will understand that the various illustrative logic blocks, modules, circuits, and steps described herein can be implemented in hardware, software, or a combination of both. To clearly illustrate this interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps are described above in the form of sets of functions. Whether such sets of functions are implemented in hardware or software depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art may implement the described sets of functions in different ways for each specific application, but such design decisions should not be construed as departing from the scope of this application.

[0179] The various illustrative logic blocks, modules, and circuits described in this application may be implemented or performed using a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. The general-purpose processor may be a microprocessor, but in alternatives, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors cooperating with a DSP core, or any other such configuration.

[0180] The steps of the methods or algorithms described in this application may be embodied directly in hardware, in a software module executed by a processor, or in a combination of both. The software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read and write information to / from the storage medium. In an alternative, the storage medium may be integrated into the processor. The processor and storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In an alternative, the processor and storage medium may reside as discrete components in the user terminal.

[0181] In one or more exemplary designs, the functionality may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functionality may be stored or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media, the latter including any medium that facilitates the transfer of a computer program from one location to another. Storage media may be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0182] The above description is merely an exemplary embodiment of this application and is not intended to limit the scope of protection of this application. The scope of protection of this application is determined by the appended claims.

Claims

1. A method performed by a node device in a communication system, comprising: Obtain the noise-interference correlation matrix corresponding to each RB among multiple resource blocks RB on the demodulation reference signal DMRS, wherein the diagonal elements of the matrix correspond to the autocorrelation values ​​of the noise and interference signals received on each antenna included in the node device, and the off-diagonal elements of the matrix correspond to the cross-correlation values ​​of the noise and interference signals received on different antennas included in the node device. Based on the noise interference correlation matrix corresponding to each RB, the first feature of each RB is obtained, wherein the first feature of each RB is related to the interference signal strength corresponding to each RB; Based on the first feature of each RB, the plurality of RBs are segmented, wherein the difference in the distribution of interference signals and / or the difference in the intensity of interference signals among the RBs belonging to the same RB segment is less than a first preset threshold; and Based on the segmentation results and the noise interference correlation matrix corresponding to each RB, a whitening scheme corresponding to each RB segment is obtained by using a second artificial intelligence / machine learning (AI / ML) model. Based on the obtained whitening scheme, whitening processing is performed on the corresponding RB segments on the data symbols.

2. The method according to claim 1, wherein, The method includes: Based on the Received Signal Strength Indication (RSSI) of the received signal, the demodulation reference signal (DMRS) for calculating the noise interference correlation matrix for data symbols is determined. Based on the determined DMRS, the noise interference correlation matrix corresponding to each RB on each data symbol is determined.

3. The method according to claim 1, wherein, Based on the noise interference correlation matrix corresponding to each RB, the first feature of each RB is obtained, including: The diagonal elements and / or some or all of the off-diagonal elements of the noise interference correlation matrix corresponding to each RB are used as the first feature corresponding to each RB.

4. The method according to claim 1, wherein, The segmentation of the plurality of RBs based on the first feature of each RB includes: Based on the first feature corresponding to each RB, the multiple RBs are segmented using a first AI / ML model.

5. The method according to claim 1, wherein, The segmentation of the plurality of RBs based on the first feature of each RB includes: The value associated with the first feature of each RB is compared with the value associated with the first feature of the previous RB of the RB. Based on the comparison results, it is determined whether each RB is the boundary of the RB segment.

6. The method according to claim 1, wherein, Based on the segmentation results and the noise interference correlation matrix corresponding to each RB, a whitening scheme corresponding to each RB segment is obtained by using a second artificial intelligence / machine learning (AI / ML) model, including: Based on the segmentation results and the noise interference correlation matrix corresponding to each RB, a second feature corresponding to each RB segment is obtained, wherein the second feature is related to the relative intensity of the noise and interference signal corresponding to each RB segment; Based on the second feature corresponding to each RB segment, a whitening scheme corresponding to each RB segment is obtained by using the second AI / ML model.

7. The method according to claim 6, wherein, The second feature corresponding to each RB segment is obtained based on the segmentation results and the noise interference correlation matrix corresponding to each RB, including: The first part of the noise interference correlation matrix corresponding to each RB is averaged, and the resulting average values ​​are used as the first average values ​​of each RB. The first average values ​​of each RB are related to the interference plus noise intensity of each RB. The second part of the noise interference correlation matrix corresponding to each RB is averaged, and the resulting average values ​​are used as the second average values ​​of each RB. The second average values ​​of each RB are related to the interference intensity of each RB. Based on the first and second average values ​​of each RB segment, the second feature corresponding to each RB segment is obtained.

8. The method according to claim 7, wherein, The first part of the elements includes some or all of the diagonal elements in the noise interference correlation matrix, and / or The second part of the elements includes some or all of the off-diagonal elements in the noise interference correlation matrix.

9. The method according to claim 7, wherein, The second feature corresponding to the RB segment includes at least one of the average, maximum, or minimum values ​​of the ratio of the first average value and the second average value of each RB included in the RB segment.

10. The method according to any one of claims 1 to 9, wherein, The first AI / ML model and / or the second AI / ML model includes at least one of the following: multilayer perceptron (MLP), convolutional neural network (CNN), recurrent neural network (RNN), and transformer.

11. The method according to any one of claims 1 to 9, wherein, The whitening scheme includes at least one of the following: IWNRB (Independent Whitening Based on RB Normalization); IWNBW (Independent Whitening Based on Signal Bandwidth Normalization); RB-based whitening (IWRB).

12. The method according to any one of claims 1 to 9, wherein, The node devices include user equipment (UE) and / or base stations.

13. A node device in a wireless communication system, comprising: A transceiver is configured to transmit and / or receive signals; as well as A controller, coupled to the transceiver and configured to perform the method as described in any one of claims 1-12.

Images