Optimized resource allocation for uplink configured grant in wireless networks

Abstract

Embodiments herein disclose a method optimizes Uplink (UL) Configured Grant (CG) allocation in a wireless network. The method involves computing, by a network entity (430), a Resource Block (RB) quality metric to select uplink frequency resources for a User Equipment (UE) (410). The network entity (430) then decides if the UE (410) is suitable for UL CG operation and selects the appropriate UL CG type based on the RB quality metric. Further, the network entity (430) shares the UL CG type and resource configuration with the Radio Resource Control (RRC) layer (1480) of the UE (410), and provides cross-layer feedback from Layer 2 to Layer 3 regarding the optimal UL CG type and resources.

Description

OPTIMIZED RESOURCE ALLOCATION FOR UPLINK CONFIGURED GRANT IN WIRELESS NETWORKS

[0001] Embodiments disclosed herein relate to wireless networks, and more particularly a method and network entity for optimizing Uplink (UL) Configured Grant (CG) allocation in a wireless network.

[0002] An Uplink (UL) Configured Grant (CG) operation is a semi-persistent allocation scheme that provides User Equipment (UEs) with periodic opportunities to transmit an uplink data without requiring a Physical Downlink Control Channel (PDCCH) grant each time. The efficiency and reliability of the UL CC operation are critical for maintaining high performance in systems supporting Ultra-Reliable Low Latency Communications (URLLC) and Massive Machine-Type Communications (mMTC). The current allocation strategies for the UL CG, whether too conservative or too aggressive, result in trade-offs between resource wastage and reliability.

[0003] If the network entity uses a conservative allocation strategy, the network entity can ensure higher reliability but may waste resources when allocations are made in regions with good channel conditions and low interference.

[0004] On the other hand, an aggressive allocation strategy risks reliability degradation, especially when uplink resources are allocated in regions with poor channel conditions and high interference.

[0005] The UL CG operations, specifically Type-1 and Type-2, are used in modern cellular communication systems (e.g., Fifth-Generation New Radio (5G NR) or the like) to provide UEs with periodic uplink resource allocations without requiring a new grant for each transmission. The Type-1 UL Configured Grant is controlled by the Radio Resource Control (RRC) layer (Layer-3), while the Type-2 is handled by the Medium Access Control (MAC) layer (Layer-2) using PDCCH addressed to a Cell Radio Network Temporary Identifier (CS-RNTI). These grants are critical for achieving URLLC and supporting applications with frequent, low-latency transmission requirements.

[0006] The Uplink Configured Grant operation allows periodic uplink transmissions using pre-configured resources, which are adjusted via parameters like Modulation and Coding Scheme (MCS), Repetition Factor (RF), Redundancy Version (RV), and others. The parameters are essential for managing the efficiency and reliability of uplink transmissions.

[0007] The settings for sending data from the UEto the network are managed by the RRC using a configuration called ConfiguredGrantConfig Information Element (IE). The configuration includes various parameters like MCS Table, MCS, RF, and RV. If the rrc-ConfiguredUplinkGrant IE is present within ConfiguredGrantConfig IE, the rrc-ConfiguredUplinkGrant IE indicates a Type-1 Configured Grant. If the rrc-ConfiguredUplinkGrant IE is not present, the rrc-ConfiguredUplinkGrant IE operates as a Type-2 Configured Grant. These settings determine some or all of the scheduling and resource allocation parameters.

[0008] In conventional systems, the configuration of uplink resources, such as the Physical Resource Block (PRB) allocation, is relatively static or based on fixed parameters determined by RRC signaling. These parameters do not typically adapt to real-time conditions, such as channel quality or interference, potentially leading to inefficiencies in resource usage. Furthermore, the existing methods do not incorporate cross-layer feedback (from Layer-2 to Layer-3) for the dynamic adjustment of UL CG parameters, leaving room for improvement.

[0009] FIG. 1 illustrates the impact of aggressive resource allocation on reliability in a wireless communication system. In another way, FIG. 1 illustrates the impact of conservative resource allocation on resource utilization in a wireless communication system. FIG. 1 consists of a table with three rows and multiple columns representing Resource Blocks (RBs) labeled from RB0 to RB16. RB Region is the first row, highlights and indicates the different RB regions without specifying further details. Channel Gain is the second row shows the channel gain levels for each RB, coded as follows: High channel gain (HIGH), Medium channel gain (MID), and Low channel gain (LOW). Interference is the third row represents the interference levels for each RB, coded as follows: Low interference (LOW), Medium interference (MID), and High interference (HIGH). The Typical Resource Quality are BEST: RB0 to RB2, and WORST: RB11 to RB13

[0010] When different RB regions have varying levels of channel gain and interference, an aggressive resource allocation strategy can lead to severely degraded reliability. This occurs if the UL CG allocation falls in the "worst quality" RB region, characterized by the lowest channel gain and highest interference. In such cases, the MCS, repetition, and redundancy version determined based on the UE's wideband performance become insufficient to guarantee the required reliability levels, despite less wastage of resources.

[0011] Also, FIG. 1 consists of a table with three rows and multiple columns representing RBs labeled from RB1 to RB16. The UL CG Region is the first row, highlights and indicates two distinct regions: (RB1 to RB8) is UL CG Region, and (RB9 to RB16) is No specific label. Channel Gain is the second row shows the channel gain levels for each RB, coded as follows High channel gain, Low channel gain, and Medium channel gain. Interference is the third row that represents the interference levels for each RB, coded as follows: High interference, Low interference, and Medium interference. Typical Resource Quality: BEST: RB0 to RB2, and WORST: RB9 to RB16.

[0012] When different RB regions have varying levels of channel gain and interference, the conservative resource allocation strategy can lead to underutilization of frequency resources. This occurs if the UL CG allocation falls in the "best quality" RB region, characterized by high channel gain and low interference. In such cases, the MCS, repetition, and redundancy version determined based on the UE wideband performance result in resource wastage due to the wideband performance being lower than the "best quality" RB region performance, even though it guarantees the required reliability levels.

[0013] FIG. 2 provides a detailed scenario where UL Configured Grant Type 1 is not suitable due to frequent fluctuations in the UE's physical layer conditions. Rapid changes in the UE's physical layer conditions make it hard to adapt transmission parameters, leading to inefficiency. Thus, UL CG Type 2 can be more adaptable to changing conditions, providing better performance in fluctuating environments. If the fluctuations are too high, it might be better to handle the UE as a normal one, using dynamic scheduling to ensure better control and reliability. Therefore, to address the following issues such as the network can aim to minimize resource usage without compromising the reliability of uplink transmission, identify the best frequency resources for the UE, determine if the UE's channel conditions are suitable for a scheduled grant and which type of configured grant is ideal, and configure the system to maintain sufficient reliability.

[0014] FIG. 2 illustrates how the network dynamically adapts to changing conditions to ensure efficient and reliable uplink transmission.

[0015] FIG. 2 provides details of the sequence diagram 200 for the UL CG Operation. At step 202, the UE experiences favorable channel conditions, which means the signal quality is good and there is minimal interference. At step 204, the radio network decides whether to allocate an UL CG to the UE based on its current conditions and requirements. At step 206, the RRC layer configures the Type 1 UL CG for the UE. This involves setting up the necessary parameters for the grant. At step 208, the RRC sends the ConfiguredGrantConfig IE to the UE, which includes the rrc-ConfiguredUplinkGrant. This IE contains the configuration details for the UL Configured Grant. At step 210, the MAC layer schedules the UL CG transmissions based on a predefined periodicity. At step 212, the Physical (PHY) layer sets up the Physical Uplink Shared Channel (PUSCH) for the UL CG Type-1. At step 214, the UE sets up the PUSCH according to the configuration received from the RRC. At step 216, the UE transmits the In-phase and Quadrature (I / Q) samples for the PUSCH to the PHY layer. At step 218, the PHY layer successfully decodes the PUSCH transmission. At step 220, the PHY layer sends the decoded PUSCH report to the MAC layer. At step 222, the MAC layer continues to schedule the UL CG transmissions based on the periodicity. At step 224, the channel conditions for the UE worsen, leading to increased interference and reduced signal quality. At step 226, the PHY layer attempts to set up the PUSCH for the UL Configured Grant Type-1 again. At step 228, the UE sets up the PUSCH according to the configuration received from the RRC. At step 230, the UE transmits the I / Q samples for the PUSCH to the PHY layer. At step 232, the PHY layer fails to decode the PUSCH transmission due to poor channel conditions. At step 234, the MAC layer continues to schedule the UL Configured Grant transmissions based on the periodicity. At step 236, the UE monitors the PDCCH scrambled with the Cell Radio Network Temporary Identifier (C-RNTI) and overwrites the Configured Grant if there is an overlap. At step 238, the RRC layer modifies or deactivates the Type-1 UL Configured Grant using RRC reconfiguration based on the current channel conditions and requirements. The sequence diagram 200 provides a comprehensive view of the steps involved in the UL CG Operation, highlighting the interactions between different layers and the UE.

[0016] Current methods allocate UL resources based on different signal quality measurements. They switch UL grant types depending on channel quality measurements. They also decide UL grant configurations based on information requests and responses from the UE. However, these methods do not include ways to determine if the UE should use UL CGs or how to calculate a quality metric for RBs.

[0017] The above information is presented as background information only to help the reader to understand the present invention. Applicants have made no determination and make no assertion as to whether any of the above might be applicable as prior art with regard to the present application.

[0018] The principal object of the embodiments herein is to disclose a system and method for optimizing Uplink (UL) Configured Grant (CG) allocation in a wireless network, so as to enhance resource utilization and transmission reliability in the wireless network.

[0019] Another object of the embodiment herein is to grade / select the uplink PUSCH frequency resources for UL configured grant operation based on past PUSCH decoding performance and PUSCH scheduling / channel parameters.

[0020] Another object of the embodiment herein is to decide whether the UE is suitable for UL configured grant operation based on the UE's channel conditions and past PUSCH decoding performance / selected uplink PUSCH frequency resources.

[0021] Another object of the embodiment herein is to decide which parameters are most suitable for the operation based on the decision for UL configured grant operation for the UE.

[0022] Another object of the embodiment herein is to provide cross layer feedback to Layer-3 from Layer-2 with the decided UL Configured Grant Type and it's suitable parameters

[0023] Another object of the embodiment herein is to utilize the Layer-2 uplink performance history of a User Equipment (UE) to provide cross-layer feedback to Layer-3, so as to ensure an optimal UL CG type and resource selection in the wireless network.

[0024] Another object of the embodiment herein is to determine whether the UE is suitable for UL CG operation, taking into account the channel conditions, interference, mobility patterns, and past Physical Uplink Shared Channel (PUSCH) decoding performance.

[0025] Another object of the embodiment herein is to decide the optimal type (Type 1 or Type 2) of UL CG to be used while optimizing UL CG allocation in the wireless network.

[0026] Another object of the embodiment herein is to determine the most suitable parameters for the UL Configured Grant operation (such as MCS, Repetition Factor (RF), RV, etc.), based on the selected frequency resources, RB quality metrics, and UE's historical performance, so as to ensure minimal resource wastage and improved transmission reliability.

[0027] Another object of the embodiment herein is to leverage Artificial Intelligence (AI) and Machine Learning (ML) capabilities for real-time decision-making in UL CG operations, improving the efficiency and accuracy of resource allocation.

[0028] Another object of the embodiment herein is to improve Key Performance Indicators (KPIs) such as UL capacity, throughput, and reliability in scenarios requiring UL CG operations, particularly in Ultra-Reliable Low Latency Communications (URLLC) and Massive Machine-Type Communications (mMTC) use cases.

[0029] Accordingly, the embodiments herein disclose methods for optimizing Uplink (UL) Configured Grant (CG) allocation in a wireless communication network. The method discloses computing, by a network entity, a Resource Block (RB) quality metric for selecting one or more uplink frequency resource regions for at least one User Equipment (UE), for the UL CG operation. The RB quality metric is computed based on a combination of a history of uplink resource allocation, one or more transmission parameters including Physical Uplink Shared Channel (PUSCH) scheduling parameters, past PUSCH decoding performance, and at least one estimated interference levels.　The method further discloses deciding, by the network entity, whether the UE is suitable for UL CG operation and whether to allocate at least one selected resource for the UL CG operation, based on the combination of the computed RB quality metric, one or more physical channel parameters including channel conditions of the UE, past PUSCH decoding performance, and the history of uplink resource allocation. The method further discloses selecting, by the network entity, a type of UL CG based on the outcome of said decision and the computed RB quality metric, wherein the type of UL CG is chosen to optimize the UL CG allocation based on factors such as UE conditions and interference levels. The method further discloses sharing, by the network entity, the type of the UL CG and the selected resource configuration with a Radio Resource Control (RRC) layer of the UE for the UL CG operation, wherein, upon receiving the type of the UL CG and the configuration, the UE selects one or more parameters most suitable for the UL CG operation based on the received configuration.

[0030] Accordingly, the embodiments herein provide a system including a network entity. The network entity is configured to compute a Resource Block (RB) quality metric for selecting one or more uplink frequency resource region for at least one User Equipment (UE) for the UL CG operation, based on a combination of a history of uplink resource allocation, one or more transmission parameters including Physical Uplink Shared Channel (PUSCH) scheduling parameters, past PUSCH decoding performance, and at least one estimated interference levels. The network entity is further configured to decide whether the UE is suitable for UL CG operation and whether to allocate at least one selected resource for the UL CG operation based on the combination of the computed RB quality metric, one or more physical channel parameters including channel conditions of the UE, past PUSCH decoding performance, and the history of uplink resource allocation. The network entity is further configured to select a type of UL CG based on the outcome of said decision and the computed RB quality metric. The type of UL CG is chosen to optimize the UL CG allocation based on factors such as UE conditions and interference levels. The network entity is further configured to share the type of the UL CG and the selected resource configuration with a Radio Resource Control (RRC) layer of the UE for the UL CG operation, wherein, upon receiving the type of the UL CG and the configuration, the UE selects one or more parameters most suitable for the UL CG operation based on the received configuration.

[0031] In an example, the regression-based batch processing model is only one way of implementing the RB quality decision model in a periodic fashion. The batch processing model indicates that the input parameters are collected over time for a period and then supplied to the model for generating the outputs. The reinforcement learning model is only one way of implementing the RB quality decision model in a real-time fashion.

[0032] These and other aspects of the embodiments herein will be better appreciated and understood when considered in conjunction with the following description and the accompanying drawings. It should be understood, however, that the following descriptions, while indicating at least one embodiment and numerous specific details thereof, are given by way of illustration and not of limitation. Many changes and modifications may be made within the scope of the embodiments herein without departing from the scope thereof, and the embodiments herein include all such modifications.

[0033] The embodiments disclosed herein are illustrated in the accompanying drawings, throughout which like reference letters indicate corresponding parts in the various figures. The embodiments herein will be better understood from the following description with reference to the drawings, in which:

[0034] FIG. 1 illustrates the impact of aggressive resource allocation on reliability in a wireless network, according to existing arts;

[0035] FIG. 2 illustrates how the network dynamically adapts to changing conditions to ensure efficient and reliable uplink transmission, according to existing arts;

[0036] FIG. 3 depicts a system for optimizing Uplink (UL) Configured Grant (CG) allocation in the wireless network, according to embodiments as disclosed herein;

[0037] FIG. 4 shows various hardware components of a User Equipment (UE) for UL CG allocation in the wireless network, according to embodiments as disclosed herein;

[0038] FIG. 5 shows various hardware components of a network entity, according to embodiments as disclosed herein;

[0039] FIG. 6 is a flowchart representing the overall model for UL CG decision-making the wireless network, according to embodiments as disclosed herein;

[0040] FIG. 7 is a flowchart depicting a method for updating the Resource Block (RB) quality metric in the wireless network, according to embodiments as disclosed herein;

[0041] FIG. 8 is a flowchart that illustrates the decision-making process for determining the type of UL CG in the wireless network, according to embodiments as disclosed herein;

[0042] FIG. 9 is a flow diagram illustrating a "UL Configured Grant Configuration Model" for UL communication in the wireless network, according to embodiments as disclosed herein; and

[0043] FIG. 10 illustrates an example neural network architecture used for predicting the RB quality metric in the wireless network using an RB quality decision model, according to embodiments as disclosed herein;

[0044] FIG. 11 illustrates an example classification model for the UL CG type decision using an Artificial Neural Network (ANN), according to embodiments as disclosed herein;

[0045] FIG. 12 illustrates an example architecture of a Random Forest Regressor used for UL CG configuration, according to embodiments as disclosed herein;

[0046] FIG. 13 provides a comparison between the current methods and the disclosed method for optimized resource allocation for UL CG Type 1, according to embodiments as disclosed herein;

[0047] FIG. 14 provides a detailed sequence diagram illustrating the process of UL CG Type 2 in the wireless network, according to embodiments as disclosed herein;

[0048] FIG. 15 is a flowchart that demonstrates the integration of AI / ML capable hardware in optimizing resource allocation for UL CG, according to embodiments as disclosed herein; and

[0049] FIG. 16 is a flowchart depicting a method 1700 for optimizing UL CG allocation in the wireless network, according to embodiments as disclosed herein.

[0050] The embodiments herein and the various features and advantageous details thereof are explained more fully with reference to the non-limiting embodiments that are illustrated in the accompanying drawings and detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein. The examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein may be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.

[0051] For the purposes of interpreting this specification, the definitions (as defined herein) will apply and whenever appropriate the terms used in singular will also include the plural and vice versa. It is to be understood that the terminology used herein is for the purposes of describing particular embodiments only and is not intended to be limiting. The terms "comprising", "having" and "including" are to be construed as open-ended terms unless otherwise noted.

[0052] The words / phrases "exemplary", "example", "illustration", "in an instance", "and the like", "and so on", "etc.", "etcetera", "e.g.," , "i.e.," are merely used herein to mean "serving as an example, instance, or illustration." Any embodiment or implementation of the present subject matter described herein using the words / phrases "exemplary", "example", "illustration", "in an instance", "and the like", "and so on", "etc.", "etcetera", "e.g.," , "i.e.," is not necessarily to be construed as preferred or advantageous over other embodiments.

[0053] Embodiments herein may be described and illustrated in terms of blocks which carry out a described function or functions.　These blocks, which may be referred to herein as managers, units, modules, hardware components or the like, are physically implemented by analog and / or digital circuits such as logic gates, integrated circuits, microprocessors, microcontrollers, memory circuits, passive electronic components, active electronic components, optical components, hardwired circuits and the like, and may optionally be driven by a firmware. The circuits may, for example, be embodied in one or more semiconductor chips, or on substrate supports such as printed circuit boards and the like.　The circuits constituting a block may be implemented by dedicated hardware, or by a processor (e.g., one or more programmed microprocessors and associated circuitry), or by a combination of dedicated hardware to perform some functions of the block and a processor to perform other functions of the block.　Each block of the embodiments may be physically separated into two or more interacting and discrete blocks without departing from the scope of the disclosure.　Likewise, the blocks of the embodiments may be physically combined into more complex blocks without departing from the scope of the disclosure.

[0054] It should be noted that elements in the drawings are illustrated for the purposes of this description and ease of understanding and may not have necessarily been drawn to scale. For example, the flowcharts / sequence diagrams illustrate the method in terms of the steps required for understanding of aspects of the embodiments as disclosed herein. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the present embodiments so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having the benefit of the description herein. Furthermore, in terms of the system, one or more components / modules which comprise the system may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the present embodiments so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having the benefit of the description herein.

[0055] The accompanying drawings are used to help easily understand various technical features and it should be understood that the embodiments presented herein are not limited by the accompanying drawings. As such, the present disclosure should be construed to extend to any modifications, equivalents, and substitutes in addition to those which are particularly set out in the accompanying drawings and the corresponding description. Usage of words such as first, second, third etc., to describe components / elements / steps is for the purposes of this description and should not be construed as sequential ordering / placement / occurrence unless specified otherwise.

[0056] The embodiments herein disclose a system and method to optimize Uplink (UL) Configured Grant (CG) allocation in a wireless network. This involves computing a Resource Block (RB) quality metric based on various parameters and deciding the suitability and type of UL CG for a User Equipment (UE). The system uses Artificial Intelligence (AI) and Machine Learning (ML) hardware to make and optimize these decisions. The decisions and parameters are shared with the Layer-3 Radio Resource Control (RRC) for a better setup. This system improves reception by choosing the best frequency resources and error recovery settings, making resource allocation more efficient and reliable.

[0057] In the proposed method, the network entity can aim to minimize resource usage without compromising the reliability of uplink transmission, identify the best frequency resources for the UE, determine if the UE's channel conditions are suitable for a scheduled grant and which type of configured grant is ideal, and configure the system to maintain sufficient reliability.

[0058] In the proposed method, the network entity can dynamically optimize resource allocation by taking into account the UE's past transmission performance and channel conditions at a given time. Artificial Intelligence (AI) and Machine Learning (ML)-enhanced techniques present a significant opportunity to address the challenges by leveraging historical data (e.g., past PUSCH decoding performance, RB quality metrics, and scheduling parameters) to optimize resource allocation and ensure the reliability of the uplink transmission. Based on the proposed method, cross-layer feedback from Layer-2 to Layer-3 can significantly improve the configuration of UL CG based on real-time data from lower layers.

[0059] Referring now to the drawings, and more particularly to FIGS. 3 through 16, where similar reference characters denote corresponding features consistently throughout the figures, there is shown at least one embodiment.

[0060] Embodiments herein ensures efficient and reliable UL CG allocation, improving the overall performance of the wireless network. A network entity calculates an RB quality metric using historical data, transmission parameters, and estimated interference levels. This metric helps in selecting the best uplink frequency resources for the UE. Based on the computed RB quality metric and physical channel parameters, the network entity decides if the UE is suitable for UL CG operation and determines whether to allocate the selected resources for UL CG. The network entity selects the appropriate type of UL CG (Type 1 or Type 2) based on the suitability decision and RB quality metric. The network entity (430) shares the UL CG type and resource configuration with the RRC layer of the UE. This ensures that the UE selects the most suitable parameters for UL CG operation. The network entity provides feedback from Layer 2 to Layer 3 regarding the optimal UL CG type and resources. This feedback loop helps in refining the configuration and improving performance. For example, a UE needs to transmit data periodically. The network entity calculates the RB quality metric using historical data and current transmission parameters. Based on this metric, it decides if the UE is suitable for UL CG operation and selects the appropriate UL CG type. The network entity then shares this configuration with the RRC layer of the UE. Throughout this process, feedback from Layer 2 to Layer 3 ensures that the configuration is continuously optimized for better performance. The cross-layer feedback from the layer-2 to the layer-3 allows the layer-3 to make an informed decision for deciding UE configuration based on the UE's operation / condition as observed in the layer-2. The layer-2 has detailed history for the UE with respect to UL scheduling, decoding and channel conditions which can be appropriately utilized to judge the reliability / efficiency of each UL CG configuration. Assisted by the reliability / efficiency of each UL CG configuration judged from Layer-2 data, Layer-3 decisions improve the performance.

[0061] FIG. 3 depicts a system 400 for optimizing Uplink (UL) Configured Grant (CG) allocation in the wireless network. In an embodiment, the system 400 includes the UE 410 and a network entity 430. The system 400 ensures efficient communication and data processing between the UE 410 and the network. The wireless network can be, for example, but is not limited to, a 4G network, a 5G network, a 6G network, an open radio access network (ORAN) network, or any other 3GPP network. The network entity 430 in the wireless network is a crucial component that manages communication between UE 410 (like smartphones) and the core network. The network entity 430 allocates network resources to ensure efficient data transmission and reception. The network entity 430 processes signal to maintain communication quality and reliability. The network entity 430 manages the movement of UEs 410 across different network areas to ensure seamless connectivity. The network entity 430 ensures secure communication by managing encryption and authentication processes. The network entity 430 maintains the quality of service by prioritizing different types of network traffic. The network entity 430 can be, for example, but not limited to, a Next Generation Node B (gNB), eNodeB (eNB), Base Station Controller (BSC), Radio Network Controller (RNC), Access Point (AP), Small Cell, Distributed Antenna System (DAS), and core network elements, such as the Mobility Management Entity (MME) in Long Term Evolution (LTE) networks. The gNB is responsible for managing communication between the UE 410 and the core network. It handles tasks such as resource allocation, scheduling, and ensuring efficient data transmission. The UE 410 can be, for example, but not limited to, a laptop, a desktop computer, a notebook, a Device-to-Device (D2D) device, a Vehicle-to-Everything (V2X) device, a smartphone, a foldable phone, a smart TV, a tablet, an immersive device, an Internet of Things (IoT) device, and any other device capable of communicating using at least one 3GPP network.

[0062] FIG. 4 shows various hardware components of the UE 410 for UL CG allocation in the wireless network. In the embodiment shown herein, the UE 410 is communicated to the network entity 430. The UE 410 comprises a processor 412, a communicator 414, and a memory 416. The processor 412 is coupled with the communicator 414, and the memory 416.

[0063] In the embodiment, the processor 412 executes instructions, process data and handles the computational tasks of the UE 410. The processor 412 can be at least one of a single processor, a plurality of processors, multiple homogeneous or heterogeneous cores, multiple Central Processing Units (CPUs) of different kinds, microcontrollers, special media, and other accelerators. The processor 412 may be an Application Processor (AP), a graphics-only processing unit such as a Graphics Processing Unit (GPU), a Visual Processing Unit (VPU), and / or an AI-dedicated processor such as a Neural Processing Unit (NPU). The processor 412 can be configured to execute instructions stored in the memory 416.

[0064] In the embodiment, the communicator 414 is configured to handle communication between the UE 410 and the network entity 430. The communicator 414 ensures that data is transmitted and received efficiently. the communicator 414 includes an electronic circuit specific to a standard that enables wired or wireless communication. The communicator 414 is configured to communicate internally between internal hardware components of the UE 410 and with external devices via one or more networks. The communicator 414 can be, for example, but not limited to, modem, Wi-Fi module, Bluetooth module, cellular radio, Near Field Communication module (NFC), ZigBee module, and satellite communication module.

[0065] The memory 416 is configured to store instructions to be executed by the processor 412. In the embodiment shown herein, the memory 416 may comprise one or more volatile and non-volatile memory components that are capable of storing data and instructions to be executed. Examples of the memory 416 can be, but are not limited to, NAND, embedded Multimedia Card (eMMC), Secure Digital (SD) cards, Universal Serial Bus (USB), Serial Advanced Technology Attachment (SATA), solid-state drive (SSD), and so on. The memory 416 may also include one or more computer-readable storage media. Examples of non-volatile storage elements may include magnetic hard discs, optical discs, floppy discs, flash memories, or forms of electrically programmable memories (EPROM) or electrically erasable and programmable (EEPROM) memories. In addition, the memory 416 may, in some examples, be considered a non-transitory storage medium. The term "non-transitory" may indicate that the storage medium is not embodied in a carrier wave or a propagated signal. However, the term "non-transitory" should not be interpreted to mean that the memory 416 is non-movable. In certain examples, a non-transitory storage medium may store data that can, over time, change (e.g., in Random Access Memory (RAM) or cache).

[0066] Although FIG. 4 shows various hardware components of the UE 410, but it is to be understood that other embodiments are not limited thereto. In other embodiments, the UE 410 may include less or more components. Further, the labels or names of the components are used only for illustrative purposes and do not limit the scope of the invention. One or more components can be combined to perform the same or substantially similar function in the UE 410.

[0067] FIG. 5 shows various hardware components of the network entity 430. The network entity 430 comprises a processor 432, a communicator 434, and a memory 436. The processor 432 is coupled with the communicator 434, and the memory 436.

[0068] In the embodiment, the processor 432 handles computational tasks required for network operations. The processor 432 processes a user request and manages the data flow to the UE 410. The memory 436 stores data and instructions for processor 432 of the network entity 430. The memory 436 holds information about active connections and user data to ensure efficient network management. The communicator 434 manages communication between the network entity 430 and the UE 410, as well as other network entities. The communicator 434 ensures that data packets are transmitted between the UE 410 and the network without interruptions.

[0069] In the embodiment, the processor 432 computes a RB quality metric for selecting one or more uplink frequency resource region for at least one UE for the UL CG operation, based on a combination of a history of UL resource allocation, one or more transmission parameters including Physical Uplink Shared Channel (PUSCH) scheduling parameters, past PUSCH decoding performance, and at least one estimated interference levels. The processor 432 computes the RB quality metric and updates in real-time using the PUSCH scheduling parameters and decoding results, which is based on at least one of, a regression-based model for batch processing updates, and a reinforcement learning model for real-time updates. The RB quality update is processed only for the RB allocation region for the PUSCH reception under consideration. The history of uplink resource allocation includes information about previous UL transmissions. The past PUSCH decoding performance are based on feedback from a receiver. The one or more transmission parameters comprise at least one of a modulation scheme, a coding rate, and a power control level, wherein the power control level refers to the transmission power used by the UE 410 when sending data to the network. The UL CG allocation is further optimized by utilizing AI / ML hardware capable of evaluating regression models, reinforcement learning models, and classification models to make real-time decisions on the grant type, resource selection, and grant configuration.

[0070] Further, the processor 432 decides whether the UE is suitable for UL CG operation and whether to allocate at least one selected resource for the UL CG operation based on the combination of the computed RB quality metric, one or more physical channel parameters including channel conditions of the UE, past PUSCH decoding performance, and the history of uplink resource allocation. The decision of whether to perform the UL CG or not and which type of UL CG to use is based on the judgment of whether the UE is stationary with respect to physical layer characteristics, determined using a classification model, and based on the signaling overhead while configuring the UL CG type. The deciding the level of channel fluctuation and considering it during the selection of the UL CG type can minimize the signaling overhead by reducing need for Layer-3 parameter reconfiguration. Additional considerations like omitting UEs that have been judged for "No UL Configured Grant Operation" due to channel fluctuations will greatly reduce unwanted alternating between UL CG configurations.

[0071] Further, the processor 432 selects a type of UL CG based on the outcome of said decision and the computed RB quality metric, wherein the type of UL CG is chosen to optimize the UL CG allocation based on factors such as UE conditions and interference levels. The type of UL CG is one of, Type 1, and Type 2. The Type 1 is configured by the RRC layer (1440), and the RRC layer (1440) is identified as Layer 3. The Type 2 is configured by the Medium Access Control (MAC) (1470) layer using Physical Downlink Control Channel (PDCCH) addressed to CS-RNTI and the MAC layer (1430) is identified as Layer 2. The configuration for the UL CG comprises at least one of selecting resource allocation, Repetition Factor (RF), Redundancy Version (RV), and Modulation and Coding Scheme (MCS). The decision of the configuration for the UL CG to be used is made based on the computed RB quality metric, the history of uplink resource allocation, one or more transmission parameters, and past PUSCH decoding performance.

[0072] Further, the processor 432 shares the type of the UL CG and the selected resource configuration with a Radio Resource Control (RRC) layer 1480 of the UE 410 for the UL CG operation, wherein, upon receiving the type of the UL CG and the configuration, the UE 410 selects one or more parameters most suitable for the UL CG operation based on the received configuration. Further, the processor 432 provides cross-layer feedback from Layer 2 to Layer 3 regarding the optimal UL CG type and the optimal resources for the determined grant type.

[0073] Embodiments herein may include one or more processors 432 for performing the functions associated with the AI / ML hardware 1650. At least one of the plurality of modules may be implemented through the AI / ML hardware 1650 for making real-time decisions on the grant type, resource selection, and grant configuration. Functions associated with the AI / ML hardware 1650 may be performed through the non-volatile memory and the volatile memory. The AI / ML hardware 1650 evaluates the quality of RBs based on historical data and real-time updates. Examples of the AI / ML hardware 1650 can include, but are not limited to, Graphics Processing Units (GPUs), Tensor Processing Units (TPUs), Field-Programmable Gate Arrays (FPGAs), Neural Processing Units (NPUs), Wafer-Scale Engines, and AI Accelerators. The predefined operating rule or AI / ML hardware 1650 is provided through training or learning.

[0074] Here, being provided through learning means that a predefined operating rule or AI / ML hardware 1650 of a desired characteristic is made by applying a learning algorithm to a plurality of learning data. The learning may be performed in a device itself in which AI / ML according to an embodiment is performed, and / or may be implemented through a separate server / system. Examples of learning algorithms include, but are not limited to, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning.

[0075] Although FIG. 5 shows various hardware components of the network entity 430, but it is to be understood that other embodiments are not limited thereto. In other embodiments, the network entity 430 may include less or more components. Further, the labels or names of the components are used only for illustrative purposes and do not limit the scope of the invention. One or more components can be combined to perform the same or substantially similar function in the network entity 430.

[0076] FIG. 6 is a flowchart 600 representing the overall model for UL CG decision-making the wireless network. The gNB Layer-2 data for the UE used for deciding UL CG configuration will influence further configuration and may modify the data received in the future due to change in configuration. The gNB Layer-3 (RRC) is supplied with the outcome of the decision models to decide the final UL CG configuration and deliver it to the UE and gNB Layer-2. At step 602, UE channel data is the starting point where the data from the UE channel is collected to gather all relevant information about the channel conditions experienced by the UE 410. At step 604, the quality of the RB is updated based on the UE channel data to ensure that the most current and accurate RB quality metrics are used for decision-making. At step 606, UL CG Type Decision is made regarding the type of UL CG based on the updated RB quality to determine whether UL CG Type 1 or Type 2 is more suitable for the current conditions. At step 608, resources and parameters for the UL CG are selected based on the type of decision to configure the UL CG with the optimal resources and parameters for efficient communication.

[0077] At step 610, the results from decoding the PUSCH are considered to evaluate the success of the uplink transmission and adjust future configurations accordingly. At step 612, parameters for scheduling the PUSCH are determined to ensure that the PUSCH is scheduled efficiently, taking into account the current network conditions. At step 614, scheduling for the UL CG PUSCH is performed to allocate the necessary resources for the UL CG PUSCH transmission. At step 616, normal scheduling for the PUSCH is also performed to handle regular PUSCH transmissions that are not part of the UL CG. At step 618, a final decision on the UL CG is made, taking into account the scheduling parameters and decoding results to finalize the UL CG configuration for optimal performance. At step 620, the RRC makes a decision regarding the UL Configured Grant to ensure that the UL CG configuration aligns with the overall network strategy and policies. At step 622, the final decision is sent to the RRC to implement the UL CG configuration in the network.

[0078] The flowchart 600 shows the interactions and dependencies between these different processes and decision points, illustrating the complexity and interrelated nature of the UL CG decision-making process in a communication system.

[0079] FIG. 7 is a flowchart 700 depicting a method for updating the RB quality metric in the wireless network. The RB Quality Metric is a metric that indicates the past relative quality of the different frequency resources (RB regions) under consideration as observed from the UE's uplink operation history at gNB Layer-2 by accounting for the uplink scheduling and decoding parameters used. It can be either be defined as a value defined within a well-defined range or set of discrete values, as needed. The flowchart shows the process of using PUSCH decoding results and scheduling parameters as inputs to a decision model, which then updates the RB quality metric. This update is specifically for the RB allocation region for the PUSCH reception under consideration. The process starts with the PUSCH decoding results, which include Cyclic Redundancy Check (CRC), Signal-to-Interference-plus-Noise Ratio (SINR), and noise plus interference estimates. The PUSCH decoding results are used to assess the quality of the received signal. The CRC is Used to detect errors in the received data. The SINR measures the quality of the received signal. The noise plus interference estimates provides an estimate of the noise and interference affecting the signal. The PUSCH scheduling parameters are essential for efficient and effective scheduling of uplink transmissions and include RB allocation, Power Headroom Report (PHR), and power control levels. These parameters are essential for efficient and effective scheduling of uplink transmissions. The RB allocation specifies the resource blocks allocated for the uplink transmission. The PHR indicates the difference between the maximum transmit power and the current transmit power. The power control levels help in adjusting the transmission power to maintain signal quality. At step 702, the PUSCH decoding results and PUSCH scheduling parameters as the inputs are fed into the RB Quality Decision Model. This can be either a regression-based model for batch processing of updates or a reinforcement learning model for real-time updates and to evaluate the quality of the RBs. The RB Quality Decision Model processes the input data to make informed decisions about the quality of the RBs. In the regression-based model for batch processing approach, the input data can be collected for a period and then a regression model can be used to update the RB quality metric for the past period. In the case of reinforcement learning model based real time updates, input data can be used in real time to update the RB quality metric without much overhead using a simple reinforcement learning model. Note that the RB Quality metric model is not limited to regression / reinforcement based models and can be implemented using conventional programming approaches. At step 704, the RB Quality Decision Model is pertaining to implementation of the model without adding much processing overhead. In the regression-based model for batch processing approach,input data can be collected for a period and then a regression model can be used to update the RB quality metric for the past period. In the case of reinforcement learning model based real time updates, input data can be used in real time to update the RB quality metric without much overhead using a simple reinforcement learning model. Note that the RB Quality metric model is not limited to regression / reinforcement based models and can be implemented using conventional programming approaches.

[0080] At step 706, based on the decision model's output, the quality of the RBs used in scheduling is updated. This ensures that the scheduling process has the most accurate and up-to-date information about the RB quality.

[0081] FIG. 8 is a flowchart 800 that illustrates the decision-making process for determining the type of UL CG in the wireless network. The decision to select the UL Configuration grant type is taken by Layer-3 (RRC). The model in FIG 8 generates the best decision, which is based on Layer-2 data / parameters, supplementing the Layer-3 decision process.

[0082] At step 802, inputs are fed into the decision model, but not limited to doppler estimate, Power Headroom Report (PHR) and power control level, beam change history, timing advance, and RB quality. The doppler estimate measures the change in frequency due to the relative motion between the transmitter and receiver. The PHR indicates the difference between the maximum transmit power and the current transmit power. The power control level helps in adjusting the transmission power. The beam change history tracks the changes in the beam direction to maintain a stable connection. The timing advance measures the time difference between the transmission and reception of signals to adjust the timing of transmissions. The RB quality assesses the signal quality of the allocated RBs. At step 804, UL CG Type Decision Model (Classification Model) takes the above inputs and determines whether to perform the UL CG operation or not. If the decision is not to perform the operation, it leads to "No UL Configured Grant Operation." If the decision is to perform the operation, the model further decides between "UL Configured Grant Type 1" and "UL Configured Grant Type 2." At step 806, there are outcomes shown as No UL Configured Grant Operation, UL Configured Grant Type 1, and UL Configured Grant Type 2. The No UL Configured Grant Operation indicates the system 400 decides not to proceed with the UL CG. The UL Configured Grant Type 1 indicates a specific type of CG operation is selected based on the inputs. The UL Configured Grant Type 2 indicates another type of CG operation is selected based on the inputs. The flowchart 800 shows the factors considered in making decisions about UL CGs, ensuring efficient and reliable communication in the network.

[0083] FIG. 9 is a flow diagram 900 illustrating the "UL Configured Grant Configuration Model" for UL communication in the wireless network. At step 902, inputs are fed into the UL Configured Grant Configuration Model, including but not limited to PUSCH decoding results, PUSCH scheduling parameters, and RB quality. The PUSCH decoding results includes the results of decoding the PUSCH, such as the CRC and the SINR. These metrics help in assessing the quality of the received signal and the correctness of the data. The PUSCH scheduling parameters includes various parameters used for scheduling the PUSCH, such as RB allocation, PHR, power control level, RV, and retransmission count. The RB quality (For RBs used in scheduling) refers to the quality metrics of the RBs that are used in scheduling. At step 904, the UL Configured Grant Configuration Model processes the inputs to make decisions regarding the configuration of the UL CG. The decision of the configuration for the UL CG to be used is made based on the computed RB quality metric and history of uplink resource allocation, transmission parameters, and decoding results. At step 906, the UL Configured Grant Configuration Model determines the optimal configuration for the uplink configured grant, ensuring efficient and reliable uplink communication based on past performance and current network conditions and includes the selection of resources for the UL CG, the RF, the RV, and the MCS. The flowchart 900 shows the factors considered in making decisions about UL CGs, ensuring efficient and reliable communication in the network.

[0084] FIG. 10 illustrates an example neural network architecture used for predicting the RB quality metric in a wireless network using the RB quality decision model, which can be either a regression-based model or a reinforcement learning model. The neural networks are one way to implement the Regression model, but it is not the only way. Classical (non-machine learning) approaches can also be used depending on the requirements keeping the inputs and the outputs same. In an embodiment (regression-based model) as shown in FIG. 10, the input layer receives several features related to the communication system, including, but not limited to PUSCH CRC, PUSCH SINR, signal-to-interference-plus-noise ratio, noise + interference estimate, power headroom, RB allocation and power control level. The PUSCH CRC is for error detection. The signal-to-interference-plus-noise ratio indicating signal quality. The noise + interference estimate is for estimation of noise and interference affecting the signal. The power headroom is difference between the maximum transmit power and the current transmit power. The RB allocation (one-hot encoded) indicates which RBs are allocated, encoded as a one-hot vector. The power control level helps in adjusting the transmission power.

[0085] The input layer is a fully connected layer that processes the input features. The network includes multiple hidden layers with Rectified Linear Unit (ReLU) activation functions. These layers are responsible for learning complex patterns and relationships in the data. The output layer uses a linear activation function to predict the RB quality metric. This layer provides the final output of the neural network.

[0086] The output of the network is the RB quality metric, which is used to assess the quality of a particular RB region. This neural network model is trained on a large and diverse sample of data to accurately predict the RB quality metric, which is crucial for optimizing the performance of the communication system.

[0087] The RB quality decision model uses a Q-learning based reinforcement learning approach for real-time updates. The RB quality decision model aims to predict the quality of RBs in the communication system. Q-learning is a type of reinforcement learning algorithm used to find the optimal action-selection policy for a given finite Markov Decision Process (MDP). The Bellman equation is used to update the Q-values, which represent the expected utility of taking a given action in a given state.

[0088] There are steps involved in the model, follow as:

[0089] a. Inputs:

[0090] PUSCH Parameters: These include various parameters related to the PUSCH, such as CRC, SINR, noise + interference estimate, power headroom, RB allocation, and power control level.

[0091] State (s): The current state of the system, represented by the PUSCH parameters.

[0092] Action (a): The action taken, which in this case is selecting a particular RB region for scheduling.

[0093] b. Quantization:

[0094] Since the range of the input parameters is finite, the continuous values of the input parameters are quantized into discrete levels. This reduces the overall size of the state space, making the model more efficient and suitable for real-time operation.

[0095] c. Q-value Update:

[0096] The Q-value for a state-action pair (s, a) is updated using the Bellman Equation:

[0097] Q(s,a) = Q(s,a) + α × (R(s,a) + γ × max(Q(s´,a´))―Q(s,a))

[0098] where: - α is the learning rate. - R(s,a)R(s, a) is the reward received after taking action (a) in state (s). - γ is the discount factor. - max(Q(s´,a´)) is the maximum Q-value for the next state s´ and all possible actions a´.

[0099] d. RB Quality Metric:

[0100] The RB quality metric is modeled as the Q-value. The Q-value represents the expected quality of a particular RB region given the current state (PUSCH parameters) and the action (selecting the RB region for scheduling).

[0101] The Q-learning based reinforcement learning approach is one way of implementing the reinforcement learning model. Classical (non- learning) approaches can also be used depending on the requirements keeping the inputs and the outputs same.

[0102] The model allows for real-time updates of the RB quality metric, ensuring that the scheduling decisions are based on the most current information. By quantizing the input parameters, the model reduces the complexity and size of the state space, resulting in faster updates and lower computational requirements. The Q-learning approach allows the model to adapt to changing network conditions and improve its predictions over time. This approach ensures that the RB quality metric is continuously updated based on real-time data, leading to more efficient and reliable communication in the network.

[0103] The Q-learning is one way of implementing the reinforcement learning model. Classical (non- learning) approaches can also be used depending on the requirements keeping the inputs and the outputs same.

[0104] FIG. 11 illustrates the example classification model for the UL CG type decision using an Artificial Neural Network (ANN). The classification model is designed to select the best decision for UL CG operation based on various input parameters. It consists of an input layer, three hidden layers with ReLU activation functions, and an output layer with Softmax activation. The classification model takes several input parameters, which are listed on the left side of FIG. 11, including but not limited to, timing advance, doppler estimate, beam change history, power headroom, RB quality metric vector, and power control level. The timing advance measures the time difference between the transmission and reception of signals to adjust the timing of transmissions. The doppler estimate measures the change in frequency due to the relative motion between the transmitter and receiver. The beam change history tracks the changes in the beam direction to maintain a stable connection. The power headroom indicates the difference between the maximum transmit power and the current transmit power. The RB quality metric vector assesses the signal quality of the allocated resource blocks. The power control level helps in adjusting the transmission power.

[0105] The input parameters are fed into the input layer, which is a fully connected layer. This layer processes the input data and passes it to the subsequent layers. The classification model contains three hidden layers, each utilizing the ReLU activation function. These layers are responsible for learning and extracting features from the input data through multiple transformations. The final layer is the output layer, which uses the Softmax activation function. This layer produces a probability distribution over the possible output classes, allowing the classification model to make a decision.

[0106] The output of the classification model is the "UL Configured Grant Type Decision," which is the selected decision for UL CG operation based on the input parameters. This neural network model is trained to process various input parameters related to the uplink communication system and make an informed decision on the type of UL CG to be used. The use of ReLU activation in the hidden layers helps in learning complex patterns, while the Softmax activation in the output layer ensures that the classification model can provide a probabilistic decision for the UL CG operation.

[0107] FIG. 12 illustrates the example architecture of a Random Forest Regressor used for UL CG configuration. The Random Forest Regressor is an ensemble learning model used to select the best set of configuration parameters for UL Configured Grant (CG) operation. It combines the outputs of multiple decision trees to make a final decision. The inputs to the model are listed and include PUSCH decoding results, PUSCH scheduling parameters, and RB quality. FIG. 12 shows multiple decision trees labeled as "Decision Tree 1," "Decision Tree 2," "Decision Tree 3," and "Decision Tree N." Each decision tree processes the inputs independently and generates its own set of outputs. The outputs from each decision tree are labeled as "OUTPUTS - 1," "OUTPUTS - 2," "OUTPUTS - 3," and "OUTPUTS - N." These outputs are intermediate results from each decision tree. The outputs from all the decision trees are combined using a technique called "Bagging". Bagging helps in reducing variance and improving the stability and accuracy of the model. The final outputs after bagging are listed in the FIG. 12 and include UL CG resource configuration (The optimal configuration of resources for the UL CG operation), RF (The number of times the transmission is repeated to ensure reliability), RV (The version of redundancy used in the transmission.), MCS (The scheme used for modulating and coding the data).

[0108] FIG. 12 demonstrates how a Random Forest Regressor can be used as a Multi Output Regression Model to select the best set of configuration parameters for UL CG configuration. The ensemble learning approach ensures that the final configuration is determined by averaging the results from multiple decision trees, leading to more robust and accurate predictions. This method ensures efficient and reliable uplink communication by optimizing the configuration parameters based on real-time data and past performance.

[0109] FIG. 13 provides a comparison between the current methods and the disclosed method for optimized resource allocation for UL CG Type 1. The next-generation Node B (gNodeB) makes a UL CG operation decision for the UE 410. The gNodeB sends a Type-1 UL CG Configuration to the UE 410. This includes the ConfiguredGrantConfigIE with rrc-ConfiguredUplinkGrant. The UE 410 schedules the UL CG based on the periodicity defined in the configuration. The UE 410 sets up the PUSCH for UL CG Type-1. The UE 410 transmits PUSCH In-phase and Quadrature samples (I / Q) samples to the gNodeB. These samples represent the I (In-phase) and Q (Quadrature) components of the signal transmitted on the PUSCH. The I / Q samples are used in digital signal processing to represent the amplitude and phase of the signal, which are essential for accurate decoding and analysis of the transmitted data. The gNodeB decodes the PUSCH samples received from the UE 410. The gNodeB generates a decoded PUSCH report based on the received samples. The process repeats based on the periodicity configured in the initial grant configuration. The UE monitors the PDCCH scrambled with Cell Radio Network Temporary Identifier (C-RNTI) and overwrites the CG if there is an overlap. The gNodeB can modify or deactivate the Type-1 UL CG using RRC reconfiguration. This approach addresses the challenges mentioned in the current methods and provides a more efficient solution for UL CG operations.

[0110] FIG. 13 provides details of the UL CG Operation for a UE 410 in a 5G network, as illustrated in the sequence diagram 1300. At step 1302, the gNodeB decides whether the UE 410 is suitable for UL CG operation based on various factors such as channel conditions, past PUSCH decoding performance, and uplink resource allocation history. At step 1304, the gNodeB configures the Type 1 UL CG and sends this configuration to the RRC layer 1480. At step 1306, the RRC layer 1480 includes the ConfiguredGrantConfig Information Element (IE) with the rrc-ConfiguredUplinkGrant and sends it to the MAC layer 1470. At step 1308, the MAC layer 1470 schedules the UL CG based on the configured periodicity. At step 1310, the MAC layer 1470 sets up the PUSCH for UL CG Type-1 and sends the configuration to the PHY (Physical) layer 1460. At step 1312, the PHY layer 1460 sets up the PUSCH and sends the configuration to the RADIO layer 1450. At step 1314, the RADIO layer 1450 transmits the PUSCH for UL CG Type-1 to the UE 410. At step 1316, the UE 410 sends the PUSCH In-phase and Quadrature (I / Q) samples back to the RADIO layer 1450. At step 1318, the RADIO layer 1450 decodes the PUSCH and sends the decoded information to the PHY layer 1460. At step 1320, the PHY layer 1460 sends the decoded PUSCH report to the MAC layer 1470. At step 1322, the MAC layer 1470 repeats the UL CG operation based on the configured periodicity. At step 1324, the MAC layer 1470 monitors the PDCCH scrambled with Cell Radio Network Temporary Identifier (C-RNTI) and overwrites the CG if there is an overlap. At step 1326, the MAC layer 1470 modifies or deactivates the Type-1 UL CG using RRC reconfiguration and sends this information to the RRC layer 1480. The detailed sequence diagram 1300 illustrates the steps involved in the UL CG process, highlighting the interactions between different layers and components in the 5G network to manage and optimize uplink data transmission.

[0111] FIG. 14 provides a detailed sequence diagram 1500 illustrating the process of UL CG Type 2 in the wireless network. It shows the interaction between different components: UE 410, RADIO 1450, PHY 1460, MAC 1470, and RRC 1480 within the gNodeB. In an example, when the particular UL CG configuration is decided for the UE, the reliability of the uplink operation also needs to be considered. Selecting the right balance of conservative / aggressive allocation will allow for efficient resource utilization without sacrificing reliability. The physical layer characteristics can be assessed from past uplink scheduling and reception parameters / results for that particular UE.

[0112] The diagram details the process from the initial decision by the gNodeB to configure the UL CG, through the setup and periodic transmission of data by the UE, to the potential modification or deactivation of the grant. The gNodeB makes a decision regarding the UL CG operation for the UE 410. The gNodeB configures the Type-2 UL CG for the UE 410. The configuration IE for the configured grant (ConfiguredGrantConfigIE (without rrc-ConfiguredUplinkGrant)) is sent without the RRC configured uplink grant. The scheduling of the UL CG is based on a periodicity configuration. The PDCCH with CS-RNTI is used for Type-2 UL CG activation. The UE 410 sets up the PUSCH for UL CG Type-2. The UE 410 transmits PUSCH In-phase and Quadrature (I / Q) samples to the gNodeB. These samples represent the I (In-phase) and Q (Quadrature) components of the signal transmitted on the PUSCH. The gNodeB decodes the PUSCH samples received from the UE 410. The gNodeB generates a decoded PUSCH report based on the received samples. The process repeats based on the periodicity configured in the initial grant configuration. The UE 410 monitors the PDCCH scrambled with C-RNTI and overwrites the CG, if there is an overlap. The PDCCH with CS-RNTI is used for the deactivation of the Type-2 UL CG. The gNodeB can modify or deactivate the Type-2 UL CG using RRC reconfiguration.

[0113] FIG. 14 provides details of the UL CG Operation for a UE 410 in a 5G network, as illustrated in the sequence diagram 1400. At step 1402, the gNodeB (gNodeB is the base station in 5G networks) makes a decision regarding the UL CG operation for the UE 410 based on various factors such as the UE's 410 channel conditions, past uplink performance, and network policies. At step 1404, the gNodeB configures the Type-2 UL CG, which is managed by the MAC layer 1470 and used for more dynamic and flexible scheduling compared to Type-1 grants, which are managed by the RRC layer 1480. At step 1406, the gNodeB sends the ConfiguredGrantConfig IE to the UE 410. This IE contains the configuration details for the UL CG. If the rrc-ConfiguredUplinkGrant IE is not present, it indicates that a Type-2 CG is being used. At step 1408, the gNodeB schedules the UL CG based on a predefined periodicity, allowing the UE 410 to transmit data regularly without needing to request resources each time. At step 1410, the gNodeB sends the PDCCH message with the CS-RNTI to the UE 410. This message contains control information necessary for the UE 410 to understand its uplink transmission schedule. The gNodeB sends another PDCCH message with CS-RNTI to activate the Type-2 UL CG. This activation allows the UE 410 to start using the configured grant for uplink transmissions. At step 1412, the UE 410 sets up the PUSCH for the Type-2 UL CG. PUSCH is the channel used by the UE 410 to send data to the gNodeB. At step 1414, the UE 410 completes the setup of the PUSCH, ensuring that it is ready to transmit data according to the configured grant parameters. At step 1416, the UE 410 transmits data on the PUSCH using the Type-2 UL CG. This transmission is done periodically as per the grant configuration. At step 1418, the UE 410 sends the I / Q samples of the PUSCH transmission. These samples are used by the gNodeB to decode the transmitted data. At step 1420, the gNodeB decodes the PUSCH transmission from the UE 410. This involves processing the I / Q samples to extract the transmitted data. At step 1422, the gNodeB generates a report based on the decoded PUSCH transmission. This report includes information about the quality and success of the transmission. At step 1424, the entire process repeats based on the configured periodicity. The UE 410 continues to transmit data at regular intervals as per the UL CG configuration. At step 1426, the UE 410 monitors the PDCCH scrambled with the C-RNTI. If there is an overlap with the configured grant, the UE 410 may overwrite the existing grant configuration to avoid conflicts. At step 1428, the gNodeB sends another PDCCH message with CS-RNTI to the UE 410. This message may contain updates or changes to the UL CG configuration. The gNodeB sends a PDCCH message with CS-RNTI to deactivate the Type-2 UL CG. This deactivation stops the periodic uplink transmissions. At step 1430, the gNodeB may modify the Type-2 UL CG using RRC reconfiguration. This involves updating the grant parameters to adapt to changing network conditions or UE 410 requirements.

[0114] FIG. 15 is a flowchart 1500 that demonstrates the integration of AI / ML capable hardware 1650 in optimizing resource allocation for UL CG, highlighting the steps involved in decoding, updating parameters, and making decisions based on ML model outputs. At step 1502, the process starts with the collection of PUSCH I / Q samples from the UE 410. At step 1504, these I / Q samples are decoded to produce a decoded PUSCH report. This step involves analyzing the signal to extract the transmitted data. At step 1506, several parameters are updated based on the decoded PUSCH report, including but not limited to timing advance, power control levels, beam change information, and Doppler estimate. Updating timing advance adjusts the timing of transmissions to ensure synchronization. Updating power control levels adjusts the transmission power to maintain signal quality. Updating beam change information tracks changes in the beam direction to maintain a stable connection. Updating Doppler estimate measures the change in frequency due to the relative motion between the transmitter and receiver.

[0115] At step 1508, the updated parameters are then fed into the ML models. The inputs include but not limited to, PUSCH scheduling parameters, PUSCH decoding parameters, power control level, timing advance, beam change frequency, and doppler estimate.

[0116] At step 1510, the ML models are evaluated and updated periodically. This includes evaluating the output of the RB quality decision model, periodically evaluating the output of the UL Configured Grant Decision Model, and periodically evaluating the output of the UL Configured Grant Configuration Model.

[0117] At step 1512, the ML models produces outputs which include determining whether to perform UL CG operation for the UE 410 and determining the optimal configuration for the UL CG.

[0118] At step 1514, based on the output from the UL Configured Grant Models, decisions are made regarding, whether to perform UL CG operation for the UE 410, the type of UL CG to be used, the parameters in the ConfiguredGrantConfig. The process is iterative, repeats as necessary to continuously optimize resource allocation, evaluating the parameters and models.

[0119] This flowchart 1500 demonstrates the integration of AI_ML_Capable_Hardware 1650 in optimizing resource allocation for UL CG, enhancing decision-making, optimizing resource allocation, and improving the overall performance and reliability of the network, and provides a more efficient solution for UL CG operations.

[0120] In another implementation, the operations are handled by a decision model processing unit instead of AI_ML_Capable_Hardware.

[0121] FIG. 15 highlights the use of specialized hardware, such as onboard accelerators and eXtensible Applications (xAPPs), to enhance AI / ML model performance in network operations. These components, deployed via the Near-Real-Time RAN Intelligent Controller (Near-RT RIC), enable quick and efficient evaluation of decision models, optimizing resource allocation and network performance. Onboard accelerators / xAPPs via Near-RT RIC can evaluate UL Configured Grant (CG) Decision Models in any AI / ML hardware blocks. The straightforward implementation of these models, which are not very deep networks, allows for periodic evaluation and updates, ensuring optimal resource allocation. This enables the Base Station (BS) to support more devices with fewer resources while maintaining reliability. Also, the machine learning models often require SIMD (Single Instruction Multiple Data) capable hardware to perform the above mentioned operations. In that scenario, the input to ML Models 1608 is handled by "Input to Decision Models" and "Output form ML Models 1612" is handled by "Output from Decision Models".

[0122] 3GPP Release-18 introduces standardized AI / ML enhancements to Radio Access Network (RAN). RAN is a part of a mobile telecommunication system that connects individual devices to other parts of a network through radio connections. The RAN is responsible for managing the radio resources and ensuring efficient communication between UE 410 (like smartphones) and the core network. With additional AI / ML capable hardware 1650, optimized scheduling and configuration become feasible. The system 400 aims to optimize UL CG allocation using Cross-Layer feedback from Layer-2 to Layer-3, leveraging AI / ML RAN capabilities.

[0123] FIG. 16 is a flowchart depicting a method 1600 for optimizing UL CG allocation in the wireless network. The operations (1602-1612) are handled by the network entity 430.

[0124] At step 1602, the method discloses computing an RB quality metric for selecting one or more uplink frequency resource regions for at least one UE (410), for the UL CG operation. The RB quality metric is based on a combination of a history of uplink resource allocation, one or more transmission parameters including PUSCH scheduling parameters, past PUSCH decoding performance, and at least one estimated interference levels. The RB quality metric is computed and updated in real time using the PUSCH scheduling parameters and decoding results, based on at least one of, a regression-based model for batch processing updates, and a reinforcement learning model for real-time updates. The RB quality update is processed only for the RB allocation region for the PUSCH reception under consideration. The history of uplink resource allocation includes information about previous UL transmissions. The past PUSCH decoding performance is based on feedback from a receiver, and wherein the one or more transmission parameters comprise at least one of a modulation scheme, a coding rate, and a power control level. The power control level refers to the transmission power used by the UE 410 when sending data to the network. The UL CG allocation is further optimized by utilizing AI / (ML) hardware capable of evaluating regression models, reinforcement learning models, and classification models to make real-time decisions on the grant type, resource selection, and grant configuration.

[0125] At step 1604, the method further discloses deciding whether the UE (410) is suitable for UL CG operation and whether to allocate at least one selected resource for the UL CG operation, based on the combination of the computed RB quality metric, one or more physical channel parameters including channel conditions of the UE (430), past PUSCH decoding performance, and the history of uplink resource allocation. The decision of whether to perform UL CG or not and which type of UL CG to use, is based on the judgement of whether the UE (410) is stationary with respect to physical layer characteristics, determined using a classification model, and based on the signalling overhead while configuring the UL CG type.

[0126] At step 1606, the method further includes selecting a type of UL CG based on the outcome of said decision and the computed RB quality metric. The type of UL CG is chosen to optimize the UL CG allocation based on factors such as UE (410) conditions and interference levels. The type of UL configured grant is one of, a Type 1 and a Type 2. The Type 1 is configured by the RRC layer 1440, wherein the RRC layer 1440 is identified as Layer 3, and the Type 2, wherein a Type 2 is configured by the MAC layer 1430 using PDCCH addressed to CS-RNTI, wherein the MAC layer 1430 is identified as Layer 2.

[0127] At step 1608, the method further includes deciding the configuration for the UL CG comprises at least one of selecting resource allocation, repetition factor, RV, and MCS. The decision of the configuration for the UL CG to be used is made based on the computed RB quality metric, the history of uplink resource allocation, one or more transmission parameters and past PUSCH decoding performance.

[0128] At step 1610, the method further includes sharing the type of the UL CG and the selected resource configuration with the RRC layer 1440 of the UE 410 for the UL CG operation, wherein, upon receiving the type of the UL CG and the configuration, the UE 410 selects one or more parameters most suitable for the UL CG operation based on the received configuration.

[0129] At step 1612, the method further includes providing cross-layer feedback from Layer 2 to Layer 3 regarding the optimal UL CG type and the optimal resources for the determined grant type.

[0130] The various actions in method 1600 may be performed in the order presented, in a different order, or simultaneously. Further, in some embodiments, some actions listed in FIG. 1 may be omitted.

[0131] Ultra-Reliable Low Latency Communications (URLLC) and Massive Machine-Type Communications (mMTC) scenarios often require reliable repeated grants for frequent data reporting. As the number of such devices increases, and all of them need uplink resources, it becomes crucial to manage these resources optimally to support more devices with fewer resources. For example, a security camera network of drones monitors an environment and needs to upload data frequently. In an automated warehouse, bots update inventory data periodically. Remote-controlled operation of devices for tactile internet requires real-time data uploads. Therefore, the system 400 improve Key Performance Indicators (KPIs) such as UL capacity, throughput, and reliability in scenarios requiring UL CG operations, particularly in URLLC and mMTC use cases.

[0132] The embodiments disclosed herein can be implemented through at least one software program running on at least one hardware device. The modules shown in FIG. 2 to FIG. 4include blocks which can be at least one of a hardware device, or a combination of hardware device and software module.

[0133] The embodiment disclosed herein methods for optimizing UL CG allocation in a wireless communication network. Therefore, it is understood that the scope of the protection is extended to such a program and in addition to a computer readable means having a message therein, such computer readable storage means contain program code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The method is implemented in at least one embodiment through or together with a software program written in e.g., Very high speed integrated circuit Hardware Description Language (VHDL) another programming language, or implemented by one or more VHDL or several software modules being executed on at least one hardware device. The hardware device can be any kind of portable device that can be programmed. The device may also include means which could be e.g. hardware means like e.g. an ASIC, or a combination of hardware and software means, e.g. an ASIC and an FPGA, or at least one microprocessor and at least one memory with software modules located therein. The method embodiments described herein could be implemented partly in hardware and partly in software. Alternatively, the invention may be implemented on different hardware devices, e.g. using a plurality of CPUs.

[0134] The foregoing description of the specific embodiments will so fully reveal the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and / or adapt for various applications such specific embodiments without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. Therefore, while the embodiments herein have been described in terms of at least one embodiment, those skilled in the art will recognize that the embodiments herein can be practiced with modification within the spirit and scope of the embodiments as described herein.

Claims

1.A method performed by a network entity for optimizing uplink (UL) configured grant (CG) allocation in a wireless network, comprising:computing a resource block (RB) quality metric to select at least one uplink frequency resource for at least one user equipment (UE) and a UL CG operation;determining whether to allocate at least one selected resource configuration for the UL CG operation, based on a combination of the computed RB quality metric and at least one physical channel parameter;selecting a type of the UL CG based on the determination; andtransmitting, to the at least one UE, a radio resource control (RRC) message indicating the selected type of the UL CG and the selected resource configuration.2.The method of claim 1, wherein the RB quality metric is computed based on a combination of a history of uplink resource allocation, one or more transmission parameters, and at least one estimated interference level, wherein the one or more transmission parameters comprise a physical uplink shared channel (PUSCH) scheduling parameter, a PUSCH channel parameter, and a past PUSCH decoding performance.3.The method of claim 1, wherein computing the RB quality metric comprises:obtaining a PUSCH decoding result comprising a cyclic redundancy check (CRC), a signal-to-interference-plus-noise ratio (SINR), and combination of noise and interference estimate;obtaining a PUSCH scheduling parameter comprising RB allocation, power headroom report (PHR), and power control level; andcomputing the RB quality metric by feeding the obtained PUSCH decoding result and the obtained PUSCH scheduling parameter to at least one of: an RB quality decision model, a regression-based batch processing model, a reinforcement learning model, a classification model, an artificial intelligence (AI) model and a machine learning (ML) model.4.The method of claim 1, wherein the at least one physical channel parameter comprises at least one channel condition of the UE, a past PUSCH decoding performance, and a history of uplink resource allocation.5.The method of claim 1, wherein the type of UL CG is selected to optimize the UL CG allocation based on a condition of the at least one UE, and an interference level.6.The method of claim 1, wherein, upon receiving the type of the UL CG and the resource configuration, the UE selects at least one parameter suitable for the UL CG operation based on the resource configuration.7.The method of claim 1, wherein the RB quality metric is computed and updated in real time using a PUSCH scheduling parameter and a PUSCH decoding result, based on at least one of:a regression-based batch processing model; anda reinforcement learning model.8.The method of claim 1, wherein the RB quality update is processed for the RB allocation region for the PUSCH reception under consideration.9.The method of claim 1, wherein the decision of whether to perform the UL CG or not and which type of the UL CG to use, is based on the judgement of whether the at least one UE is stationary with respect to physical layer characteristics, and is determined using a classification model, based on a signalling overhead while configuring the type of the UL CG.10.The method of claim 1, wherein the decision of the configuration for the UL CG to be used is made based on the computed RB quality metric, a history of uplink resource allocation, at least one transmission parameter and a past PUSCH decoding performance.11.The method of claim 10, wherein the history of uplink resource allocation includes information about previous UL transmissions, wherein the past PUSCH decoding performance is based on feedback from a receiver, and wherein the one or more transmission parameters comprise at least one of a modulation scheme, a coding rate, and a power control level, wherein the power control level refers to the transmission power used by the at least one UE while sending data to the network entity.12.The method of claim 1, wherein the configuration for the UL CG comprises at least one of: resource allocation, a repetition factor, redundancy version (RV), and modulation and coding scheme (MCS).13.The method of claim 1, wherein the type of UL configured grant is one of:a first type, wherein the first type is configured by a RRC layer; anda second type, wherein the second Type is configured by a medium access control (MAC) layer using physical downlink control channel (PDCCH) addressed to a configured scheduling-radio network temporary idnetifier (CS-RNTI).14.The method of claim 1, wherein the method comprises providing cross-layer feedback from a second layer to a third layer 3 regarding the optimal UL CG type and the optimal resources for the determined grant type, andwherein the UL configured grant allocation is further optimized by utilizing an artificial intelligence (AI) model or a machine learning (ML) model.15.A network entity, comprising:a processor; anda memory storing instructions;wherein the instructions, when executed by the processor, cause the network entity to:compute a resource block (RB) quality metric to select at least one uplink frequency resource for at least one user equipment (UE) and a UL CG operation;determinewhether to allocate at least one selected resource for the UL CG operation, based on a combination of the computed RB quality metric and at least one physical channel parameter;select a type of the UL CG based on the determination; andtransmit, to the at least one UE, a radio resource control (RRC) message indicating the selected type of the UL CG and the selected resource configuration.

Images