Coverage and capacity optimization in communication systems using artificial intelligence

Abstract

The present disclosure relates to Coverage and Capacity Optimization (CCO) in communication systems using Artificial Intelligence / Machine Learning (AI / ML). In one embodiment, the present disclosure discloses a method which comprises predicting, using a proactive CCO mechanism, a first CCO issue potentially occurring at a first time instance within a first Radio Access Network (RAN) node. The method comprises determining, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node to be applied at a second time instance for mitigating the predicted first CCO issue. The method comprises transmitting a notification to a second RAN node, about the first CCO configuration for the first RAN node to be applied at the second time instance.

Description

[0001] COVERAGE AND CAPACITY OPTIMIZATION IN COMMUNICATION SYSTEMS

[0002] USING ARTIFICIAL INTELLIGENCE

[0003] CROSS-REFERENCE TO RELATED APPLICATIONS

[0004]

[0001] This application claims priority to India Provisional Patent Application No. 202441086265, filed onNovember 8, 2024, and India Non-Provisional Patent Application No. 202441086265, filed on March 10, 2025, the entire contents of which are incorporated herein by reference.

[0005] FIELD

[0006]

[0002] The present disclosure relates to coverage and capacity optimization in communication systems using artificial intelligence.

[0007] BACKGROUND

[0008]

[0003] The information disclosed in this background section is only for enhancement of understanding of the general background of the disclosure and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already- known to a person skilled in the art.

[0009]

[0004] In a wireless communication system, achieving an optimal balance between coverage and capacity- is a critical aspect of network design and operation. The Third Generation Partnership Project (3GPP) introduced Coverage and Capacity- Optimization (CCO) as a key functionality- within a Self-Organizing Network (SON). The SON is an advanced network management framework that provides autonomous network optimization and self-healing mechanisms with minimal human intervention. CCO plays an important role within the SON and helps in providing seamless connectivity-, optimal network performance, and enhanced user experience.

[0005] Without proper CCO, networks may suffer from issues such as congestion, dropped calls, poor data speeds, and inefficient resource utilization. For example, if CCO is not implemented in a network, areas with weak signal coverage may experience frequent call drop and / or slow data rates, and devices on cell edge may struggle with weak signals and higher interference. Without CCO, handovers between cells may not be optimized leading to frequent call drops, and network resources (such as spectrum, transmission power, and hardware infrastructure) may not be efficiently utilized. Thus, a primary goal of CCO is to dynamically optimize coverage and capacity in response to dynamically changing network conditions such as varying traffic loads, changing UE distribution, but not limited thereto.

[0010] SUMMARY

[0011]

[0006] The present disclosure discloses various enhancements for the Artificial Intelligence (Al) or Machine Learning (ML) (AI / ML) based Coverage and Capacity Optimization (CCO) mechanisms. In one example, the present disclosure addresses problems related to degradation of network services and / or User Equipment (UE) performances in a communication system. To address these problems, the present disclosure discloses proactive CCO mechanisms by which neighboring RAN nodes are made aware of future CCO issues and / or future CCO configurations associated with a RAN node ahead of time, thereby preventing untended degradation of network services and / or User Equipment (UE) performances at the neighboring RAN nodes within the communication system.

[0012]

[0007] In one non-limiting embodiment, the present disclosure discloses a method which comprises predicting, using a proactive CCO mechanism, a first CCO issue potentially occurring at a first time instance within a first RAN node. The method further comprises determining, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node to be applied at a second time instance for mitigating the predicted first CCO issue. The method further comprises transmitting a notification to a second RAN node, about the first CCO configuration for the first RAN node to be applied at the second time instance.

[0013]

[0008] In one non-limiting embodiment, the present disclosure discloses an apparatus which is configured to predict, using a proactive CCO mechanism, a first CCO issue potentially occurring at a first time instance within a first RAN node. The apparatus is further configured to determine, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node to be applied at a second time instance for mitigating the predicted first CCO issue. The apparatus is further configured to transmit a notification to a second RAN node, about the first CCO configuration for the first RAN node to be applied at the second time instance.

[0014]

[0009] In one non-limiting embodiment, the present disclosure discloses a non-transitory computer readable media storing one or more computer executable instructions which, when executed by an apparatus, cause the apparatus to predict, using a proactive CCO mechanism, a first CCO issue potentially occurring at a first time instance within a first RAN node. The instructions further cause the apparatus to determine, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node to be applied at a second time instance for mitigating the predicted first CCO issue. The instructions further cause the apparatus to transmit a notification to a second RAN node, about the first CCO configuration for the first

[0015] RAN node to be applied at the second time instance. BRIEF DESCRIPTION OF THE DRAWINGS

[0016]

[0010] Features, aspects, and advantages of embodiments of the disclosure will be described below with reference to the accompanying drawings, in which like reference numerals denote like elements, and wherein:

[0017] [Oil] FIG. 1 illustrates a high-level block diagram 100 of a disaggregated architecture of a communication system.

[0018]

[0012] FIG. 2 illustrates a high-level block diagram 200 of a communication system comprising a plurality of Radio Access Network (RAN) nodes.

[0019]

[0013] FIG. 3 illustrates an example signaling procedure 300 of a reactive Coverage and Capacity Optimization (CCO) mechanism.

[0020]

[0014] FIG. 4 illustrates a high-level block diagram of another communication system 400 comprising a plurality of RAN nodes.

[0021]

[0015] FIG. 5 illustrates an example signaling procedure 500 of a proactive CCO mechanism.

[0022]

[0016] FIG. 6 illustrates a timing diagram 600 depicting various timings related to occurrences of CCO issues and application of CCO configurations.

[0023]

[0017] FIG. 7 illustrates an example signaling procedure 700 of a proactive CCO mechanism implemented in a disaggregated architecture.

[0024]

[0018] FIG. 8a-8b illustrate timing diagrams 800-1, 800-2 depicting various timings related to occurrences of CCO issues and application of CCO configurations.

[0025]

[0019] FIG. 9 illustrates a flowchart of an example method 900 of CCO in a communication system.

[0026]

[0020] FIG. 10 illustrates a flowchart of an example method 1000 of resolving conflicts between reactive and the proactive CCO mechanisms.

[0027]

[0021] FIG. 11 illustrates a block diagram 1100 of an apparatus or device. DETAILED DESCRIPTION

[0028]

[0022] The following detailed description of example embodiments refers to the accompanying drawings. The foregoing disclosure provides illustration and description, but is not intended to be exhaustive or to limit the implementations to the precise form disclosed. Modifications and variations are possible in light of the above disclosure or may be acquired from practice of the implementations. Further, one or more features or components of one embodiment may be incorporated into or combined with another embodiment (or one or more features of another embodiment). Additionally, the flowchart and description of operations provided below relate to one of the various embodiments. It should be noted that it is possible to make other embodiments that do not exactly match the flowchart and its description. It is understood that in other embodiments one or more operations may be omitted, one or more operations may be added, one or more operations may be performed simultaneously (at least in part).

[0029]

[0023] It will be apparent that systems and / or methods, described herein may be implemented in different forms of hardware, software, or a combination of hardware and software. The actual specialized control hardware or software code used to implement these systems and / or methods is not limiting of the implementations. Thus, the operation and behavior of the systems and / or methods are described herein without reference to specific software code. It is understood that software and hardware may be designed to implement the systems and / or methods based on the description herein.

[0030]

[0024] Even though particular combinations of features are recited in the claims and / or disclosed in the specification, these combinations are not intended to limit the disclosure of implementations. In fact, many of these features may be combined in ways not specifically recited in the claims and / or disclosed in the specification. Although each dependent claim listed below may directly depend on only one claim, the disclosure of implementations includes each dependent claim in combination with every other claim in the claim set.

[0031] 1025] No element, act, or instruction used herein should be construed as critical or essential unless explicitly described as such. Also, as used herein, the articles "a” and "an” are intended to include one or more items, and may be used interchangeably with “one or more.’7Also, as used herein, the terms “has,” “have,” “having,” “include,” “including.” or the like are intended to be open-ended terms. Further, the phrase “based on” is intended to mean “based, at least in part, on” unless explicitly stated otherwise. Furthermore, expressions such as “at least one of [A] and [B],” “[A] and / or [B],” or “at least one of [A] or [B]” are to be understood as including only A. only B, or both A and B.

[0032]

[0026] The foregoing disclosure provides illustration and description but is not intended to be exhaustive or to limit the implementations to the precise form disclosed. Modifications and variations are possible in light of the above disclosure or may be acquired from practice of the implementations.

[0033]

[0027] In the present disclosure, the terms like “communication system”, “system”, and “wireless communication system” have been used interchangeably throughout the specification. The terms like “RAN node” and “base station” have been used interchangeably throughout the description. The terms like “CCO configuration” and “CCO state” may be used interchangeably throughout the description. The terms like “proactive CCO mechanism” and “AI / ML based CCO mechanism” may be used interchangeably throughout the description. The terms like “AI / ML model” and “model”, may be used interchangeably throughout the description.

[0034]

[0028] In the context of present disclosure, the term “coverage” refers to a geographical area within which a wireless network provides service so that User Equipment (UEs) may establish and maintain connectivity with the wireless network. Further, the term “capacity7’ refers to an ability of the wireless network to accommodate multiple users simultaneously while maintaining adequate data rates.

[0035]

[0029] In the context of present disclosure, “Coverage and Capacity Optimization” or “CCO” refers to a function of a Self-Organizing Network (SON), which dynamically adjusts network parameters to optimize both coverage and capacity of a cellular network. One goal of the CCO is to provide an optimal balance between coverage and capacity. CCO is implemented at a network node by modifying one or more network parameters or settings such as: antenna configurations (e.g., antenna tilt, beamforming patterns, height adjustments, etc ), transmit power levels, cell selection and reselection thresholds, load balancing and scheduling policies, but not limited thereto.

[0036]

[0030] In the context of present disclosure, a “CCO issue” refers to a network problem arising due to suboptimal coverage or capacity, which requires adjustments of network parameters. CCO issues may be classified into two main categories cell coverage issues and cell capacity issues (e.g., at cell edges). The cell coverage issues may include coverage hole (e.g., poor signal coverage in a particular region of a cell), weak coverage or signal strength at an edge / boundary of the cell, overextending cell coverage (e.g., coverage of the cell extends beyond intended region thereby causing unintended interference in neighboring cells), but not limited thereto. The cell capacity issues may arise due to change in cell coverage and may result in congestion in the cell, uneven load distribution (e.g., traffic is not balanced among neighboring cells thereby leading to poor capacity utilization), but not limited thereto.

[0037]

[0031] In the context of present disclosure, “CCO configuration” or “CCO state” refers to a specific set of network parameters or settings applied to a network node to mitigate CCO issues. Depending on a type of CCO issue (i.e., coverage or capacity issue), the CCO configuration may be referred to as a “coverage configuration7’ or a “capacity configuration”. Each network node may have a default CCO configuration, which may need to be updated dynamically upon detecting a CCO issue. The CCO configuration may include one or more network parameters or settings such as: antenna configurations (e.g.. antenna tilt, beamforming patterns, height adjustments, etc.), transmit power levels, cell selection and reselection thresholds, load balancing and scheduling policies, but not limited thereto.

[0038]

[0032] FIG. 1 illustrates a high-level block diagram 100 of a disaggregated architecture of an example communication system comprising a Radio Access Network (RAN) node or a base station 102 configured to serve a geographical area or cell 104. The cell 104 may comprise at least one UE 106 and the RAN node 102 may be configured to provide wireless services to the at least one UE 106 served by the associated cell 104.

[0039]

[0033] The at least one UE 106 may be any mobile or non-mobile computing device including, but not limited to, a phone (e.g., a cellular phone or smart phone), a pager, a laptop computer, a desktop computer, a wireless handset, a portable communication device, a portable computing device (e.g., a personal data assistant), an entertainment device (e.g., a music or video device, or a satellite radio), a global positioning system device, or any other suitable computing device including a wired or wireless communications interface. In some embodiments of the present disclosure, the at least one UE 106 may be Intemet-of-Things (loT)-enabled device including, but not limited to, vehicles configured to communicate with the RAN node or a core network.

[0040]

[0034] In a disaggregated architecture, the RAN node 102 may be implemented as a 5G NR base station (gNB) 102 and may be partitioned into multiple logical network entities. For instance, the RAN node 102 may be partitioned into a central unit (gNB-CU or CU) 108 and one or more distributed units (gNB-DUs or DUs) 110. In the embodiment of FIG. 1, the CU 108 may be further partitioned into a central unit control-plane entity 114 (gNB-CU-CP or CU-

[0041] CP) and one or more central unit user-plane entities 116 (gNB-CU-Ups or CU-UPs) that may handle the control-plane and user-plane processing of the CU 108, respectively.

[0042]

[0035] The RAN node 102 may comprise one or more physical entities such as Radio Units (RUs) 112 including one or more antennas 118 for serving the at least one UE 106 in the associated cell. The CU 108 may be communicatively coupled with the one or more DUs 110 via an Fl interface. The DU 110 may be communicatively coupled with at least one RU 112 via a fronthaul interface 120. The CU-CP 114 may be communicatively coupled with each of the CU-UPs 116 via an El interface and may be further coupled with each of the DUs 110 via an Fl-C interface. Each of the DUs 110 may be communicatively coupled to each of the CU- UPs 116 via an Fl-U interface, as shown in FIG. 1.

[0043]

[0036] In one non-limiting embodiment, each DU 110 may host multiple cells. The CU-CP 114 may host one or more DUs 110 and one or more CU-UPs 116. In one example deployment, there may be one CU-CP 114 in a RAN node 102, and each DU 110 may be served by multiple CU-UPs 116. The CU-CP 114 is responsible for managing control plane protocols and procedures. For instance, the CU-CP 114 may host Packet Data Convergence Protocol - Control Plane (PDCP-C) layer and Radio Resource Control (RRC) layer, while the CU-UP 116 may host Packet Data Convergence Protocol - User Plane (PDCP-U) and Service Data Adaptation Protocol (SDAP) layers. The DU 110 may host lower layers such as Radio Link Control (RLC), Medium Access Control (MAC), and Physical (PHY) layers.

[0044]

[0037] The CU 108 may be configured to communicate with a core network 122 using a backhaul network 124. In one non-limiting embodiment of the present disclosure, the core network 122 may be a 5G core network which may utilize cloud-aligned, service-based architecture that spans across all 5G functions and interactions including authentication. security, session management etc.

[0045]

[0038] In some non-limiting embodiments, a cluster of RAN nodes or base stations 102 may be deployed within a coverage area instead of a single RAN node. The cluster of RAN nodes may comprise a plurality of RAN nodes 102 communicatively coupled with each other over the Xn interfaces for serving a plurality of cells.

[0046]

[0039] Referring to FIG. 2, which shows a high-level block diagram 200 of a communication system. As shown in FIG. 2, the communication system 200 may comprise a cluster of RAN nodes comprising a plurality of RAN nodes 102-1, 102-2 (collectively represented by reference numeral 102). Each RAN node 102 of the cluster may be similar to the RAN node 102 discussed in connection with FIG. 1 i.e., each RAN node 102 may comprise at least one CU 108 and one or more DUs 110 configured to serve one or more respective cells via one or more RUs 112. For instance, the first RAN node 102-1 may comprise a first CU 108-1 coupled with at least one first DU 110-1 over the Fl interface for serving one or more cells of the plurality of cells. Likewise, the second RAN node 102-2 may comprise a second CU 108-2 coupled with at least one second DU 110-2 over the Fl interface for serving one or more cells of the plurality of cells. It may be worth noting here that only few relevant components / units of the RAN nodes 102 are shown in FIG. 2 for the sake of simplicity. However, the present disclosure is not limited thereto and in general some or all component / units of the RAN node 102 (shown in FIG. 1) may be a part of the RAN nodes of FIG. 2. Each of the first and second RAN nodes may comprise a base station (gNB).

[0047]

[0040] As discussed in the background section, a primary goal of CCO is to dynamically optimize coverage and capacity' in response to dynamically changing network conditions. However, coverage and capacity' requirements often conflict with each other and achieving a balance between the coverage and capacity has always been challenging. For instance, increasing a transmission power of a network node may extend coverage of the network node but this often leads to higher inter-cell interference which negatively impacts capacity of neighboring cells.

[0048]

[0041] When a network node detects a CCO issue (i.e.. either a coverage issue or capacity issue), the network node executes a CCO operation and applies anew CCO configuration (e.g., a new coverage configuration). The network node must ensure that adjacent network nodes are aware of the change (i.e., the new coverage configuration) to prevent unintended senice degradation. In one example, the network node may send a list of affected cells and their modified Synchronization Signal Blocks (SSBs) with updated configuration (e.g., modified coverage) to the neighboring network nodes, so that the neighboring network nodes may take necessary actions based on the modified coverage of the network node.

[0049]

[0042] In order to detect and resolve / mitigate CCO issues in a communication system (e.g., in the communication system 200), various CCO mechanisms are used. Each RAN node 102 be configured to detect and resolve / mitigate CCO issues. Upon detecting a CCO issue, a RAN node (e.g., the first RAN node 102-1) may be configured to autonomously adjust within and switch among various CCO configurations each comprising one or more network parameters. When a change of network parameters is executed in the first RAN node 102-1, the first RAN node 102-1 may notify neighboring RAN nodes (e.g., the second RAN node 102-2) with a list of affected cells and Synchronization Signal Blocks (SSBs) and including modified network parameters.

[0050]

[0043] FIG. 3 illustrates an example signaling procedure 300 of a reactive CCO mechanism implemented in a disaggregated architecture of the communication system 200 of FIG. 2.

[0044] In Step 1 (SI), the first CU 108-1 of the first RAN node 102-1 detects a first CCO issue

[0051] (i. e. , either a coverage issue or a capacity issue) and affected cells and / or SSBs of the first RAN node 102-1.

[0052]

[0045] In Step 2 (S2), the first CU 108-1 may send an information element (e.g., a CCO Assistance Information IE) comprising a type of the detected first CCO issue and a list of the affected cells / SSBs to the first DU 110-1. The information element may be sent within a gNB- CU CONFIGURATION UPDATE message over the Fl interface.

[0053]

[0046] In Step 3 (S3), the first DU 110-1 may use the received CCO Assistance Information IE to take corrective measures. Specifically, the first DU 110-1 may determine a first CCO configuration or CCO state for the affected cells / SSBs of the first RAN node 102-1 based on the detected first CCO issue. The corrective actions taken by the first DU 110-1 may include applying the determined first CCO configuration to mitigate the first CCO issue.

[0054]

[0047] In Step 4 (S4), the first DU 110-1 may send an information element (e.g. a Coverage Modification Notification IE) having the first CCO configuration for the affected cells / SSBs of the first DU 110-1 to the first CU 108-1. The Coverage Modification Notification IE may be sent within a gNB-DU CONFIGURATION UPDATE message over the Fl interface. Since modifying CCO configuration at one RAN node may affect neighboring nodes, the first RAN node 102-1 notifies the second RAN node 102-2 regarding change of its CCO configuration.

[0055]

[0048] In Step 5 (S5), the first CU 108-1 may send an information element (e.g., a Coverage Modification List IE) having information related to the first CCO issue (e.g., a type of the first CCO issue) and the first CCO configuration for the affected cells / SSBs of the first DU 110-1 to the second RAN node 102-2 to allow the second RAN node 102-2 to selectively take corrective measures. Here, selectively taking the corrective measures means that the second RAN node 102-2 may or may not take any corrective measures depending on one or more circumstances prevalent at the second RAN node 102-2. Specifically, the first CU 108-1 may send the Coverage Modification List IE to the second CU 108-2. The Coverage Modification List IE may be sent within an NG-RAN NODE CONFIGURATION UPDATE message over the Xn interface. In one example, the Coverage Modification List IE may comprise only the information related to the first CCO configuration for the affected cells / SSBs of the first DU 110-1

[0056]

[0049] In Step 6 (S6), based on the received Coverage Modification List IE, the second RAN node 102-2 may determine a second CCO configuration for its own cells / SSBs that best matches the first CCO configuration resulting from the first CCO issue. In one example, the second RAN node 102-2 may use a predefined static mapping between CCO configurations of different RAN nodes. If there is any change in CCO configuration of the first RAN node 102- 1, the second RAN node 102-2, upon detecting the change, may automatically change its CCO configuration based on the predefined static mapping. Specifically, the second RAN node 102- 2 may automatically change its CCO configuration to the second CCO configuration which corresponds to the first CCO configuration.

[0057]

[0050] It may be noted that in the CCO mechanism 300, a reactive approach is used in which, when a RAN node 102 (the CU 108 in case of a CU-DU split architecture) detects a CCO issue which negatively impacts network and / or UE performances after the issue has already occurred. The RAN node 102 (the DU 110 in case of CU-DU split architecture) attempts to resolve or mitigate the CCO issue. Such reactive approach has several drawbacks. One such drawback is that in the reactive CCO mechanism 300, the corrective measures are taken only after detecting performance degradation or the CCO issues, which may result in sendee disruptions for network users before corrective actions are applied.

[0051] Proactive CCO mechanisms are being explored by the 3GPP to overcome the limitations of the reactive CCO mechanisms. With proactive CCO mechanisms, a more proactive approach may be used to prevent (or limit at an early stage) rise of a potential CCO issue which may degrade network (and UE) performances. Hence, the proactive CCO mechanisms may be used to proactively predict potential CCO issues before they impact the network performance. In one example, the proactive CCO mechanisms may comprise Artificial Intelligence or Machine Learning (AI / ML) based CCO mechanisms. The AI / ML-based CCO mechanisms introduce an ability to proactively predict potential CCO issues before they impact network performance.

[0058]

[0052] For supporting the AI / ML based CCO mechanisms, AI / ML models are first trained on historical data to leam patterns of CCO issues and then trained AI / ML models are deployed on each RAN node 102 for predicting potential CCO issues. The AI / ML models may additionally be trained to predict CCO configuration of the RAN nodes 102. In one example, model training may be performed in the Operations, Administration, and Maintenance (0AM) and model inferencing may be performed in the RAN node 102. In another example, model training and inferencing both may be performed in the RAN node 102. In case of CU-DU split architecture of FIG. 1-2, AI / ML model training may be performed in the 0 AM and model inferencing may be performed in the CU 108. In another implementation of the CU-DU split architecture, the AI / ML model training and inferencing both may be performed in the CU 108.

[0059]

[0053] In one non-limiting embodiment, the AI / ML models may be trained using supervised learning, unsupervised learning, reinforcement learning, and / or other machine learning methods. The AI / ML models may be trained via online training (i.e., training during operation), offline training (i.e., training prior to the operation), or a combination depending on the deployment and need. In one non-limiting embodiment, a single model is trained and then the trained model is deployed at each RAN node of the cluster. In another non-limiting embodiment a separate model is trained for each RAN node of the cluster and then trained models are deployed at respective RAN nodes of the cluster. Once the AI / ML models have been trained, the trained AI / ML models may be deployed within the cluster of RAN nodes such that one model is deployed on each RAN node of the cluster, as shown in FIG. 4.

[0060]

[0054] FIG. 4 shows a high-level block diagram 400 of a communication system, in which the proposed CCO techniques of the present disclosure may be implemented. The communication system 400 may be same as the communication system 200 except that a trained AI / ML model may be deployed at each RAN node. As shown in FIG. 4, the first CU 108-1 may comprise a first AI / ML model 402-1 and the second CU 108-2 may also comprise a second AI / ML model 402-2. An AI / ML model may be denoted with the reference numeral 402. It has been shown in FIG. 4 that the AI / ML models 402 are deployed at the CU 108. However, the present disclosure is not limited thereto and in some deployments, the AI / ML models 402 may be deployed at other entities of the RAN nodes 102 such as the DU 110, Non-Real-Time RAN Intelligent Controller (Non-RT RIC), or the Near-RT RIC depending on requirement and need.

[0061]

[0055] FIG. 5 describes an example signaling procedure 500 of implementing a proactive CCO mechanism in the example communication system 400 of FIG. 4 comprising the first and second RAN nodes, in accordance with some embodiments of the present disclosure. FIG. 6 illustrates a timing diagram 600 depicting various timings related to occurrences of CCO issues and application of CCO configurations, in accordance with some embodiments of the present disclosure.

[0062]

[0056] In Step 11 (Sil), at a time instance TO (as shown in FIG. 6), the first RAN node 102- 1 may predict, using a proactive CCO mechanism, a first CCO issue (e.g., a cell coverage issue or a cell capacity issue) potentially occurring in future within the first RAN node 102-1. The first RAN node 102-1 may also identify cells and / or SSBs of the first RAN node 102-1 which are going to be affected by the first CCO issue. Consider that the first CCO issue is predicted likely to occur at a first time instance Tl. The proactive CCO mechanism may utilize the first AI / ML model 402-1 to predict the first CCO issue that is likely to occur in the future and may also identify the affected cells and / or SSBs of the first RAN node 102-1. The first AI / ML model 402-1 may be provided with various input data to predict the first CCO issue. In one example, the input data may include measured and / or predicted radio resource status and a cunent CCO configuration used by the cells / SSBs of the first RAN node 102-1. The first AI / ML model 402- 1 may process the input data to predict the first CCO issue.

[0063]

[0057] In another example, the input data may additionally include input data from neighboring RAN nodes (e.g., second RAN node 102-2) and input data from UEs 106 served by the first RAN node 102-1. The input data from the second RAN node 102-2 may include measured and / or predicted radio resource status of the second RAN node 102-2. The input data from the UEs 106 may include UE measurement reports including cell level and beam level UE measurements, SON reports, but not limited thereto. The UE measurements reports may include Reference Signal Received Power (RSRP) measurements, Reference Signal Received Quality (RSRQ) measurements, Signal-to-Interference-plus-Noise Ratio (SINR) measurements, but not limited thereto. The SON Reports may include Radio Link Failure (RLF) data, Coverage Enhancement Factor (CEF), but not limited thereto. In this example, the first AI / ML model 402-1 may process the input data from the first RAN node 102-1, the input data from the neighboring RAN nodes, and the input data from the UEs 106 to predict the first CCO issue.

[0064]

[0058] The first RAN node 102-1 may infer, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node 102-1 to mitigate the predicted first CCO issue before its occurrence. The first CCO configuration for the first RAN node 102-1 is scheduled to be applied in future e.g., at a second time instance T2, as shown in FIG. 6. Consider an example, where the first CCO issue is predicted to occur at 15thminute from TO (i.e., the first time instance Tl=15t11minute). Hence, to mitigate the first CCO issue before its occurrence, the first CCO configuration may be scheduled to be applied at 11thminute from TO (i.e., the second time instance T2= 11thminute). Thus, the second time instance T2 may be before the first time instance T1 (i.e., the first CCO configuration is applied before the predicted occurrence of the first CCO issue) for proactive mitigation of the first CCO issue and minimize network disruptions and service degradation. However, in one implementation, the first time instance T1 may be before the second time instance T2 allowing the first CCO configuration to be applied reactively after detecting the first CCO issue.

[0065]

[0059] In one approach, first RAN node 102-1 may be configured to utilize the first AI / ML model 402-1 to automatically infer the first CCO configuration based on real-time input data, along with the second time instance T2. Instead of relying on the predefined mappings, the first AI / ML model 402-1 analyzes various data to dynamically recommend the first CCO configuration and the second time instance T2. The first RAN node 102-1 may determine / infer the first CCO configuration based on the first CCO issue and other local information. For instance, the first AI / ML model 402-1 or other technique may consider local information and the first CCO issue for recommending the first CCO configuration. In another approach, each RAN node 102 may maintain a predefined mapping table between CCO issues and corresponding CCO configurations. When the first CCO issue is predicted, the first RAN node 102-1 may retrieve an appropriate CCO configuration from the mapping table depending on the first CCO issue. In some implementations, a hybrid approach utilizing the AI / ML model and the mapping table may be used to improve accuracy.

[0060] Since any configuration adjustments in the first RAN node 102-1 may impact neighboring RAN nodes e.g., second RAN node 102-2, the first RAN node 102-1 proactively informs the second RAN node 102-2 regarding the first CCO configuration.

[0066]

[0061] In Step 12 (S12), the first RAN node 102-1 may transmit a notification or an information element (e.g., a Coverage Modification List IE) to the second RAN node 102-2, informing the second RAN node 102-2 about the first CCO configuration for the first RAN node 102-1 to be applied at the second time instance T2. The notification transmitted to the second RAN node 102-2 selectively enables the second RAN node 102-2 to determine a second CCO issue potentially occurring within the second RAN node 102-2 at a third time instance T3 and a second CCO configuration to be applied at a fourth time instance T4 at the second RAN node 102-2. In other words, the notification enables the second RAN node 102-2 to determine, based at least on the first CCO configuration applied at the second time instance, the second CCO issue (e g., a cell coverage issue or a cell capacity' issue) potentially occurring within the second RAN node 102-2 and the second CCO configuration to be applied at the second RAN node 102-2. The Coverage Modification List IE may be sent within an NG-RAN NODE CONFIGURATION UPDATE message over the Xn interface. The notification or information element transmitted to the second RAN node 102-2 may comprise information related to the first CCO configuration. The notification transmitted to the second RAN node 102-2 may comprise additional information such as information related to the first CCO issue (e.g., a ty pe of the first CCO issue), and / or the timing information of the first and second time instances Tl, T2.

[0067]

[0062] Based on the received notification, the second RAN node 102-2 may determine e.g., using the predefined static mapping whether any changes are required in its existing CCO configuration to match the updated first CCO configuration of the first RAN node 102-1. Upon detecting that the changes are required in the existing CCO configuration, the second RAN node 102-2 may determine the second CCO configuration that matches the first CCO configuration and apply the second CCO configuration at an appropriate time (e.g., at the second time instance T2). In one example, the second RAN node 102-2 performs its own computations to infer a potential CCO issue and associated CCO configuration to be applied at the second RAN node 102-2 to avoid the potential CCO issue. In another example, the second RAN node 102-2 may infer a CCO configuration to be applied at the second RAN node 102-2 (without inferring its own potential CCO issue).

[0068]

[0063] In Step 13 (S13). the second RAN node 102-2, based on the received notification, may selectively infer a second CCO issue potentially occurring within the second RAN node 102-2 at the third time instance T3. In one example, the second RAN node 102-2 may utilize the second AI / ML model 402-2 to predict the second CCO issue that is likely to occur in the future and may affect cells and / or SSBs of the second RAN node 102-2. The second AI / ML model 402-1 may predict the second CCO issue and its time of occurrence (T3) based on one or more of: the first CCO configuration, the first CCO issue, timing information of first and second time instances, or other input data. The other input data may include measured and / or predicted radio resource status and a current CCO configuration used by the cells / SSBs of the second RAN node 102-2, input data from neighboring RAN nodes, and input data from UEs served by the second RAN node 102-2, as discussed in connection with Step 11.

[0069]

[0064] Based on the predicted second CCO issue that is likely to occur at the third time instance T3, the second RAN node 102-2 may selectively infer a second CCO configuration for the second RAN node 102-2 to mitigate the predicted second CCO issue before its occurrence. The second CCO configuration for the second RAN node 102-2 is scheduled to be applied in future e.g., at the fourth time instance T4, as shown in FIG. 6. Consider an example, where the second CCO issue is predicted to occur at 14thminute from TO (i.e., the third time instance T3=14thminute). Hence, to mitigate the second CCO issue before its occurrence, the second CCO configuration may be scheduled to be applied at 10thminute from TO (i.e., the fourth time instance T4= 10thminute). Thus, the fourth time instance T4 may be before the third time instance T3 (i.e.. the second CCO configuration is applied before the predicted occurrence of the second CCO issue) for proactive mitigation of the second CCO issue and minimize network disruptions and service degradation. However, in one implementation, the third time instance T3 may be before the fourth time instance T4 allowing the second CCO configuration to be applied reactively after detecting the second CCO issue. Further, the first and second time instances may be different from the third and fourth time instances respectively. However, the present disclosure is not limited thereto and in one implementation, the some or all time instances associated with the first RAN node 102-1 may match with corresponding time instances associated with the first RAN node 102-2. It may be noted that the second CCO configuration may be predicted / determined in the similar manner as that of first CCO configuration.

[0070]

[0065] FIG. 7 describes an example signaling procedure 700 of the proactive CCO mechanism implemented in a disaggregated architecture of the example communication system 400 of FIG. 4, in accordance with some embodiments of the present disclosure.

[0071]

[0066] In Step 21 (S21), at a time instance TO, the first CU 108-1 of the first RAN node 102- 1 (or the first base station) may employ a proactive CCO mechanism using the first AI / ML model 402-1 to predict the first CCO issue potentially occurring in future within the first RAN node 102-1. The first RAN node 102-1, using the first AI / ML model 402-1, may predict that the first CCO issue is expected to occur at the first time instance T1 and may identify the affected cells and / or SSBs of the first DU 110-1 which are going to be affected by the first CCO issue. It may be noted that the first CU 108-1 may predict the first CCO issue in the similar manner as discussed in Step 11 of FIG. 5. Consider an example, where the first CCO issue is predicted to occur at 15thminute from TO (i.e., the first time instance Tl=15thminute).

[0067] In Step 22 (S22), the first CU 108-1 may send an information element (e.g., a CCO Assistance Information IE) comprising information associated with the predicted first CCO issue (e.g., a type of the predicted first CCO issue) and a list of the affected cells / SSBs to the first DU 110-1. The information element may be sent within a gNB-CU CONFIGURATION UPDATE message over the F 1 interface.

[0072]

[0068] In Step 23 (S23), the first DU 110-1 may use the received CCO Assistance Information IE to take corrective measures. Specifically, the first DU 110-1 may infer, based at least on the predicted first CCO issue, the first CCO configuration for the first DU 110-1 to mitigate the predicted first CCO issue before the predicted occurrence. The first CCO configuration is scheduled to be applied in future e.g., at the second time instance T2, as shown in FIG. 6. Consider an example, where the first CCO configuration may be scheduled to be applied at 11thminute from TO (i.e., the second time instance T2= 11thminute). The DU 110-1 may also utilize the first AI / ML model 402-1 to infer the first CCO configuration. To implement this possibility7, the DU 110-1 may be provisioned with the first AI / ML model 402-1 which is trained to recommend CCO configurations. In another example, the first CU 108-1 may predict the first CCO configuration and the second time instance T2 using the first AI / ML model 402- 1 and provide information of the first CCO configuration and the second time instance T2 to the first DU 110-1 over the Fl interface. It may be noted that the prediction / determination of the second CCO configuration may be similar to the one discussed in Step 11 of FIG. 5.

[0073]

[0069] In Step 24 (S24), the first DU 110-1 may send an information element (e.g. a Coverage Modification Notification IE) having the first CCO configuration for the affected cells / SSBs of the first DU 110-1 to the first CU 108-1. The Coverage Modification Notification IE may be sent within a gNB-DU CONFIGURATION UPDATE message over the Fl interface. The Coverage Modification Notification IE may additionally include the timing information of the second time instance T2.

[0074]

[0070] In Step 25 (S25). the first CU 108-1 may send an information element (e.g., a Coverage Modification List IE) having information related to the first CCO configuration for the affected cells / SSBs of the first DU 110-1 to the second CU 108-2. The Coverage Modification List IE may be sent within an NG-RAN NODE CONFIGURATION UPDATE message over the Xn interface. In one example, the Coverage Modification List IE may also comprise information related to the first CCO issue (e.g.. a type of the first CCO issue) and the timing information of the first and second time instances Tl, T2.

[0075]

[0071] In Step 26 (S26), the second CU 108-2 of the second RAN node 102-2 (or the second base station) may selectively infer the second CCO issue (which may occur at a third time instance T3 due to application of the first CCO configuration at the first DU 110-1) and affected cells and / or SSBs of the second DU 110-2. The second RAN node 102-2 may utilize the second AI / ML model 402-2 to predict the second CCO issue that is likely to occur in the future at the third time instance T3. Consider an example, where the second CCO issue is predicted to occur at 14thminute from TO (i.e., the third time instance T3=14thminute). The second AI / ML model 402-1 may predict the second CCO issue and its time of occurrence (T3) based on one or more of: the first CCO configuration, the first CCO issue, timing information of first and second time instances, and / or the other input data. It may be noted that the prediction / determination of the second CCO issue may be similar to the one discussed in Step 13 of FIG. 5.

[0076]

[0072] In Step 27 (S27), the second CU 108-2 may send an information element (e.g., a CCO Assistance Information IE) comprising a type of the determined second CCO issue and the affected cells / SSBs to the second DU 110-2. The information element may be sent within a gNB-CU CONFIGURATION UPDATE message over the Fl interface.

[0077]

[0073] In Step 28 (S28), the second DU 110-2 may use the received CCO Assistance Information IE to take corrective measures. Specifically, the second DU 110-2 may infer the second CCO configuration for the affected cells / SSBs of the second RAN node 102-2 based on the second CCO issue. The second CCO configuration for the second RAN node 102-2 is scheduled to be applied in future e.g., at a fourth time instance T4. In one example, the second CCO configuration and the fourth time instance T4 may be predicted using the second AI / ML model 402-2. Consider an example, where the fourth time instance is predicted as T4= 10thminute from TO). It may be noted that the prediction / determination of the second CCO configuration may be similar to the one discussed in Step 13 of FIG. 5.

[0078]

[0074] In Step 29 (S29), the second DU 110-2 may send an information element (e.g. a Coverage Modification Notification IE) having the second CCO configuration for the affected cells / SSBs of the second DU 110-2 to the second CU 108-2. The Coverage Modification Notification IE may be sent within a gNB-DU CONFIGURATION UPDATE message over Fl.

[0079]

[0075] In this manner, the proactive CCO mechanisms helps in early identification of potential CCO issues before the CCO issues actually impact network performance. In the proactive CCO mechanisms, instead of reacting to degraded network conditions, the RAN nodes 102 may proactively apply CCO configurations thereby, preventing service disruptions and minimizing network performance degradation. Further, the AI / ML models facilitate dynamic recommendation of optimal CCO configuration based on real-time network data, traffic patterns, and interference levels. A RAN node 102 proactively informs neighboring RAN nodes about its predicted CCO issues and planned CCO configurations. This optimal coordination among the different RAN nodes 102 helps the neighboring RAN nodes 102 to determine and apply their own CCO configuration at the appropriate time, preventing performance degradation and service disruptions in the neighboring RAN nodes 102.

[0080]

[0076] In some communication systems, both proactive and reactive CCO mechanisms may operate simultaneously. In such communication systems, a potential conflict may occur between the reactive and the proactive CCO mechanisms. For example, consider a scenario where the first CCO configuration of the first DU 110-2 is to be applied at the second time instance T2, and in the vicinity of the second time instance T2, the reactive CCO mechanism detects a third (new) CCO issue potentially occurring at the first RAN node 102-1. Such scenarios of conflict may arise when the first CCO issue was incorrectly predicted at the first RAN node 102-1 and / or a predicted time of occurrence (T2) of the first CCO issue was inaccurate. Thus, there is a possibility that the prediction of the first CCO issue with the use of proactive CCO mechanism is inaccurate in terms of the first CCO issue being predicted or the timing of the first CCO issue being predicted. The present disclosure provides techniques to address these conflicts, as discussed in the forthcoming paragraphs.

[0081]

[0077] It may be noted that the underlying principle governing the conflict resolution process is “detection has a higher priority7than prediction"’ because the detection is based on actual or real-time network conditions. Hence, actions taken based on real-time detection overwrite actions previously planned based on prediction. Such mechanism prevents unnecessary7or incorrect network optimizations.

[0082]

[0078] The first RAN node 102-1 may continuously monitor network conditions using a reactive CCO mechanism to detect CCO issues in real time. Consider that the first RAN node 102-1 detects, using the reactive CCO mechanism, the third CCO issue occurring at a particular time instance (e.g., a fifth time instance T5) within the first RAN node, as shown in FIG. 8a. The reactive CCO mechanism operates independently from the proactive CCO mechanism. which had already predicted the first CCO issue occurring at the first time instance Tl. Exemplarily, the reactive CCO mechanism could be conventional CCO detection mechanism. Upon detecting the third CCO issue, the first RAN node 102-1 may determine whether the detection of the third CCO issue conflicts with the prediction of the first CCO issue occurring at the first time instance Tl.

[0083]

[0079] In order to determine whether the detection of the third CCO issue conflicts with the prediction of the first CCO issue, the first RAN node 102-1 may establish a predefined time window (referred to as “cohesion time window”) based on at least one of the first to fourth time instances (i.e., based on at least one of Tl, T2, T3, or T4). The cohesion time window may be defined as a bounded time frame within which a RAN node 102 evaluates whether a detected CCO issue conflicts with a predicted CCO issue. If the newly detected third CCO issue at occurring at the first time instance T5 falls within the predefined time window, a conflict is suspected. How ever, if the fifth time instance T5 falls outside of the predefined time window then there is no conflict, the third CCO issue is treated as an independent CCO issue, and both proactive and reactive CCO mechanisms may function separately. The predefined window could be hysteresis period before which a detection of new7CCO issue could be considered valid. A conflict may arise or may be detected when (a) the first CCO issue (predicted to occur at Tl) is of a different type or nature than the third CCO issue (detected at T5) and / or (b) the predicted first time instance Tl is different from the detected fifth time instance T5.

[0084]

[0080] Upon determining that the detection of the third CCO issue conflicts with the prediction of the first CCO issue, the first RAN node 102-1 may initiate resolution of the conflict between the detection of the third CCO issue and the prediction of the first CCO issue. Upon detecting the conflict betw een the predicted first CCO issue and the reactively detected third CCO issue, the first RAN node 102-1 determines whether: the first CCO configuration (derived from the prediction at Tl) has been applied in first RAN node 102-1 and the second CCO configuration (derived from the prediction at T3) has been applied in second RAN node 102-2. There are two possible cases. In a first case, neither the first nor the second RAN node has applied their respective CCO configurations, as shown in the example timing diagram 800-1 of FIG. 8a. In a second case, at least one of the first or the second RAN node has applied its respective CCO configuration, as shown in the timing diagram 800-2 of FIG. 8b.

[0085]

[0081] In the first case shown in FIG. 8a, upon determining that neither the first RAN node 102-1 nor the second RAN node 102-2 has applied their respective CCO configurations, the predicted first CCO configuration is considered obsolete as it was based on an inaccurate prediction of the first CCO issue and / or the time of occurrence of the first CCO issue. The first RAN node 102-1 may discard or cancel the first CCO configuration and instead determine, based at least on the detected third CCO issue, a third CCO configuration to be applied at the first RAN node 102-1. The first RAN node 102-1 then transmits another notification to the second RAN node 102-2 comprising information associated with the determined third CCO configuration. The other notification enables the second RAN node 102-2 to selectively discard the second CCO configuration and apply a fourth CCO configuration, based on the third CCO configuration. To support this, the notification further comprises information instructing the second RAN node 102-2 to discard or cancel the second CCO configuration and determine or infer a fourth CCO issue and a fourth CCO configuration based at least on the determined third CCO configuration.

[0086]

[0082] The first RAN node 102-1 sends the notification in the form of an updated Coverage Modification List IE over the Xn interface. The notification comprises information about the newly determined third CCO configuration, instructions to the second RAN node 102-2 to discard the previously determined second CCO configuration (since it was based on an inaccurate prediction), instructions to compute the fourth CCO configuration, based on the newly formulated third CCO configuration. The first RAN node 102-1 applies the third CCO configuration and the second RAN node applies the fourth CCO configuration.

[0087]

[0083] In case of disaggregated architecture, upon determining that none of the first and second RAN nodes has applied their respective CCO configurations, the first DU 110-1 determines the third CCO configuration based on the third CCO issue and informs the same to the first CU 108-1. The first CU 108-1 may provide information to the second CU 108-2 related to cancellation of the predicted first CCO issue and the first CCO configuration. The first CU 108-1 also includes information related to the currently detected third CCO issue and the third CCO configuration applied by the first DU 110-1. The second CU 108-2 discards the predicted second CCO issue and derives the new fourth CCO issue based on the information received from the first CU 108-1 and informs the same to the second DU 110-2. The second DU 110-2 discards the second CCO configuration (if the second DU 110-2 has not applied the second CCO configuration or reverses changes of the second CCO configuration (if the second DU 110-2 has applied the second CCO configuration). The second DU 110-2 then determines and applies a fourth CCO configuration (in response to the latest information received from the second CU 108-2) and informs the same to the second CU 108-2.

[0088]

[0084] In the second case, upon determining that the first RAN node 102-1 and / or the second RAN node 102-2 has applied their respective CCO configurations, the first RAN node 102-1 may determine whether the newly detected third CCO issue is a valid detection or an invalid detection. Here, in order to determine validity of the detection, the first RAN node 102-1 may establish a predefined waiting time-window (or “hysteresis period”). The hysteresis period allows the system to stabilize after a recently applied CCO configuration before reacting to new detections and prevents premature reconfigurations that might be triggered due to temporary fluctuations in network conditions. It may be noted that the predefined waiting timewindow starts at the moment when either the first RAN node 102-1 applies the first CCO configuration or the second RAN node 102-2 applies the second CCO configuration, whichever is earliest. If a CCO issue persists after the hysteresis period, then such CCO issue may be treated as a new CCO issue and may be treated by taking one or more actions, as discussed in the forthcoming paragraphs.

[0089]

[0085] Depending on whether the third CCO issue is detected within or outside of the predefined waiting time-window, the first RAN node 102-1 determines whether the detection is a false CCO issue (invalid detection) or a legitimate or valid CCO issue (valid detection). The first RAN node 102-1 determines that the detection of the third CCO issue is an invalid detection when the time T5 of the detection falls within the predefined waiting time-window and determines the detection of the third CCO issue as a valid detection when the time T5 of the detection falls beyond the predefined waiting time-window. In other words, the first RAN node 102-1 determines the detected third CCO issue as a false CCO issue when T5 falls within the predefined waiting time-window and determines the detected third CCO issue as a valid CCO issue when T5 falls beyond the predefined waiting time-window.

[0090]

[0086] The first RAN node 102-1, upon determining the detection of the third CCO issue as the invalid detection, disregards the third CCO issue or cancels the detected third CCO issue and does not apply any configuration changes in respect of the third CCO issue.

[0091]

[0087] Upon determining that the detection of the third CCO issue is the valid detection (i.e., the CCO issue is persistent and not a temporary fluctuation caused by the previously applied configurations), the first RAN node 102-1 may take one or more corrective actions to prevent network performance degradation. In one example, the first RAN node 102-1 may be configured to facilitate reversing of configuration changes already applied by at least one of the first and second RAN nodes. Alternatively or additionally, the first RAN node 102-1 may be configured to prevent pending configuration changes from being applied by at least one of the first and second RAN nodes.

[0092]

[0088] It may be noted that there may be sub-cases under the second case depending on which RAN node has applied the CCO configuration and which RAN node has detected the new CCO issue.

[0093]

[0089] In a first sub-case of the second case, consider that the first DU 110-1 has already applied the predicted first CCO configuration and the first CU 108-1 detects a new CCO issue that results in a possible conflict. As shown in the timing diagram 800-2 of FIG. 8b, the first CU 108-1 detects anew CCO issue (e.g., the third CCO issue detected at the fifth time instance T5), despite the recently applied first CCO configuration. The first CU 108-1 may determine, based on the predefined waiting time-window, whether the newly detected third CCO issue is valid or invalid. Upon determining that the detection of the third CCO issue is an invalid detection, the first CU 108-1 may disregard the third CCO issue. Upon determining that the detection of the third CCO issue is a valid detection, the first CU 108-1 may notify the first DU 110-1 of the newly detected third CCO issue and provides additional information to reverse the recently applied first CCO configuration. The first DU 110-1 may undo or reverse the recently applied first CCO configuration. The first DU 110-1 may determine and apply a third CCO configuration based on the third CCO issue and informs the same to the first CU 108-1. The first CU 108-1 may provide information to the second CU 108-2 related to cancellation of the predicted first CCO issue and the first CCO configuration. The first CU 108-1 also includes information related to the currently detected third CCO issue and the third CCO configuration applied by the first DU 110-1.

[0090] The second CU 108-2 discards the predicted second CCO issue and derives a new fourth CCO issue based on the information received from the first CU 108-1, and informs the same to the second DU 110-2. The second DU 110-2 discards the second CCO configuration (if the second DU 110-2 has not applied the second CCO configuration or reverses changes of the second CCO configuration (if the second DU 110-2 has applied the second CCO configuration). The second DU 110-2 then determines and applies a fourth CCO configuration (in response to the latest information received from the second CU 108-2) and informs the same to the second CU 108-2.

[0094]

[0091] In a second sub-case of the second case, consider that the first DU 110-1 has already applied the predicted first CCO configuration and second CU 108-2 detects a new CCO issue that results in a possible conflict. The newly detected CCO issue at second CU 108-2 might be a direct consequence of configuration changes made at the first DU 110-1. Hence, the second CU 108-2 may determine, based on the predefined waiting time-window, whether the newly detected CCO issue is valid or invalid. Upon determining that the detection of the new CCO issue is an invalid detection, the second CU 108-2 may disregard the newly detected CCO issue. Upon determining that the newly detected CCO issue is a valid detection, the second CU 108- 2 may take one or more corrective measures such as, but not limited to, discarding the second CCO configuration (if the second DU 110-2 has not applied the second CCO configuration), reversing changes of the second CCO configuration (if the second DU 110-2 has applied the second CCO configuration), applying a new CCO configuration corresponding to the newly detected CCO issue. In one example, the second CU 108-2 may instruct the first RAN node 102-1 to reverse changes of the first CCO configuration, and determine and apply a new CCO configuration.

[0092] In a third sub-case of the second case, consider that the second DU 110-2 has already applied the predicted second CCO configuration and the first CU 108-1 detects a new CCO issue that results in a possible conflict. Here, the first CU 108-1 is unaware that the second DU 110-2 has already applied the second CCO configuration. Hence, conflict resolution process of the first case (as discussed above) may be followed. Briefly, the first CU 108-1 determines whether the newly detected CCO issue is valid or invalid. Upon determining that the detection of the third CCO issue is a valid detection, the first CU 108-1 may undertake or facilitate one or more corrective measures such as, but not limited to, discarding the first CCO configuration (if first CCO configuration has not been applied), reversing changes of the first CCO configuration (if first CCO configuration has been applied), reversing changes of the second CCO configuration, determining and applying new CCO configurations at the first and second DUs.

[0095]

[0093] In a fourth sub-case of the second case, consider that the second DU 110-2 has already applied the predicted second CCO configuration and the second CU 108-2 detects a new CCO issue that results in a possible conflict. The first CU 108-1 determines whether the newly- detected CCO issue is valid or invalid. Upon determining that the detection of the third CCO issue is a valid detection, the first CU 108-1 may undertake or facilitate one or more corrective measures such as, but not limited to, reversing changes of the second CCO configuration, determining and applying new CCO configurations at the second DU 110-2. In one example, the second CU 108-2 may instruct the first RAN node 102-1 to perform one or more of: reverse changes of the first CCO configuration (if first CCO configuration has been applied), discard the first CCO configuration (if the first CCO configuration has been applied), and determine and apply a new CCO configuration. In one implementation of the fourth sub-case, the communication system may be configured such that the fourth time instance T4 is greater than the second time instance T2.

[0096]

[0094] The above discussed techniques provide effective mechanisms of handling potential conflicts, thereby ensuring that incorrectly applied CCO configurations are reversed and the communication system may re-optimize itself based on the real-time detected CCO issues rather than outdated CCO issue predictions. In this manner, the techniques of the present disclosure prevent degradation of system performance and maintain overall stability in the communication system.

[0097]

[0095] A skilled person would appreciate that the above mentioned cases and / or sub-case are just few examples of handling conflicts of predicted and detected CCO issues. Hence, the present disclosure should not be considered as limited to the above described cases / sub-cases and there may be other similar cases and / or sub-cases of handling conflict between predicted and detected CCO issues.

[0098]

[0096] Further, only two RAN nodes 102 have been shown in the communication systems disclosed in the present disclosure. However, the techniques of the present disclosure are equally applicable even when there are more than two RAN nodes 102 in the communication systems. Further, in the present disclosure it has been shown that there is only DU 110 served by a CU 108. However, the present disclosure is not limited thereto and the techniques of the present disclosure are equally applicable even when each RAN node 102 is serving more than one DUs 110.

[0099]

[0097] Referring now to FIG. 9, a flow chart is described illustrating an example method 900 performed by a RAN node or base station 102 (e.g., by the first RAN node 102-1) of CCO in a communication system, according to an embodiment of the present disclosure.

[0098] The method 900 may include, at block 902, predicting, using a proactive CCO mechanism, a first CCO issue potentially occurring at a first time instance T1 within a first RAN node 102-1. At block 904, the method 900 may include determining, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node 102-1 to be applied at a second time instance T2 for mitigating the predicted first CCO issue.

[0100]

[0099] At block 906. the method 900 may include transmitting a notification to a second RAN node 102-2, about the first CCO configuration for the first RAN node to be applied at the second time instance.

[0101]

[0100] Referring now to FIG. 10, a flowchart is described illustrating an example method 1000 performed by a RAN node or base station 102 (e.g., by the first RAN node 102-1) for resolving conflicts between reactive and the proactive CCO mechanisms, according to an embodiment of the present disclosure.

[0102]

[0101] The method 1000 may include, at block 1002, detecting, using a reactive CCO mechanism, a third CCO issue occurring at a particular time instance (e.g., the fifth time instance T5) within the first RAN node 102-1. At block 1004, the method 1000 may include determining whether the detection of the third CCO issue conflicts with the prediction of the first CCO issue.

[0103]

[0102] At block 1006, the method 1000 may include upon determining that the detection of the third CCO issue conflicts with the prediction of the first CCO issue, resolving the conflict between the detection of the third CCO issue and the prediction of the first CCO issue.

[0104]

[0103] FIG. 11 illustrates a block diagram 1100 of an apparatus or device, in accordance with some embodiments of the present disclosure. As shown in FIG. 11, the apparatus 1100 may include a processor 1110, a memory 1120, a storage component 1130, an input component 1140, an output component 1150, a communication interface 1160, a bus 1170, but not limited thereto.

[0105] 1104] The processor 1110, as used herein, means any type of computational circuit that may comprise hardware elements and software elements. The processor 1110 may be embodied as a multi-core processor, a single core processor, or a combination of one or more multi-core processors and / or one or more single core processors, a distributed processing system, or the like. The processor 1110 may be a Central Processing Unit (CPU), a graphics processing unit (GPU), an accelerated processing unit (APU), an application-specific integrated circuit (ASIC), or another type of processing component.

[0106]

[0105] The memory 1120 includes a non-transitory computer readable medium. The memory 1120 includes a random-access memory (RAM), a read only memory’ (ROM), and / or another type of dynamic or static storage device (e.g., a flash memory, a magnetic memory', and / or an optical memory) that stores information and / or instructions for use by processor 1110. The memory' 1120 comprises machine-readable instructions which are executable by the processor 1110. These machine-readable instructions when executed by the processor 1110 cause the processor 1110 to perform one or more method steps of an embodiment described in the present disclosure.

[0107]

[0106] The storage component 1130 stores information and / or software related to the operation and use of the apparatus 1100. For example, the storage component 1130 may include a hard disk (e.g., a magnetic disk, an optical disk, a magneto-optic disk, and / or a solid-state disk), a compact disc (CD), a digital versatile disc (DVD), a floppy disk, a cartridge, a magnetic tape, and / or another type of non-transitory computer-readable medium, along with a corresponding drive.

[0107] The input component 1140 is configured to receive information, such as user input. For example, the input component 1140 may include, but not be limited to, a touch screen display, a keyboard, a keypad, a mouse, a button, a switch, and / or a microphone. Additionally, or alternatively, the input component 1140 may include a sensor for sensing information (e.g., a global positioning system (GPS), an accelerometer, a gyroscope, and / or an actuator).

[0108]

[0108] The output component 1150 is configured to provide output information from the apparatus 1100. For example, the output component 1150 may be. but not limited to, a display, a speaker, an instruction device to an external device, and / or one or more light-emitting diodes (LEDs).

[0109]

[0109] The communication interface 1160 is an interface that provides a communication connection to other devices, such as external devices and internal devices. The connection by the communication interface 1160 can be a wired connection, a wireless connection, or a combination of wired and wireless connections, and can be a direct connection or an indirect connection via a communication network that exists between the apparatus 1100 and other devices. In other words, the standard of the communication interface 1160 is not limited.

[0110]

[0110] The bus 1170 acts as an interconnect between the processor 1110, the memory' 1120, the storage component 1130, the input component 1140, the output component 1150, and the communication interface 1160 of the apparatus 1100. The bus 1170 may include a wired interconnection or a wireless interconnection.

[0111] [Hl] The number and arrangement of components shown in FIG. 11 are provided as an example. In practice, the apparatus 1100 may include additional components, fewer components, different components, or differently arranged components than those shoyvn in FIG. 11. Additionally, or alternatively, a set of components (e.g., one or more components) of the apparatus 1100 may perform one or more functions described as being performed by another set of components of the apparatus 1100. Further, one or more method steps described in any of the embodiments may be performed utilizing a plurality of apparatuses 1100 in communication with one another.

[0112]

[0112] In one non-limiting embodiment, the apparatus 1100 may be used to implement some or all functions of any entity including UEs. various network entities of the RAN, various entities of the core network, but not limited thereto. Specifically, the apparatus 1100 may implement the functionalities of the RAN nodes 102 (specifically, the CU 108, the DU 110).

[0113]

[0113] Additional Description:

[0114]

[0114] A. Discussion of AI / ML based CCO

[0115]

[0115] 1. Introduction

[0116] • RAN3#125-bis meeting agreed that:

[0117] • Timing information: Introducing separate time information for future coverage state and predicted CCO issue.

[0118] • Time for future coverage state: The point in time when the future coverage state will be applied.

[0119] • Time for predicted CCO issue: The point in time when the CCO issue is predicted to happen.

[0120]

[0116] 2. Discussion

[0121] • In the legacy CCO mechanism:

[0122] • Step 1 : The gNB-CU detects the CCO issue (i.e., either coverage issue or cell edge capacity issue) and affected cells / SSBs;

[0123] • Step 2: The gNB-CU sends the CCO Assistance Information IE (within the GNB-CU CONFIGURATION UPDATE message over Fl) containing the CCO issue type and affected cells / SSBs to the gNB-DU; Step3: The gNB-DU uses the received CCO Assistance Information IE to determine the new coverage state (CCO state) for the affected gNB-DU’s cells / SSBs;

[0124] • Step 4: The gNB-DU sends the Coverage Modification Notification IE (within the GNB-DU CONFIGURATION UPDATE message over Fl) containing the new CCO state for the affected gNB-DU’s cells / SSBs to the gNB-CU;

[0125] • Step 5: The gNB-CU sends the Coverage Modification List IE (within the NG-RAN NODE CONFIGURATION UPDATE message over Xn) containing the detected CCO issue type and the new CCO state for the affected gNB-DU’s cells / SSBs to the neighbour gNB to allow for the neighbour gNB to determine a new CCO state for its own cells / SSBs that best matches the new CCO state of the gNB-CU’s cells / SSBs resulting from the detected CCO issue.

[0126] • For the AIML-based CCO mechanism:

[0127] • - Step 0: AIML training entity generates predicted CCO issue

[0128] • - Step 1 : gNB-CU predicts the CCO issue.

[0129] • - Step 2: gNB-CU sends the predicted CCO issue to gNB-DU.

[0130] • - Step 3: gNB-DU generates the future coverage status based on the predicted

[0131] CCO issue and other local information, whether only local information is used can be further discussed.

[0132] • - Step 4: gNB-DU sends the future coverage status to gNB-CU.

[0133] • - Step 5: gNB-CU sends the future coverage status to neighbour gNBs (say gNB2).

[0134] • - Step 6: gNB2-CU uses the predicted issue, the future coverage state, the respective timing information, and infers its own CCO issue • - Step 7: gNB2-CU sends this information to gNB2-DU which decides whether any changes are required in its own coverage / capacity configuration

[0135]

[0117] Observation 1 : Since gNB2-CU gets the information about the predicted CCO issue, and the future CCO state of gNBl ahead of the prediction time and the updated configuration time, it is likely that it will deduce its own CCO issue ahead of that time. And similarly, it is likely that gNB2-DU will also recommend its own future CCO state ahead of the time.

[0136] Therefore, the timing information agreed in the previous meeting needs to extend to gNB2 also, wherein the predicted CCO issue and future CCO state timing of gNBl and gNB2 might differ.

[0137]

[0118] Proposal 1: The present disclosure proposes that the timing information agreed in the previous meeting needs to extend to gNB2 also, wherein the predicted CCO issue and future CCO state timing of gNBl and gNB2 might differ.

[0138] • Assuming that at time tO, gNB 1 -CU predicts the CCO issue to occur at time tO+x (which is similar to T1 of FIG. 6). gNBl-CU sends the predicted issue with the predicted time to the gNBl -DU. gNBl -DU generates the future coverage status based on the predicted CCO issue and other local information and decides the time tO+y (which is similar to T2 of FIG. 6) when this future coverage status will be applied. Based on this, gNB2-CU predicts the CCO issue to occur at tO+p (which is similar to T2 of FIG. 3) and gNB-DU recommends the future CCO state to be applied at tO+q (which is similar to T4 of FIG. 6) based on the input from gNBl- CU.

[0139]

[0119] It is to be noted that there may be a scenario where a potential conflict might occur between the legacy and the AI / ML based CCO mechanisms. It is understood that both the mechanisms will co-exist, therefore, the present disclosure analyses the interaction between the two and avoids possibility7of scenarios where the tw o might conflict. • Consider the scenario where the future CCO state of gNBl-DU is to be applied at time tO+y, and in the vicinity of that moment of time, the legacy CCO mechanism detects a CCO issue. This could happen is the prediction at gNBl due was not accurate, or the prediction time of the CCO issue was not accurate.

[0140] 1120] Case 1 : The proposed “future” CCO state has not been applied by either gNBl or gNB2

[0141] • gNBl-DU has not yet applied the proposed configuration change which it had derived in response to the predicted CCO issue, however, that change might no longer be valid.

[0142] • gNBl-DU determines the new coverage state and informs gNBl-CU.

[0143] • gNBl-CU informs gNB2-CU of cancellation of the earlier predicted CCO issue and the future coverage state. It also includes the information of currently detected CCO issue and the CCO state applied by the gNBl-DU in response to the CCO issue detected by gNBl-CU.

[0144] • gNB2-DU derives a new CCO issue based on the information received from the gNBl- CU and informs gNB2-DU of the change.

[0145] • gNB2-DU determines the new coverage state in response to the latest information and informs the gNB2-CU.

[0146]

[0121] Case 1: The proposed “future” CCO state has been applied by either gNBl or gNB2

[0147] • gNBl-DU has not yet applied the proposed configuration change which it had derived in response to the predicted CCO issue, however, that change might no longer be valid.

[0148] • gNBl-DU determines the new coverage state and informs gNBl-CU.

[0149] • gNBl-CU informs gNB2-CU of cancellation of the earlier predicted CCO issue and the future coverage state. It also includes the information of currently detected CCO issue and the CCO state applied by the gNBl -DU in response to the CCO issue detected by gNBl -CU.

[0150] • gNB2-CU derives a new CCO issue based on the information received from the gNBl- CU and informs gNB2-DU of the change. • gNB2-DU determines the new coverage state in response to the latest information and informs the gNB2-CU

[0151]

[0122] Case 2a: If gNB 1 -DU has applied the “future” CCO state configuration and a new CCO issue is detected at gNBl-CU

[0152] • gNBl-CU informs the detected CCO issue to gNB2-DU with an additional information to reverse the recently applied change to the CCO configuration.

[0153] • The rest of the flow will follow the legacy mechanism.

[0154] • This will result in gNB2-CU and gNB2-DU to also cancel the prediction based CCO configuration, and follow the legacy mechanism.

[0155]

[0123] Case 2b: If gNBl-DU has applied the “future” CCO state configuration and a new CCO issue is detected at gNB2-CU

[0156] • This might require a careful handling. It is possible that the CCO issue has been detected at gNB2-DU in response to the CCO state configuration at gNBl-DU.

[0157] • RAN3 could recommend a hysteresis period after a CCO state configuration change during which no CCO detection is possible.

[0158]

[0124] Case 3a: If gNB2-DU has applied the “future” CCO state configuration and a new CCO issue is detected at gNBl-CU

[0159] • gNBl-CU will not be aware that the gNB2-DU has applied the future CCO state configuration.

[0160] • Therefore, the flow will be same as Case 1 .

[0161]

[0125] Case 3b: If gNB2-DU has applied the “future” CCO state configuration and a new CCO issue is detected at gNB2-CU This might require a careful handling. It is possible that in response to the configuration change at gNB2-DU, this issue has been detected, but it’s possible that gNBl-DU has not yet applied its own CCO configuration change.

[0162] • RAN3 could recommend that tO+q is greater than tO+y then gNB2-DU could potentially treat this like a legacy CCO issue.

[0163] • There are other permutations / combinations of the above, and which may be studied carefully to avoid comer cases which might result in frequency changes of configuration and an unstable system behavior.

[0164]

[0126] Proposal 2: RAN3 to consider the interaction of legacy CCO mechanism and the AI / ML based CCO mechanism and decide on recommendations to avoid conflicts between the two.

[0165]

[0127] 3 Conclusion

[0166] • Observation 1: Since gNB2-CU gets the information about the predicted CCO issue, and the future CCO state of gNBl ahead of the prediction time and the updated configuration time, it is likely that it will deduce its own CCO issue ahead of that time. And similarly, it is likely that gNB2-DU will also recommend its own future CCO state ahead of the time.

[0167] • Proposal 1 : It is proposed that the timing information agreed in the previous meeting needs to extend to gNB2 also, wherein the predicted CCO issue and future CCO state timing of gNB 1 and gNB2 might differ.

[0168] • Proposal 2: RAN3 to consider the interaction of legacy CCO mechanism and the AI / ML based CCO mechanism and decide on recommendations to avoid conflicts between the two.

[0169]

[0128] B. Conflict between legacy CCO and Al / ML-based CCO

[0170]

[0129] 1. Introduction

[0171] • RAN3#125-bis meeting agreed that: Timing information: Introducing separate time information for future coverage state and predicted CCO issue.

[0172] • Time for future coverage state: The point in time when the future coverage state will be applied.

[0173] • Time for predicted CCO issue: The point in time when the CCO issue is predicted to happen.

[0174]

[0130] 2. Discussion

[0175] • In the legacy CCO mechanism:

[0176] • Step 1 : The gNB-CU detects the CCO issue (i.e., either coverage issue or cell edge capacity issue) and affected cells / SSBs;

[0177] • Step 2: The gNB-CU sends the CCO Assistance Information IE (within the GNB-CU CONFIGURATION UPDATE message over Fl) containing the CCO issue type and affected cells / SSBs to the gNB-DU;

[0178] • Step3: The gNB-DU uses the received CCO Assistance Information IE to determine the new coverage state (CCO state) for the affected gNB-DU’s cells / SSBs;

[0179] • Step 4: The gNB-DU sends the Coverage Modification Notification IE (within the GNB-DU CONFIGURATION UPDATE message over Fl) containing the new CCO state for the affected gNB-DU’s cells / SSBs to the gNB-CU;

[0180] • Step 5: The gNB-CU sends the Coverage Modification List IE (within the NG-RAN NODE CONFIGURATION UPDATE message over Xn) containing the detected CCO issue type and the new CCO state for the affected gNB-DU’s cells / SSBs to the neighbour gNB to allow- for the neighbour gNB to determine a new CCO state for its own cells / SSBs that best matches the new' CCO state of the gNB-CU’s cells / SSBs resulting from the detected CCO issue. For the AIML-based CCO mechanism:

[0181] • Step 0: AIML training entity generates predicted CCO issue

[0182] • Step 1 : gNB-CU predicts the CCO issue.

[0183] • Step 2: gNB-CU sends the predicted CCO issue to gNB-DU.

[0184] • Step 3: gNB-DU generates the future coverage status based on the predicted CCO issue and other local information, whether only local information is used can be further discussed.

[0185] • Step 4: gNB-DU sends the future coverage status to gNB-CU.

[0186] • Step 5: gNB-CU sends the detected CCO issue and the new CCO state for the affected gNB-DU’s cells / SSBs to neighbour gNBs (say gNB2) along with the timing information.

[0187] • Step 7: gNB2-CU sends this information to gNB2-DU which decides its own coverage / capacity configuration (future CCO state) in response to gNBl-DU’s future CCO state.

[0188]

[0131] Observation 1: Since gNB2 gets the information about the predicted CCO issue, and the future CCO state of gNBl ahead of the CCO issue prediction time and future CCO state time, it is likely that it will infer its own future CCO state ahead of the CCO issue prediction and future CCO state time of gNBl.

[0189] • Since there is time lag between the prediction and the predicted time of CCO issue, there are chances that in the interim another CCO issue could get predicted or detected by the same gNB, but the neighbouring gNB, or by a third gNB altogether. Some of these cases are listed below, it is not an exhaustive list, but broad principles have been described for addressing a conflict arising between predicted issues and detected (or another predicted) issue.

[0190] • Notel : While, it is implementation dependent, it is also assumed that gNBl-DU might apply the future CCO state little before the predicted time of the CCO issue. This will avoid the CCO issue (for accurate predictions) from being detected also. • Note2: While, it is implementation dependent, it is also assumed that gNB2-DU might apply the future CCO state at the same (or similar) time as the time when gNB2-DU applies the future CCO state.

[0191] • For simplicity of discussion, assuming the following notations:

[0192] • tO = current time

[0193] • tO+p = the time at which the CCO issue is supposed to occur

[0194] • tO+a = the time at which gNBl-DU and gNB2-DU apply their respective future CCO state

[0195] • For added simplicity, it may be assumed that tO+p and tO+a are close together.

[0196] • With the above assumptions in place, following are some of the scenarios where conflicts could occur:

[0197]

[0132] Case 1: gNBl detects an issue ahead of the predicted CCO issue time (issue is detected between tO and tO+p)

[0198]

[0133] Case 2: gNB2 detects an issue ahead of the future CCO state time of gNBl and gNB2 (issue is detected between tO and tO+a)

[0199]

[0134] Case 3: gNBx detects an issue ahead of the predicted CCO issue time (issue is detected between tO and tO+p)

[0200]

[0135] First principle relies on the fact that detected issue is more important than the predicted issue. For each case of detected CCO issue, the present disclosure proposes that the CCO state in response to the detected CCO issue overrides the actions in response to the predicted CCO state.

[0201]

[0136] Proposal 1 The present disclosure proposes that the CCO state in response to the detected CCO issue overrides the actions in response to the predicted CCO state.

[0202]

[0137] For Case 1. if gNBl-CU detects an issue between tO and tO+p: In this case, the present disclosure assumes that gNBl-DU has not yet applied the proposed configuration change which it had derived in response to the predicted CCO issue, however, that change might no longer be valid. The legacy steps may follow:

[0203] • gNBl-CU informs the gNBl-DU of the detected CCO issue.

[0204] • gNBl-DU determines and applies the new CCO state and informs gNBl-CU

[0205] • gNBl-CU informs gNB2-CU of currently detected CCO issue and the CCO state applied by the gNBl-DU in response to the CCO issue detected by gNBl-CU.

[0206] • gNB2-CU forwards the information received from the gNBl-CU to gNB2-DU

[0207] • gNB2-DU determines and applies the new CCO state in response to the latest information and informs the gNB2-CU

[0208] • After this, gNBl-CU reviews the earlier prediction, and if it is still valid for time tO+p, then it takes no further action. This implies that gNB 1 -DU and gNB2-DU will proceed to apply their respective future CCO states at tO+a. However, this is an unlikely case. At the time of review if gNBl-CU changes its prediction, then:

[0209] • gNBl-CU informs gNBl-DU of the cancellation of the earlier prediction over Fl.

[0210] • It also informs gNB2-DU of the same over Xn interface, which forw ards it to gNB2- DU over F 1.

[0211] • Both gNB 1 -DU and gNB2-DU cancel their future CCO states and inform the respective gNB-CUs.

[0212]

[0138] For Case 2, if gNB2-CU detects an issue between tO and tO+p:

[0213] In this case, it is assumed that gNBl-DU and gNB2-DU has not yet applied the proposed configuration changes for the CCO states which had been derived in response to the predicted

[0214] CCO issue at gNBl. The legacy steps may follow:

[0215] • gNB2-CU informs the gNB2-DU of the detected CCO issue. • gNB2-DU determines and applies the new CCO state and informs gNB2-CU

[0216] • gNB2-CU informs gNBl-CU of currently detected CCO issue and the CCO state applied by the gNB2-DU in response to the CCO issue detected by gNB2-CU.

[0217] • gNBl-CU forwards the information received from the gNB2-CU to gNBl-DU

[0218] • gNBl-DU determines and applies the new CCO state in response to the latest information and informs the gNBl-CU.

[0219] • After this, gNBl-CU reviews the earlier prediction, and if it is still valid for time tO+p, then it takes no further action. This implies that gNB 1 -DU and gNB2-DU will proceed to apply their respective future CCO states at tO+a. However, this may be an unlikely case. At the time of review if gNBl-CU changes its prediction, then:

[0220] • gNBl-CU informs gNBl-DU of the cancellation of the earlier prediction over Fl.

[0221] • It also informs gNB2-DU of the same over Xn interface, which forwards it to gNB2- DU over Fl.

[0222] • Both gNB 1 -DU and gNB2-DU cancel their future CCO states and inform the respective gNB-CUs.

[0223]

[0139] For Case 3, if gNBx-CU detects an issue between tO and tO+p

[0224] In this case, it is assumed that gNBl-DU and gNB2-DU has not yet applied the proposed configuration changes for the CCO states which had been derived in response to the predicted CCO issue at gNBl. The legacy steps may follow between gNBx and the neighbouring gNBs.

[0225]

[0140] Case 3a: If the neighbouring gNB in this case is gNB2, then the following steps may take place:

[0226] • gNBx-CU informs the gNBx-DU of the detected CCO issue.

[0227] • gNBx-DU determines and applies the new CCO state and informs gNBx-CU • gNBx-CU informs gNB2-CU of currently detected CCO issue and the CCO state applied by the gNBx-DU in response to the CCO issue detected by gNBx-CU.

[0228] • gNB2-CU forwards the information received from the gNBx-CU to gNB2-DU

[0229] • gNB2-DU determines and applies the new CCO state in response to the latest information and informs the gNB2-CU.

[0230] • Since there is a pending future CCO state action at gNB2-DU based on the prediction of CCO issue by gNBl-CU, the present disclosure proposes that gNB2-CU informs gNBl-CU of the change in the CCO state of gNB2-DU. After this. gNBl-CU reviews the earlier prediction, and if it is still valid for time tO+p. then it takes no further action. This implies that gNBl-DU and gNB2-DU will proceed to apply their respective future CCO states at tO+a. At the time of review, if gNBl-CU changes its prediction, then the steps are followed as if a new CCO issue has been predicted, and the older CCO issue has been cancelled.

[0231] • gNBl-CU informs gNBl-DU of the cancellation of the earlier prediction over Fl.

[0232] • It also informs gNB2-DU of the same over Xn interface, which forwards it to gNB2- DU over Fl.

[0233] • Additionally, the change in prediction is treated the same as a new prediction of CCO issue at gNB 1.

[0234]

[0141] Case 3a: If the neighbouring gNB in this case is gNBl, then the following steps may take place:

[0235] • gNBx-CU informs the gNBx-DU of the detected CCO issue.

[0236] • gNBx-DU determines and applies the new CCO state and informs gNBx-CU

[0237] • gNBx-CU informs gNBl-CU of currently detected CCO issue and the CCO state applied by the gNBx-DU in response to the CCO issue detected by gNBx-CU.

[0238] • gNBl-CU forwards the information received from the gNBx-CU to gNBl-DU gNBl -DU determines and applies the new CCO state in response to the latest information and informs the gNBl-CU.

[0239] • After this, gNB 1 -CU reviews the earlier prediction, and if it is still valid for time tO+p, then it takes no further action. This implies that gNB 1 -DU and gNB2-DU will proceed to apply their respective future CCO states at tO+a. However, this may be an unlikely case. At the time of review, if gNBl-CU changes its prediction, then the steps are followed as if a new CCO issue has been predicted, and the older CCO issue has been cancelled.

[0240] • gNB 1 -CU informs gNB 1 -DU of the cancellation of the earlier prediction over F 1.

[0241] • It also informs gNB2-DU of the same over Xn interface, which forwards it to gNB2- DU over Fl.

[0242] • Additionally, the change in prediction is treated the same as a new prediction of CCO issue at gNBl.

[0243]

[0142] Proposal 2 RAN3 to discuss the various scenarios possible due to conflict in predicted and detected CCO issues and invite proposals for solutions of the same.

[0244] The present disclosure discloses scenarios above based on the assumption that the future CCO state has not been applied by gNBl -DU or gNB2-DU in case a conflict occurs. It is part of future study to take that scenario also into consideration and propose solutions to tackle it.

[0245]

[0143] 3. Conclusion

[0246] • Observation 1: Since gNB2 gets the information about the predicted CCO issue, and the future CCO state of gNBl ahead of the CCO issue prediction time and future CCO state time, it is likely that it will infer its own future CCO state ahead of the CCO issue prediction and future CCO state time of gNBl.

[0247] • Proposal 1 The present disclosure proposes that the CCO state in response to the detected CCO issue overrides the actions in response to the predicted CCO state. • Proposal 2 RAN3 to discuss the various scenarios possible due to conflict in predicted and detected CCO issues and invite proposals for solutions of the same.

[0248]

[0144] Example Items:

[0249]

[0145] Item 1. A method comprising: predicting, using a proactive Coverage and Capacity Optimization (CCO) mechanism, a first CCO issue potentially occurring at a first time instance within a first Radio Access Network (RAN) node; determining, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node to be applied at a second time instance for mitigating the predicted first CCO issue; and transmitting a notification to a second RAN node for determining, about the first CCO configuration for the first RAN node to be applied at the second time instance .

[0250]

[0146] Item 2. The method of item 1, further comprising: detecting, using a reactive CCO mechanism, a third CCO issue occurring at a particular time instance time instance within the first RAN node; determining whether the detection of the third CCO issue conflicts with the prediction of the first CCO issue; and upon determining that the detection of the third CCO issue conflicts with the prediction of the first CCO issue, resolving the conflict between the detection of the third CCO issue and the prediction of the first CCO issue.

[0251]

[0147] Item 3. The method of item 2, wherein determining that the detection of the third CCO issue conflicts with the prediction of the first CCO issue comprises: determining that the particular time instance falls within a predefined time window; and determining that at least one of following conditions is met: the first CCO issue differs from the third CCO issue; and the first time instance differs from the particular time instance.

[0252]

[0148] Item 4. The method of item 2 or 3, wherein resolving the conflict between the detection of the third CCO issue and the prediction of the first CCO issue comprises: upon determining that neither the first RAN node nor the second RAN has applied their respective CCO configurations: discarding the first CCO configuration and determining, based at least on the detected third CCO issue, a third CCO configuration to be applied at the first RAN node; and transmitting another notification to the second RAN node comprising information associated with the determined third CCO configuration, wherein the other notification enables the second RAN node to selectively discard the second CCO configuration and apply a fourth CCO configuration, based on the third CCO configuration.

[0253]

[0149] Item 5. The method of any of items 2-4, wherein resolving the conflict between the detection of the third CCO issue and the prediction of the first CCO issue comprises: upon determining at least one of: the first RAN node has applied the first CCO configuration, and the second RAN node has applied the second CCO configuration, performing following: determining that the detection of the third CCO issue as an invalid detection when a time of the detection falls within a predefined waiting time-window, wherein the predefined waiting time-window starts application of the first or second CCO configuration; and determining the detection of the third CCO issue as a valid detection when the time of the detection falls beyond the predefined waiting time-window.

[0254]

[0150] Item 6. The method of item 5, further comprising: upon determining the detection of the third CCO issue as the valid detection, performing at least one of: reversing configuration changes already applied by at least one of the first RAN node and the second RAN node; and preventing pending configuration changes from being applied by at least one of the first RAN node and the second RAN node.

[0255]

[0151] Item 7. The method of any of items 1-6, wherein the proactive CCO mechanism comprises an Artificial Intelligence (Al) or Machine learning (ML) based CCO mechanism, and wherein the first or the second CCO issue comprises at least one of: a cell coverage issue and a cell capacity issue.

[0152] Item 8. The method of any of items 1-7, wherein the notification transmitted to the second RAN node selectively enables the second RAN node to determine a second CCO issue potentially occurnng within the second RAN node at a third time instance and a second CCO configuration to be applied at a fourth time instance at the second RAN node, and wherein the first time instance differs from the third time instance and the second time instance differs from the fourth time instance.

[0256]

[0153] Item 9. The method of any of items 1-8. wherein the first RAN node comprises a first base station and the second RAN node comprises a second base station each configured to implement a disaggregated RAN architecture, and wherein the first base station comprises a Central Unit (CU) communicatively coupled with a plurality of Distributed Unit (DUs).

[0257]

[0154] Item 10. The method of item 9, further comprising: predicting, by the CU, the first CCO issue potentially occurring at the first time instance, wherein the first CCO issue is configured to affect at least one cell served by at least one DU of the plurality of DUs; transmitting, by the CU, information associated with the predicted first CCO issue and the affected at least one cell to the at least one DU; determining, by the at least one DU, the first CCO configuration for the affected at least one cell based at least on the predicted first CCO issue; transmitting, by the at least one DU, information associated with the first CCO configuration to the CU; and transmitting, by the CU, the notification to the second RAN node for determining the second CCO issue and the second CCO configuration that best matches the first CCO configuration resulting from the predicted first CCO issue.

[0258]

[0155] Item 11. An apparatus configured to: predict, using a proactive Coverage and Capacity Optimization (CCO) mechanism, a first CCO issue potentially occurring at a first time instance within a first Radio Access Network (RAN) node; determine, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node to be applied at a second time instance for mitigating the predicted first CCO issue; and transmit a notification to a second RAN node for determining, about the first CCO configuration for the first RAN node to be applied at the second time instance.

[0259]

[0156] Item 12. The apparatus of item 11. further configured to: detect, using a reactive CCO mechanism, a third CCO issue occurring at a particular time instance within the first RAN node; determine whether the detection of the third CCO issue conflicts with the prediction of the first CCO issue; and upon determining that the detection of the third CCO issue conflicts with the prediction of the first CCO issue, resolve the conflict between the detection of the third CCO issue and the prediction of the first CCO issue.

[0260]

[0157] Item 13. The apparatus of item 12. wherein to determine that the detection of the third CCO issue conflicts with the prediction of the first CCO issue, the apparatus is configured to: determine that the particular time instance falls within a predefined time window; and determine that at least one of following conditions is met: the first CCO issue differs from the third CCO issue; and the first time instance differs from the particular time instance.

[0261]

[0158] Item 14. The apparatus of item 12 or 13, wherein to resolve the conflict between the detection of the third CCO issue and the prediction of the first CCO issue, the apparatus is configured to: upon determining that neither the first RAN node nor the second RAN has applied their respective CCO configurations: discard the first CCO configuration and determining, based at least on the detected third CCO issue, a third CCO configuration to be applied at the first RAN node; and transmit another notification to the second RAN node comprising information associated with the determined third CCO configuration, wherein the other notification enables the second RAN node to selectively discard the second CCO configuration and apply a fourth CCO configuration, based on the third CCO configuration.

[0159] Item 15. The apparatus of any of items 12-14, wherein to resolve the conflict between the detection of the third CCO issue and the prediction of the first CCO issue, the apparatus is configured to: upon determining at least one of: the first RAN node has applied the first CCO configuration, and the second RAN node has applied the second CCO configuration, perform following: determine that the detection of the third CCO issue as an invalid detection when a time of the detection falls within a predefined waiting time-window, wherein the predefined waiting time- window starts application of the first or second CCO configuration; and determine the detection of the third CCO issue as a valid detection when the time of the detection falls beyond the predefined waiting time-window.

[0262]

[0160] Item 16. The apparatus of item 15, further configured to: upon determining the detection of the third CCO issue as the valid detection, perform at least one of: reverse configuration changes already applied by at least one of the first RAN node and the second RAN node; and prevent pending configuration changes from being applied by at least one of the first RAN node and the second RAN node.

[0263]

[0161] Item 17. The apparatus of any of items 11-16, wherein the proactive CCO mechanism comprises an Artificial Intelligence (Al) or Machine learning (ML) based CCO mechanism, and wherein the first or the second CCO issue comprises at least one of: a cell coverage issue and a cell capacity issue.

[0264]

[0162] Item 18. The apparatus of any of items 11-17, wherein the notification transmitted to the second RAN node selectively enables the second RAN node to determine a second CCO issue potentially occurring within the second RAN node at a third time instance and a second CCO configuration to be applied at a fourth time instance at the second RAN node, and wherein the first time instance differs from the third time instance and the second time instance differs from the fourth time instance.

[0163] Item 19. The apparatus of any of items 1 1-18, wherein the first RAN node comprises a first base station and the second RAN node comprises a second base station each configured to implement a disaggregated RAN architecture, and wherein the first base station comprises a Central Unit (CU) communicatively coupled with a plurality of Distributed Unit (DUs).

[0265]

[0164] Item 20. A non-transitory computer readable media storing one or more computer executable instructions which, when executed by an apparatus, cause the apparatus to: predict, using a proactive Coverage and Capacity Optimization (CCO) mechanism, a first CCO issue potentially occurring at a first time instance within a first Radio Access Network (RAN) node; determine, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node to be applied at a second time instance for mitigating the predicted first CCO issue; and transmit a notification to a second RAN node, about the first CCO configuration for the first RAN node to be applied at the second time instance.

[0266]

[0165] It may be noted here that the subj ect matter of some or all embodiments described with reference to Figures 1-8 may be relevant for the methods 1000, 1100 and the same is not repeated for the sake of brevity. The language used in the specification has been principally selected for readability7and instructional purposes, and it may not have been selected to delineate or circumscribe the inventive subject matter. It is therefore intended that the scope of the disclosure be limited not by this detailed description, but rather by any claims that issue on an application based here on. Accordingly, the embodiments of the present disclosure are intended to be illustrative, but not limiting, of the scope of the disclosure, which is set forth in the appended claims.

Claims

We Claim:

1. A method comprising: predicting, using a proactive Coverage and Capacitv Optimization (CCO) mechanism, a first CCO issue potentially occurring at a first time instance within a first Radio Access Network (RAN) node; determining, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node to be applied at a second time instance for mitigating the predicted first CCO issue; and transmitting a notification to a second RAN node, about the first CCO configuration for the first RAN node to be applied at the second time instance.

2. The method of claim 1, further comprising: detecting, using a reactive CCO mechanism, a third CCO issue occurring at a particular time instance within the first RAN node; determining whether the detection of the third CCO issue conflicts with the prediction of the first CCO issue; and upon determining that the detection of the third CCO issue conflicts with the prediction of the first CCO issue, resolving the conflict between the detection of the third CCO issue and the prediction of the first CCO issue.

3. The method of claim 2, wherein determining that the detection of the third CCO issue conflicts with the prediction of the first CCO issue comprises: determining that the particular time instance falls within a predefined time window; anddetermining that at least one of following conditions is met: the first CCO issue differs from the third CCO issue; and the first time instance differs from the particular time instance.

4. The method of claim 2. wherein resolving the conflict between the detection of the third CCO issue and the prediction of the first CCO issue comprises: upon determining that neither the first RAN node nor the second RAN has applied their respective CCO configurations: discarding the first CCO configuration and determining, based at least on the detected third CCO issue, a third CCO configuration to be applied at the first RAN node; and transmitting another notification to the second RAN node comprising information associated with the determined third CCO configuration, wherein the other notification enables the second RAN node to selectively discard the second CCO configuration and apply a fourth CCO configuration, based on the third CCO configuration.

5. The method of claim 2, wherein resolving the conflict between the detection of the third CCO issue and the prediction of the first CCO issue comprises: upon determining at least one of: the first RAN node has applied the first CCO configuration, and the second RAN node has applied the second CCO configuration, performing following: determining that the detection of the third CCO issue as an invalid detection when a time of the detection falls within a predefined waiting time-window, whereinthe predefined waiting time-window starts application of the first or second CCO configuration; and determining the detection of the third CCO issue as a valid detection when the time of the detection falls beyond the predefined waiting time-window.

6. The method of claim 5, further comprising: upon determining the detection of the third CCO issue as the valid detection, performing at least one of: reversing configuration changes already applied by at least one of the first RAN node and the second RAN node; and preventing pending configuration changes from being applied by at least one of the first RAN node and the second RAN node.

7. The method of claim 1, wherein the proactive CCO mechanism comprises an Artificial Intelligence (Al) or Machine learning (ML) based CCO mechanism, and wherein the first or the second CCO issue comprises at least one of: a cell coverage issue and a cell capacity7issue.

8. The method of claim 1, wherein the notification transmitted to the second RAN node selectively enables the second RAN node to determine a second CCO issue potentially occurring within the second RAN node at a third time instance and a second CCO configuration to be applied at a fourth time instance at the second RAN node, and wherein the first time instance differs from the third time instance and the second time instance differs from the fourth time instance.

9. The method of claim 1, wherein the first RAN node comprises a first base station and the second RAN node compnses a second base station each configured to implement a disaggregated RAN architecture, and wherein the first base station comprises a Central Unit (CU) communicatively coupled with a plurality of Distributed Unit (DUs).

10. The method of claim 9, further comprising: predicting, by the CU, the first CCO issue potentially occurring at the first time instance, wherein the first CCO issue is configured to affect at least one cell served by at least one DU of the plurality of DUs; transmitting, by the CU, information associated with the predicted first CCO issue and the affected at least one cell to the at least one DU; determining, by the at least one DU, the first CCO configuration for the affected at least one cell based at least on the predicted first CCO issue; transmitting, by the at least one DU, information associated with the first CCO configuration to the CU; and transmitting, by the CU, the notification to the second RAN node for determining the second CCO issue and the second CCO configuration that best matches the first CCO configuration resulting from the predicted first CCO issue.

11. An apparatus configured to: predict, using a proactive Coverage and Capacity Optimization (CCO) mechanism, a first CCO issue potentially occurring at a first time instance within a first Radio AccessNetwork (RAN) node;determine, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node to be applied at a second time instance for mitigating the predicted first CCO issue; and transmit a notification to a second RAN node, about the first CCO configuration for the first RAN node to be applied at the second time instance.

12. The apparatus of claim 11 , further configured to: detect, using a reactive CCO mechanism, a third CCO issue occurring at a particular time instance wi thin the first RAN node; determine whether the detection of the third CCO issue conflicts with the prediction of the first CCO issue; and upon determining that the detection of the third CCO issue conflicts with the prediction of the first CCO issue, resolve the conflict between the detection of the third CCO issue and the prediction of the first CCO issue.

13. The apparatus of claim 12, wherein to determine that the detection of the third CCO issue conflicts with the prediction of the first CCO issue, the apparatus is configured to: determine that the particular time instance falls within a predefined time window; and determine that at least one of follow ing conditions is met: the first CCO issue differs from the third CCO issue; and the first time instance differs from the particular time instance.

14. The apparatus of claim 12, wherein to resolve the conflict between the detection of the third CCO issue and the prediction of the first CCO issue, the apparatus is configured to:upon determining that neither the first RAN node nor the second RAN has applied their respective CCO configurations: discard the first CCO configuration and determining, based at least on the detected third CCO issue, a third CCO configuration to be applied at the first RAN node; and transmit another notification to the second RAN node comprising information associated with the determined third CCO configuration, wherein the other notification enables the second RAN node to selectively discard the second CCO configuration and apply a fourth CCO configuration, based on the third CCO configuration.

15. The apparatus of claim 12, wherein to resolve the conflict between the detection of the third CCO issue and the prediction of the first CCO issue, the apparatus is configured to: upon determining at least one of: the first RAN node has applied the first CCO configuration, and the second RAN node has applied the second CCO configuration, perform following: determine that the detection of the third CCO issue as an invalid detection when a time of the detection falls within a predefined w aiting time-window, wherein the predefined waiting time-window starts application of the first or second CCO configuration; and determine the detection of the third CCO issue as a valid detection when the time of the detection falls beyond the predefined waiting time-window.

16. The apparatus of claim 15, further configured to:upon determining the detection of the third CCO issue as the valid detection, perform at least one of: reverse configuration changes already applied by at least one of the first RAN node and the second RAN node; and prevent pending configuration changes from being applied by at least one of the first RAN node and the second RAN node.

17. The apparatus of claim 11. wherein the proactive CCO mechanism comprises an Artificial Intelligence (Al) or Machine learning (ML) based CCO mechanism, and wherein the first or the second CCO issue comprises at least one of: a cell coverage issue and a cell capacity issue.

18. The apparatus of claim 11 , wherein the notification transmitted to the second RAN node selectively enables the second RAN node to determine a second CCO issue potentially occurring within the second RAN node at a third time instance and a second CCO configuration to be applied at a fourth time instance at the second RAN node, and wherein the first time instance differs from the third time instance and the second time instance differs from the fourth time instance.

19. The apparatus of claim 11, wherein the first RAN node comprises a first base station and the second RAN node comprises a second base station each configured to implement a disaggregated RAN architecture, and wherein the first base station comprises a Central Unit (CU) communicatively coupled w ith a plurality of Distributed Unit (DUs).

20. A non-transitory computer readable media storing one or more computer executable instructions which, when executed by an apparatus, cause the apparatus to: predict, using a proactive Coverage and Capacity Optimization (CCO) mechanism, a first CCO issue potentially occurring at a first time instance within a first Radio Access Network (RAN) node; determine, based at least on the predicted first CCO issue, a first CCO configuration for the first RAN node to be applied at a second time instance for mitigating the predicted first CCO issue; and transmit a notification to a second RAN node, about the first CCO configuration for the first RAN node to be applied at the second time instance.

Images