Techniques for generating a digital twin model to perform energy saving in mobile communication network

Abstract

Disclosed herein is a method and an apparatus (202) for generating a digital twin model (204) to perform energy saving in a mobile communication network (124). The method comprises processing data of network entities and user equipment (UEs) to determine network coverage dataand generating an initial digital twin model (204) with initial location estimates of the UEs based on the network coverage dataand network entity data. The method comprises determining locations of the UEs by dynamically updating the initial location estimates to reduce error between data and modelled data of the UEs, determined using the initial digital twin model (204) and generating the digital twin model (204) by updating the locations of the UEs in the initial digital twin model (204). The method comprises performing one or more energy saving decisions of switching off one or more network entities based on coverage indicators using the digital twin model (204).

Description

[0001] TECHNIQUES FOR GENERATING A DIGITAL TWIN MODEL TO PERFORM ENERGY SAVING IN MOBILE COMMUNICATION NETWORK

[0002] FIELD OF THE DISCLOSURE

[0003] The present disclosure in general relates to mobile communication and more particularly, to techniques (e.g., a method and an apparatus) for generating a digital twin model for performing energy saving in a mobile communication network.

[0004] BACKGROUND

[0005] Open Radio Access Network (O-RAN) is a type of mobile communication network that enables open communication interfaces between one or more of Open Control Units (O-CUs), Open Distributed Units (O-DUs), and Open Radio Units (O-RUs)and a Service Management and Orchestration (SMO) Framework. The Open Control Units (O-CUs), Open Distributed Units (O-DUs) and Open Radio Units (O-RUs) may be together referred to herein as ORAN gNodeB (gNB) components.

[0006] The ORAN architecture may also comprise few control components such as, but not limited to, a non-real time RAN Intelligent Controller (Non-RT RIC), and a near real time RIC (Near RT RIC) to control functionalities of the ORAN gNB components. The near RT RIC may directly control near real time mobile communications between a plurality of User Equipment (UEs) and the plurality of O-RUs. Alternatively, the non-RT RIC may monitor real time mobile communications and provide higher layer policies to optimize functions of the plurality of O- RUs. The higher layer policies may be related to different RAN management and optimization techniques, for example, but not limited to, energy saving techniques, traffic prediction techniques and the like.

[0007] In general, the non-RT RIC may monitor the mobile communications based on data received from the near RT RIC. However, the data may be provided in the form of aggregated data related to the plurality of UEs, such as, but not limited to, statistics of the plurality of UEs, without identifying data associated with each UE. Thus, providing higher layer policies based on such vague or ambiguous data and implementing the policies in the near RT RIC and other network elements may not be reliable and may lead to adverse effects on the mobile communication network.

[0008] One method of addressing the problem may be by modelling the mobile communication network using a digital twin model to simulate impact of the higher layer policies. Such modelling may require accurate data related to positions and receiver characteristics of each UE, clutter and terrain information and channel models at deployment.

[0009] However, collecting position and receiver characteristics of each UE and providing the collected data to the SMO may be data intensive, bandwidth intensive as well as may involve privacy concerns. Furthermore, prediction of network coverage data may result in processor intensive solution and collection of clutter information may not accurately model non- stationary clutter in urban environments. Moreover, a digital twin model thus generated based on the inaccurate data may not be able to assist the non-RT RIC to provide reliable higher layer policies such as, energy saving techniques.

[0010] Thus, there is a need for techniques for efficiently generating the digital twin model for the mobile communication network.

[0011] The information disclosed in this background of the disclosure section is only for enhancement of understanding of the general background of the disclosure and should not be taken as acknowledgment or any form of suggestion that this information forms prior art already known to a person skilled in the art.

[0012] SUMMARY

[0013] Disclosed herein is a method for generating a digital twin model to perform energy saving in a mobile communication network. The method comprising processing network entity data of a plurality of network entities and signal quality data of a plurality of User Equipment (UEs) of the mobile communication network using a neural network. The method comprising determining network coverage data associated with the plurality of network entities using the neural network based on the processing. The method comprising generating an initial digital twin model corresponding to the mobile communication network based on the network entity data and the network coverage data. The method comprising determining locations of the plurality of UEs served by the plurality of network entities based at least on the network entity data and the signal quality data using an optimization model and the initial digital twin model. Further, the method comprising generating the digital twin model by updating the initial digital twin model with the locations of the plurality of UEs and performing energy saving in the mobile communication network using the digital twin model.

[0014] Also disclosed herein is an apparatus to generate a digital twin model to perform energy saving in a mobile communication network. The apparatus comprising a memory and a processor coupled with the memory. The processor is configured to process network entity data of a plurality of network entities and signal quality data of a plurality of User Equipment (UEs) of the mobile communication network using a neural network. The processor is configured to determine network coverage dataassociated with the plurality of network entities using the neural network based on the processing. The processor is configured to generate an initial digital twin model corresponding to the mobile communication network based on the network entity data and the network coverage data. The processor is configured to determine locations of the plurality of UEs served by the plurality of network entities based at least on the network entity data and the signal quality data using an optimization model and the initial digital twin model. The processor is configured to generate the digital twin model by updating the initial digital twin model with the locations of the plurality of UEs and perform energy saving in the mobile communication network using the digital twin model.

[0015] The foregoing summary is illustrative only and is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description.

[0016] BRIEF DESCRIPTION OF DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the figures to reference like features and components. Some embodiments of device or system and / or methods in accordance with embodiments of the present subject matter are now described, by way of example only, and with reference to the accompanying figures, in which:

[0018] Fig- 1 illustrates an exemplary Open Radio Access Network (ORAN) architecture of a mobile communication network;

[0019] Fig. 2a illustrates an exemplary system architecture for generating a digital twin model to perform energy saving in a mobile communication network, in accordance with an embodiment of the present disclosure;

[0020] Fig. 2b illustrates an exemplary digital twin model to perform energy saving in the mobile communication network, in accordance with an embodiment of the present disclosure;

[0021] Fig- 3 illustrates an exemplary block diagram of Digital twin model Generation and Energy saving System (DGES) to generate the digital twin model to perform energy saving in the mobile communication network, in accordance with an embodiment of the present disclosure;

[0022] Fig- 4 illustrates an exemplary flowchart of a method to generate the digital twin model to perform energy saving in the mobile communication network, in accordance with an embodiment of the present disclosure;

[0023] Fig. 5 illustrates an exemplary flowchart of a method to perform energy saving in the mobile communication network using the digital twin model, in accordance with an embodiment of the present disclosure; and

[0024] Fig- 6 illustrates an exemplary flowchart of a method to generate the digital twin model to perform energy saving in the mobile communication network, in accordance with another embodiment of the present disclosure.

[0025] The figures depict embodiments of the disclosure for purposes of illustration only. One skilled in the art will readily recognize from the following description that alternative embodiments of the structures and methods illustrated herein may be employed without departing from the principles of the disclosure described herein.

[0026] DETAILED DESCRIPTION

[0027] In the present document, the word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment or implementation of the present subject matter described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

[0028] While the disclosure is susceptible to various modifications and alternative forms, specific embodiment thereof has been shown by way of example in the drawings and will be described in detail below. It should be understood, however, that it is not intended to limit the disclosure to the particular forms disclosed, but on the contrary, the disclosure is to cover all modifications, equivalents, and alternative falling within the scope of the disclosure.

[0029] The terms “comprises”, “comprising”, or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a setup, device or method that comprises a list of

[0030] 5 components or steps does not include only those components or steps but may include other components or steps not expressly listed or inherent to such setup or device or method. In other words, one or more elements in a device or system or apparatus proceeded by “comprises. . . a” does not, without more constraints, preclude the existence of other elements or additional elements in the device or system or apparatus.

[0031] The present disclosure relates to various methods and apparatuses for generating a digital twin model for performing energy savings in a mobile communication network. The apparatus receives network entity data and signal quality data from the near RT RIC. The network entity data comprises data related to a plurality of network entities of the mobile communication network. The network entity may be defined as a base station or an O-RU serving a cell or a sector of the cell. For example, the network entity data comprises locations and configuration of the plurality of network entities, terrain information and clutter information. The signal quality data comprises probabilistic distribution of values of one or more signal quality parameters of a plurality of User Equipment (UEs) associated with the plurality of network entities. The one or more signal quality parameters comprise Channel Quality Indicator (CQI), Received Signal Strength Indicator (RS SI), Reference Signal Received Power (RSRP), throughput and the like.

[0032] Further, the apparatus also determines network coverage data associated with the plurality of network entities based on the terrain information and the clutter information. The apparatus analyses the signal quality data and the network entity data using a neural network model to determine the network coverage data. Thereafter, the apparatus generates an initial digital twin model based on the network coverage data and the network entity data and may estimate modelled signal quality data of the plurality of UEs using the initial digital twin model. Further, an optimization model may determine an error between the modelled signal quality data and the signal quality data and estimates locations of the plurality of UEs based on the error. The optimization model updates the initial digital twin model with the estimated locations of the plurality of UEs to obtain an intermediate digital twin model.

[0033] Further, the intermediate digital twin model may re-estimate the modelled signal quality data, and the optimization model re-estimates the locations of the plurality of UEs until the error is reduced to a minimum threshold value. When the error is reduced to the minimum value, the estimated locations of the UEs corresponding to the minimum threshold value of the error may be stored in the digital twin thus generating the digital twin model for the mobile communication network. Thus, the apparatus of the present disclosure generates an efficient digital twin model by determining accurate locations of the UEs and thereby providing reliable policy decisions such as energy savings.

[0034] The apparatus further performs energy saving operations using the digital twin model. In one embodiment, the apparatus may determine coverage indicators for a plurality of cells and / or sectors of the cells and may assign priorities to the plurality of cells and / or the sectors based on the coverage indicators. The coverage indicator indicates a number of UEs that are within a coverage range of (i.e., served by) a cell / sector and are also within a coverage range of a neighboring cell / sector. The more the number of UEs, the higher is the priority assigned to the cell / sector. Further, the apparatus virtually switches off each cell / sector using the digital twin model in the order of the priorities and verifies if modelled signal quality data of the cell / sector is within a threshold range upon switching off. If the modelled signal quality data is within the threshold range, the apparatus proceeds to switch off the cell / sector upon handing off the plurality of UEs to corresponding neighbor cells / sectors. On the other hand, if the modelled signal quality data does not fall within the threshold range, the apparatus proceeds to virtually switch off a next prioritized cell / sector. Thus, the apparatus of the present disclosure enables optimized selection of cells / sectors to switch off / on thereby reducing power consumption.

[0035] In the following detailed description of the embodiments of the disclosure, reference is made to the accompanying drawings that form a part hereof, and in which are shown by way of illustration specific embodiments in which the disclosure may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the disclosure, and it is to be understood that other embodiments may be utilized and that changes may be made without departing from the scope of the present disclosure. The following description is, therefore, not to be taken in a limiting sense.

[0036] Fig- 1 illustrates an exemplary Open Radio Access Network (ORAN) architecture of a mobile communication network.

[0037] As shown in Fig- 1, the exemplary ORAN architecture 100 of the mobile communication network comprises a plurality of elements to provide mobile communication services to a plurality of User Equipment (UEs) 102, 103 and 104. The plurality of elements may comprise, but not limited to, a Service Management and Orchestration (SMO) Framework 106, a Near Real Time RAN Intelligent Controller (Near RT RIC) 108 and an ORAN gNodeB (gNB) 110. The ORAN gNB 110 may comprise, but not limited to, an ORAN Centralized Unit (O-CU) 112, at least one ORAN Distributed Unit (O-DU) 114 and a plurality of ORAN nodes including a plurality of ORAN Radio Units (O-RUs) or radio nodes 116, 117 and 118. Each of the plurality of radio nodes 116-118 may serve a cell or a sector of the cell. Each of the plurality of O-RUs 116-118 that may serve the cell, or the sector of the cell may be termed herein also as a network entity. The plurality of components illustrated in the ORAN architecture 100 is only for illustrative purposes and cannot be construed as limiting in any manner. The ORAN architecture 100 may comprise one or more other components (that have not been depicted in Fig. 1) as specified in technical standards of ORAN alliance. For example, the ORAN architecture 100 may comprise at least one ORAN eNodeB and an ORAN cloud component coupled with the SMO framework 106 via 01 interface.

[0038] The SMO framework 106 may be communicatively coupled with the component of the ORAN gNB 110 via 01 interface, such as, but not limited to, O-CU 112, O-DU 114, and each of the plurality of network entities 116-118 (the 01 interface between the SMO framework 106 and the network entity 116 and the network entity 117 has not been depicted in Fig. 1 for the sake of clarity). The SMO framework 106 may be communicatively coupled with the Near -RT RIC 108 via Al interface. The Near RT RIC 108 may be communicatively coupled with the O-CU 112 and the at least one O-DU 114 via E2 interface. The at least one O-DU 114 may be communicatively coupled with the one or more of the plurality of network entities 116-118 via Open Front Haul (O-FH) interface. The SMO Framework 106 may comprise a non-RT RIC 120, that further comprises a plurality of applications termed as rApps 122.

[0039] The plurality of UEs 102, 103 and 104 may be connected to or served by the plurality of network entities 116, 117 and 118 respectively to obtain mobile communication network (MCN) services. Hence, the plurality of UEs 102-104 may also be referred to herein as served UEs 102-104 corresponding to the plurality of network entities 116-118, respectively. The plurality of UEs 102-104 may communicate with the respective plurality of network entities 116-118 to wirelessly transmit and / or receive mobile communication signals. Each of the plurality of UEs 102-104 may be stationary or moving while communicating with the respective network entity 116-118. Each of the plurality of network entities 116-118 may operate as serving cells for the plurality of UEs 102-104, respectively. Alternatively, one or more of the plurality of network entities 116-118 may serve as neighboring cells to one or more of the plurality of UEs 102-104. For example, the network entity 117 may serve as a neighboring cell to one or more of the plurality of UEs 102 and / or one or more of the plurality of UEs 104 and may serve as a serving cell for the plurality of UEs 103.

[0040] 8 Accordingly, each of the plurality of UEs 102-104 may experience different communication channels / communication environments during the communication, that may be associated with a plurality of signal quality parameters. The communication channels may also be referred to herein as channels. The plurality of signal quality parameters may comprise, but not limited to, Channel Quality Indicator (CQI), Received Signal Strength Indicator (RSSI), Reference Signal Received Power (RSRP), Signal to Interference Ratio (SINR), Reference Signal Received Quality (RSRQ), throughput and the like.

[0041] The SMO framework 106 may be configured to provide management and optimization of the plurality of components of the ORAN architecture 100 based on complex and dynamic communication environments. The SMO framework 106 may supervise lifecycle management of network services, coordinate between various components of the ORAN architecture 100, and may automate network operations. The SMO framework 106 may continuously monitor network performance and obtain real-time communication data for analysis and may enforce policies related to network operations, security, and compliance. The SMO framework 106 may also perform any other operation as per the technical standards of the ORAN alliance.

[0042] The SMO framework 106 may also comprise the non-RT RIC 120 that may enable non-RT control and optimization of the plurality of network entities 116-118 and other elements of the ORAN gNB 110. The non-RT RIC 120 may also be configured to perform Artificial Intelligence / Machine Learning (AI / ML) workflow training and updates, policy -based guidance of applications / features in near RT- RIC 108. The non-RT RIC 120 may provide configuration management, device management, fault management, performance management, and lifecycle management for the plurality of components (for example, the plurality of network entities 116-118) of the ORAN architecture 100.

[0043] The non-RT RIC 120 may enable non-real time control (ranging for time durations of more than 1 second) of the plurality of network entities 116-118. The non-RT RIC 120 may be configured to receive communication data from the near RT RIC 108 via Al interface, analyze the communication data and determine network optimization techniques for operation of the plurality of network entities 116-118. The communication data may comprise, but not limited to, data related to the plurality of UEs 102-104 and the plurality of network entities 116-118.

[0044] The non-RT RIC 120 may comprise, but not limited to, a plurality of applications, a non-RT RIC framework functionality and SMO internal interface. The plurality of applications, also known as rApps 122, may be designed to run on the non-RT RIC 120 to design the network optimization techniques for execution in the near RT RIC 108. The rApps 122 may communicate with the non-RT RIC framework via R1 interface. The rApps 122 may comprise, but not limited to, Radio Frequency (RF) signal predictor, cell utilization predictor, UE Quality of Experience (QoE) predictor, traffic predictor, anomaly detector, energy saver and the like. The list of rApps 122 has only been provided for illustrative purposes and cannot be construed as limiting in any manner. It may be appreciated that the non-RT RIC 120 may comprise any number of rApps 122 and other types of rApps 122 as per the technical specifications of ORAN alliance.

[0045] The near RT RIC 108 may be configured to perform various functions related to controlling and optimizing the functionality of the plurality of network entities 116-118. The near RT RIC 108 may be designed to enable network optimization actions that may be performed between 10 milliseconds to 1 second to complete. The near RT RIC 108 may also be configured to monitor data and actions of the plurality of RAN nodes. Further, the near RT RIC 108 may provide the monitored data and actions to the non-RT RIC 120 for analysis and providing optimized decisions such as, but not limited to, energy saving decisions.

[0046] The ORAN gNB 110 may be defined as a network node that may be configured to perform functions such as, but not limited to, transfer of user data, mobility control, RAN sharing, positioning, session management and the like. Fig. 1 illustrates a Split 7.2x disaggregated architecture of the ORAN gNB 110 comprising the O-CU 112, the at least one 0-DU 114 and the plurality of network entities 116-118. In other embodiments, the ORAN gNB 110 may be disaggregated in any other form as per the technical standards of the ORAN alliance. The architecture of the ORAN gNB 110 has only been provided as an illustration and cannot be construed as limiting in any manner. For example, the ORAN gNB 110 may comprise a greater number of O-DUs 114-1, 114-2 and 114-3 each connected to each of the plurality of network entities 116, 117 and 118 respectively.

[0047] The O-CU 112 may be configured with a Radio Resource Control (RRC) layer, a Packet Data Convergence Protocol (PDCP) layer and a Service Data Adaptation Protocol (SDAP) layer. The O-CU 112 may comprise O-CU-Control Plane (O-CU-CP) and one or more O-CU-User Planes (O-CU-UP). The O-CU-CP may host the RRC and control plane part of the PDCP layer. Each of the O-CU-UPs may host user plane functions of the PDCP layer and the SDAP layer for the plurality of UEs 102 -104.

[0048] The at least one 0-DU 114 may be configured with Radio Link Control (RLC), Medium Access

[0049] 10 Layer (MAC) and higher Physical (PHY) layers to provide communication services to the plurality of UEs 102 and 104 via the plurality of network entities 116-118. The at least one O- DU 114 may be communicating with the plurality of network entities 116-118 via Open Fron Haul (O-FH) interface. Fig. 1 illustrates a non-limiting embodiment where the O-DU 114 is communicating with the plurality of network entities 116-118. In other embodiments, the at least one O-DU 114 may comprise an O-DU communicating with the network entity 116 and another O-DU communicating with the network entities 117 and 118. The at least one O-DU 114 may perform one or more functions as designated by the technical standards of the ORAN alliance.

[0050] The plurality of network entities 116-118 may be configured with lower physical layer and RF processing, based on lower layer functional split, to provide communication with the plurality of UEs 102 -104. Each of the plurality of RUs 116-118 may provide a signal coverage across a cell or a sector of a cell. In few examples, each cell may be divided into three 120° sectors or six 60° sectors.

[0051] For the purposes of this disclosure, the network established between the plurality of UEs 102- 104 and the plurality of network entities 116-118 may be referred to herein as a mobile communication network 124 or MCN 124. The MCN 124 may comprise the plurality of network entities 116-118, the plurality of UEs 102-104 and a communication environment between the plurality of network entities 116-118 and the plurality of UEs 102-104. The communication environment may comprise a plurality of communication channels between the plurality of network entities 116-118 and the plurality of UEs 102-104.

[0052] One or more problems envisaged in the above-described architecture are detailed herein.

[0053] The non-RT RIC 120 may monitor communication data within the mobile communication network and may determine the network optimization techniques for the plurality of network entities 116-118. The communication data may comprise signal quality data corresponding to a plurality of UEs (for e.g., plurality of UEs 102) served by each serving network entity (for e.g., the network entity 116) and network entity data related to the plurality of network entities 116-118. For the sake of explanation, the one or more problems envisaged may be explained using a non-limiting example of the communication data related to the plurality of UEs 102 served by the serving network entity 116 and the neighboring network entity 117.

[0054] The signal quality data may comprise, but not limited to, distribution of the signal quality parameters (for example, CQI, RSSI, RSRP, and throughput) across the plurality of UEs 102,

[0055] 11 a count of the plurality of UEs 102 and location hints on UE positions and the like. Thus, the signal quality data received from the near RT RIC 108 may be generic in nature and may not specify data of each UE individually. In particular, the locations of plurality of UEs i.e., receivers are essential to provide accurate network optimization techniques. Hence, the generic signal quality data may be insufficient for the non-RT RIC 120 to determine accurate network optimization techniques, such as, but not limited to, energy saving techniques, for operation of the plurality of network entities 116-118. Further, to store the communication data, the non- RT RIC 120 may face challenges as the communication data may be received at short time intervals but includes huge volumes of data.

[0056] Furthermore, the network optimization techniques provided by the non-RT RIC 120 based on such unspecific data, may be unreliable. Such unreliable network optimization techniques, when implemented, may result in adverse effects on the MCN 124. For example, the non-RT RIC 120 may analyze the probabilistic distribution of CQI values and may incorrectly determine that corresponding network entity 116 may be switched off. Upon switching off the network entity, each of the plurality of UEs may need to stay connected to the MCN 124 even when the network entity 116 is switched off. However, the generic signal quality data is insufficient to ensure the connectivity post switching off the network entity. Consequently, one or more of the plurality of UEs 102 may lose connectivity to the ORAN gNB 110 upon switching off the network entity 116.

[0057] One method of addressing the problem may comprise modelling the mobile communication network using a digital twin model to simulate impact of the network optimization techniques. Such modelling may require accurate data related to each of the plurality of UEs 102, for example, positions and receiver characteristics of each of the plurality of UEs 102, terrain and clutter information related to the network entity 116. However, obtaining the data related to position and receiver characteristics of each of the plurality of UEs 102 and providing the obtained data to the SMO 106 may be data and bandwidth intensive as amount of data is voluminous and consumes high bandwidth for transmission. In addition, obtaining the position data may involve data privacy issues. Furthermore, prediction of channel models using the terrain information may result in processor intensive solution as the terrain information requires heavy memory resources to store and update the data. Moreover, clutter information may not accurately model non-stationary clutter in urban environments. Consequently, the digital twin model thus generated based on the generic data may be inefficient in assisting the non-RT RIC 120 to provide reliable network optimization decisions such as, but not limited to, energy

[0058] 12 saving techniques.

[0059] Embodiments of the present disclosure address one or more problems illustrated above.

[0060] Fig. 2a illustrates an exemplary system architecture for generating a digital twin model to perform energy saving in the MCN 124 in accordance with an embodiment of the present disclosure.

[0061] As shown in Fig. 2a, the exemplary architecture 100 may comprise a Digital twin Generation and Energy saving System (DGES) 202 within the SMO framework 106 to generate a digital twin model 204 for performing energy saving in the MCN 124. The DGES 202, also referred to herein as an apparatus, may comprise the non-RT RIC 120 and a digital twin system 203 communicatively coupled with each other. The non-RT RIC 120 may communicate with the digital twin system 203 via the one or more rApps 122 through an rApp interconnect bus interface. In some other embodiments, the non-RT RIC 120 may be capable of communicating with the DGES 202 through one or more other network interfaces.

[0062] The architecture 200 may also comprise a communication environment database 205 which may be communicatively coupled with the DGES 202. The DGES 202 may be connected to the near RT RIC 108 via the non-RT RIC 120 through the Al interface to receive the communication data required to generate the digital twin model 204.

[0063] The digital twin system 203 may comprise one or more components including, but not limited to, the digital twin model 204, a digital twin database 206, a neural network model 208 and an optimization model 210. The one or more rApps 122 of the non-RT RIC 120 may comprise, but not limited to, a coverage predictor rApp 212 and an energy saving rApp 214.

[0064] The digital twin system 203 may receive the communication data from the non-RT RIC 120 and may store the communication data in the digital twin database 206. The communication data may comprise the signal quality data of each of the plurality of UEs 102, 103, 104 and network entity data corresponding to the plurality of network entities 116-118. The network entity data may be defined as data associated with the plurality of network entities 116-118. The network entity data may comprise one or more physical characteristics including, but not limited to, locations, powers, and configurations of the plurality of network entities 116-118. The network entity data may also comprise one or more channel characteristics including, but not limited to, terrain information, weather information and clutter information related to the plurality of channels between the plurality of network entities 116-118 and corresponding

[0065] 13 plurality of UEs 102, 103 and 104, respectively. In some embodiments, the terrain and clutter information may be retrieved from the communication environment database 205.

[0066] The digital twin system 203 may be configured to analyze the communication data using the neural network model 208 and the optimization model 210 to generate the digital twin model 204 as may be explained in further detail below. Further, the digital twin system 203 may monitor the communication data in real time, update the digital twin database 206 and may dynamically update the digital twin model 204 based on the real time communication data.

[0067] The digital twin model 204 may be defined as a virtual representation of the MCN 124. The digital twin model 204 may comprise a virtual mobile communication network including, but not limited to, a plurality of virtual network entities corresponding to the plurality of network entities 116-118, a plurality of virtual UEs corresponding to the plurality of UEs 102, 103 and 104 and a virtual communication environment corresponding to the communication environment of the MCN 124. The digital twin model 204 may model the MCN 124 virtually so that the non-RT RIC 120 may monitor and measure an impact of one or more energy saving techniques (for example, switching off the network entity 117) in the virtual communication network of the digital twin model 204 and verify whether the one or more energy saving techniques provide any advantage over existing techniques. Based on the verification, the non- RT RIC 120 may determine whether to implement the one or more energy saving techniques in the MCN 124.

[0068] The communication environment database 205 may be defined as a database that may store the terrain information and clutter information related to the plurality of network entities 116-118 of the MCN 124. In one embodiment, the communication environment database 205 may comprise data related to maps of geographical locations associated with the MCN 124 to provide the terrain information such as, hills, valleys, roads, and the like and the clutter information. In an example, the communication environment database 205 may retrieve data from one or more applications related to maps. In some embodiments, the communication environment database 205 may be implemented within the digital twin system 203.

[0069] The digital twin database 206 may be defined as a database that stores the communication data received from the non-RT RIC 120. In an embodiment, the digital twin database 206, when implemented within the non-RT RIC 120, may receive the communication data from the near RT-RIC 108. The digital twin database 206 may be capable of receiving communication data at a frequency at which the near RT-RIC 108 provides to the non-RT RIC 120 and storing the

[0070] 14 communication data. The digital twin database 206 may dynamically update the communication data in real time. The digital twin database 206 may also store processing information of the digital twin model 204 as the digital twin model 204 runs one or more energy saving operations in the virtual mobile communication network.

[0071] The neural network model 208, also referred to herein as a neural network 208, may be trained to analyze the communication data for each of the plurality of network entities 116-118 and may determine the network coverage data of the plurality of network entities 116-118.

[0072] The optimization model 210 may be trained to estimate locations of each of the plurality of UEs 102, 103 and 104 by iteratively comparing the signal quality data received from the non- RT RIC 120 and modelled signal quality data generated by the digital twin model 204. Further, the optimization model 210 may update the digital twin model 204 with optimum locations of the plurality of UEs 102, 103 and 104, which may be explained in more detail below.

[0073] Upon generating the digital twin model 204, the non-RT RIC 120 may use the digital twin model 204 to determine an effect of implementing one or more energy saving techniques on the virtual mobile communication network. The coverage predictor rApp 212 may determine a coverage indicator for each of the plurality of network entities 116-118. The coverage indicator may indicate a count of one or more of the plurality of UEs 102, 103 and 104 that may be served by at least one neighboring network entity in addition to the serving network entity. Further, the energy saving rApp 214 may receive the coverage indicators of each of the plurality of network entities 116-118 from the coverage predictor rApp 212 and may assign priorities to each of the plurality of network entities 116-118 based on the coverage indicator. The energy saving rApp 214 may access the digital twin model 204 and may determine an effect of an energy saving operation such as, but not limited to, switching off each of the plurality of network entities 116-118 based on the assigned priorities, using the digital twin model 204. If the effect is constructive, the non-RT RIC 120 may instruct the near RT RIC 108 to implement the energy saving operation on the each of the plurality of network entities 116-118.

[0074] Thus, the architecture 200 efficiently generates the digital twin model by predicting the UE locations and dynamically updating the UE locations based on the signal quality data. Consequently, the architecture 200 also facilitates improved estimation of UE locations by reducing error data between the signal quality data and the modelled signal quality data. Moreover, the architecture 200 provides a secure method of estimating the UE locations without actually obtaining / using the UE locations from the UEs and / or network entities,

[0075] 15 thereby ensuring the data privacy or data security. Further, the architecture 200 facilitates optimized energy saving decisions for each network entity by determining the amount of coverage received by the served UEs from the one or more neighboring network entities.

[0076] Thus, the architecture 200 utilizes the digital twin model, to pre-emptively ensure that each of the served UEs receive MCN services when the virtual network entity is switched off even before the network entity is switched off in real world. Upon implementing the energy saving operation, the architecture 200 also monitors real world impact of the energy saving operation (for e.g., upon switching off each of the network entity) and may retain or modify the decision based on the monitoring. Thus, the digital twin model 204 may enable the non-RT RIC 120 to take informed decisions regarding energy saving by mitigating any adverse effects on the MCN 124.

[0077] Fig. 2b illustrates an exemplary digital twin model 204 to perform energy saving in the MCN in accordance with an embodiment of the present disclosure.

[0078] As shown in Fig. 2b, the digital twin model 204 may be defined as the virtual representation of the MCN 124, also referred to herein as a virtual MCN (VMCN) 250. The VMCN 250 may comprise a plurality of virtual UEs (VUEs) VI 02, VI 03, VI 04, a plurality of virtual network entities (VNEs) VI 16, VI 17, and VI 18 and a virtual communication environment, each corresponding to the plurality of UEs 102, 103 and 104, the plurality of network entities 116- 118 and the communication environment respectively of the MCN 124. The virtual communication environment may comprise a plurality of virtual channels corresponding to the plurality of channels.

[0079] Fig- 3 illustrates an exemplary block diagram of DGES 202 to generate the digital twin model to perform energy saving in the MCN 124 in accordance with an embodiment of the present disclosure.

[0080] As shown in Fig. 3, the DGES 202 may comprise, without limiting to, a processor 302, a user interface 304, the digital twin system 203 and the non-RT RIC 120. The digital twin system 203 may comprise the digital twin database 206, and a plurality of modules 306. The non-RT RIC 120 may comprise the coverage predictor rApp 212, the energy saving rApp 214 and memory 308.

[0081] The processor 302 may be any hardware processing system such as, but not limited to, a microprocessor, a microcontroller, an Application Specific Integrated Circuit (ASIC), a Field

[0082] 16 Programmable Gate Array (FPGA) or System on Chip (SOC), or any other type of processing system. The user interface 304 may comprise at least an input device and at least an output device to interact with one or more users. The plurality of modules 306 may include network coverage estimation module 310, the digital twin generation module 312, the digital twin model

[0083] 204 and a UE location estimation module 314. The plurality of modules 306 may be implemented using hardware, and / or software, or partly by hardware and partly by software or firmware. In some embodiments, the plurality of modules 306 may also be configured within the processor 302.

[0084] The network coverage estimation module 310 may comprise the neural network 208 and the UE location estimation module 314 may comprise the optimization model 210. The digital twin database 206 may comprise, but not limited to, communication data 316, network entity data 316a, signal quality data 316b, network coverage data network coverage data 318, UE location estimates 320, modelled signal quality data 322 and error data 324. The memory 308 may comprise, but not limited to, coverage indicators 326, priorities 328, and any other temporary data generated by the one or more rApps 122.

[0085] The operation of each module 306 is explained herein considering an example of a network entity (for example, the network entity 116), the plurality of UEs 102 served by the network entity 116 and the communication environment between the network entity 116 and the plurality of UEs 102. It may be appreciated that the example has been only considered for the sake of explanation but cannot be construed as limiting in any manner. The example may be extended to any number of network entities 116-118 of the MCN 124.

[0086] The network coverage estimation module 310 may be configured to process the communication data 316 received from the non-RT RIC 120 and / or the communication environment database

[0087] 205 using the neural network 208 and determine network coverage data 318 associated with the network entity 116. In one embodiment, the neural network 208 may comprise a Deep Neural Network (DNN). The neural network 208 may be trained with a plurality of training datasets comprising the communication data and the corresponding network coverage data 318 associated with a plurality of network entities and corresponding plurality of UEs. The network coverage data 318 may represent an intensity of network coverage at a plurality of locations within an area of the MCN 124.

[0088] For example, the network coverage data 318 may include, but not be limited to, a heat map indicating intensity of coverage across each point in the area of the MCN 124. In this example,

[0089] 17 the network coverage may be higher at a location nearer to a network entity and may be lower at another location away from the network entity. In another example, the network coverage may be lesser in a valley or a hill or when there is a clutter of trees or buildings at a location within the area of the MCN 124.

[0090] Upon training, the neural network 208 may receive the communication data 316 of the network entity 116 and the corresponding plurality of UEs 102. The communication data 316 may comprise, but not limited to, the network entity data 316a corresponding to the network entity 116 and the signal quality data 316b corresponding to the plurality of UEs 102.

[0091] The network entity data 316a may comprise locations and configurations of the plurality of network entities 116-118, one or more parameters of the neighboring network entities, terrain information and clutter information. The locations indicate geographical locations of the plurality of network entities 116-118. The configuration may indicate characteristics of software and / or hardware elements of each of the plurality of network entities 116-118. The neighboring network entities may be defined as one or more network entities that may act as secondary cells for one or more of the plurality of UEs 102. For example, the neighboring network entity may serve the one or more of the plurality of UEs 102 when the network entity 116 is switched off or during a handoff. The one or more parameters of the neighboring network entities may comprise signal quality data 316b of the plurality of UEs served by the one or more parameters of the neighboring network entities.

[0092] The terrain information may be defined as topological data related to terrain between the plurality of UEs 102 and the network entity 116 (for example, hills, valleys, and the like). The clutter information may be defined as objects (for example, buildings, trees, and other structures) that are present on the Earth’s surface but are not part of the terrain. The terrain and clutter information is essential to analyze the impact of the environment on signal propagation between the plurality of UEs 102 and the network entity 116.

[0093] The signal quality data 316b may comprise, but not limited to, distribution of the signal quality parameters (for example, CQI, RSSI, RSRP, and throughput) across the plurality of UEs 102, a count of the plurality of UEs 102 and location hints on locations of the plurality of UEs 102 and the like. For example, the location hints may include density of the plurality of users in a coverage area of the network entity 116. As indicated above, the signal quality data 316b is a generic data related generally to the plurality of UEs 102, without specifying data related to each UE of the plurality of UEs 102. The neural network 208 may analyze the communication

[0094] 18 data 316 and may generate network coverage data 318 associated with the plurality of UEs 102.

[0095] The digital twin generation module 312 may receive the network coverage data 318 and the network entity data 316a to generate an initial digital twin model 204. The initial digital twin model 204 may be defined as an instance of the digital twin model 204 indicating an initial VMCN 250. The initial VMCN 250 may include a virtual representation of the network coverage data 318 over the plurality of network entities. For example, the initial VMCN 250 may comprise the virtual representation of the MCN 124 including the heat map of the network coverage and the plurality of network entities 116-118 and the virtual representation of the plurality of UEs 102-104 at random locations as the actual locations of the plurality of UEs 102-104 are unavailable. To generate the initial digital twin model 204, the digital twin generation module 312 may generate an individual digital twin model for each component of the MCN 124 and further establish connections between two or more of these individual digital twin models based on real time communication. The digital twin generation module 312 may determine initial UE location estimates 320 of the plurality of UEs 102, 103 and 104 based on the communication data 316. The digital twin generation module 312 may generate individual digital twin models comprising VUEs VI 02, VI 03 and VI 04 using the initial UE location estimates 320, also referred to herein as initial location estimates 320.

[0096] The initial digital twin model 204 may represent an initial estimate of the MCN 124 based on the initial location estimates 320 of the plurality of UEs 102, 103 and 104 as real -world location estimates of the plurality of UEs 102, 103 and 104 may be unavailable. The initial digital twin model 204 may estimate modelled signal quality data 322 related to the plurality of VUEs V102, V103 and V104 based on the initial VMCN 250. The modelled signal quality data 322 may be defined as the signal quality data associated with the VUEs VI 02, VI 03 and VI 04. The modelled signal quality data 322 may be used by the optimization model 210 to determine locations of the plurality of VUEs VI 02, VI 03 and VI 04.

[0097] The UE location estimation module 314 may be configured to determine the locations of the plurality of UEs 102, 103 and 104 based at least on the communication data 316 using the optimization model 210 and the initial digital twin model 204. The optimization model 210 may comprise an AI / ML model that may be trained to determine the locations of the plurality of UEs 102, 103 and 104 by analyzing the communication data 316 and modelled signal quality data 322. In one embodiment, the optimization model 210 may comprise a Particle Swarm

[0098] 19 Optimization (PSO) model. Further, the optimization model 210 may compare the modelled signal quality data 322 and the signal quality data 316b to determine error data 324. The error data 324 may indicate deviation between the signal quality data 316b of the MCN 124 and the modelled signal quality data 322 of the initial VMCN 250. The error data 324 may comprise a plurality of error values corresponding to the one or more signal quality parameters.

[0099] To determine the plurality of error values for signal quality parameters in the form of a probabilistic distribution, the optimization model 210 may use a characteristic such as, divergence in measure theory. The signal quality parameters with probabilistic distribution may comprise, but not limited to, the CQI, RSSI, RSRP, and RSRQ. In one example, the divergence may be a Jensen-Shannon divergence (JS-divergence). For single-value parameter, such as, but not limited to, throughput, the optimization model 210 may evaluate a ratio of modelled signal quality parameter and real-value of the signal quality parameter to determine the error.

[0100] Further, the optimization model 210 may compare the error data 324 with error threshold data and may generate updated location estimates 320 of the plurality of VUEs V102-V104 to minimize difference between the error data 324 and the error threshold data. The error threshold data may comprise a plurality of threshold error values for each of the one or more signal quality parameters.

[0101] In another embodiment, the optimization model 210 may combine the plurality of error values to generate a final error value. In one embodiment, the optimization model 210 may combine the plurality of error values based on predefined weights assigned to each of the one or more signal quality parameters. The optimization model 210 may compare the final error value with a threshold error value and evaluate a difference between the final error value and the threshold error value. Further, the optimization model 210 may generate updated location estimates 320 of the plurality of VUEs V102-V104 to minimize the difference.

[0102] If the error data 324 exceeds the error threshold data, the UE location estimation module 314 may update the initial location estimates 320 of one or more of the plurality of VUEs VI 02- V104 of the initial digital twin model 204 with the updated location estimates 320. For example, the UE location estimation module 314 may update the individual digital twin models of one or more of the plurality of VUEs V102-V104 that exceed the error threshold data and may accordingly update one or more other components of the initial digital twin model 204.

[0103] The updated digital twin model 204 may further operate the updated VMCN 250 and may generate updated modelled signal quality data 322. The optimization model 210 may further

[0104] 20 receive the updated modelled signal quality data 322 from the updated digital twin model 204 and determine the error data 324 comparing the updated modelled signal quality data 322 with the signal quality data 316b. The optimization model 210 may dynamically determine the updated location estimates 320 and update the initial digital twin model 204 based on the error data 324 until the error data 324 is less than or same as the error threshold data. Further, the updated location estimates 320 corresponding to the error data 324 less than or same as the error threshold data may be referred to as the locations of the plurality of VUEs V102-V104. Thus, the digital twin generation module 312 may generate the digital twin model 204 by updating the initial digital twin model 204 with the locations of the plurality of VUEs VI 02- V104. The energy saving rApp 214 may be configured to use the digital twin model 204 to perform energy saving on the MCN 124.

[0105] Further, the non-RT RIC 120 may monitor the MCN 124 at real-time by communicating with the near RT-RIC 108. The non-RT RIC 120 may comprise at least a traffic predictor rApp and an anomaly detector rApp (not shown in Fig. 2 or Fig. 3) to monitor the MCN 124. The traffic predictor rApp may provide network traffic prediction across the plurality of network entities 116-118 by leveraging advanced machine learning models. The traffic predictor rApp may facilitate to proactively anticipate demand spikes and allocate resources, minimize congestion and prevent bottlenecks. The anomaly detector rApp may observe and group possible performance anomalies within the MCN 124. The anomalies may be related to cell performance, configuration, and alarms. The anomaly detector rApp may facilitate automatic identification of different cell issue patterns.

[0106] The coverage predictor rApp 212 may be configured to determine a coverage indicator 326 corresponding to each of the plurality of network entities 116-118. The coverage indicator 326 of a network entity may indicate a count of the plurality of UEs served by the network entity and that may be within a coverage range of one or more neighboring network entities. The plurality of UEs that may be located within a coverage range of the network entity and served by the network entity may also be referred to herein as served UEs. For example, the plurality of UEs 102 of the network entity 116 may be referred to as served UEs.

[0107] A count of a subset of the plurality of UEs 102 that may be located within the coverage range of the neighboring network entity 117 may be referred to as the coverage indicator 326. For example, the coverage indicator 326 may be expressed as a percentage, such as, 50%. the coverage indicator 326 may indicate a level of coverage received by the plurality of UEs (for e.g., UEs 102) from the neighboring network entities (for e.g., network entity 117) in case the serving network entity (for e.g., network entity 116) is switched off. Thus, the coverage indicator 326 may provide information required for energy saving performed by the energy saving rApp 214.

[0108] Further, the coverage predictor rApp 212 may be configured to assign priorities 328 to the plurality of network entities 116-117 based on the plurality of coverage indicators 326. The priority of the network entity may be proportional to the coverage indicator of the network entity. For example, a network entity with a high coverage indicator may have a high priority, whereas a network entity with a low coverage indicator may have a low priority. The coverage predictor rApp 212 may also generate a coverage graph for the plurality of network entities 116-117. The coverage graph may be defined as a visual representation of an amount and an extent of coverage for each of the plurality of UEs 102-104. For example, the coverage graph may be an image including a plurality of pixels, each pixel representing strength of the signal at the particular location.

[0109] The energy saving rApp 214 may be configured to perform one or more energy saving operations based on the assigned priorities 328. The one or more energy saving operations may comprise, but not limited to, switching off one or more of the plurality of network entities 116- 118 upon virtually implementing the one or more energy saving operations using the digital twin model 204.

[0110] To determine the one or more energy saving operations, the energy saving rApp 216 may select a first network entity (for e.g., network entity 118) among the plurality of network entities 116- 118 based on descending order of the priorities 328. Further, the energy saving rApp 214 may virtually switch off a first VNE (for e.g., VNE VI 18), corresponding to the first network entity, using the digital twin model 204. In response to virtually switching off the first VNE, the digital twin model 204 may estimate the impact of the switching off using the modelled signal quality data 322 of the plurality of virtual served UEs (for e.g., VUEs VI 04). The plurality of virtual served UEs may correspond to the served UEs of the first network entity.

[0111] The energy saving rApp214 may be configured to compare the modelled signal quality data 322 with a predefined threshold range. The predefined threshold range may include threshold values for each of the plurality of signal quality parameters. The predefined threshold range may indicate minimum acceptable levels of the plurality of signal quality parameters of the plurality of the UES 102-104 required for reliable communication with the MCN 124. Upon comparison, if the modelled signal quality data 322 exceeds the predefined threshold range, each virtual served UE is connected to one or more virtual neighboring network entities to receive MCN services. In response, the energy saving rApp 216 may determine an energy saving decision to switch off the first network entity.

[0112] On the other hand, upon comparison, if the modelled signal quality data 322 is less than the predefined threshold range, each virtual served UE is connected to one or more virtual neighboring network entities to receive MCN services. In response, the energy saving rApp 214 may be configured to select a second network entity among the plurality of network entities based on the descending order of priorities to perform energy saving.

[0113] In response to the energy saving decision, the coverage predictor rApp 212 may update the coverage graph based on switching off the first network entity. For example, the pixels within the image of the coverage graph may be updated representing updated signal strength that may have reduced when the first network entity has been switched off.

[0114] Further, the non-RT RIC 120 may send a first message indicating the energy saving decision to the near RT RIC 108. The near RT RIC 108 may receive the first message and in response may send one or more second messages to the first network entity and the one or more neighboring network entities to hand off the served UEs to the one or more neighboring network entities. Upon receipt of the one or more second messages, the first network entity and the one or more neighboring network entities may initiate hand off and may successfully hand off the served UEs to the one or more neighboring network entities. Thereafter, the near RT RIC 108 may transmit a message to the first network entity to switch off. Thus, the first network entity may switch off upon receiving the message and may result in energy saving.

[0115] Upon switching off the first network entity, the energy saving rApp 214 may analyse an impact of switching off the first network entity on the MCN 124 in real time. Accordingly, the energy saving rApp 214 may receive current traffic data and predicted traffic data from the traffic predictor rApp and coverage indicators from the coverage predictor rApp 212. Further, the energy saving rApp 214 may receive anomaly data from the anomaly detector rApp. If the anomaly data indicates one or more anomalies, the energy saving rApp 214 may not initiate performing energy saving operations. Alternatively, if the anomaly data indicates that no anomaly is predicted, the energy saving rApp 214 may predict locations of the plurality of UEs 102-104 based on the coverage indicators, current traffic data and predicted traffic data. Further, the energy saving rApp 214 may compute current signal quality data 316b. If the current signal quality data 316b is deteriorating compared to previous signal quality data 316b, the energy saving rApp 214 may initiate switching on the first network entity. On the other hand, if the current signal quality data 316b is improving compared to the previous signal quality data 316b, the energy saving rApp 214 may select a second network entity based on a descending order of the priorities 328 to determine switching off the second network entity. Thus, the energy saving rApp 214 may dynamically determine one or more energy saving decisions for each of the plurality of network entities 116-118 based on the descending order of the priorities 328 using the digital twin model 204.

[0116] Thus, the DGES 202 efficiently generates the digital twin model by predicting the UE locations and dynamically updating the UE locations based on the signal quality data. Consequently, the DGES 202 also facilitates improved estimation of UE locations by reducing error data between the signal quality data and the modelled signal quality data. Moreover, the DGES 202 provides a secure method of estimating the UE locations without actually obtaining / using the UE locations from the UEs and / or network entities, thereby ensuring the data privacy or data security. Further, the DGES 202 facilitates optimized energy saving decisions for each network entity by determining the amount of coverage received by the served UEs from the one or more neighboring network entities.

[0117] Thus, the DGES 202, enables the non-RT RIC 120 to utilize the digital twin model, to preemptively ensure that each of the served UEs receive MCN services when the VNE is switched off even before the network entity is switched off in real world. Upon implementing the energy saving operation, the DGES 202 also monitors real world impact of the energy saving operation (for e.g., upon switching off each of the network entity) and may retain or modify the decision based on the monitoring. Thus, the digital twin model 204 may enable the non-RT RIC 120 to take informed decisions regarding energy saving by mitigating any adverse effects on the MCN 124.

[0118] Fig- 4 illustrates an exemplary flowchart of a method to generate the digital twin model to perform energy saving in the MCN 124, in accordance with an embodiment of the present disclosure.

[0119] At block 402, the DGES 202 may process the network entity data 316a of the plurality of network entities 116-118 and the signal quality data 316b of the plurality of UEs 102-104 of

[0120] 24 the MCN 124 using the neural network 208.

[0121] At block 404, the DGES 202 may determine network coverage data 318 of the plurality of channels associated with the plurality of network entities 116-118 based on the processing.

[0122] At block 406, the DGES 202 may generate an initial digital twin model 204 corresponding to the MCN 124 based on the network coverage data 318 and the network entity data 316a. The initial digital twin model 204 may include a virtual representation of the MCN 124 comprising the plurality of VNEs VI 16- VI 18 and the plurality of VUEs V102-V104 at initial UE location estimates.

[0123] At block 408, the DGES 202 may estimate the modelled signal quality data 322 corresponding to the plurality of VUEs V102-V104 located at initial location estimates using the initial digital twin model 204.

[0124] At block 410, the DGES 202 may determine the error data 324 between the modelled signal quality data 322 and the signal quality data 316b using the optimization model 210 and may generate a final error value by analyzing the error data 324.

[0125] At block 412, the DGES 202 may determine whether the error data 324 exceeds the error threshold data. If the error data 324 exceeds the error threshold data, the DGES 202 may proceed to block 414. Alternatively, if the error data 324 is less than or same as the error threshold data, the DGES 202 may proceed to block 416.

[0126] At block 414, the DGES 202 may update the initial location estimates of the plurality of VUEs V102-V104 and may proceed to block 408. The DGES 202 continues to iterate between the blocks 408 and 414 until the final error value reduces to a value less than or equal to the threshold error.

[0127] At block 416, the DGES 202 may determine the updated location estimates of the plurality of VUEs when final error value is less than or equal to the threshold error. The DGES 202 may generate the digital twin model 204 by updating the initial digital twin model 204 with the updated location estimates of the plurality of VUEs V102-V104. The updated location estimates may be referred to as the locations of the plurality of VUEs V102-V104.

[0128] Fig- 5 illustrates an exemplary flowchart of a method to perform energy saving in the MCN 124 using the digital twin model 204 in accordance with an embodiment of the present

[0129] 25 disclosure.

[0130] At block 502, the DGES 202 may determine coverage indicators 326 of the plurality of network entities 116-118.

[0131] At block 504, the DGES 202 may assign priorities 328 to the plurality of network entities 116- 118 based on the coverage indicators 326.

[0132] At block 506, the DGES 202 may select a network entity among the plurality of network entities 116-118 based on the priorities 328. The DGES 202 may select the network entity with the highest priority or next highest priority.

[0133] At block 508, the DGES 202 may verify whether maximum number of iterations has been completed. The maximum number of iterations may indicate a total number of the plurality of network entities. In other embodiment, the maximum number of iterations may also be proportional to a number of energy saving operations, a threshold amount of energy saved, or a maximum number of network entities switched off and the like. In some embodiments, the maximum number of iterations may be completed when an anomaly is detected by the anomaly detector rApp. If the maximum number of iterations has been completed, the DGES 202 may terminate the method of performing energy saving at block 509. Alternatively, if the maximum number of iterations has not been completed, the DGES 202 may proceed to block 510.

[0134] At block 510, switch off a VNE corresponding to the selected network entity.

[0135] At block 512, the DGES 202 may determine the modelled signal quality data 322 upon switching off the VNE.

[0136] At block 514, the DGES 202 may determine whether the modelled signal quality data 322 exceeds a predefined threshold range. If the modelled signal quality data 322 exceeds the predefined threshold range, the DGES 202 may proceed to block 516. Alternatively, if modelled signal quality data 322 is less than the predefined threshold range, the DGES 202 may proceed to block 506 to select another network entity with priority less than the network entity in the descending order of the priorities 328.

[0137] At block 516, the DGES 202 may switch off the network entity as the modelled signal quality data 322 exceeds the predefined threshold range indicating that the served UEs may still be connected to the MCN 124 even if the network entity is switched off and hence switching off

[0138] 26 the network entity may not have any adverse effect on the MCN 124 thereby facilitating energy saving.

[0139] At block 518, the DGES 202 may determine the signal quality data 316b of the served UEs of the network entity upon switching off the network entity.

[0140] At block 520, the DGES 202 may determine whether the signal quality data 316b exceeds the predefined threshold range. If the signal quality data 316b exceeds the predefined threshold range, the DGES 202 may proceed to block 506 to select another network entity for switching off. Alternatively, if signal quality data 316b is less than the predefined threshold range, the DGES 202 may proceed to re-iterate blocks 508-518.

[0141] During re-iteration at blocks 508-518, the DGES 202 may switch on the VNE corresponding to the network entity using the digital twin model 204, determine the modelled signal quality data 322 upon switching on the VNE and may determine whether the modelled signal quality data 322 exceeds the predefined threshold range. If the modelled signal quality data 322 exceeds the predefined threshold range, the DGES 202 may switch on the network entity. Alternatively, if the modelled signal quality data 322 is within the predefined threshold range even when the VNE is virtually switched on, the DGES 202 may determine that a performance of the VMCN 250 is deteriorating due to switching off a previous network entity. In this case, the DGES 202 may switch on the network entity as well as proceeds to block 506 to perform the blocks 506-512 for the previous network entity.

[0142] Fig- 6 illustrates an exemplary flowchart of a method to generate the digital twin model to perform energy saving in the MCN 124 in accordance with another embodiment of the present disclosure.

[0143] At block 602, the DGES 202 may process the network entity data 316a of the plurality of network entities 116-118 and the signal quality data 316b of the plurality of UEs 102-104 of the MCN 124 using the neural network 208.

[0144] At block 604, the DGES 202 may determine network coverage data 318 of the plurality of channels associated with the plurality of network entities 116-118 based on the processing.

[0145] At block 606, the DGES 202 may generate an initial digital twin model 204 corresponding to the MCN 124 based on the network coverage data 318 and the network entity data 316a. The initial digital twin model 204 may include a virtual representation of the MCN 124 comprising the plurality of VNEs VI 16- VI 18 and the plurality of VUEs V102-V104 at initial UE location estimates.

[0146] At block 608, the DGES 202 may determine locations of the plurality of UEs 102-104 served by the plurality of network entities 116-118 based at least on the network entity data 316a and the signal quality data 316b using the optimization model 210 and the initial digital twin model 204.

[0147] At block 610, the DGES 202 may generate the digital twin model 204 by updating the initial digital twin model 204 with the locations of the plurality of UEs 102-104.

[0148] At block 612, the DGES 202 may perform energy saving in the MCN 124 using the digital twin model 204.

[0149] The methods 400-600 may be described in the general context of computer executable instructions. Generally, computer executable instructions can include routines, programs, objects, components, data structures, procedures, modules, and functions, which perform specific functions or implement specific abstract data types. The order in which the methods 400-600 are described is not intended to be construed as a limitation, and any number of the described method blocks can be combined in any order to implement the method. Additionally, individual blocks may be deleted from the methods 400-600 without departing from the scope of the subject matter described herein. Furthermore, the method can be implemented in any suitable hardware, software, firmware, or combination thereof.

[0150] The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Alternatives (including equivalents, extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope of the disclosed embodiments. Also, the words "comprising," "having," "containing," and "including," and other similar forms are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise.

[0151] Finally, the language used in the specification has been principally selected for readability and instructional purposes, and it may not have been selected to delineate or circumscribe the inventive subject matter. Accordingly, the disclosure of the embodiments of the disclosure is intended to be illustrative, but not limiting, of the scope of the disclosure.

[0152] With respect to the use of substantially any plural and / or singular terms herein, those having skill in the art can translate from the plural to the singular and / or from the singular to the plural as is appropriate to the context and / or application. The various singular / plural permutations may be expressly set forth herein for sake of clarity.

[0153] 29

Claims

We Claim:

1. A method for generating a digital twin model (204) to perform energy saving in a mobile communication network, the method comprising: processing network entity data of a plurality of network entities (116-118) and signal quality data of a plurality of User Equipment (UEs) (102-104) of the mobile communication network (124) using a neural network (208); determining network coverage data associated with the plurality of network entities (116-118) using the neural network (208) based on the processing; generating an initial digital twin model (204) corresponding to the mobile communication network (124) based on the network entity data and the network coverage data; determining locations of the plurality of UEs (102-104) served by the plurality of network entities (116-118) based at least on the network entity data and the signal quality data using an optimization model (210) and the initial digital twin model (204); generating the digital twin model (204) by updating the initial digital twin model (204) with the locations of the plurality of UEs (102-104); and performing energy saving in the mobile communication network (124) using the digital twin model (204).

2. The method of claim 1, wherein each network entity is a radio node serving one of a cell of the mobile communication network (124) or a sector of the cell of the mobile communication network (124), wherein the mobile communication network (124) comprises the plurality of network entities (116-118), the plurality of UEs (102-104) and a communication environment between the plurality of network entities (116-118) and the plurality of UEs (102-104), wherein the network entity data indicates one or more physical characteristics and one or more channel characteristics of the plurality of network entities (116-118), wherein the signal quality data indicates one or more signal quality parameters associated with the plurality of UEs (102-104), and wherein the digital twin model (204) represents a virtual mobile communication network (124) corresponding to the mobile communication network (124), wherein the virtual mobile communication network (124) includes a plurality of virtual networkentities (V116-C118), a plurality of UEs (V102-V104), and a plurality of virtual channels.

3. The method of claim 1, wherein determining the locations of the plurality of UEs (102- 104) comprises: determining initial location estimates of the plurality of UEs (102-104) based on the network entity data and the signal quality data using the optimization model (210); estimating modelled signal quality data by operating the initial digital twin model (204) with a plurality of UEs (VI 02- VI 04), corresponding to the plurality of UEs (102-104), located at the initial location estimates; determining error data between the modelled signal quality data and the signal quality data using the optimization model (210); generating updated location estimates of the plurality of UEs (102-104) upon determining that the error data exceeds error threshold data; iteratively operating the initial digital twin model (204) with the plurality of UEs (102-104) located at the updated location estimates to estimate the signal quality data, and generating updated location estimates until the error data is less than or same as the error threshold data; and determining the initial location estimates, or the updated location estimates, corresponding to the error data that is less than or same as the error threshold data as the locations of the plurality of UEs (102-104).

4. The method of claim 1, wherein performing energy saving using the digital twin model (204) comprises: determining a plurality of coverage indicators corresponding to the plurality of network entities (116-118), wherein a coverage indicator of a network entity indicates a count of served UEs within a coverage range of one or more neighboring network entities, wherein the served UEs fall within the coverage range of the network entity; assigning priorities to the plurality of network entities (116-118) based on the plurality of coverage indicators; selecting a first network entity among the plurality of network entities (116- 118) based on descending order of the priorities;virtually switching off a first virtual network entity using the digital twin model (204) based on the priorities, wherein the first virtual network entity is a virtual network entity corresponding to the first network entity; determining modelled signal quality data of a plurality of virtual served UEs in response to virtually switching off the first virtual network entity, wherein the plurality of virtual served UEs correspond to the served UEs of the first network entity; and switching off the first network entity upon determining that the modelled signal quality data exceeds a predefined threshold range.

5. The method of claim 4, wherein a priority of the network entity is proportional to the coverage indicator of the network entity.

6. The method of claim 4, wherein switching off the first network entity by: handing off served UEs of the first network entity to the one or more neighboring network entities based on the determination; and switching off the first network entity upon handing off.

7. The method of claim 1, wherein performing energy saving in the mobile communication network (124) using the digital twin model (204) comprising: determining that modelled signal quality data is less than a predefined threshold range, upon virtual switching off a first virtual network entity; and selecting a second network entity among the plurality of network entities (116- 118) based on priorities to perform energy saving, wherein a priority of the second network entity is different from a priority of a first network entity.

8. An apparatus (202) to generate a digital twin model (204) to perform energy saving in a mobile communication network (124), the apparatus (202) comprising: a memory; and a processor coupled with the memory, wherein the processor is configured to: process network entity data of a plurality of network entities (116-118) and signal quality data of a plurality of User Equipment (UEs) (102-104) of the mobile communication network (124) using a neural network (208);determine network coverage data associated with the plurality of network entities (116-118) using the neural network (208) based on the processing; generate an initial digital twin model (204) corresponding to the mobile communication network (124) based on the network entity data and the network coverage data; determine locations of the plurality of UEs (102-104) served by the plurality of network entities (116-118) based at least on the network entity data and the signal quality data using an optimization model (210) and the initial digital twin model (204); generate the digital twin model (204) by updating the initial digital twin model (204) with the locations of the plurality of UEs (102-104); and perform energy saving in the mobile communication network (124) using the digital twin model (204).

9. The apparatus (202) of claim 8, wherein each network entity is a radio node serving one of a cell of the mobile communication network (124) or a sector of the cell of the mobile communication network (124), wherein the mobile communication network (124) comprises the plurality of network entities (116-118), the plurality of UEs (102-104) and a communication environment between the plurality of network entities (116-118) and the plurality of UEs (102-104), wherein the network entity data indicates one or more physical characteristics and one or more channel characteristics of the plurality of network entities (116-118), wherein the signal quality data indicates one or more signal quality parameters associated with the plurality of UEs (102-104), and wherein the digital twin model (204) represents a virtual mobile communication network (124) corresponding to the mobile communication network (124), wherein the virtual mobile communication network (124) includes a plurality of virtual network entities (V116-C118), a plurality of UEs (V102-V104), and a plurality of virtual channels.

10. The apparatus (202) of claim 8, wherein to determine the locations of the plurality of UEs (102-104), the processor is configured to:determine initial location estimates of the plurality of UEs (102-104) based on the network entity data and the signal quality data; estimate modelled signal quality data by operating the initial digital twin model (204) with a plurality of UEs (VI 02- VI 04), corresponding to the plurality of UEs (102-104), located at the initial location estimates; determine error data between the modelled signal quality data and the signal quality data using the optimization model (210); generate updated location estimates of the plurality of UEs (102-104) upon determining that the error data exceeds error threshold data; iteratively operate the initial digital twin model (204) with the plurality of UEs (102-104) located at the updated location estimates to estimate the signal quality data, and generating updated location estimates until the error data is less than or same as the error threshold data; and determine the initial location estimates, or the updated location estimates, corresponding to the error data that is less than or same as the error threshold data as the locations of the plurality of UEs (102-104).

11. The apparatus (202) of claim 8, wherein to perform energy saving using the digital twin model (204), the processor is configured to: determine a plurality of coverage indicators corresponding to the plurality of network entities (116-118), wherein a coverage indicator of a network entity indicates a count of served UEs within a coverage range of one or more neighboring network entities, wherein the served UEs fall within the coverage range of the network entity; assign priorities to the plurality of network entities (116-118) based on the plurality of coverage indicators; select a first network entity among the plurality of network entities (116-118) based on descending order of the priorities; virtually switch off a first virtual network entity using the digital twin model (204) based on the priorities, wherein the first virtual network entity is a virtual network entity corresponding to the first network entity; determine modelled signal quality data of a plurality of virtual served UEs in response to virtually switching off the first virtual network entity, wherein the plurality of virtual served UEs correspond to the served UEs of the first network entity; andswitch off the first network entity upon determining that the modelled signal quality data exceeds a predefined threshold range.

12. The apparatus (202) of claim 11, wherein a priority of the network entity is proportional to the coverage indicator of the network entity.

13. The apparatus (202) of claim 11, wherein to switch off a first network entity, the processor is configured to: hand off served UEs of the first network entity to the one or more neighboring network entities based on the determination; and switch off the first network entity upon handing off.

14. The apparatus (202) of claim 8, wherein to perform energy saving in the mobile communication network (124) using the digital twin model (204), the processor is configured to: determine that modelled signal quality data is within a predefined threshold range, upon virtual switching off a first virtual network entity; and select a second network entity among the plurality of network entities (116-118) based on priorities to perform energy saving, wherein a priority of the second network entity is different from a priority of a first network entity.

Images