Method and system for detecting anomaly cells in a network

Abstract

The present disclosure provides a method (500) and a system (108) for detecting anomaly cells in a network (106). The method includes extracting performance metrics for network cells. Additionally, the cell radius for each network cell is determined using morphological data, and cells are categorized within the determined cell radius based on building data. Clusters are formed, comprising network cells sharing common environmental and morphological features. Within each cluster, performance metrics are evaluated, and cells exhibiting anomalous behaviour are flagged as the anomaly cells based on this evaluation. This approach leverages spatial, environmental, and performance data to improve the accuracy of network anomaly detection.

Description

METHOD AND SYSTEM FOR DETECTING ANOMALY CELLS IN A NETWORKRESERVATION OF RIGHTS

[0001] A portion of the disclosure of this patent document contains material, which is subject to intellectual property rights such as, but are not limited to, copyright, design, trademark, Integrated Circuit (IC) layout design, and / or trade dress protection, belonging to JIO PLATFORMS LIMITED or its affiliates (hereinafter referred as owner). The owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent files or records, but otherwise reserves all rights whatsoever. All rights to such intellectual property are fully reserved by the owner.TECHNICAL FIELD

[0002] The present disclosure relates generally to the field of telecommunications. More particularly, the present disclosure relates to a method and a system for detecting one or more anomaly cells in a network.DEFINITION

[0003] As used in the present disclosure, the following terms are generally intended to have the meaning as set forth below, except to the extent that the context in which they are used to indicate otherwise.

[0004] The term ‘Network Cells’ used hereinafter in the specification refers to the fundamental coverage units of a communication network, each served by a base station or access point that provides wireless connectivity to a user equipment (UE) within a defined geographical area. Each network cell operates on specific radio frequencies and is responsible for handling communication sessions, resource allocation, and mobility management of connected users. The network cells may include, but are not limited to, macro cells, micro cells, pico cells, and femto cells, which differ in coverage area, transmission power, and deployment environment (e.g., urban, suburban, or rural).

[0005] The term ‘Anomaly Cell’ used hereinafter in the specification refers to a specific cell within a network that exhibits abnormal performance compared to peer cells in the network.

[0006] The term ‘KPI’ used in this specification refers to a Key Performance Indicator (KPI). A KPI is a measurable metric that is used to evaluate the performance of the cells in the network. It may include metrics such as signal strength, call drop rates, data throughput, and similar.

[0007] The term ‘ISD’ used hereinafter in the specification refers to inter-site distance, which represents physical distance between two adjacent network sites or base stations within a cellular network. The ISD is a key morphological parameter that influences cell coverage area, signal strength, frequency reuse, and network planning.

[0008] These definitions are in addition to those expressed in the art.BACKGROUND

[0009] The following description of related art is intended to provide background information pertaining to the field of the disclosure. This section may include certain aspects of the art that may be related to various features of the present disclosure. However, it should be appreciated that this section be used only to enhance the understanding of the reader with respect to the present disclosure, and not as admissions of prior art.

[0010] In telecommunications, ensuring reliable and high-quality service delivery is essential for network operators. One of the key aspects of optimizing service delivery is the ability to monitor and manage the performance of individual network cells within a network. The network cells are the fundamental coverage units of a cellular communication network, each served by a base station or access point that provides wireless connectivity to user equipment (UE) within a defined geographicalarea. Each network cell operates on specific radio frequencies and is responsible for handling communication sessions, resource allocation, and mobility management of connected users. Over time, these cells may experience performance issues due to various factors such as increased traffic load, environmental interference, signal interference, equipment degradation, and the like. These performance issues may occur in cells, which might show abnormal performance metrics, such as unexpected drops in signal strength, increased call drop rates, or slower data speeds, ultimately leading to poor user experience. Identifying these anomalies in cells is essential for maintaining network quality and ensuring that users have a reliable connection.

[0011] Traditional approaches for identifying these anomaly cells mostly rely on manual monitoring and reactive interventions by network operators, such as when users report performance issues or the network operators detect these anomaly cells through routine network checks. However, these traditional methods are often inefficient as they do not allow for proactive identification of network cells at risk of underperforming. Moreover, the traditional approach may lead to delayed response and prolonged service disruptions, increasing user complaints and network inefficiency.

[0012] There is, therefore, a need in the art to provide a method and a system that can mitigate the disadvantages of the prior art.SUMMARY OF THE DISCLOSURE

[0013] In an exemplary embodiment, a method for detecting one or more anomaly cells in a network is described. The method includes extracting, by an extracting unit, one or more performance metrics associated with a plurality of network cells. The method further includes determining, by a processing unit, a cell radius of each of the plurality of network cells based on morphological data associated with the plurality of network cells. The method further includes categorizing, by the processing unit, the plurality of network cells into one or more cell categories within thedetermined cell radius based on building data associated with the plurality of network cells. The method further includes forming, by the processing unit, one or more clusters of the plurality of network cells based on the one or more cell categories. Each cluster includes one or more network cells having a common set of environmental and morphological characteristics. The method further includes evaluating, by the processing unit, the one or more performance metrics of the one or more network cells within each cluster and flagging the one or more network cells as the one or more anomaly cells based on the evaluation.

[0014] In some embodiments, the method further includes generating, by a processing unit, a report comprising a list of the flagged one or more anomaly cells and the corresponding one or more performance metrics.

[0015] In some embodiments, extracting the one or more performance metrics includes determining, by the processing unit, a median value of each of one or more performance metrics for each network cell based on historical data collected over a predefined time period.

[0016] In some embodiments, the one or more performance metrics include at least one of signal quality, connectivity, data throughput, resource utilization, or fault indicators of the plurality of network cells.

[0017] In some embodiments, evaluating the one or more performance metrics of the one or more network cells within each cluster includes comparing, by the processing unit, the one or more performance metrics of each network cell within each cluster with a corresponding baseline performance threshold, selecting, by the processing unit, the one or more network cells when corresponding performance metrics associated with each of the one or more network cells are below the corresponding baseline performance threshold, the selected one or more network cells are identified as the one or more anomaly cells that are exhibiting low performance;and selecting, by the processing unit, the one or more network cells when corresponding performance metrics associated with each of the one or more network cells are above the corresponding baseline performance threshold, the selected one or more network cells are identified as the one or more anomaly cells that are exhibiting over-utilization.

[0018] In some embodiments, the morphological data comprises at least one of the inter-site distance between each of the plurality of network cells, a type of the plurality of network cells, and environmental information associated with each network cell.

[0019] In some embodiments, the building data comprises at least one of building type, number of floors, number of flats, building height, building density, rooftop structures, building materials, and building orientation.

[0020] In another exemplary embodiment, a system for detecting one or more anomaly cells in a network is described. The system includes an extraction unit and a processing unit. The extraction unit is configured to extract one or more performance metrics associated with a plurality of network cells. The processing unit is configured to determine a cell radius of each of the plurality of network cells based on morphological data associated with the plurality of network cells. The processing unit is further configured to categorize the plurality of network cells into one or more cell categories within the determined cell radius based on building data associated with the plurality of network cells. The processing unit is further configured to form one or more clusters of the plurality of network cells based on the one or more cell categories. Each cluster includes one or more network cells having a common set of environmental and morphological characteristics. The processing unit is further configured to evaluate the one or more performance metrics of the one or more network cells within each cluster and flag the one or more network cells as the one or more anomaly cells based on the evaluation.

[0021] In an exemplary embodiment, a computer program product comprising a non-transitory computer-readable medium is disclosed. The medium includes instructions that, when executed by one or more processors, cause the one or more processors to perform a method for detecting one or more anomaly cells in a network is described. The method includes extracting, by an extracting unit, one or more performance metrics associated with a plurality of network cells. The method further includes determining, by a processing unit, a cell radius of each of the plurality of network cells based on morphological data associated with the plurality of network cells. The method further includes categorizing, by the processing unit, the plurality of network cells into one or more cell categories within the determined cell radius based on building data associated with the plurality of network cells. The method further includes forming, by the processing unit, one or more clusters of the plurality of network cells based on the one or more cell categories. Each cluster includes one or more network cells having a common set of environmental and morphological characteristics. The method further includes evaluating, by the processing unit, the one or more performance metrics of the one or more network cells within each cluster and flagging the one or more network cells as the one or more anomaly cells based on the evaluation.

[0022] The foregoing general description of the illustrative embodiments and the following detailed description thereof are merely exemplary aspects of the teachings of this disclosure, and are not restrictive.OBJECTIVES OF THE PRESENT DISCLOSURE

[0023] Some of the objectives of the present disclosure, which at least one embodiment herein satisfies, are as follows:

[0024] An objective of the present disclosure is to provide a system and a method for detecting anomaly cells in a network.

[0025] Another objective of the present disclosure is to provide a comprehensive approach for detecting a list of anomaly cells in the network, which will maintain the network's health and efficiency and ensure a high-quality experience for users.

[0026] Yet another objective of the present disclosure is to provide an optimized approach and analysis results to detect anomaly cells based on morphological-based cell radius.

[0027] Yet another objective of the present disclosure is to filter out the anomaly cells in the network using artificial intelligence (Al) techniques or machine learning models.

[0028] Yet another objective of the present disclosure is to provide a feedbackbased mechanism for refining the anomaly detection process.

[0029] Another objective of the present disclosure is to provide a scalable and automated anomaly detection framework adaptable to diverse network environments.

[0030] Other objectives and advantages of the present disclosure will be more apparent from the following description, which is not intended to limit the scope of the present disclosure.BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWING

[0031] The accompanying drawings, which are incorporated herein, and constitute a part of this disclosure, illustrate exemplary embodiments of the disclosed methods and systems in which like reference numerals refer to the same parts throughout the different drawings. Components in the drawings are not necessarily to scale; emphasis is instead being placed upon clearly illustrating the principles of the present disclosure. Some drawings may indicate the components using block diagramsand may not represent the internal circuitry of each component. It will be appreciated by those skilled in the art that disclosure of such drawings includes disclosure of electrical components, electronic components, or circuitry commonly used to implement such components.

[0032] FIG. 1 illustrates an exemplary network architecture in which or with a system configured for detecting one or more anomaly cells in a network may be implemented, in accordance with embodiments of the present disclosure.

[0033] FIG. 2 illustrates an exemplary block diagram of the system configured for detecting the one or more anomaly cells in the network, in accordance with embodiments of the present disclosure.

[0034] FIG. 3 illustrates an exemplary process flow for detecting the one or more anomaly cells in the network, in accordance with an embodiment of the present disclosure.

[0035] FIG. 4 illustrates another exemplary process flow for detecting the one or more anomaly cells in the network, in accordance with an embodiment of the present disclosure.

[0036] FIG. 5 illustrates an exemplary flow diagram of a method for detecting the one or more anomaly cells in the network, in accordance with an embodiment of the present disclosure.

[0037] FIG. 6 illustrates an exemplary computer system in which or with which the embodiments of the present disclosure may be implemented.

[0038] The foregoing shall be more apparent from the following more detailed description of the disclosure.LIST OF REFERENCE NUMERALS100 - Network Architecture102 - User(s)104 - User Equipments (UEs)106 - Network 108 - System200 -Block Diagram202 - Processor(s)204 - Memory206 - Interface(s) 208 - Extracting Unit210 - Processing Unit212 - Database300 - Process flow400 - Process Flow 500 - Method Flow600 - Computer System610 - External Storage Device620 - Bus630 - Main Memory640 - Read Only Memory650 - Mass Storage Device660 - Communication Port(S)670 - ProcessorDETAILED DESCRIPTION

[0039] In the following description, for the purposes of explanation, various specific details are set forth in order to provide a thorough understanding of embodiments of the present disclosure. It will be apparent, however, that embodiments of the present disclosure may be practiced without these specific details. Several features described hereafter can each be used independently of one another or with any combination of other features. An individual feature may not address any of the problems discussed above or might address only some of the problems discussed above. Some of the problems discussed above might not be fully addressed by any of the features described herein. Example embodiments of the present disclosure are described below, as illustrated in various drawings in which like reference numerals refer to the same parts throughout the different drawings.

[0040] The ensuing description provides exemplary embodiments only, and is not intended to limit the scope, applicability, or configuration of the disclosure. Rather, the ensuing description of the exemplary embodiments will provide those skilled in the art with an enabling description for implementing an exemplary embodiment. It should be understood that various changes may be made in the function and arrangement of elements without departing from the spirit and scope of the disclosure as set forth.

[0041] Specific details are given in the following description to provide a thorough understanding of the embodiments. However, it will be understood by one of the ordinary skill in the art that the embodiments may be practiced without thesespecific details. For example, circuits, systems, networks, processes, and other components may be shown as components in block diagram form in order not to obscure the embodiments in unnecessary detail. In other instances, well-known circuits, processes, algorithms, structures, and techniques may be shown without unnecessary detail in order to avoid obscuring the embodiments.

[0042] Also, it is noted that individual embodiments may be described as a process that is depicted as a flowchart, a flow diagram, a data flow diagram, a structure diagram, or a block diagram. Although a flowchart may describe the operations as a sequential process, many of the operations can be performed in parallel or concurrently. In addition, the order of the operations may be re-arranged. A process is terminated when its operations are completed but could have additional steps not included in a figure. A process may correspond to a method, a function, a procedure, a subroutine, a subprogram, etc. When a process corresponds to a function, its termination can correspond to a return of the function to the calling function or the main function.

[0043] The word “exemplary” and / or “demonstrative” is used herein to mean serving as an example, instance, or illustration. For the avoidance of doubt, the subject matter disclosed herein is not limited by such examples. In addition, any aspect or design described herein as “exemplary” and / or “demonstrative” is not necessarily to be construed as preferred or advantageous over other aspects or designs, nor is it meant to preclude equivalent exemplary structures and techniques known to those of ordinary skill in the art. Furthermore, to the extent that the terms “includes,” “has,” “contains,” and other similar words are used in either the detailed description or the claims, such terms are intended to be inclusive like the term “comprising” as an open transition word without precluding any additional or other elements.

[0044] Reference throughout this specification to “one embodiment” or “an embodiment” or “an instance” or “one instance” means that a particular feature, structure, or characteristic described in connection with the embodiment is included inat least one embodiment of the present disclosure. Thus, the appearances of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0045] The terminology used herein is to describe particular embodiments only and is not intended to limit the disclosure. As used herein, the singular forms “a”, “an”, and “the” are intended to include the plural forms as well, unless the context indicates otherwise. It will be further understood that the terms “comprises” and / or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. As used herein, the term “and / or” includes any combinations of one or more of the associated listed items. It should be noted that the terms “mobile device”, “user equipment”, “user device”, “communication device”, “device” and similar terms are used interchangeably for the purpose of describing the invention. These terms are not intended to limit the scope of the invention or imply any specific functionality or limitations on the described embodiments. The use of these terms is solely for convenience and clarity of description. The invention is not limited to any particular type of device or equipment, and it should be understood that other equivalent terms or variations thereof may be used interchangeably without departing from the scope of the invention as defined herein.

[0046] While considerable emphasis has been placed herein on the components and component parts of the preferred embodiments, it will be appreciated that many embodiments can be made and that many changes can be made in the preferred embodiments without departing from the principles of the disclosure. These and other changes in the preferred embodiment as well as other embodiments of the disclosurewill be apparent to those skilled in the art from the disclosure herein, whereby it is to be distinctly understood that the foregoing descriptive matter is to be interpreted merely as illustrative of the disclosure and not as a limitation.

[0047] In conventional techniques, network operators typically relied on manual monitoring, visual dashboard inspections, or periodic field reports to identify anomaly cells. These reactive approaches depended heavily on operator intervention when users reported degraded service quality or when performance deviations were observed during scheduled network audits. Such methods are inefficient and timeconsuming, as they do not support proactive identification of cells likely to underperform. The delay in detecting and addressing cell-level issues often leads to extended service disruptions, inconsistent network optimization, and increased user dissatisfaction. Additionally, manual monitoring methods are prone to human error and are difficult to scale across large, heterogeneous network environments.

[0048] To address the limitations of conventional techniques, the present disclosure provides a method and a system for detecting anomaly cells in a network using an automated and intelligent analysis framework. The system employs data- driven clustering, morphology-based cell radius estimation, and performance metric comparison to identify underperforming and overutilized cells in real time. The system enables network operators to perform large-scale classification of network cells efficiently, eliminating the dependency on manual monitoring. The method can be easily scaled to classify thousands of cells across diverse geographic regions, including dense urban, suburban, and rural areas. The automated nature of the classification process ensures consistent and unbiased results, significantly reducing human error and operational effort. Furthermore, by proactively identifying anomaly cells before critical service degradation occurs, the method enhances overall network reliability, optimizes maintenance planning, and ensures superior quality of service for end users.

[0049] Hereinafter, exemplary embodiments of the present disclosure will be described with reference to the accompanying drawings.

[0050] FIG. 1 illustrates an exemplary network architecture 100 in which or with which a system 108 configured for detecting one or more anomaly cells in a network 106 may be implemented, in accordance with embodiments of the present disclosure.

[0051] In an embodiment, the network architecture 100 may include one or more user equipment (UEs) 104-1, 104-2... 104-N associated with one or more users 102-1, 102-2... 102-N in an environment. A person of ordinary skill in the art will understand that one or more users 102-1, 102-2... 102-N may be individually referred to as the user 102 and collectively referred to as the users 102. Further, the user 102 may correspond to a network administrator or a network service provider. Similarly, a person of ordinary skill in the art will understand that one or more UEs 104-1, 104- 2... 104-N may be individually referred to as the UE 104 and collectively referred to as the UEs 104. Although three UEs 104 are depicted in FIG. 1, however, any number of the UEs 104 may be included without departing from the scope of the ongoing description.

[0052] In another implementation, the UE 104 may function as smart devices operating in a smart environment, for example, an Internet of Things (loT) system. In such an embodiment, the UE 104 may include, but is not limited to, smart phones, smart watches, smart sensors (e.g., mechanical, thermal, electrical, magnetic, etc.), networked appliances, networked peripheral devices, networked lighting system, communication devices, networked vehicle accessories, networked vehicular devices, smart accessories, tablets, smart television (TV), computers, smart security system, smart home system, other devices for monitoring or interacting with or for the users 102 and / or entities, or any combination thereof. A person of ordinary skill in the art will appreciate that the UE 104 may include, but is not limited to, intelligent, multi-sensing, network- connected devices, which can integrate seamlessly with each other and / or with a central server or a cloud- computing system or any other device that is network-connected.

[0053] In an embodiment, the UE 104 may include, but is not limited to, any electrical, electronic, electro-mechanical, or an equipment, or a combination of one or more of the above devices such as virtual reality (VR) devices, augmented reality (AR) devices, laptop, a general-purpose computer, desktop, personal digital assistant, tablet computer, mainframe computer, or any other computing device, wherein the UE 104 may include one or more in-built or externally coupled accessories including, but not limited to, a visual aid device such as a camera, an audio aid, a microphone, a keyboard, and input devices for receiving input from the user 102 or the entity such as touch pad, touch enabled screen, electronic pen, and the like. A person of ordinary skill in the art will appreciate that the UE (104) may not be restricted to the mentioned devices and various other devices may be used.

[0054] In FIG. 1, the UE 104 may communicate with the system 108 via a telecommunication network 106 (interchangeably referred to as a network 106). In order to establish communication, initially, the telecommunication network 106 is configured to receive a connection request from the UE 104. In response to receiving the connection request, the telecommunication network 106 is configured to send an acknowledgment of the connection request to the UE 104. Further, a plurality of signals is transmitted in response to the connection request. Based on the connection request, the sessions are created in the telecommunication network 106. In an embodiment, the network 106 includes at least one of the 4G network, the 5G network, the 6G network, or the like. The telecommunication network 106 may enable the UE 104 to communicate with other devices in the network architecture 100 and / or with the system 108.

[0055] The network 106 may include a wireless card or some other transceiver connection to facilitate this communication. In another embodiment, the network 106 may be implemented as, or include any of a variety of different communication technologies such as a wide area network (WAN), a local area network (LAN), a wireless network, a mobile network, a Virtual Private Network (VPN), an internet, an intranet, a public network, a private network, a packet- switched network, a circuit- switched network, an ad hoc network, an infrastructure network, a Public-Switched Telephone Network (PSTN), a cable network, a cellular network, a satellite network, a fiber optic network, or some combination thereof. In another embodiment, the telecommunication network 106 includes, by way of example but not limitation, at least a portion of one or more networks having one or more nodes that transmit, receive, forward, generate, buffer, store, route, switch, process, or a combination thereof, etc. one or more messages, packets, signals, waves, voltage or current levels, some combination thereof, or so forth.

[0056] In an implementation, the system 108 may be a dedicated anomalydetection server deployed in an artificial intelligence (Al) environment and accessible to the UE 104 via the network 106. In one embodiment, the UE 104 may represent an administrator terminal operated by user 102. The UE 104 may include a client application or a browser- based console through which the user 102 may authenticate themselves to access the system 108. To access the system 108, the user 102 may perform authentication using one or more credentials and security techniques, including but not limited to username-password, certificate-based authentication, or multi-factor authentication. Once authenticated, the system 108 may assign the UE 104 role-based permissions with available actions (such as view-only, acknowledge, remediate, export, or push-change). Once the UE 104 initiates access to the system 108, the system 108 begins the process of detecting the one or more anomaly cells. In an aspect, the user 102 may set analysis parameters (such as time window, key performance indicators (KPIs), thresholds, morphology profiles, etc) in the system 108Y1 via the UE 104. Once the detection process is complete, the UE 104 may request periodic or real-time reports from the system 108.

[0057] Although FIG. 1 shows exemplary components of the network architecture 100, in other embodiments, the network architecture 100 may include fewer components, different components, differently arranged components, or additional functional components than depicted in FIG. 1. Additionally, or alternatively, one or more components of the network architecture 100 may perform functions described as being performed by one or more other components of the network architecture 100.

[0058] FIG. 2 illustrates an exemplary block diagram of the system 108 configured for detecting the one or more anomaly cells in the network 106, in accordance with an embodiment of the present disclosure.

[0059] In an embodiment, the system 108 may include one or more processor(s) 202. The one or more processor(s) 202 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, logic circuitries, and / or any devices that process data based on operational instructions. Among other capabilities, the one or more processor(s) 202 may be configured to fetch and execute computer-readable instructions stored in a memory 204 of the system 108. The memory 204 may be configured to store one or more computer-readable instructions or routines in a non-transitory computer readable storage medium, which may be fetched and executed to create or share data packets over a network service. The memory 204 may include any non-transitory storage device including, for example, volatile memory such as a Random- Access Memory (RAM), or a non-volatile memory such as an Erasable Programmable Read Only Memory (EPROM), a flash memory, and the like.

[0060] In an embodiment, the system 108 may include an interface(s) 206. The interface(s) 206 may include a variety of interfaces, for example, interfaces for data input and output devices (I / O), storage devices, and the like. The interface(s) 206 may facilitate communication through the system 108. The interface(s) 206 may also provide a communication pathway for one or more components of the system 108. Examples of such components include, but are not limited to, an extracting unit 208, a processing unit 210 and a database 212.

[0061] In an implementation, the processing unit 210 may be implemented as a combination of hardware and programming (for example, programmable instructions) to implement one or more functionalities of the system 108. In examples described herein, such combinations of hardware and programming may be implemented in several different ways. For example, the programming for the processing unit 210 may be processor-executable instructions stored on a non- transitory machine-readable storage medium and the hardware for the processing unit 210 may include a processing resource (for example, the one or more processors 202) to execute such instructions. In one implementation, the extracting unit 208, the processing unit 210 and the processor(s) 202 may be operatively coupled to each other via the interface(s) 206 to enable seamless data acquisition, decoding, and dispatching of received information for further handling or storage.

[0062] In an embodiment, the system 108 is configured to detect the one or anomaly cells in the network 106. Herein, the one or more anomaly cells refer to network cells that exhibit unusual or abnormal performance compared to their peer cells. The one or more anomaly cells may indicate various abnormal performance metrics, such as unexpected drops in signal strength, increased call drop rates, or slower data speeds. Identifying these anomalies in the cells is essential for maintaining network quality and ensuring that users have a reliable connection. Additionally,monitoring and addressing anomaly cells is essential for maintaining the health and efficiency of the network 106, ensuring a high-quality user experience.

[0063] In an embodiment, the system 108 may be a network server embedded with one or more artificial intelligence (Al) or machine learning (ML) techniques or algorithms to identify the one or more anomaly cells within the network 106. In an aspect, accurate and comprehensive data from different network monitoring tools, such as a network management system (NMS), may be fed as an input into the system 108 by the user 102. The NMS is a centralized software platform and serves as a data source from which the system 108 extracts key performance metrics for analysis and anomaly detection. Based on the input data, the system 108 may analyze and identify the one or more anomaly cells using the one or more Al or ML-based algorithms.

[0064] In an embodiment, in order to detect the one or more anomaly cells, the extracting unit 208 is configured to extract one or more performance metrics associated with a plurality of network cells. The one or more performance metrics refer to quantitative parameters or Key Performance Indicators (KPIs) that represent the operational efficiency, service quality, and performance status of each of the plurality of network cells within the network 106. In a more elaborate way, the one or more performance metrics (i.e., the KPIs) provide insight into how each network cell performs relative to its expected behaviors or objectives. In an aspect, the one or more performance metrics associated with the plurality of network cells may include, but are not limited to, a signal quality of each network cell, connectivity of each network cell, data throughput, resource utilization, or fault indicators of the plurality of network cells. The one or more performance metrics may be extracted from the network monitoring systems (NMS), operation support systems (OSS), or performance management databases such as the database 212. In an aspect, for extracting the one or more performance metrics, the processing unit 210 is configured to determine a median value of each of one or more performance metrics for each network cell based onhistorical data collected over a predefined time period. In an example, the historical data is the historical one or more performance metric values (i.e., KPI) regarding each cell’s performance. Using the median value of each KPI helps to make informed decisions based on a more reliable representation of performance, which is essential for performance evaluation, resource allocation, and strategic planning.

[0065] In an aspect, the historical data may be continuously gathered and stored in the database 212 from multiple network sources, such as performance management servers or network data collectors. During extraction, the processing unit 210 retrieves this stored historical data corresponding to each network cell and organizes one or more values associated with each performance metric in a chronological or numerical order for a selected time period (for example, daily, weekly, or monthly intervals).

[0066] In an example, the one or more performance metrics, such as Reference Signal Received Power (RSRP), Reference Signal Received Quality (RSRQ), Signal- to-Interference-plus-Noise Ratio (SINR), throughput, latency, Call Setup Success Rate (CSSR), Handover Success Rate (HOSR), packet loss ratio, resource block utilization, and cell availability, may be extracted from the database 212. Once extracted, the processing unit 210 performs a computation to determine the median value for each performance metric corresponding to each network cell. The processing unit 210 may organize the extracted performance metrics values for a defined analysis period, such as the last 7 days, 15 days, or one month. These values are then sorted in ascending numerical order for each performance metric. After sorting, the median is identified as the central value in the ordered dataset.

[0067] For example, considering a scenario of a network cell where the historical data (i.e., KPI ) of the extracted RSRP values (in dBm) over ten observation periods are: -90, -89, -88, -87, -87, -86, -84, -83, -80, -70. When arranged in ascending order (-90, -89, -88, -87, -87, -86, -84, -83, -80, -70), the dataset contains ten values (an even number). Hence, the median is computed as the average of the 5th and 6thvalues, i.e.,-86.5 dBm. This median RSRP value of -86.5 dBm represents the typical signal strength experienced by users in that network cell.

[0068] In an embodiment, the processing unit 210 is configured to determine a cell radius of each of the plurality of network cells based on morphological data associated with the plurality of network cells. The cell radius is the effective coverage distance or area around each network cell within which the plurality of network cells can provide reliable wireless communication to the UE 104. The purpose of determining the cell radius is to estimate the coverage area of each network cell for accurate network planning and optimization, and enable identification of overlapping or under-served areas, which is essential for detecting performance anomalies. The morphological data (also referred to as surrounding morphology) associated with the plurality of network cells refers to the physical and environmental characteristics of the geographical area in which each of the network cells is deployed. In an aspect, the morphological data may be collected from the NMS, Geographical Information Systems (GIS), digital map databases, and external data repositories.

[0069] In an aspect, the morphological data may include, but are not limited to, inter-site distance (ISD) between each of the plurality of network cells, the type of the plurality of network cells, and environmental information associated with each network cell. These factors collectively influence how radio signals propagate, attenuate, or reflect within the network cell’s coverage area. The ISD refers to the physical separation distance between two or more sites (such as cell towers or base stations) within the network 106 and serves as a key parameter for estimating the cell radius (coverage boundary) of each network cell. The type of network cell indicates the deployment category of each network cell, such as macro cell, micro cell, pico cell, or femto cell, each having distinct transmission power levels and coverage capabilities. The environmental information represents characteristics and surrounding physicalobstructions, including building density, building height, and topography, all of which influence signal propagation and attenuation.

[0070] Using these morphological data collectively, the processing unit 210 determines the cell radius by applying one or more predefined propagation models and scaling factors. For example, for dense urban areas, which are the areas with high population density and numerous buildings, the cell radius may be approximated as the inter-site distance, for mixed urban (MU) or suburban (SU) areas, the areas that are less densely populated than dense urban zones, the effective cell radius may be 0.8 times the ISD, and for rural (RU) areas, with much lower population density, the cell radius may be 0.6 times the ISD.

[0071] In an embodiment, the processing unit 210 is configured to categorize the plurality of network cells into one or more cell categories within the determined cell radius based on building data associated with the plurality of network cells. The building data may include, but are not limited to, building type, number of floors, number of flats, building height, building density, rooftop structures, building materials, and building orientation. Such categorization enables the system 108 to differentiate between different propagation environments that directly influence radio signal behavior. For example, the building data helps the processing unit 210 classify the plurality of network cells into the one or more categories such as: dense urban cells that are characterized by high-rise buildings, high building density, and reflective materials (e.g., glass or metal facades), leading to higher multipath propagation and signal attenuation, urban residential cells that are characterized by medium-height apartments and moderate building density, resulting in partial signal blockage and shadowing effects, suburban cells consisting of low-rise buildings and open spaces, where signal propagation is relatively less obstructed and rural or open area cells where few or no large buildings exist, allowing free-space propagation and larger effective cell coverage, indoor cells that are deployed in small coverage areas and dedicated in-building solutions such as femtocells or distributed antenna systems (DAS). The building data used for such categorization may be obtained from city planning databases, GIS maps, or network optimization tools that integrate three-dimensional (3D) building models. Further, the obtained building data may be stored in the database 212 and continuously updated by the network operator (i.e., the user 102) based on network survey inputs.

[0072] In an aspect, a machine learning (ML) approach is employed to enhance the categorization or classification of the plurality of network cells. The machine learning approach leverages the building data and cell radius to create more accurate cell categories. Examples of ML approaches that may be employed include, but are not limited to, a feature extraction approach, a K-means clustering approach, and a normalization approach. In an example, in a machine learning (ML) model, the building data, which includes the types of buildings (residential, commercial, industrial), the number of floors in the building, the number of flats in the building, etc., and the determined cell radius, is given as input. Further, buildings within the determined cell radius are categorized into buckets based on the number of floors. This bucketization process captures the distribution of building types within the determined cell radius of each network cell. Further, the ML model performs feature extraction, where for each network cell, the proportional area occupied by buildings in each bucket within the cell radius is calculated and treated as input features for the ML model. Additionally, a normalization process is applied to these features to scale them to a consistent range, to ensure that the feature extraction ML algorithm processes the features correctly.

[0073] In an embodiment, the processing unit 210 is configured to form one or more clusters of the plurality of network cells based on the one or more cell categories. In an aspect, each cluster includes one or more network cells having a common set of environmental and morphological characteristics. The one or more clusters refer to thegroup of network cells that share common characteristics. The one or more network cells are considered to have a common set of morphological and environmental characteristics if it exhibits common in features such as cell radius, inter-site distance (ISD), type of cell (macro, micro, pico, femto), building distribution, building heights, building density, terrain type (urban, suburban, rural), presence of obstacles, building materials, etc. For example, two cells located in a dense urban environment with high- rise buildings, similar ISD, and cell radius would be grouped into the same cluster, whereas a suburban cell with low-rise residential buildings and a larger coverage radius would belong to a different cluster. To form these clusters, the processing unit 210 may employ an unsupervised machine learning algorithm, such as the K-Means clustering, where a feature vector is first constructed for each cell using normalized values of the morphological and environmental characteristics. Further, a predefined or dynamically determined number of clusters, k, is selected, and the K-Means algorithm initializes k centroids in a feature space. Each network cell is then assigned to the cluster and updated until a maximum number of iterations is reached.

[0074] In an embodiment, the processing unit 210 is configured to evaluate the one or more performance metrics of the one or more network cells within each cluster. In order to evaluate, the processing unit 210 is configured to compare the one or more performance metrics of each network cell within each cluster with a corresponding baseline performance threshold. The baseline performance threshold of each performance metric refers to a reference value or expected standard for that metric, against which the current performance of the one or more network cells within each cluster is compared to detect anomalies or deviations. In an aspect, the baseline performance threshold may be derived from historical data, industry standards, or it may be defined by the network operator. The corresponding baseline performance threshold for each performance metric is regularly updated to adapt to changing network conditions and traffic patterns.

[0075] In an aspect, the processing unit 210 is configured to select the one or more network cells when corresponding performance metrics associated with each of the one or more network cells are below the corresponding baseline performance threshold. The selected one or more network cells are identified as the one or more anomaly cells that are exhibiting low performance. Thus, the one or more network cells identified below the corresponding baseline performance threshold require further investigation to determine the root cause of their poor performance. Subsequently, the processing unit 210 is configured to select the one or more network cells when corresponding performance metrics associated with each of the one or more network cells are above the corresponding baseline performance threshold, where the selected one or more network cells are identified as the one or more anomaly cells that are exhibiting over-utilization. Thus, the one or more network cells identified above the corresponding baseline performance threshold experience high traffic volumes and may require additional capacity to meet demand.

[0076] In an exemplary scenario, a cluster of dense urban cells in the network 106 is formed by the processing unit 210, and the baseline performance threshold of the corresponding performance metric of each network cell is defined based on historical data and operator standards. The baseline thresholds for the cluster may include: a throughput of 50 Mbps, a call drop rate (CDR) of 2%, a handover success rate (HOSR) of 98%, a packet loss ratio of 1%, a SINR of -5 dB, and a resource block utilization of 80%. If a particular cell within this cluster exhibits the throughput of 30 Mbps, the call drop rate of 5%, the HOSR of 95%, the packet loss ratio of 3%, the SINR of -8 dB, and the resource block utilization of 90%, the processing unit 210 compares each measured performance metric against the corresponding baseline threshold. The observed deviations include the lower throughput, higher call drop rate, reduced handover success, increased packet loss, weaker signal quality, and higher resource utilization, indicating a performance anomaly, potentially caused by network congestion, interference, hardware issues, or misconfiguration. Conversely, if thenetwork cell performs above the baseline performance threshold in any metrics, for example, resource block utilization exceeding 85% consistently, it indicates an overutilized or high-traffic network cell. In an embodiment, based on the evaluation, the processing unit 210 is configured to flag the one or more network cells as the one or more anomaly cells. The term flagging refers to the process of marking or labeling specific network cells that exhibit abnormal performance behavior when compared with the corresponding baseline performance threshold. Based on the evaluation, the selected one or more network cells are flagged as potential anomaly candidates and are prioritized for further analysis, diagnostic investigation, or corrective action by the network operators. For example, if the baseline threshold for the HOSR within a particular cluster is 98%, and a specific cell consistently reports 92%, the system 108 flags that cell as the anomaly cell. The flagging mechanism allows the network operator to quickly isolate problem areas, streamline troubleshooting, and maintain optimal network performance.

[0077] In an embodiment, the processing unit 210 is configured to generate an automatic report that includes a list of flagged or identified one or more anomaly cells and the corresponding one or more performance metrics, i.e., the information about the KPIs. The generated report provides a comprehensive overview of the one or more anomaly cells and their characteristics. The processing unit 210 organizes the identified one or more anomaly cells and the corresponding one or more performance metrics in a structured manner, including for each anomaly cell: the cell identifier (ID), cluster association, cell type, cell radius, measured KPI values, corresponding baseline thresholds, deviation from baseline, and potential cause of anomaly. The report may be generated in digital formats such as Portable Document Format (PDF), Comma- separated values (CSV), or Excel. Additionally, the report may include visual representations, such as tables, charts, heatmaps, or graphs, to depict performance deviations across network cells. For example, a heatmap may illustrate cells with high call drop rates in red and cells performing within thresholds in green. Further, thegenerated report may be utilized by the user 102 to quickly identify, analyze, and take corrective actions for underperforming or overutilized network cells.

[0078] In an embodiment, the system 108 may employ a feedback mechanism to refine the detection and analysis of the one or more anomaly cells. Based on the feedback, the system 108 may update thresholds, and retrain the ML model to improve the precision of cell categorization and anomaly detection. For example, if a cell previously flagged as anomalous is confirmed to be performing normally after maintenance or optimization, the system 108 adjusts the baseline performance threshold or the feature weightage used in the ML model to prevent false positives in future analyses. The iterative feedback loop ensures continuous improvement of the system 108, which enables more accurate, adaptive, and context-aware network performance monitoring and reduces errors in identifying underperforming or overutilized network cells over time.

[0079] Although FIG. 2 shows exemplary components of the system 108, in other embodiments, the system 108 may include fewer components, different components, differently arranged components, or additional functional components than depicted in FIG. 2. Additionally, or alternatively, one or more components of the system 108 may perform functions described as being performed by one or more other components of the system 108.

[0080] FIG. 3 illustrates an exemplary process flow 300 for detecting the one or more anomaly cells in the network 106, in accordance with an embodiment of the present disclosure. FIG. 3 is explained in conjunction with FIGS. 1 and 2.

[0081] In an aspect, the process flow 300 may be executed by the system 108 (as described with reference to FIG. 2) for detecting or identifying the one or more anomaly cells within the network 106. The system 108 may be implemented as a centralized or distributed server equipped with the one or more Al modules and MLengines configured to analyze network performance data. The system 108 integrates data-driven analytics, Al-based pattern recognition, and morphological analysis to ensure a comprehensive and intelligent anomaly detection process.

[0082] At step 302, the detection of the one or more anomaly cells within the network 106 begins with the identification of anomaly cells, which are cells exhibiting abnormal performance compared to their peers. In an aspect, the user 102 may initiate the anomaly cell detection process by providing a command through a user interface or command terminal associated with the system 108. The input parameters provided by the user 102 may include, but are not limited to, a specific geographical region, cluster identifier, time window for analysis, network technology type (e.g., LTE, NR / 5G), and one or more KPI categories to be evaluated. Additionally, the user 102 may define custom performance thresholds, select data sources, or specify preferred machine learning model configurations to tailor the anomaly detection process. Upon receiving and authenticating the user input, the system 108 triggers the detection workflow for identifying the plurality of network cells within the specified parameters.

[0083] Once the detection process is initiated, further at step 304, the system 108 proceeds to extract one or more performance metrics (i.e., KPIs) associated with the plurality of network cells from a network database, such as the database 212. In an embodiment, the system 108 may retrieve the KPI data from one or more network management platforms, such as the NMS, the OSS, or other performance monitoring entities integrated within the network 106. The extracted KPIs may include, but are not limited to, RSRP, SINK, throughput, HOSR, latency, call setup success rate, packet loss ratio, resource block utilization, etc. These performance metrics may be periodically collected, timestamped, and stored in the database 212 in a structured format, such as tabular or time-series data.

[0084] At step 306, the system 108 performs the morphology-based cell radius identification process begins to refine the analysis further. To further refine theanalysis, the cell radius is determined based on the surrounding morphology. This step considers the density of the urban environment and its impact on signal propagation. The cell radius is determined based on the surrounding morphology (referred to as the morphological data). As depicted in the process flow 300, the morphology-based cell radius identification is shown in three sub-steps: 308, 310, and 312. When the intersite distance between the neighboring cells is relatively small, the system 108 may determine that the corresponding cells are deployed in a dense urban environment, thereby assigning a smaller cell radius (for instance, between 300 to 500 meters). For example, for dense urban cells (or areas with high population density and numerous buildings), as shown in sub-step 308 the cell radius is equal to the inter-site distance, When the inter-site distance is large and the cell type is macro, the system 108 may determine a larger coverage radius (for instance, between 2 to 5 kilometers), as typically observed in suburban or rural environments. For example, for mixed urban (MU) and suburban (SU) cells (or areas that are less densely populated than dense urban zones), as shown in sub-step 310, the cell radius is reduced to 0.8 times the ISD. This adjustment accounts for lower user density and fewer obstacles.

[0085] Further, when the inter-site distance is significantly large, and the surrounding morphological data indicate low building density, open terrain, and minimal physical obstructions, the system 108 may identify the plurality of network cells as rural (RU) cells which means areas with much lower population density. In such cases, the system 108 assigns a larger cell radius. For example, for rural cells, as shown at sub-step 312, the cell radius is further reduced to 0.6 times the ISD. This reflects the increased distance between users and the need for a broader coverage area due to fewer cell sites.

[0086] At step 314, predictive model-based cell categorization is performed. The cells are categorized based on the type of buildings in their coverage area. Building types can include residential, commercial, industrial, number of floors, number of flats,etc. The machine learning (ML) model is employed to enhance the classification of cells based on their morphological types. As depicted in the process flow 300, the ML model performs cell categorization in a plurality of sub-steps.

[0087] At sub-step 316, each coverage boundary of each network is created based on a predetermined coverage radius (i.e., the cell radius).

[0088] At sub-step 318, the building data within each cell’s coverage boundary is integrated into the ML model. The building data includes details such as the geographic location of buildings, the total area covered by the building, building type, number of floors, number of flats, building height, building density, rooftop structures, building materials, building orientation etc.

[0089] At sub-step 320, the buildings within the cell’s coverage area are categorized into groups or buckets based on the number of floors. This bucketization allows to capture the distribution of building types within each cell's coverage area.

[0090] At sub-step 322, for each network cell, the area covered by each bucket within the coverage boundary is calculated and treated as features for the ML model.

[0091] At sub-step 324, the ML model processes the extracted features to perform the normalization process. The normalization process adjusts the scale of the extracted features, making them comparable and preventing any one feature from disproportionately influencing the ML model.

[0092] At sub-step 326, in the normalized features an unsupervised machine learning algorithm such as the K-Means clustering algorithm is applied to form the one or more clusters. For example, the K-Means clustering algorithm is applied to the normalized features to identify the optimal clusters that represent the one or more cell categories. K-Means is a popular unsupervised learning algorithm that partitions data into K distinct clusters based on feature similarity.

[0093] At sub-step 328, the trained ML model is saved and used for future cell classifications, ensuring consistency and accuracy in the categorization process.

[0094] Upon application of the unsupervised ML algorithm, at step 330, the ML model identifies the look-alike (or similar) network cells based on building statistics and morphological category within the cell radius (i.e., the cell coverage area),

[0095] At step 332, the categorized look-alike network cells are extracted based on the pre-defined thresholds (referred to as the baseline performance threshold). During this step, the system 108 analyzes the performance metrics (i.e., KPIs) of each network cell within the formed clusters and compares them with the corresponding baseline thresholds. Based on this comparison, the system 108 identifies bottomperforming network cells, which exhibit performance metrics below the expected threshold values indicating potential anomalies such as congestion, interference, or configuration errors. Additionally, the system 108 determines top-performing network cells that show abnormally high utilization or resource consumption, which may signify emerging capacity issues or load imbalance.

[0096] At step 334, the network cells identified in step 332 are flagged as anomaly cells. This step involves marking the detected network cells as potentially anomalous for further diagnostic analysis and corrective action.

[0097] At step 336, the system 108 automatically generates a report listing the identified anomaly cells along with detailed KPI information, cluster identifiers, deviation levels from the baseline, and possible causes provided by the ML model. The generated report may be presented in a tabular or dashboard format, allowing network operators to visualize trends, severity levels, and spatial distribution of anomalies for efficient decision-making.

[0098] At step 338, the feedback mechanism is implemented and is provided in step 308 to continuously refine and retrain the ML model based on the analysisresults, operational outcomes, and corrective actions taken by administrators. This closed-loop feedback process ensures that the ML model dynamically adapts to evolving network conditions and improves its anomaly detection accuracy over time.

[0099] At step 340, the anomaly detection process concludes, and the generated report is reviewed by network administrators or automated network management tools for further analysis, root cause diagnosis, or execution of remedial actions such as reconfiguration, optimization, or capacity enhancement.

[0100] FIG. 4 illustrates another exemplary process flow 400 for detecting the one or more anomaly cells in the network 106, in accordance with an embodiment of the present disclosure. FIG. 4 is explained in conjunction with FIGS. 1, 2 and 3.

[0101] At step 402, the system 108 initiates the process of detecting the plurality of network cells in the network 106. In an aspect, the user 102 may provide a command via the user interface or the command terminal provided by the system 108 (as described with reference to FIGS. 2 and 3) to initiate the process of detecting the one or more anomaly cells. The input received from the user 102 may include parameters such as a specific geographical region, cluster identifier, time duration, network technology type (e.g., LTE, 5G), or key performance indicator (KPI) category to be analyzed. Additionally, the user 102 may specify custom threshold values, data sources, or ML model selection preferences to guide the anomaly detection process. The system 108 authenticates the user 102 and, based on the input, the system 108 initiates the process of detecting the plurality of network cells.

[0102] Once the detection process is initiated, at step 404, the system 108 proceeds to extract the one or more performance metrics (i.e., the KPIs) associated with the plurality of network cells from the database 212. In an example, the KPIs such as the RSRP, SINK, throughput, HOSR, latency, call setup success rate, packet loss ratio,resource block utilization, etc., may be stored in the database 212 and are extracted by the system 108.

[0103] At step 406, the system 108 determines the cell radius for each of the plurality of network cells based on the surrounding morphology, i.e., the morphological data. For example, based on the inter-site distance between each of the plurality of network cells, the type of the plurality of network cells, and the environmental information associated with each network cell, the system 108 may determine the cell radius, i.e., the coverage area.

[0104] At step 408, the system 108 categorizes the plurality of network cells based on the determined cell radius and the building statistics (referred to as the building data in the FIG. 2). For example, using the building statistics such as building type, number of floors, number of flats, building height, building density, rooftop structures, building materials, and building orientation within the determined cell radius, the system 108 classifies the plurality of network cells into the one or more cell categories.

[0105] At step 410, the system 108 forms the one or more clusters or groups the one or more network cells among the plurality of network cells based on the one or more cell categories. For example, the system 108 identifies the one or more network cells that exhibit common or similar characteristics based on the one or more categories and groups the look-alike (or similar) network cells.

[0106] Further, within each formed group or cluster, the system 108 performs evaluation by comparing the one or more performance metrics of each network cells within each cluster with the corresponding baseline performance threshold. Further, at step 412, the system 108 extracts the one or more anomaly cells from the group based on the evaluation.

[0107] At step 414, the system 108 generates the report listing the identified or flagged anomaly cells with details about their performance KPIs.

[0108] At step 416, a feedback loop from step 414 to step 402 is implemented to refine the system 108 based on findings and operational outcomes.

[0109] At step 418, the anomaly cells detection process is completed and marked as ‘End’. The network operators may review the report for further analysis or action.

[0110] FIG. 5 illustrates an exemplary flow diagram 500 of a method for detecting the one or more anomaly cells in the network 106, in accordance with an embodiment of the present disclosure. FIG. 5 is explained in conjunction with FIG. 2.

[0111] The method 500 provides a comprehensive approach to cell anomaly detection, providing a powerful tool for network operators to identify and address performance issues. By leveraging data analysis, machine learning, and morphological considerations, the method 500 enables proactive identification of anomalous cells, leading to improved network performance and user experience.

[0112] At step 502, the method 500 includes extracting, by the extracting unit 208, the one or more performance metrics associated with the plurality of network cells. The one or more performance metrics may include, but are not limited to, signal quality, connectivity, data throughput, resource utilization, or fault indicators of the plurality of network cells. In an aspect, extracting the one or more performance metrics includes determining, by the processing unit 210, the median value of each of one or more performance metrics for each network cell based on historical data collected over a predefined time period.

[0113] At step 504, the method 500 includes determining, by the processing unit 210, the cell radius of each of the plurality of network cells based on themorphological data associated with the plurality of network cells. The morphological data may include, but is not limited to, the inter-site distance between each of the plurality of network cells, the type of the plurality of network cells, and the environmental information associated with each network cell.

[0114] At step 506, the method 500 includes categorizing, by the processing unit 210, the plurality of network cells into one or more cell categories within the determined cell radius based on building data associated with the plurality of network cells. The building data may include, but is not limited to, building type, number of floors, number of flats, building height, building density, rooftop structures, building materials, and building orientation.

[0115] At step 508, the method 500 includes evaluating, by the processing unit 210, the one or more performance metrics of the one or more network cells within each cluster. In an aspect, for evaluating the one or more performance metrics of the one or more network cells within each cluster includes the processing unit 210 is configured to compare the one or more performance metrics of each network cell within each cluster with the corresponding baseline performance threshold. Further, the processing unit 210 is configured to select the one or more network cells when corresponding performance metrics associated with each of the one or more network cells are below the corresponding baseline performance threshold, where the selected one or more network cells are identified as the one or more anomaly cells that are exhibiting low performance. Further, the processing unit 210 is configured to select the one or more network cells when corresponding performance metrics associated with each of the one or more network cells are above the corresponding baseline performance threshold, where the selected one or more network cells are identified as the one or more anomaly cells that are exhibiting over-utilization.

[0116] At step 510, the method 500 includes flagging, by the processing unit 210, the one or more network cells as the one or more anomaly cells based on theevaluation. In an aspect, the processing unit 210 is configured to generate a report comprising a list of the flagged one or more anomaly cells and the corresponding one or more performance metrics.

[0117] FIG. 6 illustrates an example computer system 600 in which or with which the embodiments of the present disclosure may be implemented.

[0118] As shown in FIG. 6, the computer system 600 may include an external storage device 610, a bus 620, a main memory 630, a read-only memory 640, a mass storage device 650, a communication port(s) 660, and a processor 670. A person skilled in the art will appreciate that the computer system 600 may include more than one processor and communication ports. The processor 670 may include various modules associated with embodiments of the present disclosure. The communication port(s) 660 may be any of an RS-232 port for use with a modem-based dialup connection, a 10 / 100 Ethernet port, a Gigabit or 10 Gigabit port using copper or fiber, a serial port, a parallel port, or other existing or future ports. The communication ports(s) 660 may be chosen depending on a network, such as a Local Area Network (LAN), Wide Area Network (WAN), or any network to which the computer system 600 connects.

[0119] In an embodiment, the main memory 630 may be Random Access Memory (RAM), or any other dynamic storage device commonly known in the art. The read-only memory 640 may be any static storage device(s) e.g., but not limited to, a Programmable Read Only Memory (PROM) chip for storing static information e.g., start-up or basic input / output system (BIOS) instructions for the processor 670. The mass storage device 650 may be any current or future mass storage solution, which can be used to store information and / or instructions. Exemplary mass storage solutions include, but are not limited to, Parallel Advanced Technology Attachment (PATA) or Serial Advanced Technology Attachment (SATA) hard disk drives or solid-state drives (internal or external, e.g., having Universal Serial Bus (USB) and / or Firewire interfaces).

[0120] In an embodiment, the bus 620 may communicatively couple the processor(s) 670 with the other memory, storage, and communication blocks. The bus 620 may be, e.g. a Peripheral Component Interconnect PCI) / PCI Extended (PCI-X) bus, Small Computer System Interface (SCSI), Universal Serial Bus (USB), or the like, for connecting expansion cards, drives, and other subsystems as well as other buses, such a front side bus (FSB), which connects the processor 670 to the computer system 600.

[0121] In another embodiment, operator, and administrative interfaces, e.g., a display, keyboard, and cursor control device may also be coupled to the bus 620 to support direct operator interaction with the computer system 600. Other operator and administrative interfaces can be provided through network connections connected through the communication port(s) 660. Components described above are meant only to exemplify various possibilities. In no way should the aforementioned exemplary computer system 600 limit the scope of the present disclosure.

[0122] In an exemplary embodiment, a computer program product comprising a non-transitory computer-readable medium is disclosed. The medium includes instructions that, when executed by one or more processors, cause the one or more processors to perform a method for detecting one or more anomaly cells in a network is described. The method includes extracting, by an extracting unit, one or more performance metrics associated with a plurality of network cells. The method further includes determining, by a processing unit, a cell radius of each of the plurality of network cells based on morphological data associated with the plurality of network cells. The method further includes categorizing, by the processing unit, the plurality of network cells into one or more cell categories within the determined cell radius based on building data associated with the plurality of network cells. The method further includes forming, by the processing unit, one or more clusters of the plurality of network cells based on the one or more cell categories. Each cluster includes one ormore network cells having a common set of environmental and morphological characteristics. The method further includes evaluating, by the processing unit, the one or more performance metrics of the one or more network cells within each cluster and flagging the one or more network cells as the one or more anomaly cells based on the evaluation.

[0123] The present disclosure provides a technical advancement in the field of network performance management and optimization within telecommunication networks by introducing a system (i.e., an Al-based anomaly cell detection system) for accurately identifying and classifying anomalous or underperforming network cells. Unlike conventional network monitoring tools that rely on static KPI thresholds and manual analysis, the system leverages morphological data, building information, and machine learning (ML) techniques to dynamically detect performance irregularities. By extracting and analyzing real-time and historical KPI data, determining cell radius based on morphology, categorizing cells using building data, forming clusters through unsupervised learning, and applying baseline performance comparison, the system ensures automated, data-driven, and scalable anomaly detection. The method enables network operators to proactively identify performance degradation, capacity bottlenecks, and coverage anomalies, thereby improving network reliability, user experience, and operational efficiency.

[0124] While the foregoing describes various embodiments of the invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof. The scope of the invention is determined by the claims that follow. The invention is not limited to the described embodiments, versions or examples, which are included to enable a person having ordinary skill in the art to make and use the invention when combined with information and knowledge available to the person having ordinary skill in the art.

[0125] The method and system of the present disclosure may be implemented in a number of ways. For example, the methods and systems of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order for the steps of the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless specifically stated otherwise. Further, in some embodiments, the present disclosure may also be embodied as programs recorded in a recording medium, the programs including machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.

[0126] While considerable emphasis has been placed herein on the preferred embodiments, it will be appreciated that many embodiments can be made and that many changes can be made in the preferred embodiments without departing from the principles of the disclosure. These and other changes in the preferred embodiments of the disclosure will be apparent to those skilled in the art from the disclosure herein, whereby it is to be distinctly understood that the foregoing descriptive matter to be implemented merely as illustrative of the disclosure and not as limitation.ADVANCEMENTS OF THE PRESENT DISCLOSURE

[0127] The present disclosure described herein above has several technical advantages as follows:

[0128] The present disclosure offers an optimized approach to detect anomaly cells in the network.

[0129] The present disclosure provides an optimized machine learning modelbased classification approach to group network cells exhibiting abnormal performance.

[0130] The present disclosure performs a comparative analysis of cell key performance indicators (KPIs) to filter out anomaly cells among similar cells.

[0131] The present disclosure offers Al and ML-based approaches that can be easily scaled to classify thousands of cells across diverse geographic regions.

[0132] The present disclosure ensures consistent results by automating the classification process, thus reducing human error.

[0133] The present disclosure provides a comprehensive approach to cell anomaly detection, providing a powerful tool for network operators to identify and address performance issues. By leveraging data analysis, machine learning, and morphological considerations, the module enables proactive identification of anomalous cells, leading to improved network performance and user experience.

[0134] The present disclosure offers adaptability by incorporating the ML model that can be retrained with new data, allowing adaptation to changes in network infrastructure or urban development

Claims

CLAIMS1. A method (500) for detecting one or more anomaly cells in a network (106), the method (500) comprising: extracting (502), by an extracting unit (208), one or more performance metrics associated with a plurality of network cells; determining (504), by a processing unit (210), a cell radius of each of the plurality of network cells based on morphological data associated with the plurality of network cells; categorizing (506), by the processing unit (210), the plurality of network cells into one or more cell categories within the determined cell radius based on building data associated with the plurality of network cells; forming (508), by the processing unit (210), one or more clusters of the plurality of network cells based on the one or more cell categories, wherein each cluster comprises one or more network cells having a common set of environmental and morphological characteristics; evaluating (510), by the processing unit (210), the one or more performance metrics of the one or more network cells within each cluster; and flagging (512), by the processing unit (210), the one or more network cells as the one or more anomaly cells based on the evaluation.

2. The method (500) as claimed in claim 1, comprising: generating, by a processing unit (210), a report comprising a list of the flagged one or more anomaly cells and the corresponding one or more performance metrics.

3. The method (500) as claimed in claim 1, wherein extracting the one or more performance metrics comprises determining, by the processing unit, a medianvalue of each of one or more performance metrics for each network cell based on historical data collected over a predefined time period.

4. The method (500) as claimed in claim 1, wherein the one or more performance metrics comprise at least one of signal quality, connectivity, data throughput, resource utilization, or fault indicators of the plurality of network cells.

5. The method (500) as claimed in claim 1, wherein evaluating the one or more performance metrics of the one or more network cells within each cluster comprising: comparing, by the processing unit (210), the one or more performance metrics of each network cell within each cluster with a corresponding baseline performance threshold; selecting, by the processing unit (210), the one or more network cells when corresponding performance metrics associated with each of the one or more network cells are below the corresponding baseline performance threshold, wherein the selected one or more network cells are identified as the one or more anomaly cells that are exhibiting low performance; and selecting, by the processing unit (210), the one or more network cells when corresponding performance metrics associated with each of the one or more network cells are above the corresponding baseline performance threshold, wherein the selected one or more network cells are identified as the one or more anomaly cells that are exhibiting over-utilization.

6. The method (500) as claimed in claim 1, wherein the morphological data comprises at least one of the inter-site distance between each of the plurality of network cells, a type of the plurality of network cells, and environmental information associated with each network cell.

7. The method (500) as claimed in claim 1, wherein the building data comprises at least one of building type, number of floors, number of flats, building height, building density, rooftop structures, building materials, and building orientation.

8. A system (108) for detecting one or more anomaly cells in a network (106), the system (108) comprising: an extracting unit (208) configured to extract one or more performance metrics associated with a plurality of network cells; a processing unit (210) configured to: determine a cell radius of each of the plurality of network cells based on morphological data associated with the plurality of network cells; categorize the plurality of network cells into one or more cell categories within the determined cell radius based on building data associated with the plurality of network cells; form one or more clusters of the plurality of network cells based on the one or more cell categories, wherein each cluster comprises one or more network cells having a common set of environmental and morphological characteristics; evaluate the one or more performance metrics of the one or more network cells within each cluster; and flag the one or more network cells as the one or more anomaly cells based on the evaluation.

9. The system (108) as claimed in claim 8, wherein the processing unit (210) is configured to generate a report comprising a list of the flagged one or more anomaly cells and the corresponding one or more performance metrics.

10. The system (108) as claimed in claim 8, wherein for extracting the one or more performance metrics, the processing unit (210) is configured to determine a median value of each of one or more performance metrics for each network cell based on historical data collected over a predefined time period.

11. The system (108) as claimed in claim 8, wherein the one or more performance metrics comprise at least one of signal quality, connectivity, data throughput, resource utilization, or fault indicators of the plurality of network cells.

12. The system (108) as claimed in claim 8, wherein for evaluating the one or more performance metrics of the one or more network cells within each cluster the processing unit (210) is configured to: compare the one or more performance metrics of each network cell within each cluster with a corresponding baseline performance threshold; select the one or more network cells when corresponding performance metrics associated with each of the one or more network cells are below the corresponding baseline performance threshold, wherein the selected one or more network cells are identified as the one or more anomaly cells that are exhibiting low performance; and select the one or more network cells when corresponding performance metrics associated with each of the one or more network cells are above the corresponding baseline performance threshold, wherein the selected one or more network cells are identified as the one or more anomaly cells that are exhibiting over-utilization.

13. The system (108) as claimed in claim 8, wherein the morphological data comprises at least one of the inter-site distance between each of the plurality of network cells, a type of the plurality of network cells, and environmental information associated with each network cell.

14. The system (108) as claimed in claim 8, wherein the building data comprises at least one of building type, number of floors, number of flats, building height, building density, rooftop structures, building materials, and building orientation.

15. A computer program product comprising a non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to execute a method (500) for detecting one or more anomaly cells in a network (106), the method (500) comprising: extracting (502), by an extracting unit (208), one or more performance metrics associated with a plurality of network cells; determining (504), by a processing unit (210), a cell radius of each of the plurality of network cells based on morphological data associated with the plurality of network cells; categorizing (506), by the processing unit (210), the plurality of network cells into one or more cell categories within the determined cell radius based on building data associated with the plurality of network cells; forming (508), by the processing unit (210), one or more clusters of the plurality of network cells based on the one or more cell categories, wherein each cluster comprises one or more network cells having a common set of environmental and morphological characteristics; evaluating (510), by the processing unit (210), the one or more performance metrics of the one or more network cells within each cluster; and flagging (512), by the processing unit (210), the one or more network cells as the one or more anomaly cells based on the evaluation.

Images