FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
(See section 10; rule 13)
TITLE OF THE INVENTION
SYSTEM AND METHOD FOR IDENTIFYING ROOT CAUSE OF USER LEVEL CONGESTION IN
WIRELESS NETWORK
APPLICANT
JIO PLATFORMS LIMITED of Office-101, Saffron, Nr. Centre Point, Panchwati 5 Rasta, Ambawadi, Ahmedabad - 380006, Gujarat, India; Nationality : India
The following specification particularly describes
the invention and the manner in which
it is to be performed

RESERVATION 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 5 dress protection, belonging Jio Platforms Limited (JPL) 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.
10 FIELD OF DISCLOSURE
[0002] The present disclosure generally relates to a wireless
telecommunications network. More particularly, the present disclosure relates to a system and a method for automatically identifying a root cause of user level congestion in a wireless network.
15 BACKGROUND OF DISCLOSURE
[0003] 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 20 to enhance the understanding of the reader with respect to the present disclosure, and not as admissions of prior art.
[0004] Network congestion is a common problem in wireless networks that
leads to reduced quality of service for users. The network congestion occurs when network nodes or links are carrying more data than they can handle, resulting in 25 issues like queueing delay, packet loss, and the blocking of new connections. User data congestion, in particular, can be experienced when transferring data over the
2

control or user plane, and happens when the number of packets being transmitted exceeds the network's packet handling capacity.
[0005] Identifying the root cause of user-level congestion is crucial for
maintaining a high quality of service and user experience. Currently, to identify 5 congestion at the Subscription Permanent Identifier (SUPI) level, which corresponds to the individual user level, conventional systems rely on the generalization of cell-level congestion data. However, in many cases, congestion is only identified reactively when a user reports an issue like slow data speeds or dropped calls. This means congestion often goes undetected until it has already 10 significantly degraded the user experience.
[0006] Therefore, there remains a need for a system and method that can
automatically identify the root cause of user-level congestion in wireless networks in real-time.
SUMMARY
15 [0007] One embodiment of the present subject matter relates to a system for
automatically identifying a root cause of user level congestion in a wireless network. The system may comprise a memory and one or more processors configured to execute instructions stored in the memory. The one or more processors may execute the instructions related to a correlation engine for receiving
20 data related to the wireless network from one or more data consumers (DCs). The one or more data consumers may include a Session Management Function (SMF), an Access and Mobility Management Function (AMF), an Application Function (AF), an xProbe, and a vProbe. The system may comprise a Network data analytics function (NWDAF), which may include the correlation engine and an artificial
25 intelligence/machine learning (AI/ML) engine.
[0008] The correlation engine may correlate the received data on a
timestamp basis to derive correlated data for a particular Subscription Permanent Identifier (SUPI), cell, and time slot. The one or more processors may execute the
3

instructions related to the artificial intelligence/machine learning (AI/ML) engine for training an AI/ML model using the correlated data. The AI/ML engine may then identify the root cause of the user level congestion using the trained AI/ML model, which may determine whether the congestion is due to cell loading or a failure in 5 any core network node. The trained model may also forecast congestion for a number of SUPIs in a particular cell and time window.
[0009] A user interface may provide real-time User Equipment (UE)
mobility analytics visualization of the identified root cause at the SUPI level. The user interface may also communicate closed-loop reporting and actions to data 10 consumers based on the visualization. The AMF and SMF may update their respective policies based on forecasted UE mobility trends included in the closed-loop reporting. One potential benefit of this system is enabling proactive identification and resolution of user-level wireless network congestion issues by leveraging AI/ML techniques and diverse data sources.
15 [0010] In another embodiment of the present subject matter relates to a
method for automatically identifying a root cause of user level congestion in a wireless network. The method may involve receiving by a correlation engine data related to the wireless network from one or more data consumers. The one or more data consumers may include a Session Management Function (SMF), an Access
20 and Mobility Management Function (AMF), an Application Function (AF), an xProbe, and a vProbe.
[0011] The correlation engine may correlate the received data on a
timestamp basis to derive correlated data for a particular Subscription Permanent Identifier (SUPI), cell, and time slot. An artificial intelligence/machine learning 25 (AI/ML) engine may train an AI/ML model using the correlated data. The AI/ML engine may then identify the root cause of the user level congestion using the trained AI/ML model, determining whether the congestion is due to cell loading or a failure in any core network node. Data consumers may subscribe for SUPI-based subscription of User Equipment (UE) mobility analytics, and the correlation engine
4

may collect subscription details from the AMF for a requested cell and time slot, Radio Access Network (RAN) data from the xProbe, and SUPI-based core network data from the vProbe. The trained AI/ML model may forecast a number of SUPIs in a particular cell and time window. A user interface may provide real-time UE 5 mobility analytics visualization of the identified root cause at a SUPI level and may communicate closed-loop reporting with forecasted UE mobility trends to data consumers, enabling them to take appropriate actions. The AMF and SMF may update their policies based on the forecasted trends. One potential advantage of this method is providing a data-driven, automated approach to pinpointing and 10 proactively addressing the underlying causes of user-perceived wireless network congestion.
DEFINITION
[0012] For the purposes of this disclosure, certain technical terms used in
throughout disclosure and claims are defined as follows:
15 [0013] "User level congestion" refers to network congestion experienced by
individual users or user equipment (UE) in a wireless network, as identified by their Subscription Permanent Identifier (SUPI). In an example, the user level congestion may be a condition within a network where individual users experience delays, reduced performance, or service interruptions due to various factors.
20 [0014] "Correlation engine" is a component of the system that receives data
from various sources and correlates the data based on timestamps to create a unified dataset for analysis. The data received by the correlation engine may include user-specific data from an Access and Mobility Management Function (AMF), core network data from vProbe, and Radio Access Network (RAN) data from xProbe.
25 The user-specific data may include data related to sessions and Quality of Service (QoS) parameters. The core network data may include data related to network traffic and sessions. The RAN data or RAN network data may include data related to signal strength, signal quality, and interference level.
5

[0015] "Artificial intelligence/machine learning (AI/ML) engine" is a
component of the system (108) that utilizes AI/ML algorithms to train models and identify patterns or insights from the correlated data.
[0016] "AI/ML model" refers to a mathematical model trained using AI/ML
5 techniques to learn patterns and relationships from input data and make predictions or decisions based on new data.
[0017] "Root cause" refers to the underlying reason or main factor
contributing to the user level congestion in the wireless network.
[0018] "Real-time User Equipment (UE) mobility analytics visualization"
10 refers to the presentation of insights and information related to user level congestion and UE mobility patterns in a visual format, updated continuously as the new data becomes available.
[0019] "Subscription Permanent Identifier (SUPI)" is a globally unique
identifier assigned to each subscriber in a 5G network, used to identify and track 15 individual users or UE for the purpose of analytics and congestion management.
[0020] "Wireless network" refers to a communication network that enables
wireless communication between devices, such as cellular networks (e.g., 5G, LTE), Wi-Fi networks, or any other type of wireless network.
[0021] "Session Management Function (SMF)": A network function
20 responsible for managing sessions and allocating IP addresses to UE.
[0022] "Access and Mobility Management Function (AMF)": A network
function responsible for registration, connection, and mobility management of UE.
[0023] "Application Function (AF)": A network function that interacts with
applications and provides services to UE.
6

[0024] "xProbe": A probe or data collection agent installed in the Radio
Access Network (RAN) to collect data related to radio conditions, signal strength, interference, and other RAN-specific metrics.
[0025] "vProbe": A probe or data collection agent installed in the core
5 network to collect data related to user sessions, bearer contexts, and Quality of Service (QoS) parameters.
[0026] "NWDAF" stands for "Network Data Analytics Function," a
network function introduced in 5G networks to collect and analyze data from various sources and provide insights to other network functions and external 10 consumers. The correlation engine and the AI/ML engine are components of the NWDAF, working together to process data, identify insights, and provide analytics related to user level congestion in the wireless network.
[0027] "Timestamp basis" refers to the process of aligning and correlating
data from different sources based on the timestamps associated with each data point, 15 ensuring a synchronized view of events and metrics across the network.
[0028] "Training" in the context of AI/ML refers to the process of using
historical data to adjust the parameters of an AI/ML model, enabling it to learn patterns and relationships in the data.
[0029] "Cell" refers to a geographical area covered by a specific base station
20 or radio access point in a wireless network.
[0030] "Data consumers" refer to entities, such as network operators,
service providers, or applications, that subscribe to and utilize the insights and analytics generated by the system for various purposes, such as network optimization, troubleshooting, or service enhancement. "Data consumers" can also 25 act as data source to provide the data to network functions. Data Consumers such as SMF/AF subscribe for SUPI based subscription of UE mobility analytics.
OBJECTS OF THE PRESENT DISCLOSURE
7

[0031] Some of the objects of the present disclosure, which at least one
embodiment herein satisfies, are as listed herein below.
[0032] It is an object of the present disclosure to provide a system and a
method for automatically identifying root cause of user level congestion in a 5 wireless network.
[0033] It is an object of the present disclosure to provide a system and a
method that includes a correlation engine and an Artificial Intelligence/Machine Learning (AI/ML) engine to detect user level congestion.
[0034] It is an object of the present disclosure to provide a system and a
10 method that includes the correlation engine to correlate user specific data received from an Access and Mobility Management Function (AMF), core network data received from vProbe, and Radio Access Network (RAN) data received from xProbe on a time stamp basis and identify root cause of the congestion.
[0035] It is an object of the present disclosure to provide a system and a
15 method that includes the AI/ML engine to decide based on cell loading capacity and to perform an auto Root Cause Analysis (RCA) of whether congestion is actually due to cell loading or due to failure in any core network node.
[0036] It is an object of the present disclosure to provide a system and a
method that includes the AI/ML model to identify core network failures across 20 Network Functions (NFs) to identify exact cause for end-to-end UE mobility experience.
BRIEF DESCRIPTION OF DRAWINGS
[0037] The accompanying drawings, which are incorporated herein, and
constitute a part of this disclosure, illustrate exemplary embodiments of the 25 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 instead being placed upon clearly illustrating the
8

principles of the present disclosure. Some drawings may indicate the components using block diagrams and 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 the disclosure of electrical components, electronic components 5 or circuitry commonly used to implement such components.
[0038] FIG. 1 illustrates an exemplary network architecture, in accordance
with embodiments of the present disclosure.
[0039] FIG. 2 illustrates an exemplary block diagram of a system, in
accordance with embodiments of the present disclosure.
10 [0040] FIG. 3 illustrates another exemplary block diagram of the system
architecture, in accordance with embodiments of the present disclosure.
[0041] FIG. 4 illustrates an exemplary computer system in which or with
which embodiments of the present disclosure may be implemented.
[0042] FIG. 5 illustrates a flowchart of a method, in accordance with
15 embodiments of the present disclosure.
[0043] The foregoing shall be more apparent from the following more
detailed description of the disclosure.
LIST OF REFERENCE NUMERALS
100 – Network Architecture
20 102-1, 102-2…102-N – User (s)
104-1, 104-2…104-N – User Equipment (s)
106 –Network
108 –System
9

202 – One or more processor(s)
204 – Memory
206 –Interface(s)
208 – Processing Engine (s) 5 210 – Database
212 – Correlation Engine
214 – AI/ML Engine
216 – Other Engine (s)
300 – block diagram of system architecture 10 410 – External Storage Device
420 – Bus
430 – Main Memory
440 – Read Only Memory
450 – Mass Storage Device 15 460 – Communication Port
470 – Processor
500 – Method
BRIEF DESCRIPTION OF THE INVENTION
[0044] In the following description, for the purposes of explanation, various
20 specific details are set forth in order to provide a thorough understanding of
10

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 5 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 10 drawings.
[0045] 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 15 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.
[0046] Specific details are given in the following description to provide a
thorough understanding of the embodiments. However, it will be understood by one
20 of ordinary skill in the art that the embodiments may be practiced without these specific 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
25 unnecessary detail in order to avoid obscuring the embodiments.
[0047] 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
11

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 5 function, its termination can correspond to a return of the function to the calling function or the main function.
[0048] 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
10 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
15 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.
[0049] Reference throughout this specification to “one embodiment” or “an
embodiment” or “an instance” or “one instance” means that a particular feature,
20 structure, or characteristic described in connection with the embodiment is included in at 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
25 in any suitable manner in one or more embodiments.
[0050] The terminology used herein is to describe particular embodiments
only and is not intended to be limiting 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
12

“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. 5 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 10 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.
15 [0051] As used herein, an “electronic device”, or “portable electronic
device”, or “user device” or “communication device” or “user equipment” or “device” refers to any electrical, electronic, electromechanical, and computing device. The user device is capable of receiving and/or transmitting one or parameters, performing function/s, communicating with other user devices, and
20 transmitting data to the other user devices. The user equipment may have a processor, a display, a memory, a battery, and an input-means such as a hard keypad and/or a soft keypad. The user equipment may be capable of operating on any radio access technology including but not limited to IP-enabled communication, Zig Bee, Bluetooth, Bluetooth Low Energy, Near Field Communication, Z-Wave, Wi-Fi,
25 Wi-Fi direct, etc. For instance, the user equipment may include, but not limited to, a mobile phone, smartphone, virtual reality (VR) devices, augmented reality (AR) devices, laptop, a general-purpose computer, desktop, personal digital assistant, tablet computer, mainframe computer, or any other device as may be obvious to a person skilled in the art for implementation of the features of the present disclosure.
13

[0052] Further, the user device may also comprise a “processor” or
“processing unit” includes processing unit, wherein processor refers to any logic circuitry for processing instructions. The processor may be a general-purpose processor, a special purpose processor, a conventional processor, a digital signal 5 processor, a plurality of microprocessors, one or more microprocessors in association with a DSP core, a controller, a microcontroller, Application Specific Integrated Circuits, Field Programmable Gate Array circuits, any other type of integrated circuits, etc. The processor may perform signal coding data processing, input/output processing, and/or any other functionality that enables the working of 10 the system according to the present disclosure. More specifically, the processor is a hardware processor.
[0053] As portable electronic devices and wireless technologies continue to
improve and grow in popularity, the advancing wireless technologies for data transfer are also expected to evolve and replace the older generations of 15 technologies. In the field of wireless data communications, the dynamic advancement of various generations of cellular technology are also seen. The development, in this respect, has been incremental in the order of second generation (2G), third generation (3G), fourth generation (4G), and now fifth generation (5G), and more such generations are expected to continue in the forthcoming time.
20 [0054] 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
25 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 is to be interpreted merely as illustrative of the disclosure and not as a limitation.
14

[0055] At present, the process of identifying user level congestion in a
wireless network can be challenging and often relies on generalized cell level congestion analysis or manual investigation when users report issues such as slow data speeds or call failures. This reactive approach leads to a suboptimal user 5 experience and inefficient resolution of congestion problems. The present disclosure addresses these challenges by providing a system and method for automatically identifying the root cause of user level congestion in a wireless network using a Network Data Analytics Function (NWDAF) equipped with artificial intelligence and machine learning (AI/ML) capabilities. By correlating 10 user-specific data from various sources and training an AI/ML model to analyze the correlated data, the present disclosure enables the accurate identification of congestion causes and proactive measures to mitigate congestion, ultimately enhancing the user experience and network performance.
[0056] The present disclosure serves the purpose of improving the
15 efficiency and effectiveness of congestion management in wireless networks. The
system and method provided by the present disclosure enable network operators to
leverage the power of data analytics and AI/ML techniques to gain insights into
user-specific congestion issues and their root causes. By automating the process of
data collection, correlation, and analysis, the present disclosure empowers network
20 operators to make informed decisions and take targeted actions to alleviate
congestion, optimize resource allocation, and enhance overall network
performance. The real-time visualization of user level congestion analytics at a
Subscription Permanent Identifier (SUPI) level further enables network operators
to prioritize and address critical congestion issues promptly, leading to improved
25 user satisfaction and reduced churn.
[0057] The present disclosure relates to a system and a method for
automatically identifying the root cause of user level congestion in a wireless network. The system receives data related to the wireless network from various data consumers including a Session Management Function (SMF), an Access and 30 Mobility Management Function (AMF), an Application Function (AF), an xProbe,
15

and a vProbe. A correlation engine within the system correlates the received data on a timestamp basis to derive correlated data. An AI/ML engine then trains an AI/ML model using the correlated data and identifies the root cause of the user level congestion using the trained model. The system provides real-time User Equipment 5 (UE) mobility analytics visualization of the root cause of the user level congestion at a SUPI level through a user interface. By leveraging the correlated data and the trained AI/ML model, the system enables the accurate identification of congestion causes, such as cell loading or core network node failures, and facilitates proactive congestion management in the wireless network.
10 [0058] The various embodiments throughout the disclosure will be
explained in more detail with reference to FIG. 1- FIG. 5.
[0059] FIG. 1 illustrates an exemplary network architecture (100), in
accordance with embodiments of the present disclosure. As illustrated in FIG. 1, one or more computing devices or user equipment (UE) (104-1, 104-2…104-N)
15 may be connected to a proposed system (108) through a network (106). The computing devices (104-1, 104-2…104-N) may be collectively referred to as computing devices (104) or UEs (104) and individually referred to as a computing device (104) or UE (104). One or more users (102-1, 102-2…102-N) may provide requests to the system (108) and may be collectively referred to as users (102) and
20 individually referred to as a user (102).
[0060] The computing device (104) or UE (104) may include, but is not
limited to, mobile devices, laptops, smartphones, virtual reality (VR) devices, augmented reality (AR) devices, general-purpose computers, desktops, personal digital assistants, tablet computers, and mainframe computers. The computing 25 device (104) may also include in-built or externally coupled accessories such as visual aid devices (e.g., cameras), audio aids, microphones, keyboards, touchpads, touch-enabled screens, electronic pens, and other input devices for receiving input from the user (102).
16

[0061] The network (106) may include at least a portion of one or more
networks with nodes that transmit, receive, forward, generate, buffer, store, route, switch, process, or perform a combination thereof on messages, packets, signals, waves, voltage or current levels, or other data. The network (106) may include 5 wireless networks, wired networks, the Internet, intranets, public networks, private networks, packet-switched networks, circuit-switched networks, ad hoc networks, infrastructure networks, Public-Switched Telephone Networks (PSTN), cable networks, cellular networks, satellite networks, fiber optic networks, or a combination thereof.
10 [0062] The system (108) for automatically identifying a root cause of user
level congestion in a wireless network comprises a Network Data Analytics Function (NWDAF) equipped with a correlation engine and an artificial intelligence/machine learning (AI/ML) engine. The correlation engine receives data related to the wireless network from one or more data consumers, including user-15 specific data from an Access and Mobility Management Function (AMF), core network data from vProbe, and Radio Access Network (RAN) data from xProbe. The correlation engine correlates the received data on a timestamp basis to derive correlated data. The AI/ML engine trains an AI/ML model using the correlated data and identifies the root cause of the user level congestion using the trained AI/ML 20 model. The trained AI/ML model determines whether the user level congestion is due to cell loading or a failure in any core network node and can forecast a number of Subscription Permanent Identifiers (SUPIs) likely to face congestion in a particular cell and a particular time window. The system (108) provides real-time User Equipment (UE) mobility analytics visualization of the root cause of the user 25 level congestion at a SUPI level through a user interface.
[0063] The network architecture (100) shown in FIG. 1 is exemplary and
may include fewer, different, differently arranged, or additional functional components in other embodiments. Additionally, one or more components of the network architecture (100) may perform functions described as being performed by 30 other components of the network architecture (100).
17

[0064] FIG. 2 illustrates an example block diagram of the system (108) for
automatically identifying a root cause of user level congestion in a wireless network, in accordance with an embodiment of the present disclosure.
[0065] Referring to FIG. 2, the system (108) includes one or more
5 processors (202) implemented as microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, logic circuitries, and/or devices that process data based on operational instructions. The one or more processors (202) are configured to fetch and execute computer-readable instructions stored in a memory (204) of the system (108). The memory 10 (204) comprises a non-transitory computer-readable storage medium that stores instructions or routines, which can be fetched and executed to create or share data packets over a network service. The memory (204) may include volatile memory, such as random-access memory (RAM), or non-volatile memory, such as erasable programmable read-only memory (EPROM) or flash memory.
15 [0066] The system (108) includes one or more interfaces (206) for data input
and output devices (I/O), storage devices, and other components. The interfaces (206) facilitate communication within the system (108) and provide communication pathways for components such as processing engines (208) and a database (210).
[0067] The processing engines (208) are implemented as a combination of
20 hardware and programming, such as processor-executable instructions stored on a non-transitory machine-readable storage medium and executed by a processing resource. The system (108) may comprise the machine-readable storage medium storing the instructions and the processing resource to execute the instructions, or the machine-readable storage medium may be separate but accessible to the system 25 (108) and the processing resource. In some examples, the processing engines (208) may be implemented by electronic circuitry.
[0068] The processing engines (208) include a correlation engine (212), an
artificial intelligence/machine learning (AI/ML) engine (214), and other engines (216). The other engines (216) typically serve specific functions within the system
18

(108). The other engines (216) may include a data ingestion engine, an input/output engine, and a notification engine. The data ingestion engine may collect and import data from various sources into the system (108). The input/output engine may manage the data flow into and out of the system (108). The notification engine may 5 manage and send notifications or alerts within the system (108).
[0069] The correlation engine (212) receives data related to the wireless
network from one or more data consumers, including user-specific data from an Access and Mobility Management Function (AMF), core network data from vProbe, and Radio Access Network (RAN) data from xProbe. The correlation 10 engine (212) correlates the received data on a timestamp basis to derive correlated data.
[0070] In an example, the correlation engine (212) may receive data from
multiple data sources such as the AMF, xProbe, and vProbe for same timeframe (timeslot) and specific cell. Each data source may provide information about events
15 or measurements that occur within the designated timeframe. In example, the timeframe may be defined based on various parameters such as duration (e.g., hourly, daily), specific time of day, or based on operational requirements (e.g., peak hours). The correlation engine (212) may correlate the received data on a timestamp basis to ensure that the data collected from different sources is synchronized within
20 the same timeframe for accurate correlation.
[0071] The AI/ML engine (214) trains an AI/ML model using the correlated
data. In an implementation, the correlated data is fed into the AI/ML model. The AI/ML model is trained to analyze the correlated data and predict future congestion scenarios. In examples, the training of the AI/ML model involves learning from 25 past instances where congestion occurred and correlating the instances with specific network conditions including cell loading, core network behavior, device specific issues. In an implementation, the AI/ML engine (214) may extract relevant features from the correlated data. These features may include cell loading metrics (from xProbe data), core network performance metrics (from vProbe data), historical
19

trends, and possibly environmental factors (time of day, location, etc.). Further, in an implementation, the AI/ML model is trained using the correlated data (historical data) to recognize patterns that indicate whether congestion at the SUPI level is likely due to high cell loading (indicating RAN congestion), core network failures 5 (indicating issues with network functions like SMF, AF, etc.), and user device-related issues.
[0072] In an implementation, the AI/ML engine (214) identifies the root
cause of the user level congestion using the trained AI/ML model. The trained AI/ML model determines whether the user level congestion is due to cell loading
10 or a failure in any core network node and can forecast a number of Subscription Permanent Identifiers (SUPIs) likely to face congestion in a particular cell and a particular time window. According to an example, the trained AI/ML model can perform real-time analysis to determine the root cause of congestion when a user reports slow data speeds or failed calls. The AI/ML engine (214) may be configured
15 to distinguish between congestion due to cell loading issues (for example, when there is high demand or insufficient capacity at the radio cell level), core network issues (for example, failure or bottlenecks within the network nodes, and device-specific issues (for example, problems related to the user's device or its interaction with the network).
20 [0073] In an example, in real-time scenario, when a user reports issues, such
as slow data speed or call failures. The AI/ML engine (214) may evaluate the current state of the cell and the performance of core network functions. The AI/ML engine (214) compares these with historical data/patterns to determine the root cause of congestion.
25 [0074] Although FIG. 2 shows exemplary components of the system (108),
the system (108) may include fewer, different, differently arranged, or additional functional components in other embodiments. Additionally, one or more components of the system (108) may perform functions described as being performed by other components of the system (108).
20

[0075] In one embodiment a system (108) and method for automatically
identifying a root cause of user level congestion in a wireless network may be disclosed. The system (108) may comprise a memory (204) and one or more processors (202) configured to execute instructions stored in the memory (204). The 5 system (108) may be implemented as a Network Data Analytics Function (NWDAF) within a wireless network architecture, such as a 5G network, to provide advanced analytics and insights into network performance and user experience.
[0076] In an embodiment, the system (108) may include a correlation
engine (212) that receives data related to the wireless network from one or more
10 data consumers. These data consumers may include, but are not limited to, a Session Management Function (SMF), an Access and Mobility Management Function (AMF), an Application Function (AF), an xProbe for collecting Radio Access Network (RAN) data, and a vProbe for collecting core network data. The term “data consumers” can be interchangeably used as term “data sources” as the data
15 consumers receive and transmit data or information form the Session Management Function (SMF), an Access and Mobility Management Function (AMF), an Application Function (AF), an xProbe for collecting Radio Access Network (RAN) data, and a vProbe. The correlation engine (212) may be configured to collect and process data from these various consumers to gain a comprehensive view of the
20 network's performance and user experience.
[0077] The correlation engine (212) may correlate the received data on a
timestamp basis to derive correlated data. This correlation process may involve aligning and synchronizing data from different consumers based on their timestamps, allowing for a unified analysis of network events and user activities. 25 By correlating data from multiple consumers, the system (108) may gain a holistic understanding of the network's behaviour and identify potential causes of user level congestion.
[0078] In an embodiment, the system (108) may further include an artificial
intelligence/machine learning (AI/ML) engine (214). The AI/ML engine (214) may
21

be responsible for training an AI/ML model using the correlated data provided by the correlation engine (212). The AI/ML model may employ various machine learning algorithms, such as deep learning, reinforcement learning, or unsupervised learning, to analyze patterns, anomalies, and trends in the network data. By training 5 on historical data and continuously adapting to new data, the AI/ML model may become increasingly accurate in identifying the root causes of user level congestion.
[0079] Once trained, the AI/ML model may be used by the AI/ML engine
(214) to identify the root cause of user level congestion in real-time. The model may analyze the correlated data, considering factors such as network load, resource 10 utilization, user mobility patterns, and device characteristics, to determine the underlying reasons for congestion experienced by individual users or groups of users. This analysis may enable the system (108) to pinpoint whether the congestion is due to cell loading, a failure in a core network node, or other factors.
[0080] The system (108) may provide a user interface (206) that presents
15 real-time User Equipment (UE) mobility analytics visualization of the root cause of
user level congestion. This visualization may be presented at a Subscription
Permanent Identifier (SUPI) level, allowing network operators and administrators
to drill down into individual user experiences and understand the specific factors
contributing to their congestion. The user interface (206) may display relevant
20 metrics, such as throughput, latency, packet loss, and signal strength, along with
contextual information about the user's location, device, and application usage. In
an example, a geographical map may be displayed that shows areas where
congestion is reported. In some examples, a dashboard showing the performance of
different network slices relevant to SUPIs may be provided. Further, in some
25 examples a timeline graph indicating congestion events reported by SUPIs may be
displayed.
[0081] In an embodiment, the system (108) may support SUPI-based
subscription of UE mobility analytics. The SUPI-based subscription may refer to a method where network services or analytics are subscribed to and managed based
22

on individual subscriber identities. Data consumers, such as network operators, service providers, or third-party applications, may subscribe to receive analytics and insights specific to individual users or groups of users. The correlation engine (212) may be configured to collect subscription details from the AMF for a 5 requested cell and a requested time slot, enabling targeted analysis and reporting based on the subscribed details. UE mobility analytics refers to the process of monitoring, analyzing, and deriving insights from the movement patterns and behavior of user equipments within a network.
[0082] The correlation engine (212) may further collect at least one of RAN
10 data from the xProbe for the requested cell and time slot and SUPI-based core network data from the vProbe. This targeted data collection may allow the system (108) to perform fine-grained analysis and identify congestion issues specific to particular cells, time periods, or user groups. By correlating data from multiple sources at a granular level, the system (108) may provide highly accurate and 15 actionable insights.
[0083] The trained AI/ML model may not only identify the root cause of
user level congestion but also forecast the number of SUPIs likely to face congestion in a particular cell and time window. This predictive capability may enable proactive measures to be taken to mitigate congestion before it occurs. By 20 anticipating potential congestion hotspots and identifying at-risk users, the system (108) may help network operators optimize resource allocation, adjust network parameters, or deploy additional capacity to ensure a seamless user experience.
[0084] The user interface (206) may further facilitate closed-loop reporting
and actions based on the real-time UE mobility analytics visualization. Network 25 operators and administrators may receive automated alerts, notifications, and recommendations for addressing congestion issues. These closed-loop mechanisms may enable quick and effective resolution of congestion problems, minimizing the impact on user experience.
23

[0085] The closed-loop reporting may include forecasted UE mobility
trends, which can be leveraged by data consumers to take one or more appropriate actions. For example, the AMF and SMF may update their respective policies based on the predicted mobility patterns to ensure a seamless user experience. This may 5 involve dynamically adjusting resource allocation, optimizing handover procedures, or implementing load balancing techniques to distribute traffic evenly across the network. The forecasted UE mobility trends refer to predictions or projections of how UEs are expected to move within a cellular network over a specified period. This forecasting is typically based on historical data analysis, 10 predictive analytics, and machine learning models trained on past mobility patterns and other relevant network data.
[0086] In an embodiment, the system (108) may be integrated with existing
network management and orchestration frameworks, such as Self-Organizing Networks (SON) or Network Functions Virtualization (NFV) platforms. This 15 integration may allow for automated activation of congestion mitigation measures based on the insights provided by the AI/ML model. For instance, the system (108) may trigger the deployment of additional virtual network functions (VNFs) or the reconfiguration of radio resources to alleviate congestion in specific areas of the network.
20 [0087] The system (108) may also provide APIs or integration points for
third-party applications and services to consume the UE mobility analytics and insights. This may enable the development of solutions and use cases that leverage the real-time congestion information to optimize application performance, enhance user experience, or offer personalized services based on network conditions.
25 [0088] The AI/ML model used by the system (108) may be continuously
trained and updated based on new data and feedback from the network. This ongoing learning process may allow the model to adapt to changing network conditions, user behaviour, and traffic patterns over time. The system (108) may
24

employ techniques such as transfer learning or incremental learning to efficiently update the model without requiring a complete retraining process.
[0089] In addition to identifying the root cause of user level congestion, the
system (108) may also provide insights into the overall health and performance of 5 the wireless network. It may monitor key performance indicators (KPIs) such as throughput, latency, packet loss, and resource utilization across different network segments and layers. This holistic view of network performance may enable operators to identify bottlenecks, optimize network configurations, and plan capacity expansions based on data-driven insights.
10 [0090] The system (108) may also incorporate security measures to protect
sensitive user data and prevent unauthorized access to the UE mobility analytics. This may include encryption of data in transit and at rest, secure authentication and authorization mechanisms, and compliance with relevant privacy regulations such as General Data Protection Regulation (GDPR) or California Consumer Privacy
15 Act (CCPA). The system (108) may also implement data anonymization techniques to ensure that individual user identities are protected while still allowing for meaningful analysis and insights.
[0091] In an embodiment, the system (108) may provide a role-based access
control (RBAC) mechanism to govern access to the UE mobility analytics and
20 visualization. The RBAC is a mechanism that restricts system access based on the roles of individual users. According to the RBAC mechanism, only individuals with the necessary permissions can access specific resources or perform certain actions within a system. For example, different user roles, such as network administrators, customer support representatives, or marketing analysts, may have different levels
25 of access and permissions based on their responsibilities and data requirements. This may ensure that sensitive information is only accessible to authorized personnel and that data privacy is maintained.
[0092] The user interface (206) of the system (108) may provide intuitive
and interactive visualizations of the UE mobility analytics. It may include
25

dashboard-style views that display key metrics, trends, and alerts in real-time. Users may be able to drill down into specific regions, cells, or time periods to gain more detailed insights. The user interface (206) may also provide data export and reporting capabilities, allowing users to generate custom reports or integrate the 5 analytics data with other business intelligence tools.
[0093] The system (108) may also support the integration of external data
sources to enrich the UE mobility analytics. For example, it may incorporate weather data, event data, or social media data to provide additional context and insights into user behaviour and network performance. By combining multiple data 10 sources, the system (108) may enable more comprehensive and accurate analysis of user level congestion and its root causes.
[0094] FIG. 3 illustrates an example architecture (300) of the system (108)
for automatically identifying a root cause of user level congestion in a wireless network, in accordance with an embodiment of the present disclosure. The 15 architecture (300) may depict the interactions and data flow between various components of the system (108) and external entities.
[0095] In step 1, one or more data consumers, such as a Session
Management Function (SMF) or an Application Function (AF), may subscribe for SUPI-based subscription of UE mobility analytics. This subscription may indicate 20 their interest in receiving analytics and insights related to specific users or groups of users identified by their Subscription Permanent Identifiers (SUPIs). The subscription details may include the desired level of granularity, frequency of updates, and specific metrics or KPIs of interest.
[0096] In step 2, a correlation engine (212) of the system (108) may collect
25 SUPI-based subscription information from the Access and Mobility Management Function (AMF) for the requested cell and requested time slot. The terminology the correlation engine (212) and a NWDAF back-end (BE) module may be used interchangeably. The information may include details about the users' devices, their location, and their associated network resources. Similarly, the correlation engine
26

(212) may also collect Radio Access Network (RAN) data from the xProbe for that particular cell and time slot. The xProbe may provide data related to radio conditions, signal strength, interference, and other RAN-specific metrics. Additionally, the correlation engine (212) may collect SUPI-based core network 5 data from the vProbe. The SUPI-based core network data may refer to the network data that is specific to an individual subscriber. The SUPI-based core network data may provide insights into the network usage and behavior of the subscriber. Examples of SUPI-based core network data may include information about the users' sessions, bearer contexts, and quality of service (QoS) parameters.
10 [0097] In step 3, the correlation engine (212) may perform analytics by
correlating the data received from the AMF, xProbe, and vProbe for at least one of a particular SUPI, cell, and time slot. This correlation may involve aligning the data based on timestamps, aggregating metrics, and identifying patterns or anomalies. The correlation engine (212) may then provide the analyzed data to its AI/ML
15 model of an AI/ML engine (214) for training purposes. The AI/ML model may use this historical data to learn patterns and relationships between different network parameters and user experience. Based on this training, the AI/ML model may develop the capability to forecast the number of SUPIs that are likely to face congestion in a particular cell and time window in the future.
20 [0098] In step 4, the correlation engine (212) may provide real-time
analytics visualization in the user interface (UI) for the exact reason of congestion at the SUPI level (user level). SUPI level refers to analyzing, monitoring, and managing network activities at the level of individual subscriber identifiers, enabling detailed insights into user behavior and network performance. The
25 visualization may indicate whether the congestion is due to issues with the UE device itself, congestion happening at the cell level, or problems in the core 5G network. This granular insight may help network operators and administrators quickly identify the root cause of congestion and take appropriate remedial actions.
27

[0099] In step 5, through the UI, closed-loop reporting and actions may be
communicated to the data consumers. The closed-loop reporting and actions refer to a process where information gathered from the system (108) is used to take immediate or near-immediate corrective actions or optimizations. For example, the 5 system (108) may provide closed-loop reporting of the forecasted UE mobility trends, enabling the consumers to take proactive measures. The AMF and SMF may leverage this information to update their respective policies and optimize network resources to ensure a seamless user experience. This closed-loop feedback mechanism may allow for continuous improvement and adaptation of the network 10 based on real-time analytics and predictions.
[00100] The architecture (300) may enable the system (108) to collect and
process data from various sources, including the AMF, xProbe, and vProbe, to gain a comprehensive view of the network and user experience. By correlating this data at a granular level, the system (108) may accurately identify the root cause of user 15 level congestion and provide actionable insights to data consumers.
[00101] The AI/ML model employed by the system (108) may continuously
learn and improve its forecasting capabilities based on the historical data provided by the correlation engine (212). This ongoing learning process may allow the model to adapt to changing network conditions and user behaviours over time, ensuring 20 accurate predictions and proactive congestion management.
[00102] The user interface (UI) of the system (108) may provide intuitive
visualizations and dashboards that present the real-time analytics data in a meaningful and actionable format. Network operators and administrators may use the UI to drill down into specific SUPIs, cells, or time slots to investigate congestion 25 issues and identify the underlying causes. The UI may also facilitate the configuration of closed-loop reporting and actions, enabling data consumers to specify their desired outcomes and receive automated recommendations or triggers based on the analytics insights.
28

[00103] The closed-loop reporting and actions communicated through the UI
may enable a proactive and automated approach to congestion management. For example, if the AI/ML model predicts that a particular cell is likely to experience congestion in the near future, the system (108) may automatically trigger actions 5 such as resource allocation adjustments, load balancing, or the activation of additional network capacity. These proactive measures may help prevent or mitigate congestion before it impacts the user experience.
[00104] Furthermore, the closed-loop reporting may provide valuable
feedback to the data consumers, such as the AMF and SMF, enabling them to 10 optimize their policies and configurations based on the forecasted UE mobility trends. This feedback loop may ensure that the network remains responsive and adaptive to changing user demands and network conditions.
[00105] In another embodiment, the present disclosure also provides a
computer program product comprising a non-transitory computer-readable medium
15 having a computer-readable program stored thereon. The computer-readable program, when executed by one or more processors (202), causes the processor(s) (202) to perform operations for automatically identifying a root cause of user level congestion in a wireless network. The operations include receiving, by a correlation engine (212), data related to the wireless network from one or more data consumers,
20 and correlating, by the correlation engine (212), the received data on a timestamp basis to derive correlated data. An artificial intelligence/machine learning (AI/ML) engine (214) then trains an AI/ML model using the correlated data and identifies a root cause of user level congestion in the wireless network using the trained AI/ML model. Finally, a user interface (206) provides real-time analytics visualization of
25 the root cause of the user level congestion at a Subscription Permanent Identifier (SUPI) level. This computer program product enables the implementation of the system and method for automatically identifying the root cause of user level congestion in a wireless network on various computing devices and platforms.
29

[00106] FIG. 4 illustrates an example computer system (400) in which or
with which the embodiments of the system (108) for automatically identifying a root cause of user level congestion in a wireless network may be implemented.
[00107] As shown in FIG. 4, the computer system (400) may include an
5 external storage device (410), a bus (420), a main memory (430), a read-only memory (440), a mass storage device (450), communication port(s) (460), and a processor (470). The computer system (400) may include multiple processors and communication ports to support the functionality of the system (108). The processor (470) may include various modules associated with the embodiments of the system 10 (108), such as the correlation engine (212), the AI/ML engine (214), and other components.
[00108] The communication port(s) (460) may include various interfaces for
connecting the computer system (400) to external devices and networks. These ports may include, but are not limited to, an RS-232 port for use with a modem-15 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 choice of communication port(s) (460) may depend on the network to which the computer system (400) connects, such as a Local Area Network (LAN), Wide Area Network (WAN), or any other network.
20 [00109] In an embodiment, the main memory (430) may be Random Access
Memory (RAM) or any other dynamic storage device commonly known in the art. The main memory (430) may be used to store data and instructions that are actively being processed by the processor (470). The read-only memory (440) may be any static storage device(s), such as a Programmable Read Only Memory (PROM) chip,
25 used for storing static information, such as start-up or basic input/output system (BIOS) instructions for the processor (470).
[00110] The mass storage device (450) may be used to store large amounts
of data and instructions for the system (108). It may include current or future mass storage solutions, such as Parallel Advanced Technology Attachment (PATA) or
30

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).
[00111] In an embodiment, the bus (420) may communicatively couple the
5 processor(s) (470) with the other memory, storage, and communication blocks within the computer system (400). The bus (420) may be implemented using various technologies, such as a Peripheral Component Interconnect (PCI)/PCI Extended (PCI-X) bus, Small Computer System Interface (SCSI), Universal Serial Bus (USB), or the like. The bus (420) may also include other buses, such as a front 10 side bus (FSB), which connects the processor (470) to the computer system (400).
[00112] The computer system (400) may provide the necessary computing
resources and infrastructure to support the functionality of the system (108) for automatically identifying the root cause of user level congestion in a wireless network. The processor (470) may execute the instructions stored in the memory 15 (430) and mass storage device (450) to perform the correlation, AI/ML model training, and analytics visualization tasks described in the embodiments of the system (108).
[00113] The communication port(s) (460) may enable the computer system
(400) to connect to external data sources/consumers, such as the AMF, xProbe, and 20 vProbe, to collect the necessary data for analysis. The communication port(s) (460) may also facilitate the transmission of analytics insights, closed-loop reporting, and actions to the data consumers.
[00114] The main memory (430) and mass storage device (450) may store
the data collected from various sources, as well as the intermediate results and 25 outputs generated by the system (108). The read-only memory (440) may store the static information required for the initial setup and configuration of the system (108).
31

[00115] Optionally, operator and administrative interfaces, e.g., a display,
keyboard, joystick, and a cursor control device, may also be coupled to the bus (420) to support direct operator interaction with the computer system (400). Other operator and administrative interfaces may be provided through network 5 connections connected through the communication port (460). Components described above are meant only to exemplify various possibilities. In no way should the aforementioned exemplary computer system (400) limit the scope of the present disclosure.
[00116] FIG. 5 illustrates a flowchart of a method for automatically
10 identifying a root cause of user level congestion in a wireless network, in accordance with an embodiment of the present disclosure. The method may be performed by a system (108) comprising a correlation engine (212), an artificial intelligence/machine learning (AI/ML) engine (214), and a user interface (206).
[00117] In step 502, the correlation engine (212) receives data related to the
15 wireless network from one or more data consumers. The data consumers may include, but are not limited to, a Session Management Function (SMF), an Access and Mobility Management Function (AMF), an Application Function (AF), an xProbe for collecting Radio Access Network (RAN) data, and a vProbe for collecting core network data. One or more data consumers may subscribe for SUPI-20 based subscription of User Equipment (UE) mobility analytics. The data consumers may include entities such as network operators, service providers, or third-party applications that are interested in receiving analytics and insights related to specific users or groups of users identified by their Subscription Permanent Identifiers (SUPIs). The correlation engine (212) collects subscription details from the AMF 25 for a requested cell and a requested time slot. These subscription details may include information about the users' devices, their location, and the associated network resources. Additionally, the correlation engine (212) collects RAN data from the xProbe for the requested cell and time slot, as well as SUPI-based core network data from the vProbe.
32

[00118] In step 504, the correlation engine (212) correlates the received data
on a timestamp basis to derive correlated data. The correlation process may involve aligning the data based on timestamps, aggregating metrics, and identifying patterns or anomalies. The correlated data may be derived for a particular SUPI, cell, and 5 time slot, providing a granular view of the network and user experience.
[00119] In step 506, the AI/ML engine (214) trains an AI/ML model using
the correlated data. The training process may involve feeding the correlated data into the AI/ML model and adjusting the model's parameters to optimize its performance. The AI/ML model may learn patterns and relationships between 10 different network parameters and user experience, enabling it to identify the root cause of user level congestion.
[00120] In step 508, the AI/ML engine (214) identifies the root cause of the
user level congestion using the trained AI/ML model. The AI/ML model may analyze the correlated data and determine whether the congestion is due to cell 15 loading, a failure in any core network node, or other factors. The model may also forecast a number of SUPIs likely to face congestion in a particular cell and time window.
[00121] In step 510, the user interface (206) provides real-time UE mobility
analytics visualization of the root cause of the user level congestion at a SUPI level.
20 The visualization may present the analytics insights in a meaningful and actionable format, enabling network operators and administrators to quickly identify and address congestion issues. The user interface (206) may also communicate closed-loop reporting and actions to the data consumers based on the real-time analytics visualization. The closed-loop reporting, which includes forecasted UE mobility
25 trends, enables the data consumers to take appropriate actions. For example, the AMF and SMF may update their respective policies based on the forecasted mobility trends to ensure a seamless user experience. This may involve optimizing resource allocation, adjusting network parameters, or implementing load balancing techniques.
33

[00122] The method may continuously loop back to step 502 to receive
updated data from the data consumers and repeat the process of correlation, AI/ML model training, root cause identification, and analytics visualization. This iterative approach allows the system (108) to adapt to changing network conditions and user 5 behaviours over time, providing ongoing insights and recommendations for effective congestion management.
[00123] In another embodiment, the present disclosure also describes a user
equipment (104) communicatively coupled to a system (108) for automatically identifying a root cause of user level congestion in a wireless network via a network
10 (106). The communicative coupling involves transmitting, by the user equipment (104), data related to the user equipment (104) to the system (108). The system (108) comprises a memory (204) and one or more processors (202) configured to execute instructions stored in the memory (204) to perform steps of a method for automatically identifying a root cause of user level congestion in the wireless
15 network. The method includes receiving data related to the wireless network from one or more data consumers, correlating the received data on a timestamp basis to derive correlated data, training an AI/ML model using the correlated data, identifying the root cause of user level congestion using the trained AI/ML model, and providing real-time User Equipment (UE) mobility analytics visualization of
20 the root cause of the user level congestion at a Subscription Permanent Identifier (SUPI) level. This user equipment (104) and its communicative coupling to the system (108) enable the collection of user-specific data and the application of the system and method for automatically identifying the root cause of user level congestion experienced by the user equipment (104) in the wireless network.
25 [00124] 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
30 present disclosure are not limited to the order specifically described above unless
34

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 5 medium storing a program for executing the method according to the present disclosure.
[00125] 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 10 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.
15 ADVANTAGES OF THE PRESENT DISCLOSURE
[00126] The system and method of the present disclosure enable the
automatic detection of a root cause of user level congestion in a wireless network. By identifying the underlying factors contributing to congestion, network operators can take targeted actions to mitigate congestion and improve user experience.
20 [00127] The present disclosure leverages a Network Data Analytics Function
(NWDAF) equipped with a correlation engine and an Artificial Intelligence/Machine Learning (AI/ML) model to detect user level congestion. The NWDAF's advanced analytics capabilities, combined with the power of AI/ML, enable accurate and timely identification of congestion issues at a granular user
25 level.
[00128] The correlation engine correlates user-specific data received from an
Access and Mobility Management Function (AMF), core network data received from vProbe, and Radio Access Network (RAN) data received from xProbe on a
35

timestamp basis. This correlation of data from multiple sources provides a comprehensive view of the network and user experience, enabling the identification of the root cause of congestion.
[00129] The AI/ML model in the present disclosure decides, based on cell
5 loading capacity, to perform an auto Root Cause Analysis (RCA) of whether congestion is actually due to cell loading or due to failure in any core network node. This intelligent analysis helps pinpoint the specific factors contributing to congestion, whether it is related to radio access network capacity or core network issues.
36

We Claim:
1. A system (108) for automatically identifying a root cause of user level
congestion in a wireless network, the system (108) comprising:
a memory (204); and
5 one or more processors (202) configured to execute instructions
stored in the memory (204) to:
receive, by a correlation engine (212), data related to the wireless network from one or more data consumers;
correlate, by the correlation engine (212), the received data
10 on a timestamp basis to derive correlated data;
train, by an artificial intelligence/machine learning (AI/ML) engine (214), an AI/ML model using the correlated data;
identify, by the AI/ML engine (214), the root cause of the user level congestion using the trained AI/ML model; and
15 provide, by a user interface (206), a visualization of the root
cause of the user level congestion.
2. The system (108) of claim 1, wherein the one or more data consumers
include a Session Management Function (SMF), an Access and Mobility
Management Function (AMF), an Application Function (AF), an xProbe,
20 and a vProbe.
3. The system (108) of claim 1, wherein the system (108) comprises a Network
data analytics function (NWDAF), wherein the Network data analytics
function (NWDAF) comprises the correlation engine (212) and the artificial
intelligence/machine learning (AI/ML) engine (214).
37

4. The system (108) of claim 1, wherein the one or more data consumers are
subscribed for SUPI-based subscription of User Equipment (UE) mobility
analytics.
5. The system (108) of claim 2, wherein the correlation engine (212) is further
5 configured to collect atleast one of: Radio Access Network (RAN) data from
the xProbe and collect SUPI-based core network data from the vProbe.
6. The system (108) of claim 1, wherein the correlation engine (212) is
configured to correlate the received data for atleast one of: a particular
SUPI, cell, and time slot.
10 7. The system (108) of claim 1, wherein the trained AI/ML model is
configured to determine whether the user level congestion is due to cell loading or a failure in any core network node.
8. The system (108) of claim 1, wherein the trained AI/ML model is
configured to forecast congestion for a number of SUPIs in a particular cell
15 and a particular time window.
9. The system (108) of claim 1, wherein the user interface (206) is further
configured to communicate closed-loop reporting and actions to the one or
more data consumers.
10. The system (108) of claim 9, wherein the closed-loop reporting including
20 forecasted UE mobility trends that enable the one or more data consumers
to take one or more actions.
11. The system (108) of claim 10, wherein the one or more data consumers
update their respective policies based on the forecasted UE mobility trends.
12. A method for automatically identifying a root cause of user level congestion
25 in a wireless network, the method comprising:
38

receiving (502), by a correlation engine (212), data related to the wireless network from one or more data consumers;
correlating (504), by the correlation engine (212), the received data on a timestamp basis to derive correlated data;
5 training (506), by an artificial intelligence/machine learning
(AI/ML) engine (214), an AI/ML model using the correlated data;
identifying (508), by the AI/ML engine (214), the root cause of the user level congestion using the trained AI/ML model; and
providing (510), by a user interface (206), a visualization of the root
10 cause of the user level congestion.
13. The method of claim 12, wherein the one or more data consumers include a Session Management Function (SMF), an Access and Mobility Management Function (AMF), an Application Function (AF), an xProbe, and a vProbe.
15 14. The method of claim 12, further comprising: subscribing, by the one or more
data consumers, for SUPI-based subscription of User Equipment (UE) mobility analytics.
15. The method of claim 13, further comprising: collecting, by the correlation
engine (212), atleast one of: Radio Access Network (RAN) data from the
20 xProbe ; and collecting, by the correlation engine (212), SUPI-based core
network data from the vProbe.
16. The method of claim 12, wherein correlating the received data on the
timestamp basis comprises deriving the correlated data for atleast one of: a
particular SUPI, cell, and time slot.
25 17. The method of claim 12, wherein identifying the root cause of the user level
congestion using the trained AI/ML model comprises determining whether
39

the user level congestion is due to cell loading or a failure in any core network node.
18. The method of claim 12, further comprising: forecasting, by the trained
AI/ML model, congestion for a number of SUPIs in a particular cell and a
5 particular time window.
19. The method of claim 12, further comprising: communicating, via the user
interface (206), closed-loop reporting and actions to the one or more data
consumers.
20. The method of claim 19, wherein the closed-loop reporting includes
10 forecasted UE mobility trends that enable the one or more data consumers
to take one or more actions.
21. The method of claim 20, further comprising: updating, by the one or more
data consumers, their respective policies based on the forecasted UE
mobility trends.
15 22. A user equipment/ computing device (104) communicatively coupled to a
system (108), the communicative coupling comprising: transmitting, by the user equipment (104), data related to the user equipment (104) to the system (108), wherein the system (108) comprises: a memory (204); and one or more processors (202) configured to execute instructions stored in the
20 memory (204) to perform steps of a method for automatically identifying a
root cause of user level congestion in a wireless network as claimed in claim 11.