Search

EP-4740387-A1 - SYSTEM AND METHOD FOR AUTO-TUNING NETWORK CONFIGURATION

EP4740387A1EP 4740387 A1EP4740387 A1EP 4740387A1EP-4740387-A1

Abstract

The present disclosure relates to a method and a system for configuration auto-tuning in a network system. The method encompasses fetching, by a fetching unit [206] via a service communication proxy - performance Artificial Intelligence (SCP-pAI) [306], the set of network statistics from the SCP [110] at predetermined intervals, wherein the set of network statistics is determined at the SCP [110]; analysing, by an analysis unit [208] using a trained SCP-pAI [306], the determined set of network statistics to determine an optimal network system configuration for prevailing network conditions; comparing, by a comparator [210], the determined optimal network system configuration with a current network system configuration; and in case of a mismatch between the determined optimal network system configuration and the current network system configuration, automatically adjusting, by an adjusting unit [212], the network system configuration based on the comparison.

Inventors

  • BISHT, SANDEEP
  • PANDEY, PRASHANT
  • YADAV, RAVINDRA
  • BHATNAGAR, AAYUSH
  • SINHA, ANURAG
  • Ansari, Ezaj

Assignees

  • Jio Platforms Limited

Dates

Publication Date
20260513
Application Date
20240611

Claims (20)

  1. 1. A method [400] for configuration auto-tuning in a network system, the method comprising: fetching, by a fetching unit [206] via a service communication proxy - performance Artificial Intelligence (SCP-pAI) [306], a set of network statistics from the SCP [110] at predetermined intervals, wherein the set of network statistics is determined at the SCP [110]; analysing, by an analysis unit [208] using a trained SCP-pAI [306], the determined set of network statistics to determine an optimal network system configuration for prevailing network conditions; comparing, by a comparator [210], the determined optimal network system configuration with a current network system configuration; and in case of a mismatch between the determined optimal network system configuration and the current network system configuration, automatically adjusting, by an adjusting unit [212], the network system configuration based on the comparison.
  2. 2. The method as claimed in claim 1, wherein the set of network statistics comprises at least one of Round-trip Time (RTT), internet control message protocol (ICMP) ping packet drop ratio, and available bandwidth.
  3. 3. The method as claimed in claim 1, wherein the method comprises displaying, via a display unit [214], a network system configuration recommendation.
  4. 4. The method as claimed in claim 3, wherein if the network system configuration recommendation is provided, an administrator is alerted, enabling consent-based tuning of the network system configuration.
  5. 5. The method as claimed in claim 3, wherein the network system configuration comprises any or a combination of transmission control protocol (TCP)-related kernel parameters, application-related timeouts, and TCP connection counts.
  6. 6. The method as claimed in claim 4, wherein the network system configuration recommendation provided to the administrator details differences between the current network system configuration and the optimal network system configuration, facilitating informed decision-making.
  7. 7. The method as claimed in claim 1, wherein the method comprises training, by a training unit [202], the SCP-pAI [306] with a set of network conditions against a set of respective network system configurations and corresponding set of Key Performance Indicator (KPI) outcomes.
  8. 8. The method as claimed in claim 7, wherein the set of network conditions corresponds to at least one of values of RTT, packet drop ratios, and different TCP connection counts.
  9. 9. The method as claimed in claim 1, wherein the SCP-pAI [306] uses stored historical set of network statistics in conjunction with current statistics to predict optimal network system configuration.
  10. 10. A system [200] for configuration auto-tuning in a network system, the system comprises: a fetching unit [206] configured to fetch, via a service communication proxy - performance Artificial Intelligence (SCP-pAI) [306], a set of network statistics from the SCP at predetermined intervals, wherein the set of network statistics is determined at the SCP [110]; an analysis unit [208] configured to analyse, using a trained SCP-pAI [306], the determined set of network statistics to determine an optimal network system configuration for prevailing network conditions; a comparator [210] configured to compare the determined optimal network system configuration with a current network system configuration; and in case of a mismatch between the determined optimal network system configuration and the current network system configuration, an adjusting unit [212] is configured to automatically adjust the network system configuration based on the comparison.
  11. 11. The system as claimed in claim 10, wherein the set of network statistics comprises at least one of Round-trip Time (RTT), internet control message protocol (ICMP) ping packet drop ratio, and available bandwidth.
  12. 12. The system as claimed in claim 10, wherein the system comprises a display unit [214] configured to display a network system configuration recommendation.
  13. 13. The system as claimed in claim 12, wherein if the network system configuration recommendation is provided, an administrator is alerted, enabling consent-based tuning of the network system configuration.
  14. 14. The system as claimed in claim 12, wherein the network system configuration comprises any or a combination of transmission control protocol (TCP)-related kernel parameters, application-related timeouts, and TCP connection counts.
  15. 15. The system as claimed in claim 13, wherein the network system configuration recommendation provided to the administrator details differences between the current network system configuration and the optimal network system configuration, facilitating informed decision-making.
  16. 16. The system as claimed in claim 10, wherein the system comprises a training unit [202] configured to train the SCP-pAI [306] with a set of network conditions against a set of respective network system configurations and corresponding set of Key Performance Indicator (KPI) outcomes.
  17. 17. The system as claimed in claim 16, wherein the set of network conditions corresponds to at least one of values of RTT, packet drop ratios, and different TCP connection counts.
  18. 18. The system as claimed in claim 10, wherein the SCP-pAI [306] uses stored historical set of network statistics in conjunction with current statistics to predict optimal network system configuration.
  19. 19. A user equipment (UE) [102] for configuration auto-tuning in a network system, said UE [102] comprising: a receiving unit [602] configured to receive network system configuration recommendation; a display unit [604] configured to display the network system configuration recommendation; the receiving unit [602] configured to receive user input for the displayed network system configuration recommendation; and a transmitting unit [606] configured to transmit the network system configuration recommendation .
  20. 20. The UE [102] as claimed in claim 19, wherein the UE [102] is associated with an administrator.

Description

SYSTEM AND METHOD FOR AUTO-TUNING NETWORK CONFIGURATION FIEED OF THE DISCEOSURE [0001] The present disclosure relates generally to the field of wireless communication systems. More particularly, the present disclosure relates to methods and systems for auto-tuning network configuration. BACKGROUND [0002] 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. [0003] Wireless communication technology has rapidly evolved over the past few decades, with each generation bringing significant improvements and advancements. The first generation of wireless communication technology was based on analog technology and offered only voice services. However, with the advent of the second-generation (2G) technology, digital communication and data services became possible, and text messaging was introduced. 3G technology marked the introduction of high-speed internet access, mobile video calling, and location-based services. The fourth-generation (4G) technology revolutionized wireless communication with faster data speeds, better network coverage, and improved security. Currently, the fifth-generation (5G) technology is being deployed, promising even faster data speeds, low latency, and the ability to connect multiple devices simultaneously. With each generation, wireless communication technology has become more advanced, sophisticated, and capable of delivering more services to its users. [0004] Existing techniques in network management and configuration tuning faces several challenges. One of the primary issues is the lack of adaptability in existing systems, which are not equipped to automatically adjust to sudden changes in network conditions such as increased Round-trip Time (RTT), congestion, and packet drop ratios. Existing systems are typically optimized for normal operating conditions and fail to adapt when anomalies occur. Consequently, network administrators are required to manually re-tune the system configurations to adapt to the new conditions, a process that is time-consuming, labour-intensive, and prone to human error. Additionally, existing systems often lack effective mechanisms for continuously monitoring network performance metrics in real time, limiting their ability to detect and respond to changes promptly. Many existing techniques use static configuration parameters that do not account for the dynamic nature of network conditions, leading to suboptimal performance and inefficiencies. Furthermore, existing systems typically do not employ artificial intelligence to anticipate network changes and adjust configurations proactively. Managing and optimizing network configurations can be complex, especially in large-scale networks, and prior art solutions may not provide intuitive tools or interfaces for simplifying this process. Lastly, effective re-tuning of network configurations often requires in-depth knowledge and expertise, which may not be readily available in all organizations. [0005] The proposed system aims to solve these problems by providing a more adaptive, automated, and efficient approach to network configuration management, minimizing the impact of network disruptions on performance and reducing the reliance on manual intervention. OBJECTS OF THE INVENTION [0006] Some of the objects of the present disclosure, which at least one embodiment disclosed herein satisfies are listed herein below. [0007] It is an object of the present disclosure to provide a method and system for auto-tuning network configuration. [0008] It is another object of the present disclosure to provide a method and system for auto-tuning network configuration that enables the network system to adapt automatically to changes in network conditions such as Round-trip Time (RTT), congestion, and packet drop ratios, ensuring optimal performance even during disruptions. [0009] It is another object of the present disclosure to provide a method and system for auto-tuning network configuration that reduces the need for manual reconfiguration by network administrators, thereby saving time and reducing the potential for human error. [0010] It is another object of the present disclosure to provide a method and system for auto-tuning network configuration that incorporates mechanisms for continuous real-time monitoring of network performance metrics, allowing for timely detection and response to changes in network conditions. [0011] It is another object of the present disclosure to provide a method and system for auto-tuning network configuration that employs dynamic configuration parameters that adjust according to the current network conditions, ensuring mo