NetAware AI

NetAware AI is a realtime network data monitoring and management solution that improves base station electricity consumption.

Comment

NetAwareAI

NetAware AI is a realtime network data monitoring and management solution that improves base station electricity consumption.

Inspiration

The inspiration for your project NetAwareAI stems from the climate change problem. The demand for 5G networks around the world have significantly increased due to the increase in data traffic due to the proliferation of connected devices around the world. With more data being managed results more energy being used to manage the data. We acknowledged the need for a strategic optimization algorithm that can help efficiently allocate energy to certain network nodes in order to make sure sufficient energy is present for effective data processing. We also aim to educate the upcoming generation about the importance of the energy consumption problem and it's impact on climate change.

How we built it?

Our tech stack includes:

  • Next.js 14 for the WebApp hosted on vercel
  • Clerk for a seamless and drop-in OAuth Provider
  • AWS Kinesis, Glue, and SageMaker for ML Pipeline and Real-time Data Ingestion
  • AWS Bedrock with Claude Anthropic FM fine-tuned to offer network insights
  • AWS Web API Gateway interacting with AWS Lambda to allow for initiating MLOps
  • AWS S3 Bucket for centralized data storage of Elastic Container Registry instance of Sage Maker, Real-Time Ingestion from Kinesis, and instance of AWS Bedrock.

Challenges we ran into

Some challenges we faced along the way was figuring out all the AWS Policies and Configurations for each of the AWS Services/technologies.

  • AWS Policies and Configurations for each of the AWS Services/technologies.
  • Limitation to stay within the AWS EcoSystem to allow for real-time data ingestion from telecom providers that can be supported using insights from Bedrock.

Accomplishments that we're proud of

  • Effectively utilized and integrated a collection of AWS products to build a stable product framework.
  • Designed a web application using Next.js that included authorization sign-on (powered by Clerk)
  • Fine-tuned a Machine Learning model from an open-source library to improve accuracy and get different results

Built our product with Responsible AI guidelines in mind:

  1. Fairness : Users are secured using an Authentication service and a secure API gateway
  2. Explainability : Uses Bedrock to create explanations as to why certain decisions were taken and statistics were generated by our model.
  3. Privacy and security : We implement Clerk as our authorization tool in order to protect the access of this data to personnel qualified to analyze and utilize the data.

What we learned

  • Throughout this project, we learned how to develop effective and ergonomic web applications through Next JS (React apps) and combining it with multiple AWS tools and applications.

What's next for NetAware AI

  • The goal of our product in the coming future involves steady function and deployment, but more extensively increasing the geographic range throughout which this product can be used.
Share this project:

Updates