RANGPT

An Intent-based GenAI platform to

Improve Operational Efficiency

What is Operational Efficiency?

The swift and efficient way to manage massive volumes of data for quick experimentation and improvements to service, network, and anomaly detection.

3 Challenges to Operational Efficiency

1

Too Much Data

Telecom networks generate a LOT of data. Operators are painfully aware that this data can be used to better design and better manage their networks and possibly even generate new streams of reveneue. However, today this is easier said than done.

2

Disjointed Analysis

At most operators, it takes two teams to generate insights from data: 1) the first team knows how to ask the right questions, and then 2) the second team writes the appropriate big-data code to answer the questions based on data.

But this process is time-consuming and prevents quick experimentation and testing of ideas.

3

Slow Experimentation

Operators are looking for ways to put their data to use quickly and efficiently. They're also increasingly looking to use a data-driven approach for a) Anomaly detection, b) Network Optimization, and c) Quality of Service Optimizations.

Using network data to gain quick insights and drive improvements via iterative experimentation requires massive improvements in operational efficiency.

RANGPT

Aira's Solution for Operational Efficiency

RANGPT is Aira's intent-based platform that uses GenAI to help network experts deal with any issue affecting the RAN.

Intent-Based RAN Assistant for the Network

Observation

Observation Opportunity

Telco data is diverse and spread across multiple databases, e.g., Performance Metrics, Configuration Settings, and User-Equipment metrics. Accessing & visualizing this data and deriving insights requires specialized query scripts (SQL, Snowflake, etc.). The Network and SW teams must interact.

Observation using RANGPT

RANGPT understands the intent behind natural-language visualization queries and can retrieve appropriate data from the underlying database.

RANGPT then automatically writes scripts that run on the data to generate requested visualizations.

Analysis Opportunity

Telco operators have many pre-written tools and routines that operate on data to derive insights. These scripts are rigid and usually require the network data to be pre-processing before being fed into the scripts. Any ill-formed data (e.g., missing columns, different format than expected, etc.) must be handled manually.

Analysis using RANGPT

RANGPT understands how to use existing scripts, knowing the input arguments that a particualr script takes, the format of the data, etc. RANGPT can translate a user's intent about using a particular script and do all the necessary pre-processing and organization to use the script. It can also use the script/tool as part of a larger script.

Analysis

Control Opportunity

Every EMS/NMS system has its own GUI or script to control various aspects of the network.

Control using RANGPT

RANGPT can map natural language to map control requests to specific NMS API calls.

Automation Opportunity

Today, writing automations for the network can take many months. It's a long, complex task with rigid process. RANGPT reduces this duration to a few days by allowing network engineers to experiment and gain insights from their network data using a conversational, intent-based interface.

Automation using RANGPT

RANGPT understands natural-language visualization intents and can map these intents into the appropriate columns in the underlying data. It then automatically writes scripts that run on the underlying data to generate the requested visualizations.

Automation

RANGPT

Architecture

RANGPT is a software module that runs on the operator's on-prem cloud. For Observation and Analysis of the network, RANGPT only needs northbound and read-only access to data. It can get this either from:
Static / Backward looking database
OR
A dynamic database that is ingesting data via a streaming interfaces (e.g., kafka).
For control and automation, RANGPT needs southbound access to the RAN elements either from:
A SON via NMS
OR
An SMO via a Non-Real Time RIC
“VMware delivers an SMO and RIC platform that enables CSPs and ecosystem partners to harness the full power of generative AI. This RIC and SMO platform offers flexibility and functionality that enables Aira’s RANGPT to transform RAN automation and optimization across the industry. Our work with Aira aligns to our software-defined edge approach—enabling network programmability and optimizing placement of workloads at the edge.”
Padma Sudarsan
Chief of Architecture, Telco, Software-Defined Edge Division, Broadcom

RANGPT Perspectives.

PRESS RELEASE
Aira and Broadcom showcase RANGPT
February 16, 2024
INDUSTRY INSIGHT
Three Motivations for the Modern RAN
February 06, 2024
INDUSTRY INSIGHT
The AI Defined Cellular Network
January 31, 2024
Aira Technologies
20640 Third Street
Suite 200
Saratoga, CA 95070
©2024 Aira™ Technologies | All rights reserved.