Skip to content
Best Place To BuildBest Place To Build

How does 5G work? | A RARE look inside the 5G testbed facility @IITM | BP2B Labcast Ep 1

India’s 5G revolution is unfolding faster than ever, and the IIT Madras 5G Test Bed is at the centre of it. In this special Labcast series, we take you inside various such labs at the forefront of cutting-edge research. This first episode leads us right to the place where India’s first 5G phone call was made—the 5G Testbed Facility at IIT Madras. From advanced signal processing algorithms and massive MIMO setups to cutting-edge 5G transmission experiments, this lab does it all. Check out their official website here: http://www.5gtbiitm.in/ As we walk through the antennas, servers, SDRs, and network-monitoring tools, Jeeva Keshav S (we sincerely apologise for the typing error in the video), one of the researchers behind the testbed, explains how India is building indigenous telecom capability and why IIT Madras has become a critical hub for scalable, secure wireless innovation. Whether you're a tech enthusiast, an engineering student, or someone curious about how India’s 5G backbone actually works, this episode is your inside pass to the future of communication. What You’ll Learn: * The engineering behind India’s first 5G phone call and large-scale 5G transmission * Inside the antenna arrays, SDRs, and signal processing workflows used in 5G * How IIT Madras became a national hub for wireless technology innovation * The real challenges of building and testing 5G systems at scale * How India is strengthening telecom R&D and next-gen network capability Chapters: 00:23 Welcome to the Best Place to Build - Labcast 00:50 Introduction to the 5G testbed at IIT Madras 01:37 Infrastructure involved in building the indigenous 5G 05:30 Why is 5G so much faster than 4G? 08:25 Glimpse at the 5G core network system 09:40 Story of India’s 1st official 5G phone call 12:34 The teams behind the 5G innovation 14:30 How is AI/ML used in the 5G testbed facility? 17:00 Closing thoughts

Nov 27, 202517mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Inside IIT Madras’ indigenous 5G testbed: radios, core, future

  1. The video breaks down the 5G signal chain from phone-to-antenna (radio unit) to baseband processing and into a software-based 5G core that authenticates users, manages mobility, and routes traffic.
  2. It explains why 5G can be faster than 4G by using many more antenna elements (massive MIMO) and beamforming, while highlighting the calibration and synchronization challenges this creates.
  3. The IIT Madras testbed implements open interface standards (O-RAN 7.2 split) to define how processing is divided between the radio and baseband so multi-vendor components can interoperate.
  4. Researchers recount the April 2022 milestone of India’s first official 5G phone call on a completely indigenous stack, and the subsequent year of work to harden it for field reliability.
  5. The discussion connects current 5G work (mobility optimization, receiver/transmitter algorithms) to forward-looking 6G themes like AI/ML-driven positioning and reducing reference-signal overhead while scaling to massive IoT connectivity.

IDEAS WORTH REMEMBERING

5 ideas

5G speed gains are tightly linked to more antennas and smarter directionality.

Moving from a handful of antenna elements in typical 4G deployments to 16/32/64-element arrays in 5G enables beamforming and higher spatial multiplexing, concentrating energy toward users instead of broadcasting everywhere.

More antenna elements increase performance, but they also multiply calibration complexity.

Each antenna chain must be fine-tuned without disrupting live traffic, and calibration frequency depends on environment, temperature, aging, and how fast channel conditions change.

Channel estimation is a continuous, ultra-fast process in real networks.

The lab describes channel estimation happening about every 0.5 ms to track rapid variations—critical for reliable connectivity when users are moving (train/auto) and when interference/noise is present.

Precise timing sync is foundational to dense cellular deployments.

A GNSS/GPS antenna provides a common time reference so neighboring base stations remain synchronized, reducing interference and enabling coordinated transmission/reception behavior.

Open standards like O-RAN enable mixing vendors by defining processing splits.

By implementing the O-RAN 7.2 split, part of the PHY processing stays near the radio and the rest in the baseband unit, allowing interoperability while balancing compute, fronthaul needs, and performance.

WORDS WORTH SAVING

5 quotes

This happens with a periodicity of about point five milliseconds.

— Jeeva (IITM lab)

Till 4G, this core network was typically deployed in a custom hardware... But for 5G, it is completely a software-based deployment.

— Jeeva (IITM lab)

It was only in April 2022 that we made the first official 5G call.

— Jeeva (IITM lab)

We have implemented something called as O-RAN seven point two split.

— Jeeva (IITM lab)

So one issue is there is a lot of overhead on the reference signals.

— Jeeva (IITM lab)

5G transmission chain (RU → BBU → 5G core)Massive MIMO and beamformingAntenna calibration and channel estimationGNSS/GPS timing synchronization between base stationsO-RAN interoperability and 7.2 functional splitSoftware-defined 5G core network scalingIndigenous 5G milestone and testbed timelineAI/ML for 6G (centimeter-level positioning)6G challenges: reference-signal overhead, IoT scale

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome