Editor: S Vishal
Design: Hardhik
To most of us, 5G is simply one of those ubiquitous buzzwords that are heralded as one of the ‘next big things’. To some, it is merely the next iteration of the endless series of Gs — providing us with ever-increasing internet speeds for our insatiable needs. But what is it all really about? Turns out, there’s much more to it than just speed. In this article, we try to deconstruct the what, why and how of 5G, and insti’s pivotal role in its development.
“Am I aud…? Ca..*buzz*…hear me?”
“No, sir. It’s breaking a lot”
“Is it…*garbled voice*…?”
“No sir!”
Poor internet is frustrating. With almost everything going online in the middle of a pandemic, our needs for stable and far-reaching connectivity have become even more pressing. As we move into this completely new era, there’s a whole bunch of people working hard to push the limits of what we can achieve with the next generation of mobile connectivity.
To better understand what’s in store for the future and the work behind it, we reached out to Prof. Radha Krishna Ganti, who has been leading the efforts of the indigenous 5G testbed. The ambitious project was initiated in early 2018 to provide companies in India with a domestic platform to conduct research experiments on the latest generation of mobile communication.[1] Led by IIT Madras, it is a collaborative effort involving eight institutes and organisations across the country.
What do we get out of 5G?
If faster data is what you’re looking for, the next generation in mobile connectivity certainly doesn’t disappoint. Data rates in 5G are expected to be about ten times that of 4G. Prof. Ganti, however, tells us that speed is only a part of the story — there’s a lot more to it than what meets the eye.
5G has been envisioned to encompass more applications like IoT and smart-cities, rather than just browsing and telephony
He tells us that there are three main ‘usage scenarios’ that encapsulate all those requirements that the 5G specification focuses on addressing — mMTC, eMBB, and URLLC, each catering to an aspect of the future of mobile communication.
We now have an exploding number of IoT (internet of things) devices, with unique requirements of their own. Such devices often need to be low-cost, have longer battery life and are frequently placed in remote and closed locations such as basements – this also calls for far-reaching coverage. The density of such devices per unit area in the future is expected to be enormous (~1 million/km2).[3] This scenario creates the need for massive machine-type communication (mMTC).
Enhanced Mobile Broadband (eMBB) is the natural evolution of the 4G network, with higher capacity, enhanced connectivity and higher user mobility for both indoor or outdoor usage. This will ultimately enable truly immersive AR and VR applications and 360o video streaming.[2]
There is a huge buzz about futuristic technologies such as self-driving cars, remote surgeries, and a whole range of autonomous gadgets, which also need to be connected to the network. A car or surgery robot blundering because of a poor connection is nothing short of a recipe for disaster. That’s where the most game-changing additions to 5G’s capabilities, the URLLC (Ultra Reliable Low-Latency Communication) comes in. The name clearly indicates what it focuses on — reliability and low delay (latency). The networks will adapt to changing data in real-time – to guarantee a latency of one millisecond or less. [4]
Why, though, do we need a significant amount of re-engineering through elaborate hardware changes? Why isn’t it sufficient to increase the capacity of the existing 4G network? Prof. Ganti points out that every generation continually evolves and there exists a limit to how much sophistication you can achieve:
There’s only so much for which you can push the specification to. If you start a specification with data rate of 100Mbps, over 5-6 years you can tweak the specifications so that you can get upto 1 Gbps. But if you want 10 Gbps, tweaking the old specification is difficult.
The enormous number of users in India(~1.14 billion wireless subscribers[5]), add a heavy load on base stations (the cellphone towers you see). With the limitless possibilities of the internet, it’s elementary to see why better connectivity could be crucial for India’s future. Prof. Ganti helps us put this in context:
When we did last semester’s courses, a lot of students were using their mobile phones. It’s very interesting that cellular networks are being used for broadband connectivity, especially for education. This is much more relevant in India because we don’t have deep penetration of fiber and optical broadband.
The Testbed
In order to implement such cutting-edge technology, where will the required expertise come from? India has, for long, been dependent on imports for telecom equipment. To experiment with newer technologies, companies — startups in particular, require facilities for equipment testing. That’s where the indigenous 5G testbed comes in. Initiated in early 2018, the primary goal of the testbed was to build an indigenous end-to-end 5G system.
“You should be able to pick up a commercial phone and connect to the testbed. That’s the goal”, says Prof. Ganti. Building a complete communication system from scratch is no easy task. It’s not just about the antennas that you see on the top of cellphone towers (base stations). There’s the circuitry, the baseband unit (BBU) below, and a collection of those connected to the core network, which has its own set of intricate modules. Over and above the functionality, the components also need to be weatherproof — working perfectly, rain or shine.
We felt that in India, we should have the expertise to do that, and be able to help out new, smaller companies, and several other people who may require help.
The testbed development is a collaborative effort of several organisations with strengths in various areas relevant to the project — IIT Madras, IIT Kanpur, IIT Mumbai, IIT Delhi, IIT Hyderabad, IISc, SAMEER and CEWiT. For instance, the focus at IIT Madras is on the complete hardware and software development for the base station. The core network, on the other hand, is being built mainly by CEWiT and IITB. The integrated setup is to be soon tested at IIT Madras.
Several notable developments in 5G pose very different hardware requirements. One of them is the use of phased antenna arrays. 5G operates at two frequency bands, the sub-6GHz band, and the (higher) mmWave (>24GHz) band. Physics tells us that when you have higher frequencies, the signal dies down quicker with distance. This means radiating more power uniformly in all directions would be a costly affair. To save power, you use a phased array (PA), which is an array of antennas that directs the input power towards the user(s) instead, improving efficiency. The PA for the testbed, as seen in the picture, was developed by SAMEER.
With an exploding number of users and complex algorithms, there is much higher computational demand. To meet these challenges, another innovative feature is the offloading of processing from base stations over to the cloud. This introduces a significant difference in the design of the baseband unit (BBU).
What is to come out of it?
How is the testbed useful? One of the uses, of course, is as a facility to test out equipment. If a company has developed its own RRH (remote radio head) or BBU (baseband unit), they will be able to test it out on the network by replacing the existing component with theirs and even build upon the existing design, rather than beginning from scratch.
It also facilitates research. A PhD student, for example, may want to try changing the algorithm that has been implemented in the base station. The testbed has also been a source of active industry-academia collaboration, which is necessary for progress in any domain. In effect, it has been instrumental in accelerating the commercial deployment of 5G.
“The amount of learning that happens during the development (of the testbed) is huge”, says Prof. Ganti. Notably, he also mentions that the students themselves took some of the critical design decisions. The number of different fields that come together in the project is extremely vast – ranging from circuitry, FPGA development, embedded systems, software development, and even thermal engineering.
Even if you go to any company, you will never see this kind of system being built from end-to-end in one shot. Certainly not at the student level. If you do, your hair should have greyed by then!
This aspect has become one of the most important outcomes of the testbed. Several young people who had just graduated from college have been making use of this opportunity to gather valuable experience working on a project of such immense scale. “The government gave us money at the right time, and we’ve been fighting to actually build a system that takes a company a few years and billions of dollars,” he says, “It’s not an easy thing to do.”
Some insti students have also stayed back to work with the testbed after graduation. For those interested in working on the testbed, Prof. Ganti says that the primary requirement is dedication. Due to the importance and scale of the project, there are strict deadlines that they adhere to. “If your module is not delivered, it slows down the entire project”, he points out.
The testbed is likely to begin initial trials in March next year, with most of the hardware complete. “It’s kind of exciting”, he says. “So all of you will be seeing weird towers around campus, if you come back by that time!”, he chuckles. Reaching this stage was a no small accomplishment, given that they had nothing to start out with.
When we started, we had nothing. It was just a blank paper, and we said we’ll build it. How do you build hardware that you’ve never built before? We’ve come a long way.
Prof. Ganti hopes that by the end of this, they would have a complete 5G system entirely designed, developed and manufactured in India. Pointing out the important role that the telecom sector will continue to play, he urges the students to start up and make use of the testbed. “Like it or not, we will always be stuck to our mobiles, and almost all of our data flows through our mobiles. So we should have our own network”, he says.
We should have our own networks that we can trust. There are a lot of things that should fall in place before we become completely self-reliant. This is a right step in that direction.
With exciting new technologies coming up every now and then, achieving independence in developing them is indeed a matter of great pride and economic advantage. The 5G testbed is an example of how a single initiative could provide a multitude of learning opportunities, and most importantly, make a significant leap towards self-reliance.
References :
- Coming Up : 5G testbed at IIT Madras – The New Indian Express
- What is enhanced Mobile Broadband (eMBB)?
- What is mMTC in 5G NR, and how does it impact NB-IoT and LTE-M
- What is URLLC?
- TRAI – Telecom Subscription Data as on 30th June, 2020
Labs of Insti is a series where we attempt to cover the multifarious research activities that take place in insti. Comments and suggestions are always welcome, you can send them to us at [email protected].
Series by: Amrita & Vishal
