Best Place To BuildHow IIT Professors created India's own 5G Tech | Prof. Radha Krishna Ganti, EE, IITM on BP2B S2 Ep.6
CHAPTERS
Why IIT Madras is “the best place to build” + setting up the conversation
The host frames the podcast’s theme—builders at IIT Madras—and introduces the episode’s focus on wireless communication and indigenous 5G. The conversation sets expectations: understanding fundamentals first, then moving into India’s 5G testbed story and what it takes to build real systems.
- •Podcast premise: meet builders at IIT Madras and learn what it takes to build
- •Host’s perception gap: “theoretical campus” vs visible real-world building
- •Episode focus: wireless communications and India’s 5G journey
- •Roadmap: fundamentals → 5G components/testbed → scaling/standards → careers
Meet Prof. Radhakrishna Ganti: wireless researcher and Bhatnagar awardee
Prof. Ganti discusses winning the Shanti Swarup Bhatnagar Prize, including the delayed announcement and surprise factor. The host positions his work in the broader IIT Madras communications lineage and India’s telecom ambitions.
- •Professor’s background: EE faculty at IIT Madras, wireless communication
- •Bhatnagar Prize: surprise email, delay due to election timing
- •Recognition in academia: fewer “headline” awards, making it meaningful
- •Sets credibility for later discussion on standardization and national projects
Wireless communication basics: what it is and why it’s hard
A clear primer on wireless: sending data reliably without wires amid noise, interference, and fast-changing channels. Prof. Ganti uses everyday analogies—parties, stadiums, trains—to explain why reliability and capacity are difficult at scale.
- •Wireless examples: cellular (4G/5G), Wi‑Fi, Bluetooth, radio, satellite TV
- •Core challenge: reliable data transfer over a noisy, distortion-prone channel
- •Mobility complicates channels (e.g., high-speed trains streaming video)
- •Interference: many devices “talking” simultaneously like voices at a party
- •Real-world pain point: dense events where networks fail (concerts/stadiums)
What 5G actually consists of: base stations, radios, and the core network
Prof. Ganti breaks 5G into two big parts: the phone/device side and the network side. He details how base stations convert RF signals into digital packets and how the core network aggregates massive traffic, emphasizing the engineering complexity and global concentration of vendors.
- •Two halves: 5G devices + 5G network infrastructure
- •Base station anatomy: antennas, RF chips, signal processing, heavy software
- •Core network: regional aggregation, enormous data handling requirements
- •Complexity barrier: only a handful of global companies build end-to-end (Ericsson, Nokia, Samsung, Huawei, ZTE)
- •R&D reality: decades of iteration and billions in investment behind commercial gear
Building indigenous 5G: the multi-institute testbed vision
India’s lack of homegrown vendors motivates an ambitious academic consortium effort. Eight institutions split responsibilities, with IIT Madras leading radio development and final integration, aiming to recreate a deeply patented, end-to-end stack.
- •Motivation: India lacked domestic companies making full 5G infrastructure
- •Consortium model: 8 institutions; each owns a subsystem/module
- •IIT Madras roles: radio stack ownership + system integration responsibility
- •Cross-functional engineering: software, embedded, RF, antennas, hardware teams
- •Long-horizon commitment: treating it as a necessary “moonshot” despite scale
From “testbed” to near-product: TRL jump and real deployments
Originally scoped as a lab-scale test platform, the project evolves toward something close to field-ready. The team adopts industry interfaces and practices, engages in customer discovery, and ends with deployable base stations and core—used on campus and by startups.
- •Testbed intent: lab platform for testing algorithms/products by outsiders
- •Pivot: evolved toward TRL ~7–7.5 (near field-deployable)
- •Industry alignment: kept interfaces consistent with broader ecosystem practices
- •Validation approach: trade shows, stakeholder discussions, customer-style surveys
- •Outcomes: IITM campus 5G deployment + startups/companies using the facility
India’s first 5G call: the high-stakes demo moment
Prof. Ganti recounts the first 5G phone call in India made on the homegrown setup during a ministerial visit. The event becomes tense due to crowding and signal blockage, requiring quick engineering judgment (raising transmit power) to complete the demo.
- •Milestone: first 5G call in India conducted on IITM’s homegrown network
- •Setup: base station + phone in a small room at the Research Park
- •Unexpected challenge: reporters crowding blocked signal path
- •Real-time fix: student increased transmit power to ensure the call succeeded
- •Reality of demos: success followed by system instability afterward—typical in bleeding-edge builds
Why commercial rollout is hard: scaling, TRL limits, and industry handoff
Despite technical success, production-grade nationwide deployments require companies to take the technology further with major capital and engineering investment. Prof. Ganti contrasts academic capability with productization needs and notes Indian entities beginning to bridge the gap.
- •India’s current private rollouts still largely use foreign equipment
- •Academia ceiling: reaching TRL ~7.5 is impressive but not full commercialization
- •Productization needs: reliability, manufacturing, support, certification, scale
- •Emerging Indian players: C‑DOT, Tejas Networks; Reliance building base-station capability
- •Key transition: turning a proven system into a durable commercial supply chain
4G vs 5G vs 6G: what “G” means and who sets the rules
The episode explains “G” as a decade-scale generational cycle and introduces ITU’s role in spectrum harmonization and performance definitions. 5G is not a marketing label but a set of KPI targets around speed, latency, device density, and mobility.
- •“G” = generation; roughly a 10-year cycle (5G: 2020–2030; 6G: 2030–2040)
- •ITU role 1: spectrum regulation and cross-country harmonization
- •ITU role 2: defining generations via KPIs (data rate, latency, devices, etc.)
- •Bandwidth growth: ~2 MHz (2G) → 20 MHz (4G) → 100 MHz (5G)
- •Technology enablers: massive MIMO, stronger error-correction, improved signal processing
5G use-cases and verticals: eMBB, massive IoT, and ultra-reliable low-latency
Prof. Ganti maps 5G capabilities to distinct verticals: enhanced mobile broadband, massive machine-type communications, and ultra-reliable low-latency links. The conversation connects these to India-relevant examples like smart meters and industrial/factory reliability needs.
- •eMBB: high-throughput mobile broadband (streaming at high mobility)
- •Massive IoT (mMTC): millions/billions of sensors and low-data devices
- •URLLC: ultra-reliable links for factories and critical applications
- •Latency improvements: better experience for interactive apps and gaming
- •India angle: large-scale smart meter rollout as a major IoT driver
How countries shape global standards: India’s push for rural coverage (LMLC)
The discussion turns to standardization politics: countries propose requirements and negotiate what becomes part of the global spec. Prof. Ganti describes India’s effort to insert “low mobility, large cell” coverage into ITU requirements to suit rural economics and geography.
- •Standards are negotiated: national priorities influence global specifications
- •India’s need: rural coverage + price sensitivity + low rural revenue per tower
- •Proposed model: base station at Gram Panchayat covering nearby villages
- •Pushback: Western priorities favor speed/latency over coverage radius
- •Outcome: India helped push LMLC (Low Mobility, Large Cell) into standard discussions; continued work toward 6G coverage
BSNL’s 4G story: security, geopolitics, and building an Indian consortium stack
Prof. Ganti explains why BSNL’s 4G rollout lagged and how a geopolitical inflection pushed a “homegrown only” mandate. The result is a consortium approach where multiple Indian entities jointly deliver radios and core infrastructure for a national deployment.
- •BSNL delay: not purely technical—policy and political constraints mattered
- •Trigger: Galwan and national security concerns around telecom supply chains
- •Bold policy: BSNL network equipment must be entirely Indian-made
- •Execution reality: Indian vendors formed consortiums to meet scale/PoC needs
- •Named contributors: Tata group/Tejas (radio/infrastructure) + C‑DOT (core); positioning India among few countries with indigenous capability
Why choose Electrical Engineering: the “information” discipline behind the digital world
The episode pivots to careers, arguing EE is foundational to the information economy—communication, signal processing, storage, and computation. Prof. Ganti positions EE as the mathematical and conceptual base beneath modern AI/ML and digital systems.
- •EE breadth: power/machines → communications → quantum/devices and more
- •Core framing: EE as information propagation, storage, and retrieval
- •Everyday impact: phones, laptops, USB, imaging, networking—enabled by EE work
- •Signal processing as a foundation of modern technology and AI methods
- •EE-to-other-fields mobility: strong math makes transitions (ML/CS/quant) easier
Is EE “hard” and is IITM too theoretical? Bridging math to systems-building
Prof. Ganti reframes EE difficulty as rigorous training rather than gatekeeping. He describes his own shift from theory-heavy learning to practical systems work (software-defined radio), arguing that tinkering and systems constraints deepen theoretical insight—and that IITM builds extensively today.
- •Perceived hardness comes from heavy math: probability, linear algebra, calculus
- •Long runway: concepts can take years to ‘click’ into real wireless intuition
- •Theory ↔ practice loop: implementation constraints (latency/real-time) reshape algorithms
- •Myth-busting: “theory-only people can’t build” is largely false; tinkering changes that
- •IITM culture shift: increased systems work, industry collaboration, and building across departments
EE vs EEE vs ECE + interdisciplinarity + AI/ML in wireless (Shannon limits)
The closing segment clarifies degree labels across institutions, emphasizing IITs’ broader EE umbrella and the inherently interdisciplinary nature of telecom hardware. It ends with how AI/ML is influencing wireless—useful as a tool but constrained by information-theoretic limits—followed by final reflections.
- •Degree labels: EE (power/machines), EEE (+electronics), ECE (+communications); IITM groups many areas under EE
- •Interdisciplinary builds: thermal/cooling challenges in radios require mechanical engineering too
- •AI/ML in wireless: exploring ML-based coding/decoding and optimization with strong domain knowledge
- •Why wireless is uniquely testable: Shannon capacity provides firm bounds—ML must prove closeness to limits
- •Wrap-up: gratitude, lab visit invitation, and emphasis on long-horizon team dedication (10-year cycles)
