Skip to content
SIH Buddyby Ganeev Singh
Dev
All problem statements
SIH26055Worth consideringacceptance 4/5

Smart Scan strategy for Electronic Warfare

DRDO · Clean & Green Technology · Software

A clean sequential decision problem with a fair baseline and simulation explicitly sanctioned, which is a rare position for a defence statement — the risk is presentational rather than technical, so plan how you will make a general panel feel why dwelling in the right band at the right moment is hard.

Open dataset ↗

Data: J. C. Wise, Radar Emitter Database, 2024 — https://huggingface.co/datasets/alan-turing-institute/turing-synthetic-radar-dataset

What it actually is

To find hostile radar and radio transmissions you have to sweep a receiver across a very wide band, but the receiver can only listen to a small slice at a time, so where it points and when decides what it catches. Current practice sweeps on a fixed plan drawn up before the mission, which wastes time on harmless transmitters and misses new ones. The ask is a scheduler that learns where to listen next from what it has already heard.

What to build

Two connected pieces: a simulated radio environment holding ground truth on which emitters are transmitting in which band at each time step, including frequency-agile and spatially scanning emitters, together with a receiver model whose instantaneous bandwidth is much narrower than the band it must cover; and a learned scheduler deciding which band to dwell on at each step to minimise intercept time and maximise interception ratio, trained on the hits and misses the simulator returns, evaluated against the figures of merit the statement names — probability of detection and false alarm, sensitivity, average intercept rate, average reward, percentage of correct predictions and average intercept time error — and benchmarked against the conventional open-loop sweep it is meant to beat, with a separate treatment of the periodically scanning emitter case.

Smallest thing that wins the room

Run your scheduler and a fixed open-loop sweep side by side over the same simulated environment and show the intercept time curve separating, with the learned receiver locking onto a newly appearing agile emitter while the sweep is still working through empty bands.

How crowded this one gets

A guess, projected from the 2025 statements — the last year where both the submission counts and the winners were published.

Moderate120–270 teams expectedroughly 1 in 98–226 wins it

Quieter than 59% of the 226 · #93 of 226 by expected field

A normal-sized field. Your idea has to be good, not miraculous.

Why: defence, intelligence and space bodies drew small fields.

This is a guess, not a fact

Nobody has published 2026’s numbers yet. This is an analysed estimate from last year’s pattern, so please do not take it as the truth — check the live counter on the SIH portal before you decide anything. The range covers the middle half of likely outcomes, so one statement in two lands outside it. Entry closes at 500 ideas per statement, so no range goes past that — a statement that reaches the cap fills and shuts rather than drawing an unlimited crowd. The model reads only three things a team can see before choosing — software or hardware, the theme, and what kind of body posted it — and those explain about a quarter of the variation in last year’s field sizes (R² 0.25 on held-out statements). Trust the band more than the number, and the ordering more than either. It cannot see how good your idea is, which is the part that actually decides it.

The scores

The number is the shorthand. The line under it is the reason.

What you will be writing

  • restless multi-armed bandit scheduling
  • POMDP belief-state tracking of emitter activity
  • reinforcement learning with intercept reward shaping
  • RF environment simulator with per-band truth states
  • Thompson sampling / UCB baselines
  • intercept time and interception ratio evaluation harness
  • Electronic warfare
  • Sequential decision making
  • Spectrum surveillance

Prior art to read before you start

adaptive receiver scan scheduling · reinforcement learning for search allocation · emitter interception probability modelling

Analysed by Claude Opus. Every score above is a judgment call with its reasoning attached — kindly cross-check this against the official statement on the SIH portal before your team commits to it.