T1UD-MISTelecom
5G/6G cell planning & frequency reuse
Assign channels to base stations so no two interfering cells share a frequency — interference radius maps directly onto blockade radius. Textbook native fit.
Sector
Telecommunications network planning
Likely buyer
Mobile network operators; RAN planning vendors
Named precedent
Graph-coloring / MIS literature; interference-radius equivalence
Hardware gate
Orion-class today; dense metros → larger registers
Taxonomy
throughput-limited × accelerating
Live demo — adiabatic sweep on 6 cells
Ω 9.4 · δ ±12.6 rad/µs · 4000 ns · R_b 9.1 µmSix base stations; nearby cells interfere and can't reuse a channel. The largest interference-free reuse set is the MIS.
GTM talk track
'Your interference radius is literally our blockade radius. We don't model your problem — we embody it in atom spacing.'
OGSM — product operating frame
Objective
Win a lighthouse RAN-planning benchmark against an operator's incumbent tool.
Goals
- One operator runs a real metro cell cluster
- Documented spectrum-efficiency delta
Strategies
- Map cell interference graph to register geometry
- Benchmark against their current planner
Measures
- Spectrum reuse efficiency %
- Planning runtime on hard instances
OBR — outcome-based roadmap
| Horizon | Outcome we create | Buyer behavior change | Result we measure |
|---|---|---|---|
| Now | Operator sees channel assignment on a sample cluster | Prospect runs the emulated demo on their own instance data | Booked QPU-time evaluation or paid pilot |
| Next | Operator benchmarks a full metro against incumbent | Prospect co-designs a scoped benchmark against their incumbent solver | Documented crossover curve; expansion to production instances |
| Later | Operator integrates into planning cadence | Prospect standardizes on the workflow or buys an on-prem system | Recurring QPU consumption / system sale; reference case |
Fit notes (honesty gate)
Classical planners are strong; value window is dense, high-value metro instances where heuristic gaps cost spectrum.
Ready to run this on real hardware?
Emulation-verified today — the same program runs on a Pasqal QPU unchanged.