Path
Does it survive different transmitter and receiver chords?
Experimental scanner research · Gate 1
Transmit gently.
Listen completely.
PHANTOM–CONCORDE is an open research architecture for low-energy, phase-coherent microwave tomography. PHANTOM acquires many relative measurements; CONCORDE asks whether a reconstructed feature survives independent ways of looking.
The whole scanner does not exist yet. The measurement, antenna, switching, control and reconstruction technologies exist separately. This project shows one way to connect them and defines the experiment that must prove—or disprove—the arrangement.
12345678
PHYSICALREFERENCE30 mm inclusion
Eight reciprocal viewpoints · many intersecting paths · one bounded reference volume
This is not a diagnostic system, a clinical-performance claim or an invitation to scan a person. Gate 1 does not authorise human use.
Current status
Early bench-test components are awaiting delivery.
The first two-antenna experiment can begin when the early components are available. The complete eight-node scanner has not been assembled. The research pack is public now so engineers, researchers and institutions can inspect the design, find weaknesses and help make it real.
“I do not need to own it. I need to be able to use it. If people can help make it work for everyone, that matters more than ownership.”
— Alex Anderson, project initiator
The aim is not one person’s private escape from a shared problem. If the method survives testing, it should become a shared capability that people can scrutinise, improve and access fairly.
The working idea
Do not turn up the power. Improve the listening.
The design improves information quality through multiple positions, frequencies, coherent amplitude and phase measurement, calibrated coupling and repeated independent paths. Raw complex measurements remain separate from reconstruction.
A feature should weaken when it depends on one channel, one frequency, one placement or one modelling assumption. Where the evidence cannot support detail, the required result is insufficient measurement.
08Reciprocal antenna nodes
1–4GHz measurement band
≈20µW normal selected-feed power
22,512Complex values per scan
How the proposed Gate‑1 system is made
Existing parts. A new arrangement. A test before a claim.
This is a proposed engineering route, not a finished product recipe. Component substitutions and safety assumptions must be reviewed and measured by qualified specialists.
- 01Coherent measurement core
An open two-port vector network analyser supplies a controlled 1–4 GHz signal and records all four complex S-parameters.
- 02Passive power boundary
Fixed 3 dB pads sit directly at both possible transmit ports; there is no RF power amplifier.
- 03Reciprocal switch matrix
Two absorptive 12-way selectors and eight local A/B selectors route either VNA port to a selected pair while unused nodes remain terminated.
- 04Eight equal paths
Equal-length feeds connect to eight reciprocal antenna nodes arranged around an adjustable 200–400 mm reference ring.
- 05Reference and control
An RP2040-class controller enforces valid switch states while known through, load, open and short routes reveal drift or calibration failure.
- 06Evidence before reconstruction
The host stores immutable raw complex data, geometry, timing and power records before any map is reconstructed or scored.
Confirm through concordance
An image is never the proof.
The proof is whether a candidate feature remains when the system changes the relative frame used to observe it.
Direction
Does reciprocal S21/S12 evidence agree?
Frequency
Does it persist across separated parts of the primary band?
Placement
Does it return after removal, replacement and a full rebuild?
Method
Does it survive an alternative reconstruction or regularisation setting?
Integrity
Are the clock, calibration, geometry and reference paths still valid?
The first falsifiable test
Can it measure contrast before anybody calls it medicine?
With no more than 1 mW available at the source and approximately 20 µW at the selected antenna during a normal scan, Gate 1 asks whether the ring can detect, coarsely localise and classify a concealed 30 mm inclusion in a torso-scale reference volume within five seconds.
Every frozen engineering threshold passes, including blinded detection, localisation, rebuild repeatability and conservative refusal.
The complete data misses one or more thresholds. The raw result remains useful for diagnosing the limiting mechanism.
The measurement conditions left the protocol envelope, so the result cannot honestly support a performance conclusion.
Build only what the evidence earns
Four gates from signal to system.
- 01
Two-antenna link
Ask whether two reciprocal flat antennas can repeatedly detect a concealed 30 mm contrast object at tens of microwatts.
- 02
Switch coupon
Measure the real loss, leakage and phase repeatability of a reduced switching path before paying for the complete matrix.
- 03
Eight-node ring
Build the adjustable 200–400 mm reference ring only after the physical antenna and coupling link earns the next cost.
- 04
Blinded gate
Freeze the method, run 50 randomised primary trials, rebuild the system and test whether forced data loss produces “insufficient measurement”.
The line this work will not cross
Gate 1 does not scan people.
Any move toward human research would require a separate device architecture and independent electromagnetic-exposure modelling, implant compatibility, thermal and fault analysis, radio compliance, medical-device review, ethics approval and an authorised protocol.
The present work measures only physical phantoms. It makes no claim of anatomical imaging, clinical sensitivity, clinical specificity, diagnosis or human-use safety.
Open Gate‑1 research pack
Requirements, hardware and a test that can say no.
The download contains ten files: system requirements, architecture, bill of materials, executable scan generator, generated schedule, blinded protocol, two-antenna experiment and two explanatory HTML visuals.
Download the complete ZIP · 38 KB ↓See it in the public library →Independent scrutiny wanted
Help find the failure before the failure finds us.
RF engineers, microwave-imaging researchers, medical physicists, inverse-problem specialists, safety reviewers and regulatory experts are invited to challenge the assumptions, lend equipment, help test the hardware or strengthen the phantom protocol. Begin with a plain-text message through the protected manual-review route. The purpose is shared access and public benefit—not possession by one person.
Offer help or scrutiny →