Rules of Participation

What is the FMT Project

FMT (Frequency Measurement Tests) is an international project for radio amateurs and engineers, where participants compete in the accuracy of measurements of real radio signals. The main idea is to give everyone the opportunity to test their equipment and skills through three types of tasks, with results automatically converted into points and ratings.

Propagation, Doppler and what we actually measure

HF is not a laboratory cable

On skywave paths the ionosphere can shift the received carrier (Doppler). The size of this effect depends on band, path length, geometry (NVIS vs low-angle), and current conditions.

What this means for FMTLab

We do not claim that a single HF on-air carrier measurement can fully remove ionospheric effects. What you measure at the receiver is always “TX + propagation” to some extent.

FMTLab is run as a practical skills challenge under real HF conditions. All participants measure the same transmissions, but each path is different — distance, geometry and ionospheric behaviour are part of the challenge. We use percentile-based scoring to rank results within each tour, and we publish the full results openly.

Calibration sessions

We run calibration transmissions on announced frequencies before and during each month. These sessions help participants see the typical offset on their path, estimate Doppler and short-term drift trends, and apply a correction and report it transparently.

Calibration improves consistency, but it does not “turn off” the ionosphere. Doppler can change between sessions and during a tour.

FREQ vs PULSE / PPS

FREQ (unmodulated carrier) is more sensitive to Doppler and carrier wander on skywave paths. PULSE/PPS-based measurements are affected differently than an unmodulated carrier and often give a more stable estimate of timing and frequency.

Project Goals

• Develop a culture of precise radio engineering measurements.
• Provide participants with a clear training and progress system.
• Make the competition transparent: all data, formulas, and results are open.
• Unite practicing engineers who want to improve the stability of their receivers, frequency standards, and synchronization systems.

How Competitions Work

The season is divided into tours (usually every two weeks). Each tour includes three tasks:

• Carrier frequency (FREQ) — determine the carrier frequency of the signal.
• Pulse duration (PULSE) — measure the duration of the third pulse in a packet of three sinusoidal signals at 1 kHz.
  Packet structure: 1st pulse — fixed duration 100 ms, then a 100 ms pause; 2nd pulse — fixed duration 50 ms, then a 100 ms pause; 3rd pulse — fixed 25 ms in calibration sessions, but during competitions its duration varies and that is what must be measured.
  The first two pulses and the pauses have constant values and can be used to calibrate your measurement chain.
• 1 PPS offset (PPS) — determine the offset relative to the GPS 1 PPS reference pulse.
  Important: the three-pulse packet used for the PULSE metric is intentionally time-shifted by a specific offset relative to the GPS 1 PPS reference. In PPS, your task is to measure this time offset precisely.

Participants choose which metrics to provide: you can participate only in FREQ and PULSE, or in all three. A submission contains measured values, the chosen class (PRO or Hobbyist), and optionally raw data.

How Points Are Awarded

Each submitted metric is scored independently. Your final tour score is the sum of points from all metrics you submitted. For every metric, points are calculated in a clear and predictable way, based on four elements.

1. Placement within your class (Base points)

   Participants are ranked within their declared class (PRO or Hobbyist) by measurement error. Lower error always means a better position.

   The best result in a class receives the full base score (100 points). The last result is guaranteed at least 30% of the maximum points. All positions in between are distributed linearly.

   If two or more participants have the same error, they share the same rank and receive the same base points.

   This ensures that:

   • ranking always matters,
   • even in small classes, every valid measurement is rewarded.
2. Measurement quality (Quality bonus)

   Placement alone does not describe how strong a result really is. To reflect the actual difficulty of the tour, we evaluate measurement quality relative to other participants.

   For each metric, we analyze the error distribution of the current tour:

   • E10 — top 10% of results
   • E50 — median
   • E90 — bottom 10% of results

   Your error is compared to this distribution on a logarithmic scale and converted into a Quality Score from 0 to 100: results close to the best measurements score high, while results close to the weakest measurements score low. The Quality Score is then converted into a bonus (up to 10 points, depending on current rules).

   This means you are evaluated against real results from this tour, not against an abstract or ideal reference.

3. Field multiplier (number of competitors)

   Winning in a large field is harder than winning in a small one. To account for this, the score for each metric is adjusted based on the number of participants in your class.

   The adjustment is calculated as: field_multiplier = √(min(N, Nref) / Nref)

   Where:

   • N is the number of participants in your class for this metric,
   • N_ref is the full-credit threshold (currently N_ref = 10).

   In practice, with 10 or more participants, you receive full credit, while with fewer participants, the score is slightly reduced. This prevents small “mini-tours” from outweighing competitive ones, while still keeping small fields meaningful.

4. Proof and methodology bonuses

   We actively encourage transparent and verifiable measurements. You may receive additional fixed bonuses for:

   • Declared uncertainty — +5 points, awarded if your actual error does not exceed your declared uncertainty and stays within the metric’s allowed limit.
   • Raw data submission — +3 points, awarded for providing raw measurement data, which allows others to review and analyze your method.

   These bonuses reward good measurement practice, not optimistic numbers.

Final Metric Score

All components are combined as follows:

Tour Total Score

Your final tour score is calculated by summing points from all submitted metrics, taking their weights into account:

Important notes:

• Scores are summed, not averaged.
• Submitting more valid metrics gives more opportunities to score.
• Metric weights are defined by the organizers and may vary by season.

In Short

• Placement shows how you rank.
• Quality shows how strong your measurement really is.
• Field size keeps scoring fair.
• Proof bonuses reward transparency.
• More work, done properly, is rewarded.

Participant Classes

• PRO — participant with a synchronized reference oscillator (GPSDO or equivalent). Competes in all metrics.
• Hobbyist — participant without a reference oscillator. Competes in FREQ and PULSE metrics. PPS measurements can be submitted but are not scored.

What Participants Need to Do

• Conduct a measurement and submit the result before the tour deadline (all times are in UTC).
• Optionally specify your uncertainty estimate.
• Wait for the table publication — it will show your placement, points, and progress chart.

Season Standings

After each closed tour, Season Standings are recalculated — the rating tables. They apply the active version of rules (Score Rules), which specify:

• maximum points for 1st place;
• full credit threshold N_ref;
• quality bonus weight and verification bonus values.

Season results account for the best N tours (e.g., 8). We maintain several standings simultaneously:

If a participant works only with pulses, they compete in Pulse Cup and their Hobbyist rating, while all-rounders participate in Overall.

Quick Tour Example

Suppose a tour has 10 participants. Base points are distributed linearly from first to tenth place.

Why All This

To provide an opportunity to compare your equipment's accuracy with colleagues worldwide.

To create a unified standard for objective measurements in the amateur radio community.

To show progress: you can see how error decreases and stability increases.

To provide motivation — ratings, badges, achievement awards.

Simply put: FMT is not just a contest, but a precision laboratory where each participant sees their level, grows, and contributes to overall measurement quality.