FT2 Is a New Amateur Radio Digital Mode — But Should We Be Developing FT16 Instead?

This article explores the question from a signal-processing, information-theory, HF propagation, spectral-efficiency, hardware, and community-adoption perspective.

Understanding the Physics: Processing Gain and Shannon Limits

At its core, the idea of FT16 relies on one simple principle: longer coherent integration increases processing gain.

For additive white Gaussian noise (AWGN), doubling integration time theoretically provides approximately +3 dB improvement in SNR. This comes directly from energy accumulation over time:

SNR ∝ Signal Energy / Noise Energy

Since signal energy increases linearly with time while noise integrates statistically, longer symbol durations improve decodability — assuming the channel remains coherent.

However, Shannon capacity places a hard boundary on what is achievable:

C = B log2(1 + SNR)

As SNR becomes extremely negative, improvements require exponentially more integration time or stronger coding. Therefore, gains beyond 6–8 dB rapidly become inefficient unless bandwidth is reduced dramatically.

This is precisely why WSPR achieves extreme sensitivity: not only long integration (~110 s), but extremely narrow bandwidth (~6 Hz).

An FT16 implementation that keeps ~50 Hz bandwidth cannot approach WSPR-level performance indefinitely through time extension alone.

Channel Coherence on HF

HF propagation is not a static channel. It is a time-varying, frequency-selective, multipath environment.

Key constraints:

• Fading (QSB) time constants: typically 5–20 seconds

• Selective fading across narrow frequency spans

• Doppler spread in dynamic ionospheric conditions

• Rapid signal amplitude variation during greyline transitions

If the coherence time of the channel is shorter than the transmission duration, coherent integration efficiency degrades.

For example:

• 15 s integration → often within fading coherence window

• 60 s integration → likely spans multiple fading cycles

In practice, this means the theoretical +6 dB gain at 60 seconds may compress to +3–4 dB on real HF paths.

Therefore, FT16’s practical improvement is constrained by propagation physics, not algorithmic limitations.

Coding Considerations: Stronger FEC vs Longer Frames

Sensitivity improvements can be achieved through:

1. Longer integration

2. Lower symbol rate

3. Stronger forward error correction

4. Narrower bandwidth

5. Hybrid combinations

Stronger FEC (e.g., lower code rate LDPC) increases redundancy but reduces net information throughput. Message length must remain constrained to preserve practical QSO exchange timing.

FT8’s structured message format is optimized for minimal overhead. Extending redundancy further would increase robustness but reduce spectral efficiency.

FT16 would likely need:

• Lower code rate (e.g., 1/3 instead of ~1/2)

• Deeper interleaving to combat fading

• Improved time-frequency synchronization tolerance

Each addition improves sensitivity but increases decoding complexity and CPU demand.

Hardware and Stability Requirements

Longer transmissions place stricter demands on:

• Frequency stability (TCXO minimum)

• Time synchronization (sub-second NTP accuracy)

• Receiver phase noise performance

• SDR dynamic range

In practice, many stations operating FT8 today use consumer-grade oscillators with modest drift. FT16 may implicitly raise the hardware baseline.

This could limit adoption in portable or budget setups unless decoder algorithms incorporate drift compensation mechanisms.

Energy Efficiency and Portable Operation

One strong argument in favor of FT16 is energy efficiency.

Portable QRP operation is constrained by:

• Battery capacity

• Solar charging limitations

• Thermal dissipation in compact transceivers

If FT16 reduces required transmit power by 3 dB, energy consumption halves for equivalent link reliability.

However, longer transmissions also increase duty cycle per QSO attempt. Total energy per completed QSO must be considered, not just instantaneous power.

Energy per QSO becomes:

E = P × T × retransmissions

If longer transmissions reduce retransmissions due to better decodability, total energy efficiency may improve despite longer airtime.

This area deserves quantitative modeling.

Network Effect and Adoption Dynamics

Digital modes succeed not only due to technical merit but due to adoption density.

FT8 became dominant because:

• It solved a real problem (weak-signal DX)

• It was integrated into widely used software

• It had immediate global participation

• It occupied well-defined sub-bands

FT4, despite technical validity, remains secondary due to narrower use-case demand.

FT16 risks entering the same category unless it clearly differentiates itself.

Adoption depends on:

• Visible performance advantage

• Backward compatibility

• Community advocacy

• Demonstrable DX results

Without these, technical superiority alone does not ensure success.

Band-Specific Considerations

The utility of FT16 may vary by band:

160 m and 80 m:

• High noise floor

• Longer fading cycles

• Greater benefit from extra sensitivity

20 m:

• High activity density

• Moderate fading

• Mixed benefit

10 m:

• Rapid sporadic E fading

• Short coherence times

• Limited benefit from long integration

Thus, FT16 may find its niche on low bands during solar minimum.

Competitive Alternatives: Adaptive Modes

Instead of creating a separate FT16 mode, an adaptive architecture could be considered:

• Dynamic cycle length based on SNR

• Automatic fallback to longer frames

• SNR-aware coding rate adjustments

• Hybrid fast/slow response protocol

Such an approach would minimize fragmentation while enabling extended sensitivity when required.

This could represent a more forward-looking evolution than a fixed-length “FT16.”

Quantitative Link Budget Illustration

Consider:

• 20 m path

• 10,000 km

• 10 W transmit

• 0 dBi antenna

• –120 dBm noise floor

If marginal FT8 decoding occurs at –24 dB SNR, a +3 dB FT16 gain shifts the margin significantly in deep-fade scenarios.

However, if path loss variability exceeds 10 dB due to QSB, the incremental gain may not change success probability dramatically.

Statistically, the probability-of-decode curve shifts modestly rather than fundamentally.

Psychological and Operational Factors

Operator satisfaction also matters.

FT8 already faces criticism for:

• Automation

• Minimal operator interaction

• Structured exchanges

FT16 would further slow exchanges without adding conversational depth.

Unless it incorporates enhanced messaging capability, it may not satisfy those seeking more interactive digital communication (e.g., JS8Call philosophy).

Is There a Compelling Strategic Case?

The strongest arguments for FT16:

• Measurable sensitivity gain

• QRP friendliness

• Marginal DX experimentation

• Low-band weak-signal exploration

The strongest counterarguments:

• Limited incremental benefit

• Increased latency

• Spectrum occupancy concerns

• Community fragmentation risk

• Hardware stability demands

Technically, FT16 is coherent and feasible.

Strategically, it must justify its existence by delivering a clearly visible advantage under real-world conditions, not only under laboratory assumptions.

Conclusion

Developing FT16 would not violate any physical constraints, nor would it represent a trivial modification. It would meaningfully explore the sensitivity–latency trade space in weak-signal digital communication.

However, its gains are bounded by ionospheric coherence time, oscillator stability, and spectral practicality. The improvement would likely be evolutionary rather than transformative.

If FT2 represents optimization toward temporal efficiency and responsiveness, FT16 would represent optimization toward marginal decodability and QRP efficiency.

Whether that direction is worth pursuing depends less on signal theory and more on community appetite for trading speed for depth in the next phase of digital mode evolution.