Field Guide · term

Also known as: fade margin, link margin, system operating margin

Fade margin is the number of decibels by which a link’s normal received signal level exceeds the minimum level the receiver needs to work — the headroom set aside so that short-term fading does not drop the signal below threshold.1 It falls straight out of the link budget: once you know the expected received power and the receiver’s required SNR or sensitivity, the difference is your margin. A link with too little margin works in fine weather but drops out on every deep fade.

level (dB) mean received level receiver threshold fade margin
Fade margin is the gap between the mean received level and the receiver threshold; deeper margin survives deeper fades.

How it works

Fade margin is simply

  • M = P_received − P_threshold (all in dB),

where P_received is the level predicted by the link budget after free-space path loss, antenna gains, and fixed losses, and P_threshold is the receiver’s minimum usable level (its sensitivity, or the level giving the required SNR).

How much margin you need is dictated by the fading statistics of the channel and the availability you are targeting:

  • On a deeply fading Rayleigh channel, deep fades are common, so hitting even 99% availability demands roughly 20 dB of margin, and 99.9% demands about 30 dB — each extra “nine” costs about 10 dB.
  • On a gentler Rician channel with a strong line-of-sight component, the same availability needs far less margin, because deep fades are rarer.
  • On microwave links the driver is often rain fade, and margin is sized from rain-rate statistics for the target percentage of the year.

The relationship is non-linear: because fade depth grows slowly with rarity, buying the last fraction of a percent of availability costs disproportionately many decibels.

In practice

Because margin is expensive, engineers spend it wisely and combine it with techniques that lower the required margin rather than just cranking up power:

  • Antenna gain and height raise the mean received level directly and, by improving line-of-sight, push the channel toward Rician statistics.
  • Diversity (space, frequency, or polarisation) makes a deep fade on one branch unlikely to coincide with one on another, sharply cutting the margin needed for a given outage.
  • Forward error correction and interleaving let the receiver tolerate a lower instantaneous SNR, effectively lowering the threshold and so freeing up margin.
  • Adaptive coding and modulation trades data rate for robustness on the fly, keeping the link alive through a fade instead of budgeting fixed margin for it.

Relevance to SDR

Fade margin is why a scanner’s reception is reliable in some spots and marginal in others. A strong average signal from a nearby trunking site carries plenty of margin, so occasional deep fades never cross threshold and the audio stays clean. Near the fringe of coverage the margin shrinks to a few dB, and every fade produces a burst of errors or a dropout — the audible signature of running out of margin. Raising the antenna, adding gain, or reducing feedline loss all buy margin back.

GopherTrunk does not compute a link budget, but its per-frame SNR and EVM readouts are effectively a live margin meter: a comfortable SNR above the demodulator’s floor means healthy margin, while values hovering near the decode threshold warn that the link has little headroom left against the next fade.

Sources

  1. Fade margin — Wikipedia, on the difference between received level and threshold and its role in link availability. ↩

See also