A border surveillance system of fixed masts or of mobile patrols
A mast sees the same ground for ever. A patrol sees different ground for a while. For a border surveillance system the choice between fixed masts and mobile patrols is usually reported as a technology decision; it is a geometry and staffing decision.
What a fixed mast buys a border surveillance system
A mast holds a sensor at a fixed height above a fixed point, and everything that follows comes from that one fact. Its coverage is a property of the terrain in front of it and does not change: if a ridge hides a valley on the day the installation is commissioned, that valley stays hidden for the life of the installation. In exchange the mast observes continuously, needs no crew in the field, and produces returns that are comparable from one hour to the next — which is what makes correlation across time possible at all.
The number that appears in most reporting is the sensor's nominal range. It is not the coverage, and on broken ground the two are not close. The figure below separates them for a single mast on a single profile.
Read the drawing from the mast outward. The shaded wedge is what the sensor observes. The hatched wedge behind the crest is dead ground: inside the nominal range, and never seen. The dashed line is the sightline itself, and everything below it on the far side of the crest is unobservable from that position at that height. No improvement to the sensor changes any of this, because none of it is a sensor property.
Raising the mast helps, and helps less than intuition suggests. The table beside this paragraph gives the same terrain and the same sensor at two heights.
| Nominal sensor range | 8.0 km |
|---|---|
| Mast 12 m — observed arc | 4.1 km |
| Mast 12 m — ground covered | 51 % |
| Mast 24 m — observed arc | 5.4 km |
| Mast 24 m — ground covered | 68 % |
| Coverage gained by doubling height | +32 % |
What a mobile patrol buys instead
A mobile platform trades continuity for choice of viewpoint. It can be taken to the ridge the mast cannot see past, which is the only way to close dead ground without building a second mast. What it cannot do is be in two places at once, and its coverage figure is therefore a rate — ground observed per hour — rather than an area.
- Coverage is bounded by transit time, not by sensor range
- Every move costs observation: the platform is not watching while it repositions
- Returns are harder to compare over time, because the viewpoint keeps changing
- The crew requirement does not disappear with autonomy; it changes shape
The last point is the one most often lost. A supervised platform still needs an operator, a maintenance cycle, a recovery plan for the day it fails in the field, and someone empowered to act on what it reports. Published operator-to-platform ratios in European research programmes vary widely enough that quoting a single figure would be dishonest.
Where the two approaches cross over in coverage
For a given stretch of border, the two arrangements cost the same at a length that depends almost entirely on how broken the ground is. On flat terrain the mast wins early: dead ground is small, coverage approaches the nominal range, and continuity is nearly free once the structure is up. On broken terrain the crossover moves sharply toward patrols, because each additional mast buys progressively less new ground — the second mast largely re-covers what the first already saw, and only the third begins to reach behind the ridge.
This is why border authorities on the same external frontier often run both within a short distance of each other, and why a comparison that quotes one figure per approach, without naming the terrain it was measured on, cannot be checked by anyone.
Where the model breaks
Both sets of figures, for fixed masts and mobile patrols alike, assume the sensor performs to specification. It frequently does not. Fog and heavy rain reduce a thermal channel's effective range without changing any published number, and the reduction is not uniform across the arc. Around dawn and dusk, thermal crossover — the period when target and background reach the same apparent temperature — removes contrast entirely for a window that varies with season and surface.
None of that appears in a coverage diagram for a border surveillance system, including the one on this page. A geometric drawing shows what could be seen in ideal conditions. It is an upper bound, and it should be read as one.
Questions about fixed masts and mobile patrols
Is a fixed mast or a mobile patrol better for a border surveillance system?
Neither, and the question has no general answer for a border surveillance system. Fixed masts and mobile patrols fail in different ways: a mast cannot close the dead ground it was built with, and a patrol cannot watch continuously. The terrain decides which failure matters more on a given stretch, which is why the same country often runs both within fifty kilometres of each other.
How much coverage does a taller mast buy?
Less than proportion suggests. On the profile drawn on this page, doubling the mast from 12 m to 24 m raises the observed arc from 4.1 km to 5.4 km — about a third more ground for twice the height, and considerably more than twice the structure. The relationship is geometric and flattens the higher you go.
Why is mobile-patrol coverage quoted as a rate rather than an area?
Because a mobile platform observes different ground at different times rather than the same ground continuously. Its useful figure is ground observed per hour, bounded by transit time rather than by sensor range. Comparing that rate directly with a mast's area — which many side-by-side tables do — compares two different quantities.
Does autonomy remove the crew requirement for mobile patrols?
It changes its shape rather than removing it. Supervised platforms still need operators, maintenance, recovery when something fails in the field, and someone to act on a detection. What changes is the ratio of people to platforms, and the published ratios vary widely enough that no single figure can be quoted honestly. See autonomy levels.