Add to Facebook
download .zip with all pictures
> why wouldn’t the military use balloons
You misunderstand the type of balloon. [This](http://i.imgur.com/VDSlRMm.jpg) is a weather balloon. It tracks wind speed by being blown around, and reporting back where it’s been blown to. To use them to detect aircraft, you’d need *billions* of them, launched 24/7.
> this point drones as sensors
You say drone like it’s some kind of cure all, but it really isn’t. The problem is *what* sensor – a drone on its own doesn’t have any ability to detect anything. If you put an IRST on it, it’s going to cost ~$3 million in sensor alone (probably more like $10 million each) and is going to weigh somewhere about a metric ton, which isn’t exactly cheap, expendable, or easy to maintain. Radar is worse, since they’re even heavier, more expensive, and need 100x as much power to operate.
If you put a simpler sensor on it, then it can’t detect the stealth aircraft in the first place, since (unless you buy an even *bigger* drone) there’s going to be a major altitude differential => small optical target size, especially when combined with no-flash paints. Remember that you’re trying to track a target at 40,000ft – well above the altitude ceiling of even the $20 million medium altitude long endurance drones like the Predator.
The downfall of drones is that you still need to put sensors on it, and sensors are *expensive*. Drones are a *terrible* option for counter-stealth as such, because you spend waaaaaaaay too much money on a platform that requires a lot of maintenance and support and provides questionable benefit over other options e.g. ground stations.
Even worse, let’s say that each ($10 million) UAS with IRST can search an area 50km on a side perfectly (note: it can’t). Then, to defend a border 1000km long (not unreasonable), you would need 20 in the air, 24/7 (note: no overlap – if we give them 10km of overlap then we need 34 airborne). If we then assume 50% availability (that is, at any given moment 1/2th of the fleet is airborne), you then need between 40 and 68 UASes in your whole fleet, operating at a preposterous OPTEMPO. As such, to defend a single thin strip of border, you need between $400 million and $680 million in UASes (costing between $3.84 million and $6.5 million *per day* to operate, using MQ-1 demonstrated operating cost), enough to buy between 5 and 8 F-35As and operate them for 27 hours.
This is assuming the enemy isn’t shooting your drones down (hint: they are). If we then say that you need to keep the line up for 30 days, each day taking 5% attrition, you’re going through 1.5 drones per day, with a total requirement of 119 working drones, at a cost of $1.19 billion in purchase costs alone. Drones are neither cheap nor easy.