## Orbit isn't high, it's fast ##
Orbit is moving sideways so fast you miss the Earth. Something like 7.8 km/s, which is 17,500 mph. A Cessna flies at rounding error speeds in comparison.

So instead of thinking that it would be possible if the Cessna were dropped low enough, think of it being possible if the Cessna is dropped at a slow enough air speed.

## Kinetic Energy is your enemy ##
We have a Cessna strapped to an orbital craft doing 7.8 km/s (or it's not orbiting), and by the time it gets into the atmosphere to avoid burning up it needs to be going about 130 mph, or 58 m/s, which is zero by comparison to the initial speed. Our 1,000 kg Cessna thus has a relative KE of 0.5*1000*(7.8e3^2) = 30.4 Gigajoules which we need to lose.

## You can't just slow it down over many orbits ##
I think the question is suggesting letting a little bit of drag slow the Cessna down until it's at a normal speed before gliding through the atmosphere. That's nice, but it won't work. Orbiting is ballistic flight. Ballistic flight without lateral speed (i.e. once you've started to really slow the Cessna down) is just falling - your ballistic trajectory intersects the Earth's surface. There is no low-speed way to miss the Earth without lots of thrusters, which is not gliding.

## Decelerockets! ##
There is only one way at present to transfer 30 GJ of energy to a Cessna before it reaches the atmosphere (i.e. before needing heatshields) and that is a rocket.

Essentially you are going to drop the Cessna off the orbital vehicle and fire a large booster rocket with a delta-v of 7.8 km/s. That is energetically similar to launching a LEO rocket from Earth (LEO has a delta-v of 9 or so from Earth), because again it's about the speed, not the height, and you're just accelerating backwards instead of forwards.

An appropriately sized rocket might be the [Japanese Epsilon rocket][1]; it has a payload of 1200 kg for LEO which should get the Cessna and a pilot from +7.8 km/s to 0.06 km/s over the course of a couple of minutes.

## The Tyranny ##
There's one snag; the Epsilon is 91,000 kg because of the Tyranny Of The Rocket Equation - to accelerate a thing you have to stick it on a rocket, which you also have to accelerate, which takes more fuel, which also has to be accelerated, etc. So 1,200 kg payload requires a 91,000 kg rocket. 

That rocket is now a payload we have to put in orbit in the first place, on the top of a much bigger rocket.

But wait, we've lifted more before; Saturn V managed 140,000 kg. The smallest previous launch capable of that scale is the Russian [Energia rocket][2], a follow-on from the unsuccessful N1 project (Russia's Saturn V equivalent). It is about 2,400,000 kg (2.4 kt), about the size of Big Ben's tower and would cost something over $1.5bn.

## Or just put a coat on ##
I've let it get a bit silly. It should hopefully be clear that a heatshield is rather cheaper than a rocket to decelerate with, which is essentially why Shuttle was designed the way it was - decelerating by airbrake only costs you the weight of a heatshield.


  [1]: https://en.wikipedia.org/wiki/Epsilon_(rocket)
  [2]: https://en.wikipedia.org/wiki/Energia