Hmm, it's been some time since I did my semiconductor electronics class, but I'll give it a shot here:
In the electrostatic diode analysis you make the depletion approximation assumption, you assume that the depletion region ends abruptly instead of decaying at a logarithmic scale. You also have to input the boundary values at the edge of the space-charged region (SCR) to solve for the unknown constants of solving the various DE's starting from Poisson's equation. Once you do all that, you get an equation relating the width of the depletion region to the doping concentration on either side of the depleted pn-junction.
Low-level injection is the assumption that the amount of minority carriers injected into the device is >>> (much greater) than the current concentration level, but the amount of majority carriers is much smaller than the existing concentration of majority carriers. The concept of majority and minority carriers assumes of course, that you are dealing with a doped semiconductor. There are a couple of other assumptions made, which I can't quite recall at the moment. But the primary reason why the low-level injection assumption is made is to simplify the resulting DE to that of a single variable. Otherwise it becomes much more complicated.