right, this isn't much, but i'm writing it before I forget.
CMOS is better with respect to size, integration, power consumption and mass. This all is because the chip for reading, and converting data interfacing and the rest is one single integrated thing.
The power consumption (according to http://www.micron.com/innovations/imaging/advantages) is roughly 100 times less than for a CCD. The quoted figures for the CCD are a power requirement of 2-5 Watts, with large clock swings of 5 - 15 Volts. The CMOS runs on 20-50 milliwatts apparently, and can use a single 3.3 or 5 volt supply.
The major disadvantage with the CMOS is the comprimise it must make in picture quality. Basicly the better image will be obtained from a CCD. It just depends on how good an image we want.
CMOS is also meant to be better in more rugged environments, as the reduced amount of leads and solder, which are the leading causes of failure, mean that the CMOS is more likely to keep going.
Anyway, I don't know how much good this is to you, I'll keep looking. General upshot is that with the mass, power and size budgets, the CMOS is the better option. However, the better image is the CCD. I think the prices are comparable, but don't have any figures as yet.
(Also looking at flux, using http://www.gps.caltech.edu/genesis/DocumentN.html it describes the beta meteoroids as being with mass below 1.4 x 10-12 g and density of 2 g/cm3. The particles have to be from a circular orbit due the low mass. The particles are radition dominated, but this gives a lower mass limit of 10-15 g as radiation pressure drops off rapidly at this limit. The website gives an apporximation of cumulative flux, leading to a total mass flux of 1.1 x 10-17 gm-2s-1. This gave a total flux in two years of 3.5 x 10-14 gcm-2. Assuming that there is consideration given to mass distribution, just changing the time period to nine months should lead to a total fluxof 1.31 x 10-14 gcm-2. I don't see a way of converting this into impacts, as I don't know how the mass distribution varies. I also don't know the errors in the website calculation, so I can't give errors with this flux. I just knocked off a few decimal places. These estimates are based on a device pointing directly at the sun for a continuous amount of time.
The website makes frequent references to two sources, Grun et Al 1985, and Zook and Berg 1975.. Thought these might be good places to look next.)
Owen