$14\mathrm{.}3$ Consider the cutback attenuation measurement technique described by Eq$.(14\mathrm{.}2)$ using a photodetector, the power measurements are proportional to the detector output voltage. If the uncertainty in the voltage readings for the two power measurements are $+-0\mathrm{.}1$ percent each, what is the uncertainty in the attenuation accuracy? How long must the fiber be to get a sensitivity better than $+-0\mathrm{.}05$ dB$/$km$?$