Dear Bilal
As you know in the IR-UWB there is no carrier and the data is transmitted in very-short pulses (nanoseconds), thus Gaussian pulses are used because it easy to get narrow Gaussian pulses with sharp trailing and leading edges compared with square and triangle pulses. Now derivatives of Gaussian pulses are used (MONO, DOUBLET, TRIPLET..) for the following reasons:
1- The most important reason is to fit and comply with the mask specified by the FCC, as you note from the attached image, as the derivative order increase, the power spectrum of the transmitted UWB pulses fit the FCC mask and the power efficiency increased as well.
2- As you know the antenna is not efficient at DC and by using derivative the DC value is removed.