The current work studies the influence of the most widely used time-domain techniques on SNR for laser-generated ultrasound, employing a low power (150 mW) modulable diode: these methods utilize highly random pseudo-noise (PN) sequences which, after the use of the cross-correlation, underline the presence of an ultrasonic output. PN pulse trains can broaden the signal’s band through random lengths of LOW and HIGH levels (in a TTL which feeds the laser), scattering the energy on a wide range of frequencies: this might represent a disadvantage in case of controls with narrow-band probe, used in most cases for this kind of applications. From this point of view, creation of specific drive signals allows to improve significantly the control on frequency response. A SNR consistent with the ones achievable through the classic ways is highlighted, but featuring an advantageous concentration of the output frequency band.