Particulate matter (PM) is reported as dangerous and can cause respiratory and health issues. Regulations, based on PM concentration, have been implemented to limit human exposition to air pollution. An innovative system with Surface Acoustic Wave (SAW) sensors combined to 3 Lpm cascade impactor was developed in our team for real time mass concentration measurements. In this study, we compare the PM sensitivity of two types of SAW sensors. The first one consists of delay lines based on Rayleigh waves propagating on a Lithium Niobate Y-X 128° substrate and the second one is a based-on Love waves on AT-Quartz. The aerosols were generated from NaCl for PM2.5 and from Silicon carbide for PM10. The sensor’s response was compared to a reference sensor based on optical measurement. The sensitivity of the Rayleigh wave based sensor is clearly lower than the Love wave sensor for both PM. Although less sensitive, Rayleigh wave sensors are very promising for the development of self-cleaning sensors using RF power due to their high electromechanical factor. To check the performance of our system in real conditions, we tested the sensitivity to PM from cigarette smoke using Rayleigh SAW. The PM2.5 stage shows a phase shift while the PM10 does not respond. This result agrees with previous studies which report that the size of particles from cigarette smoke varies between 0.1 to 1.5 µm. A good correlation between the reference sensor’s response and the phase variation of SAW sensors was obtained.