One of the widely accepted challenges in the field of structural health monitoring (SHM) using ultrasonic guided waves is to carry an accurate damage diagnosis even in the presence of changing environmental conditions. The present paper investigates the role of ambient temperature in causing changes to the structural wave propagation in a newer perspective and presents a novel methodology to compensate its effect. The changes in the sensor signal parameters, caused by the variation in ambient temperature, are related to the changes in the physical properties of the system through a linear system identification model. A very limited data set is used to train the input-output model which has the capability to generate a large baseline space of sensor signals underlying its practical efficacy and usefulness.