Continuous temperature monitoring algorithm for SAW sensors

Stability of wireless temperature reading from a surface acoustic wave (SAW) sensor relies on sophisticated signal processing algorithms. Disturbances due to radio interference often lead to erroneous measurements. In a continuous setting we address this issue with a novel efficient algorithm. Our approach is an extension to the current state of the art methods. It is based on fitting a polynomial approximation to a sequence of past measurements. We investigated a least squares fit, a sparsity promoting fit and a filtered least squares fit. Through experiments with simulated and measured data, with temperatures of up to 300 °C, we found the latter two types showed promising results, justifying the use of the method in practice.