Denoising Temperature on Martian Surface dataset

In the present article we analyze data from twotemperature sensors of the Curiosity rover, which has beenactive in Mars since August 2012. Temperature measurementsreceived from the rover are noisy and must be processed andvalidated before being delivered to the scientific community.Currently, a simple moving average filter is used to performsignal denoising. The application of this basic algorithm is basedon the assumption that the noise is stationary and statisticallyindependent from the underlying structure of the signal, anarguable assumption in this kind of harsh environment. In thispaper, we analyze the application of two alternative methodsto process the temperature sensor measurements: the DiscreteWavelet Transform (DWT) and the Hilbert Huang Transform(HHT). We consider two different datasets, one coming fromthe current Martian measurement campaigns, and the otherfrom the Thermal Vacuum data test. These datasets allow toseparate the random noise from the interference created by othersystems. The experiments show that the moving average filterprovides useful results, however, the proposed methods allowa better fitting for all the realistic scenarios and they providethe possibility to identify and analyse other interesting signalfeatures and artifacts. The large amount of data to be processedmakes computational efficiency an important requirement in theCuriosity mission. Considering the computational cost and thefiltering performance, we propose the method based on DWT asthe most suitable for this application.