Sensor signal fusion is becoming increasingly important in many areas including medical diagnosis and classification. Clinicians/experts often do the diagnosis of stress, sleepiness, tiredness etc. based on several physiological sensor signals to achieve better accuracy in classification. This paper presents a case-based reasoning (CBR) system that offers an opportunity to classify healthy and stressed persons based on sensor signal fusion. Several sensor measurements for instance, i.e., heart rate, inter-beat-interval, finger temperature, skin conductance and respiration rate have been combined for the data level fusion using Multivariate Multiscale Entropy Analysis (MMSE) algorithm. This algorithm supports complexity analysis of multivariate biological recordings. Here, MMSE is used to formulate cases in the case-based classification system.