Calibration is the process of identifying and correcting for the systematic bias component of the error in sensor measurements. Traditionally, calibration has usually been conducted by considering a set of measurements in a single time frame and restricted to linear systems with the assumption of equal-quality sensors and single modality. The basis for the new calibration procedure is to construct a statistical error model that captures the characteristics of the measurement errors. Such an error model can be constructed either off-line or on-line. It is derived using the nonparametric kernel density estimation techniques. We propose four alternatives to make the transition from the constructed error model to the calibration model, which is represented by piecewise polynomials. In addition, statistical validation and evaluation methods such as resubstitution, is used in order to establish the interval of confidence for both the error model and the calibration model. Traces of the distance ranging measurements recorded by in-field deployed sensors are used as our demonstrative example.