ABSTRACT
This chapter seeks to classify the methods used to quantify the performance of the sensor networks based on the mathematical apparatus used to evaluate the network performance. These methods can be broadly categorised in statistics-based, information theory-based, expert recommendations, and others. Fundamentally, the model-free approach aims to minimise the variance of the measured variable, therefore, minimising the variance in the estimation. In the model-based approach, the performance of sensor networks is carried out using a catchment model. The integration of dynamic sensors in operational meteorology is limited due to the difficulties in its deployment, and lack of the theory to support its use. This situation leads to only a few examples in the literature, where only conceptual approaches and small tests have been conducted. The possibilistic framework considers that the reasoning yield from inductive knowledge might be better represented by belief functions, instead of probability distributions.
