Cylinder pressure signal provides key feedback information that allows for several engine monitoring and control capabilities. During the last few years, the use of cylinder pressure traces for advanced combustion strategies has become the focus of several research works. This is primarily due to the availability of reliable and inexpensive cylinder pressure sensors with expected durability that meets the vehicle lifetime. In this context, several approaches have been proposed to determine the engine trapped mass from the cylinder pressure measurements. A recent innovative approach for trapped charge determination based on a two‐dimensional graphical signature is proposed in a previous work. The resulting estimator uses cylinder pressure signal as a unique input and allows to deduce the trapped mass on a cycle by cycle basis in steady and transient operating conditions. It has been validated in a wide range of engine operating conditions using instrumentation and industrial cylinder pressure sensors. This paper provides a theoretical framework and in‐depth analysis for the signature based trapped mass estimator. A state‐space model for in‐cylinder conditions during the compression phase that complies with the signature modeling structure is developed. Extensive numerical investigations using an experimentally validated simulation platform are then performed. The objective is to select the optimal signature generation interval that reduces the impact of cycle to cycle fluctuations in terms of intake valve closing temperature and polytropic index. The obtained results are promising and clearly show the performance and robustness of the signature based trapped mass estimator that can provide relevant feedback information for adaptive engine control systems. It can be easily implemented for real‐time monitoring and control in industrial automotive applications.