For the development of a very high efficiency engine, the continuous monitoring of the engine operating conditions is needed. Moreover, the early detection of engine faults is fundamental in order to take appropriate corrective actions and avoid malfunctioning and failures.The in-cylinder pressure is the most direct parameter associated to the engine thermodynamic cycle. The cost and the intrusiveness of the dynamic pressure sensor and the harsh operating condition that limits its life-time, make the direct measurement of the in-cylinder pressure not suitable for mass production applications.Consequently, research is oriented on the measurement of physical phenomena linked to the thermodynamic cycle to obtain useful information for the ICE control.For turbocharged engine application, the direct connection between the thermo-dynamic and fluid-dynamic conditions at the engine cylinder exit and the turbocharger behavior suggests that turbocharger instantaneous speed could give useful information about the engine cycle. By considering diesel engines, strong attention is paid to the injectors, which operating characteristics vary with respect to the nominal one usually due to the injector individual tolerance and the time degradation. The correct operation of the injectors and the control of the injected fuel quantity, allow ensuring the right combustion process and consequently maintaining engine performance at design-level in the long term.In previous studies, a preliminary investigation of the relationship between the engine performance and the turbocharger speed of a four-stroke multi-cylinder turbo-diesel engine was carried out. It was assessed that a condition monitoring and fault detection strategy based on the turbocharger (TC) speed measurement can be defined. Both the numerical and experimental results confirm that the instantaneous TC speed is a potentially suitable indicator for the detection of the cylinder-to-cylinder injection variation.In the present study, by exploiting the calibrated numerical model, several cylinder-to-cylinder injection variations were considered in the whole engine operating range in order to define a robust injection monitoring strategy based on the instantaneous TC speed for the estimation of the actual injected fuel quantity in each cylinder.The results of the Fast Fourier Transform (FFT) processing methodology are here reported. The method is based on the frequency analysis of the instantaneous TC speed signal. By taking advantage of the periodicity of the signal, its treatment on the frequency domain could simplifies the injection monitoring strategy. Indeed, the FFT has a low computational cost and has been used successfully in many other applications.