In the present paper we present a methodology for internal heat flux assessment on Pulsating Heat Pipes (PHP) by solving an inverse heat conduction problem for parameter estimation using the Kalman filter. The proposed technique has been chosen to evaluate its accuracy allied to the low time required for real-time estimation purposes since the heat flux provides crucial information for such devices during operation. The inverse problem was solved by considering available temperature measurements on the outer wall of the pipe, and an Infrared Camera performed the acquisition. We adopted two different heat conduction models: a complete model, which is the evolution model for the temperature, and a thin-wall model, which corrects the random walk parameter for the evolution model of the heat flux. The measurement was first synthetically generated by solving the forward problem and adding random Gaussian noise with known standard deviation, emulating the measurement errors. Then the Kalman filter was tested with experimental data. To improve the solution accuracy, we performed the temperature smoothness by a Cubic Smoothing Spline with Cross Validation following the Discrepancy Principle. The results obtained were in good agreement with both synthetic data and experimental data. The technique is promising since it requires a short computational time, allowing its application for real-time heat flux estimation during pulsating heat pipes operation, which provides essential information on correct performance for such devices.