A speed and high resolution position identification algorithm, suitable for brushless drives, is presented in this paper. The algorithm, which is developed basing upon the Popov's Hyperstability Theory, guarantees the convergence of the estimated signals to their corresponding actual values by means of the measured currents. The algorithm is proposed for brushless DC (BLDC) motors which are characterized by emfs shapes different from the ideal trapezoidal ones. The identification procedure can be performed starting from the knowledge of both low resolution rotor position Hall signals and emf shape. The identification algorithm is properly tested on a BLDC drive controlled by a predictive algorithm, by performing a simulation study in the Matlab-Simulink environment. The corresponding results have highlighted the effectiveness of the proposed sensorless control system at both low and high speed operation.