The authors address the functional testing of microprocessors. A method is introduced for obtaining a minimum set of instructions that replaces the whole instruction set during testing procedure. The method is illustrated on the digital signal processor TM32010.>

A speaker independent isolated Arabic word recognition system has been successfully implemented and tested. End-point detection and voiced/unvoiced segmentation algorithms are based on energy and zero-crossing rate (ZCR) using adaptive ZCR threshold. Formants are extracted using ZCI histograms obtained from filtered speech signal [1]. The speaker independent recognition rate is about 96% for ten Arabic words.

A new ARMA model is proposed to get the PSDF of noisy random ergodic zero mean discrete time signals. In this model, the residual power which is not covered by the AR polynomial is represented by a limited order MA polynomial. A new algorithm is proposed and used in minimizing the residual power which is still not represented by the added limited MA polynomial. A comparison is made between the proposed model and that of Kay [9] which showed that for the same degree of performance, the new model needs much less storage space and computation time.