ANN based demodulator for UMTS signal measurements

The paper proposes the use of the Artificial Neural Network (ANN) to design a new demodulator architecture of the telecommunication signals. This demodulator is designed to operate into the measurement instruments of modulated signals according to the Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access, that is the third generation mobile telephone standard. It is developed on the basis of the Multi-Layer Perceptron ANN and it permits to perform in one shot both the descrambling and the despreading operations. Numerical tests performed by using signals according to the UMTS standard show that the ANN demodulator can guaranty greater accuracy than the traditional one in the case the modulated signal is affected by high level of both noise and distortion.