The 6 th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications 15-17 September 2011, Prague, Czech Republic Universal Acquisition System for Frequency Domain Parameters Measurement Linas Svilainis, Vytautas Dumbrava, Andrius Chaziachmetovas Signal Processing Department, Kaunas University of Technology, Studentu str. 50-340, LT-51368, Kaunas, Lithuania, www.sak.ktu.lt E-mail: linas.svilainis@ktu.lt, vytautas.dumbrava@ktu.lt, andrius.chaziachmetovas@ktu.lt Abstract— The universal acquisition system is presented. System is dedicated for multiple frequency domain parameters measurement: transfer, insertion loss, AC response, harmonic distortion, noise and electrical impedance. The primary application was the ultrasonic equipment AC parameters evaluation. Operation range is 20 kHz to 30 MHz. System is targeted for parameters comparison and estimation, but it is universal: all the measurements are accomplished using just one excitation and two analog reception channels. System structure and operation are presented. Modular PC104 form-factor construction allows for both local and remote PC connectivity. Setup for most common tasks and example measurement results with system is presented. Keywords— signals acquisition system; AC response; noise measurement; sine wave correlation; impedance measurement I. INTRODUCTION The parameter variation in frequency domain is one of the essential parameters of every electronic system. Problems in network parameters, distortion, AC response or noise may degrade the system performance. It is quite useful to have the frequency domain parameters monitoring of the device during the design process and during the service period. Primary goal was to design the system for ultrasonic equipment AC parameter evaluation. Therefore 20 kHz to 30 MHz frequency range was indicated as a target. System is dedicated for parameters comparison and estimation, so moderate accuracy was allowed, yet system must be universal. It should be able to measure the complex transmission AC response, signal distortion, noise and impedance of the device under test (DUT). Measurements should be collectable on host PC for further processing and storage. II. THE SYSTEM A. Operation Principle The frequency response at certain frequency can be determined by probing the system input with a single frequency continuous wave (CW) signal while measuring the input/output amplitude ratio and phase difference. In presence of noise the amplitude and phase estimation turns a more difficult task. The sine fitting using the least squares error techniques [1], Fourier transform, interpolated discrete Fourier transform, short-time Fourier Transform [2-4] or wavelet transform [5] are used for parameter estimation. These techniques are used if the sine frequency is not known or the sampling frequency value is not as accurate as would be expected. The amplitude and phase measurement can be simplified and frequency mismatch induced errors reduced if frequency is known. Sine wave correlation (SWC) can be used to get the signal amplitude and phase from the acquired data set. SWC [1] can be treated as signals sampled version s n of length N windowed by w n correlation coefficient with sine and cosine computation, normalized by window L1 norm W 1 :             1 1 0 1 1 0 2 sin 2 2 cos 2 W N w s ft f S W N w s ft f C N n n n n N n n n n               . (1) It was decided to use the excitation generator using the direct digital synthesizer (DDS). Such CW signal is in fixed proportion to system clock frequency. Requirement was put for the system to use common reference oscillator for excitation signal generation and acquisition channel analog-to-digit converters (ADC). Impedance measuring involves current and voltage measurement. It was decided to use the Autobalancing Bridge (ABB) technique [6]. B. Requirements Rectification To ensure system universality, system parameters have to be chosen to match the measurement tasks planned. In order to rectify the system parameters requirements, numerical experiments have been carried out. If additive white Gaussian noise (AWGN) with amplitude U nRMS is present over first Nyquist zone f N at sampling frequency f s then corresponding noise voltage density is [7]