| Resumo: | Performing impedance frequency response measurements with standalone embedded instruments currently presents some challenges due to the limited amount of memory available in these systems. The impedance frequency response can be obtained through a single-tone sweep approach together with a traditional sine-fitting algorithm. However, this approach is not suitable when the measurand is quickly evolving over time because in those cases the acquisition should be performed as fast as possible. This can be achieved by applying a multiharmonic stimulus and using least-squares estimation algorithms to extract the amplitude and phase of each frequency component. However, the memory requirements of these least-squares algorithms do not allow them to be implemented in standalone embedded systems. This paper presents a memory efficient least-squares fitting algorithm that enables the use of multiharmonic signals for impedance frequency response measurements in memory limited embedded systems. The measurement time and memory requirements are compared between the single-tone sweep, traditional multiharmonic and memory efficient multiharmonic approaches for different combinations of the number of harmonics and number of acquired samples. This performance evaluation is also done on impedance frequency response data, acquired and processed in a DSP based device for the single-tone and memory efficient multiharmonic measurement alternatives.
|
|---|