Partial randomization dynamic element matching (DEM) was recently introduce
d as a promising DEM technique for low harmonic distortion digital-to-analo
g conversion. The DEM technique is well suited for applications such as dir
ect digital synthesis in wireless communication systems for which low hardw
are complexity is essential in addition to low harmonic distortion. Previou
sly reported simulation results demonstrate that partial randomization DEM
greatly attenuates harmonic distortion resulting from static errors in the
analog output levels of the DAC, while offering considerable savings in har
dware compared to other DEM techniques. This paper presents the first quant
itative performance analysis of partial randomization DEM As a main result,
the minimum spurious-free dynamic range provided by the digital-to-analog
converter has been quantified as a function of its hardware complexity and
the analog output level errors.