Self-healing narrowband filters via 3D heterogeneous integration of AlN MEMS and CMOS chips

AlN MEMS CMRs offer high Q and k
_t
²
, and a center frequency (f
₀
) that can be primarily set by lithography. They also have direct 50 Ω matching capability. These properties make AlN CMRs an ideal candidate to synthesize multi-frequency bandpass filters on a single chip with low insertion loss (IL). However, their practical implementation is hindered by fabrication induced variations. To address this challenge, we propose the use of statistical element selection (SES) technique where a subset k from a bank of N nominally identical subfilters are combinatorially selected in parallel via series CMOS switches at RF input and output in order to construct a high yield, self healing filter.