In contrast, digital (CMOS-based) circuits have no quiescent currents and are driven by Moore’s law, yielding an exponential increase in computational power at an exponentially diminishing dissipation per digital operation over the last several decades. Unfortunately, Moore’s law did not contribute to any significant performance improvements in the RF power domain. In contrast, digital (CMOS-based) circuits have no quiescent currents and are driven by Moore’s law, yielding an exponential increase in computational power at an exponentially diminishing dissipation per digital operation over the last several decades. Unfortunately, Moore’s law did not contribute to any significant performance improvements in the RF power domain.
Namely, for the PAs and their pre-drivers, dedicated high-voltage device technologies are used, like laterally diffused MOS (LDMOS) or, more recently, GaN devices. In fact, more than 20 years of LDMOS/GaN technology optimization and 50 years of research to improve the analog/RF circuit performance have only yielded limited progress in TX efficiency. Comparing this progress with the exponential performance increase for digital circuitry, it falls short by orders of magnitude!
The proposed research in the DRASTIC-V2b project aims for the complete bits-in-RF-out digitalization of a wireless transmitter with a segmented LDMOS output stage to fully benefit from Moorse law also in the RF power domain. It builds on revolutionary DTX research ideas recently gained in preceding research at the Delft University of Technology. The related hardware DTX demonstrator, using these ideas, targets a totally new wideband DTX architecture that will be evaluated in a mMIMO base station test bench.