Servicenavigation

Project Details

Back

Projekt Print View

Design of Circuits and Systems for Nonvolatile Nanomagnetic Logic

Subject Area Electronic Semiconductors, Components and Circuits, Integrated Systems, Sensor Technology, Theoretical Electrical Engineering
Term from 2013 to 2017
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 229838035
 
Final Report Year 2018

Final Report Abstract

Even though computation in the ferromagnetic domain is mainly unidentified in the domain of IC design, the presented results may provide a promising technology platform for future 3D integrated circuits and systems especially when implemented in a co-processor, systolic or reconfigurable architecture. In order to point towards system level implementations, we developed computationally efficient compact-models, calibrated against our current NML technology. The models were implemented on three levels of abstraction, (1) MATLAB/SIMULINK to address the Arrhenius-type switching, (2) Verilog-A to provide the link to a de-facto industry standard implementation and (3) VHDL for digital implementation with reduced complexity of the underlying switching probability. Even more, our partners in Torino implemented a front-end called MagCAD to ease the circuit layout for NML circuits. With that, we were able to provide a platform to benchmark digital NML circuits to their CMOS counterpart. CMOS and electronic beyond-CMOS devices are heavily researched. With 3D integrated pNML devices and circuits, we were able to show, that our devices in the present state are competitive. Especially highly pipelined and systolic architectures have proven to be efficient for pNML implementations. Furthermore, we do not see a severe showstopper in this stage of technology readiness level (TRL3-TRL4), as we were able to demonstrate 3D logic operation, signal propagation by domain walls, the option for bipolar on-chip clocking with homogeneous field-clock and benefits compared to conventional CMOS in low-power and monolithic 3D integration. However, everyone familiar with ultimate-scale-integration is sceptic of paradigmchanges in digital computation, simply because MOS technology as electronic switch is such thoroughly understood. However, we think that other, non-Boolean concepts of computation are also extremely abundant. Magnetic devices are more than attractive for such implementations, as they combine very interesting characteristics like robustness against radiation, bi-stability, scaling potential, room-temperature operation combined with very rich high-frequency dynamics. Hence, it is worth to study those and exploit them for real-world applications in future.

Publications

 
 

Additional Information

© 2024 DFG Disclaimer / Copyright / Privacy Policy

Textvergrößerung und Kontrastanpassung