Systematic Design and Yield Optimization of RF and Millimeter-Wave Oscillators Using Neuro-Genetic Algorithm

This paper presents a systematic procedure to optimize the design of RF and millimeter wave (MMW) integrated oscillators, and, at the same time, to maximize their yield using a neuro-genetic algorithm for the design centering. The design optimization procedure has been explained and demonstrated using a multi-band switched-inductor Colpitts VCO as well as a 30GHz cross-coupled VCO. The significance of the layout parasitic components in MMW oscillators has been described. An extended optimization procedure has been derived based on the sensitivity analysis involving the design components as well as the estimated layout parasitic elements. The pre-fabrication design yield for the given specifications is maximized using a genetic algorithm with neural models for the active circuit. The design yield is improved up to 90% with the developed neuro-genetic procedure

[1]  M. Meehan.  Understanding and maximising yield through design centering (microwave circuits) , 1991 .

[2]  Jian Wang,et al.  Theory of cross-coupled RF oscillator for multi- and quadrature-phase signal generation , 2003, ASICON 2003.

[3]  P. Sen,et al.  Reconfigurable RFICs in Si-based technologies for a compact intelligent RF front-end , 2005, IEEE Transactions on Microwave Theory and Techniques.

[4]  Qi-Jun Zhang,et al.  Neural Networks for RF and Microwave Design , 2000 .

[5]  J. Laskar,et al.  Millimeter wave RF front end design using neuro-genetic algorithms , 2005, Proceedings Electronic Components and Technology, 2005. ECTC '05..

[6]  Wei-Liem Loh.  On Latin hypercube sampling , 1996 .

[7]  Stephen W. Director,et al.  A new methodology for the design centering of IC fabrication processes , 1989, 1989 IEEE International Conference on Computer-Aided Design. Digest of Technical Papers.