HEMT models for S-parameter and noise parameter extrapolation

@inproceedings{citeulike:2782066, abstract = {Four models have been developed and assessed for fitting the measured noise parameters up to 26 GHz and S-parameters up to 400 GHz for a commercial HEMT (high electron mobility transistor) chip. The first treats the intrinsic noise sources as uncorrelated thermal sources. The second is an extension of this, allowing a better fit to be achieved by including the distributed nature of the gate and drain electrodes using a semidistributed, sliced model. The third model neglects the distributed effect but takes into account the partial correlation of the gate and drain noise sources. This causes a larger improvement in the quality of fit, allowing the model to fit the measured data within reasonable measurement limits. The addition of the distributed effect to the correlated model gives the fourth model, which allows a further marginal improvement, but the conditioning of the problem and the accuracy of the data appear to be insufficient to allow accurate extraction of the additional parameters needed}, author = {Hickson, M. T. and Gardner, P. and Paul, D. K. }, booktitle = {Microwave Symposium Digest, 1992., IEEE MTT-S International}, citeulike-article-id = {2782066}, doi = {10.1109/MWSYM.1992.187966}, journal = {Microwave Symposium Digest, 1992., IEEE MTT-S International}, keywords = {model, noise, parameter}, pages = {277--280 vol.1}, posted-at = {2008-05-10 01:04:23}, priority = {2}, title = {HEMT models for S-parameter and noise parameter extrapolation}, url = {http://dx.doi.org/10.1109/MWSYM.1992.187966}, year = {1992} }

TY - CONF ID - citeulike:2782066 TI - HEMT models for S-parameter and noise parameter extrapolation T2 - Microwave Symposium Digest, 1992., IEEE MTT-S International JF - Microwave Symposium Digest, 1992., IEEE MTT-S International SP - 277 EP - 280 vol.1 N2 - Four models have been developed and assessed for fitting the measured noise parameters up to 26 GHz and S-parameters up to 400 GHz for a commercial HEMT (high electron mobility transistor) chip. The first treats the intrinsic noise sources as uncorrelated thermal sources. The second is an extension of this, allowing a better fit to be achieved by including the distributed nature of the gate and drain electrodes using a semidistributed, sliced model. The third model neglects the distributed effect but takes into account the partial correlation of the gate and drain noise sources. This causes a larger improvement in the quality of fit, allowing the model to fit the measured data within reasonable measurement limits. The addition of the distributed effect to the correlated model gives the fourth model, which allows a further marginal improvement, but the conditioning of the problem and the accuracy of the data appear to be insufficient to allow accurate extraction of the additional parameters needed KW - model KW - noise KW - parameter AU - Hickson, MT AU - Gardner, P AU - Paul, DK PY - 1992/// UR - http://dx.doi.org/10.1109/MWSYM.1992.187966 ER -