An Efficient Breakdown Characterization for D/T-RESURF SOI LDMOS Using Important Structure Parameters Selection Technique

Double/triple-reduced surface field (D/T-RESURF) techniques have been proposed to improve the breakdown characteristics of conventional silicon on insulator (SOI) lateral double-diffused metal–oxide–semiconductor (LDMOS). However, the modeling and design complexity also increases with these RESURF structures. In this article, to improve the electrical performance prediction efficiency and reduce the design complexity, an automatic essential structure parameters filtering technique is proposed for the breakdown performance of D/T-RESURF SOI LDMOS. Given the input structure parameters, the crucial structure parameters that significantly impact the breakdown performance will be provided. The results indicate that by removing unimportant structural parameters, the breakdown characteristics are also kept stable. Meanwhile, the model’s average application time is accelerated almost by 38.3% since the prediction model is simplified with less data. Besides, the technique also shows great prediction ability in breakdown location and breakdown voltage (BV) without model prediction accuracy degradation by removing the unimportant structure parameters. Moreover, the method can also be extended to other devices to simplify the design inputs. We believe that the approach can effectively guide and accelerate the design process.