Surrogate Based Automated Tuning of Microwave Devices and Systems

基于代理的微波设备和系统自动调谐

• 批准号: RGPIN-2020-06853
• 负责人: Laforge, Paul
• 金额: $ 2.04万
• 依托单位: University of Regina
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=717070
• 关键词: Surrogate; Based; Automated; Tuning; Microwave

The main goal of this research program is to investigate surrogate based automated tuning systems for tuning very complex microwave devices and systems to replace the existing practice of using highly skilled human experts to perform the tuning. Microwave devices are essential in the operation of communication systems including wired and 5G networks, air and ground-based radar systems, medical imaging systems and sensor applications. The majority of these devices require some form of post-fabrication tuning due to fabrication and material tolerances to meet the specific requirements of each system. The current practice is to use highly skilled human experts for very complex devices with 50+ tuning elements or to use computer aided tuning (CAT) systems for simple devices with a single tuning objective and few tuning elements. This research program proposes to investigate new CAT systems to improve the device and system performance and decrease the tuning time and associated expenses. The research work proposed in this research program is novel as surrogate based tuning algorithms implementing the implicit space mapping (ISM) technique will be investigated and implemented for the first time to fully automate the tuning of very complex devices with 50+ tuning elements, to automate the tuning of devices with discrete tuning elements, to automate the tuning of devices with permanent tuning, and to automate the tuning of amplifier circuits with multiple tuning specifications. The proposed CAT systems will make use of the well-known ISM technique, which has been traditionally used for the design and optimization of microwave devices. For these tuning systems, the fabricated microwave device will replace the fine model electromagnetic (EM) simulator structure when implementing the ISM technique. While providing training for highly qualified personnel (HQP), the main objectives of this research program are to propose, investigate, develop and validate: (1) novel space mapping surrogate models, parameter extraction and mapping techniques for tuning three dimensional microwave structures and microstrip circuits with a large number of tuning elements (50+ tuning elements), (2) novel update procedures in the space mapping (SM) technique for tuning microwave devices that make use of discrete tuning values and/or permanent, one-way tuning, and (3) novel surrogate based automated tuning systems for microwave amplifiers using multi-objective optimization techniques. The transfer of this technology to Canadian industry will result in a savings in time and expenses for training highly skilled human experts that currently perform the tuning for these types of microwave devices. The fabrication and assembly times will be reduced for devices and systems with the integration of these CAT systems enabling industry a quicker path to market for new designs and more cost effective ways for existing designs.

本研究计划的主要目标是研究基于替代的自动调谐系统，用于调谐非常复杂的微波设备和系统，以取代使用高技能的人类专家进行调谐的现有实践。