EAGER: Adaptive Performance Models of Sensing Systems for Design Space Exploration

这项研究将在互补金属氧化物半导体（CMOS）技术中实现“芯片实验室”ChemFET（化学敏感场效应晶体管）传感器阵列，其中包括具有校准和读出电路的传感器。沉积在表面上的选择性膜或涂层将增强对目标分子或离子的选择性，以实现多参数传感器阵列。至于这些生物传感器的潜在应用，低成本/低功率传感器可用于环境的众源监测。对社会的好处将是将无处不在的传感的可行性变为现实。获取的传感数据可以提供有关环境的信息，并提醒人们注意危险情况。此外，这些信息可以通过安全和环境监测、生产控制和医疗保健等应用进行挖掘和分析。最后，拟议的研究将用于创建一个研究生水平的课堂体验，并通过亚利桑那州科学实验室的免费研讨会。探索性项目工作的技术影响的长期潜力是在设计空间探索的传感系统使用的方法从设计自动化。传感器和模拟集成电路的协同设计和模型将使用性能指标进行检查。通过这项工作，传感系统的设计过程可以从完全自下而上的方法转变为从单个传感器到电路再到通信设备的整个系统的系统优化。此外，系统参数可以动态调整，以适应传感器漂移、环境因素（如温度变化、电源和网络可用性）影响下的行为和性能。通过使用预测模型和算法来管理系统行为和操作条件，可以在传感系统设计和实现中向新的范例迈进。

This research will implement a "lab on a chip" ChemFET (Chemically sensitive Field Effect Transistor) sensor array in a Complementary Metal Oxide Semiconductor (CMOS) technology that includes sensors with calibration and read-out circuitry all in a common substrate. Selective membranes or coatings deposited on the surface will enhance selectivity for a target molecule or ion to achieve a multi-parametric sensor array. As for potential applications of these biosensors, low cost/low power sensors can be used to crowd source monitoring of the environment. The benefit to society will be in terms of bringing the feasibility of ubiquitous sensing to reality. The acquired sensing data can provide information about the environment and alert people to dangerous conditions. Further, this information can be mined and analyzed with applications in safety and environmental monitoring, production control, and healthcare. Finally, the proposed research will be used to create a classroom experience at the graduate level, and a free workshop though the Arizona Science Lab.The longer term potential of technical impact of the exploratory project work is in the design space exploration for the sensing systems using methods from design automation. Co-design and models of sensors and analog integrated circuits will be examined using performance metrics. Through this work the design process for sensing systems can be transformed from a completely bottom up approach to a systematic optimization of the entire system from individual sensors to circuits to communication devices. In addition, the system parameters can be adjusted on the fly to adapt the behavior and performance under the influence of sensor drift, environmental factors such as temperature variation, power supply and network availability. Through the use of predictive models and algorithms to govern the system behavior and operating conditions, strides toward a new paradigm in sensing system design and implementation can potentially be made.