Browsing by Author "Xu, Wei"

Now showing 1 - 7 of 7

A CMOS-MEMS Thermoresistive Micro Calorimetric Flow Sensor With Temperature Compensation
(2019-10-01)
Xu, Wei
;
Ma, Shenhui
;
Wang, Xiaoyi
;
Chiu, Yi
;
Lee, Yi-Kuen
In this paper, we propose a new 1D thermoresistive micro calorimetric flow (TMCF) sensor model for moist airflow with the temperature-dependent thermophysical properties. The 1D model and the corresponding proposed equivalent circuit model enable rapid design optimization of an ambient temperature-compensated TMCF ((TMCF)-M-2) sensor integrated with the interface circuit by using a commercial 0.35-mu m 2P4M CMOS MEMS technology. The fabricated sensor shows a prominent sensitivity of 0.543 V/(m/s) and a low power consumption of less than 2.8 mW. Moreover, in comparison with the large drifting of 49% for the uncompensated TMCF sensor, the (TMCF)-M-2 sensor's output has an ultralow drifting of 0.5% with the ambient temperature variation of 22 degrees C similar to 48 degrees C. Therefore, this robust (TMCF)-M-2 sensor can be a promising Internet of Things (IoT) for smart energy-efficient buildings.
Achievable Rate Analysis and Feedback Design for Multiuser MIMO Relay with Imperfect CSI
(2014-02-01)
Peng, Zhangjie
;
Xu, Wei
;
Wang, Li-Chun
;
Zhao, Chunming
;
電機資訊學士班
This paper investigates a multiuser MIMO relay downlink system with imperfect channel estimation and limited feedback. We analyze the achievable rate loss due to channel state information (CSI) mismatch arising from both channel estimation and quantization feedback. We first derive an upper bound to characterize the effect of imperfect CSI, and then present a limited feedback strategy for both the two-hop channels in the relay system. This newly presented feedback strategy reveals the relationship among the CSI quantization levels, the transmit power, and the pilots for channel estimation. An optimized pilot design at the base station and the relay is also presented by applying the derived bounds for the two-hop system.
High Sensitivity and Wide Dynamic Range Thermoresistive Micro Calorimetric Flow Sensor With CMOS MEMS Technology
(2020-04-15)
Xu, Wei
;
Wang, Xiaoyi
;
Chiu, Yi
;
Lee, Yi-Kuen
;
電機工程學系
In this paper, a thin film based Thermoresistive Micro Calorimetric Flow (TMCF) sensor by using a 0.35 mu m 2P4M CMOS MEMS technology is designed, fabricated and tested. The developed 1.6 mu m thick thin film CMOS TMCF sensor shows a maximum normalized sensitivity of 103 mV/(m/s)/mW with respect to the input heating power and a wide dynamic flow range of 0 similar to 19.4 m/s. The high sensitivity and wide dynamic range achieved by our CMOS MEMS flow sensor enable its deployment as a promising sensing node for HVAC application in the smart energy-efficient building (EeB). Moreover, a new one-dimensional (1D) analytical sensor model that considers the three-dimensional (3D) heat loss via the end of the thin film is proposed. The new 1D model can successfully capture the thin film based TMCF sensor's response, and its prediction results show a good agreement with the experimental study. Therefore, this new 1D model will be a useful design tool for the development of a high-performance TMCF sensor with a greatly reduced footprint.
LOW-COST TEMPERATURE-COMPENSATED THERMORESISTIVE MICRO CALORIMETRIC FLOW SENSOR BY USING 0.35 mu m CMOS MEMS TECHNOLOGY
(2016)
Xu, Wei
;
Gao, Bo
;
Ma, Shenhui
;
Zhang, Anping
;
Chiu, Yi
In this paper, a low-cost Temperature-compensated Thermoresistive Micro Calorimetric Flow ((TMCF)-M-2) sensor by using a commercial 0.35 mu m 2P4M CMOS MEMS technology is reported. For nitrogen flow, the fabricated (TMCF)-M-2 sensor achieves a normalized sensitivity of 230 mV/(m/s)/mW with respect to the input heating power, which is two orders of magnitude better than the reported micro calorimetric flow sensors. The experimental results of (TMCF)-M-2 sensor under different ambient temperatures T-a of 22 degrees C similar to 48 degrees C showed the excellent temperature insensitive output with the maximum normalized variation of 0.5%. Compared to the uncompensated counterpart (49%), it demonstrates our robust (TMCF)-M-2 sensor design is applicable for the integration of on-chip CMOS electronics for the Internet of Things.
Packaging Effect on the Flow Separation of CMOS Thermoresistive Micro Calorimetric Flow Sensors
(2016-01-01)
Xu, Wei
;
Gao, Bo
;
Lee, Yi-Kuen
;
Chiu, Yi
;
電機工程學系
We systematically study the packaging effect on the performance of CMOS Thermoresistive Micro Calorimetric Flow (TMCF) sensors, including two types of sensor packaging designs: S-type and E-type. For S-type design, the experimental results indicate that not only the sensitivity of TMCF sensors was significantly changed by more than 30%, but also the flow range compared to the E-type design was dramatically reduced from 0 - 11 m/s to 0 -4.5 m/s. Comprehensive CFD simulations and the concise heater transfer analysis were performed to understand the physics behind these two different designs. Accordingly, a flow regime map (the protrusion vs the reduced Reynolds number) was constructed, where the critical boundary curve for the flow separation of CMOS TMCF sensors was determined. This flow regime map is a useful guideline for designing a well packaged flow sensor system.
Systematic study of packaging designs on the performance of CMOS thermoresistive micro calorimetric flow sensors
(2017-08-01)
Xu, Wei
;
Pan, Liang
;
Gao, Bo
;
Chiu, Yi
;
Xu, Kun
We systematically study the effect of two packaging configurations for the CMOS thermoresistive micro calorimetric flow (TMCF) sensors: S-type with the sensor chip protrusion-mounted on the flow channel wall and E-type with the sensor chip flush-mounted on the flow channel wall. Although the experimental results indicated that the sensitivity of the S-type was increased by more than 30%; the corresponding flow range as compared to the E-type was dramatically reduced by 60% from 0-11 m s(-1) to 0-4.5 m s(-1). Comprehensive 2D CFD simulation and in-house developed 3D numerical simulations based on the gaskinetic scheme were applied to study the flow separation of these two packaging designs with the major parameters. Indeed, the S-type design with the large protrusion would change the local convective heat transfer of the TMCF sensor and dramatically decrease the sensors' performance. In addition, parametric CFD simulations of the packaging designs provide inspiration to propose a novel general flow regime map (FRM), i.e. normalized protrusion d* versus reduced chip Reynolds number Re*, where the critical boundary curve for the flow separation of TMCF sensors was determined at different channel aspect ratios. The proposed FRM can be a useful guideline for the packaging design and manufacturing of different micro thermal flow sensors.
Theoretical and Experimental Investigations of Thermoresistive Micro Calorimetric Flow Sensors Fabricated by CMOS MEMS Technology
(2016-10)
Xu, Wei
;
Song, Kui
;
Ma, Shenhui
;
Gao, Bo
;
Chiu, Yi
A general 1-D model was presented to predict the characteristics of CMOS thermoresistive micro calorimetric flow (TMCF) sensor with two types of packaging, i.e., open-space type and channel type, for both gases and liquids. The 1-D model was first validated by a numerical computational fluid dynamics (CFD) model and was subsequently normalized for different fluids flow. Notably, the model proposed by Nguyen and Dotzel is a special case of our 1-D model. The normalized output of TMCF sensor is a function of normalized input parameters of Reynolds number Re and Prandtl number Pr. The scaling analysis of the sensor output, sensitivity, and power consumption was performed to optimize the design of TMCF sensors in terms of key design parameters, including the thin film thickness, the height of bottom cavity, and so on. Accordingly, three pairs of TMCF sensors were designed and fabricated by using a 0.35 mu m 2P4M CMOS microelectromechanical systems technology. The fabricated sensors showed a normalized sensitivity of 230 mV/(m/s)/mW for nitrogen gas flow, which was two orders of magnitude higher than the previous CMOS flow sensors. Therefore, the proposed 1-D model is a promising tool for the sensor\'s system-level design with the on-chip microelectronics for the Internet of Things. [2016-0098]