Micro-Electro-Mechanical System (MEMS) devices have numerous advantages including small sizes, high performance and easy integration capability, and thus have been widely used in the Internet of Things. A typical MEMS device usually includes a set of performance parameters and each parameter is sensitive to the device geometries but with different regularities and weights, thus resulting in the complexity of MEMS device design. The conventional design method is mainly based on iterative finite element simulation and optimization, which is time-consuming and inefficient. To address the above issues, a bidirectional artificial neural network (ANN) based method is explored and used as the design method by using a MEMS pressure sensor as a design example. First, a forward ANN with the geometries and performance as the input and output respectively is trained and constructed, which can accurately predict the performance. Then, an inverse ANN with the performance and geometries as the input and output respectively is also investigated. By means of tandem network, the non-uniqueness issue of the inverse ANN caused by one-to-many response from the input to the output can be well addressed. This tandem network can output the corresponding geometries instantly according to the target performance. This work demonstrates the great potential of ANN as a new and facile strategy in MEMS device design.

To achieve the desired characteristics for MEMS sensors, traditional design process obtains the geometrical parameters based on complex theoretical calculations and interactive finite element (FE) simulations, which are time-consuming and data-consuming. To solve the above problems, a data-driven bidirectional design approach based on deep learning (DL) method is introduced to improve the design efficiency of MEMS sensors in this work. By using the piezoresistive acceleration sensor as a design example, the forward artificial neural network (ANN) with the sensor geometrical parameters as the input and the sensor performance as the output is trained and realized by using 1000 groups of data collected through FE simulation. This forward ANN can accurately predict the sensor performance including the measurement range, sensitivity, and resonant frequency. In addition, the inverse ANN with the sensor performance as the input and the sensor geometrical parameters as the output is also achieved by using a tandem network. This inverse ANN can provide the geometrical parameters directly and instantly according to the target performance. Both the forward and inverse networks cost only about 6 ms for each task and the mean relative errors are less than 3%. The high efficiency and low relative error indicate that DL is a promising approach to improve the design efficiency for MEMS sensors.

Due to the increasing resolution, the duty cycle in digital pulse width modulation (DPWM) is affected by the variation of exter-nal clock frequency or temperature, and the time error becomes larger and larger, even reaches a few nanoseconds. The incre-ment of duty cycle will affect the regulation performance of the converter and the output of DPWM. In this paper, a delay-line DPWM architecture with a compensation module and a delay-adjustable unit based on Delay Locked Loop (DLL) is proposed. The delay-adjustable unit is realized by using a multiplexer and some delay paths with different delay time, which effectively reduces the influence of temperature or the frequency changes from input clock. Furthermore, a time compensation method is used to reduce the error generated by the critical path delays. A 10-bit DPWM with 781 kHz switching frequency is achieved on A-7 (xc7a100tfgg484) Xilinx FPGA, and the time error of the architecture decreases to around 500 ps. The duty cycle range is from 1.63% to 98.44%.

Bionic electronic skin can accurately sense and locate surface pressure, which is widely demanded in many fields. Traditional electronic skin design usually relies on grid-architecture sensor arrays, requiring complex grid and interconnection arrangements as well as high cost. Grid-less planar sensors can solve the problem by using electrodes only at the edges, but they usually require the use of mapping software such as electrical impedance tomography to achieve high precision. In this work, a design method of high-precision grid-less planar pressure sensors based on the back-propagation (BP) neural network is proposed. The measurement precision of this method is demonstrated to be over two orders of magnitude higher than that of a grid-structure sensor array with the same electrode distribution density. Moreover, this method can be used for irregularly-shaped and non-uniform sensors, which further reduces the manufacturing difficulty and increases the application flexibility.

Designing MEMS sensors to meet specific performance requirements is essential. Traditional approaches, which rely heavily on expert knowledge and extensive finite-element simulations, are often time-consuming. Current deep learning methods in MEMS design typically focus on finding a single feasible solution, neglecting the need to generate multiple solutions simultaneously, which is critical in practical design scenarios. This paper presents a methodology to address these limitations, introducing a hybrid network called conditional variational autoencoder and generative adversarial network (CVAE-GAN), along with a multi-solution generator (G-MS). The CVAE-GAN enables high-accuracy and high-efficiency inverse design, while the G-MS, with its tailored noise updating strategy, generates multiple distinct feasible solutions for given performance criteria. This methodology has been experimentally validated on piezoresistive MEMS accelerometer, finding the second solutions in 3.60 ± 2.46 s, with a normalized distance of 0.75 ± 0.19, improving the existing method as much as 3.63× and 7.19×, respectively. While traditional methods struggle to find more than two solutions, our G-MS can continuously output solutions according to the specified number, with the time taken to find each solution remaining nearly constant. This approach demonstrates the capability to quickly generate multiple accurate structural parameters based on desired performance, showcasing significant potential and providing valuable insights for MEMS sensor design.

This paper presents a micro-electro-mechanical systems (MEMS) piezoresistive differential micro-pressure sensor (0-9kPa) with exceptional linearity over a wide operating temperature range ( $- 55~^{\circ }$ C to $125~^{\circ }$ C). Conventional sensors suffer from linearity degradation caused by nonlinear membrane deflection under applied pressure and accumulation of thermal stress. To address these challenges, we propose a sensor structure incorporating rectangular islands and grooves. The rectangular islands enhance structural rigidity to reduce nonlinear mechanical deflection, while the grooves serve as localized stress-releasing structures to effectively mitigate thermal stress. Both finite element simulations and experimental results confirm the effectiveness of this approach. The fabricated sensor demonstrates a sensitivity of 0.65mV/V/kPa and an ultra-low nonlinearity of 0.022% FSS (full-scale span) at room temperature, and still maintains a quite low nonlinearity of 0.068% FSS at $- 55~^{\circ }$ C and 0.072% FSS at $125~^{\circ }$ C, respectively. These results demonstrate its potential for high-precision micro-pressure measurements over a wide operating temperature in demanding aerospace environments. [2025-0208]