This paper aims at solving the sparse reconstruction (SR) problem via a multiobjective evolutionary algorithm. Existing multiobjective evolutionary algorithms for the SR problem have high computational complexity, especially in high-dimensional reconstruction scenarios. Furthermore, these algorithms focus on estimating the whole Pareto front rather than the knee region, thus leading to limited diversity of solutions in knee region and waste of computational effort. To tackle these issues, this paper proposes an adaptive decomposition-based evolutionary approach (ADEA) for the SR problem. Firstly, we employ the decomposition-based evolutionary paradigm to guarantee a high computational efficiency and diversity of solutions in the whole objective space. Then, we propose a two-stage iterative soft-thresholding (IST)-based local search operator to improve the convergence. Finally, we develop an adaptive decomposition-based environmental selection strategy, by which the decomposition in the knee region can be adjusted dynamically. This strategy enables to focus the selection effort on the knee region and achieves low computational complexity. Experimental results on simulated signals, benchmark signals and images demonstrate the superiority of ADEA in terms of reconstruction accuracy and computational efficiency, compared to five state-of-the-art algorithms.

Micro-droplet generation is related to liquid dispensing technology that has potential applications in many fields. Specifically, pneumatic micro-droplet generation is controlled by a solenoid valve being briefly turned on, so that high pressure gas enters the liquid reservoir, forming a gas pressure pulse waveform P(t), forcing the liquid out through a tiny nozzle to form a micro-droplet. For each ejection, P(t) is acquired by a high speed pressure sensor, and the ejection state is obtained by machine vision methods. A prediction model based on BP neural network is established, with P(t) as input and the droplet ejection state as output. Experiments show that the BP neural network can predict the number of droplets with an accuracy higher than 99%. It is also shown that the BP neural network can improve the prediction accuracy for the position of droplets relative to the nozzle, at a given moment. Under typical working conditions, P(t) is not consistent. As a result, the ejection state is not consistent either. These prediction models may be used for real time monitoring and control of the pneumatic micro-droplet generator.

The noise-shaping (NS) successive approximation register (SAR) analog-to-digital converter (ADC) is an innovative hybrid structure that offers performance advantages. The NS-SAR ADC leverages the SAR ADC as its foundation and combines oversampling technology and noise-shaping technology found in Sigma-Delta ADC. This integration effectively combines the strengths of both structures and enhances overall performance. The ADC features a simple circuit structure, compact chip area, and high energy efficiency, which has positioned it as a prominent research area. In this paper, leveraging the TSMC 65 nm GP process, the NS-SAR ADC is designed with a power supply voltage of 1 V. This design adopts an 8-bit differential capacitor structure, operates at a sampling frequency of 16 MS/s, and achieves an oversampling rate of 16 times the desired performance indicators. Through extensive circuit post simulation verification, the SNR obtained reaches 78 dB, providing an effective bit resolution of 12.7 bits. The core chip area of the ADC measures 366 × 333 μm², while the power consumption is impressively low at 417 μW and FoMs is 168 dB.