Chip-level mass detection for micro-LED displays based on regression analysis and deep learning.

Though micro-light-emitting diode (micro-LED) displays are regarded as the next-generation emerging display technology, challenges such as defects in LED's light output power and radiation patterns are critical to the commercialization success. Here we propose an electroluminescence mass detection method to examine the light output quality from the on-wafer LED arrays before they are transferred to the display substrate. The mass detection method consists of two stages. In the first stage, the luminescent image is captured by a camera by mounting an ITO (indium-tin oxide) transparent conducting glass on the LED wafer. Due to the resistance of the ITO contact pads and on-wafer n-type electrodes, we develop a calibration method based on the circuit model to predict the current flow on each LED. The light output power of each device is thus calibrated back by multi-variable regression analysis. The analysis results in an average variation as low as 6.89% for devices predicted from luminescent image capturing and actual optical power measurement. We also examine the defective or non-uniform micro-LED radiation profiles by constructing a 2-D convolutional neural network (CNN) model. The optimized model is determined among three different approaches. The CNN model can recognize 99.45% functioning LEDs, and show a precision of 96.29% for correctly predicting good devices.

CIM-Based Smart Pose Detection Sensors.

The majority of digital sensors rely on von Neumann architecture microprocessors to process sampled data. When the sampled data require complex computation for 24×7, the processing element will a consume significant amount of energy and computation resources. Several new sensing algorithms use deep neural network algorithms and consume even more computation resources. High resource consumption prevents such systems for 24×7 deployment although they can deliver impressive results. This work adopts a Computing-In-Memory (CIM) device, which integrates a storage and analog processing unit to eliminate data movement, to process sampled data. This work designs and evaluates the CIM-based sensing framework for human pose recognition. The framework consists of uncertainty-aware training, activation function design, and CIM error model collection. The evaluation results show that the framework can improve the detection accuracy of three poses classification on CIM devices using binary weights from 33.3% to 91.5% while that on ideal CIM is 92.1%. Although on digital systems the accuracy is 98.7% with binary weight and 99.5% with floating weight, the energy consumption of executing 1 convolution layer on a CIM device is only 30,000 to 50,000 times less than the digital sensing system. Such a design can significantly reduce power consumption and enables battery-powered always-on sensors.