In this work, we present an experimental study on hydrogenated amorphous silicon (a-Si:H) junction field-effect transistors (JFETs), focusing on the gate-controlled depletion of the conductive channel. Low-voltage electrical measurements were performed on devices with different n-layer thicknesses, allowing the channel conductance to be determined as a function of the gate bias. From these data, the evolution of the undepleted channel thickness was estimated and compared with the prediction of an analytical depletion model.Three distinct regimes where needed to describe the observed curves: a linear behavior consistent with a first-order approximation of the analytical expression, an exponential decay observed at the highest depletion conditions, and a transition regime, which maintains mathematical continuity between the two previous regimes.The proposed formulation captures the main experimental features and supports the development of circuit-compatible models for signal processing in lab-on-chip applications.
This paper presents a fully embedded real-time person tracking pipeline for assistive quadrupedal robots supporting safe navigation for visually impaired users. Our approach combines a deep learning-based 2D LiDAR person detector with a lightweight multi-object tracker and integrates it into a Guide Dog Robot (GDR) navigation framework. A novel detection post-processing scheme is proposed, reducing detector latency by 52.38% compared to state-of-the-art voting-based methods while preserving accuracy. The improved latency enables the entire pipeline to operate reliably at 20 Hz on a resource-constrained mobile robotic embedded platform based on the NVIDIA Jetson Xavier NX. The experimental setup shows that our system tracks dynamic obstacles and continuously localizes the user holding the robot’s handle, enabling a dynamic safety footprint for proactive collision avoidance. Under the tested setup, the optimal configuration achieves a MOTA of 83.27% and a user tracking RMSE below 0.2 m on two custom datasets recorded with motion-capture ground truth. Real-world navigation experiments in indoor environments demonstrate effective collision prevention and smooth corrective maneuvers when the user drifts from the default following position. The modular design of the detection, tracking, and planning components ensures flexibility and ease of integration into other robotic platforms. This work contributes a scalable and efficient tracking and navigation solution for human-aware mobile robots operating in dynamic environments, supporting safer human-robot interaction in assistive contexts.
This work presents a novel MEMS-based microgripper integrating electrostatic actuation and capacitive motion sensing, both achieved through interdigitated rotary comb structures. The device features curved flexure hinges for tip motion and ten pairs of electrostatic comb drive actuators for independently controllable tip displacement on both sides within a 130 mu m x 75 mu m workspace. Four pairs of double-sided rotary combs provide real-time position feedback of both tips, achieving a sensitivity of 5.2 fF/mu m. Finite Element Method (FEM) simulations confirm a maximum tip displacement of 65 mu m under 12 V, and a torque generation of up to 3.9 x 10(-10) Nm. Mechanical stress analysis shows that the design remains well within the structural limits of silicon, ensuring long-term reliability. The proposed device enables closed-loop control of jaw position, offering a compact and robust solution for high-precision micromanipulation tasks, such as single-cell stimulation, microsurgery, and lab-on-chip biological analyses.
The demand for hundreds of tightly synchronized channels operating at tens of MSPS in ultrasound systems exceeds conventional low-voltage differential signaling links’ bandwidth, pin count, and latency. Although the JESD204B serial interface mitigates these limitations, commercial FPGA IP cores are proprietary, costly, and resource-intensive. We present ListenToJESD204B, an open-source receiver IP core released under a permissive Solderpad 0.51 license for AMD Xilinx Zynq UltraScale+ devices. Written in synthesizable SystemVerilog, the core supports four GTH/GTY lanes at 12.8 Gb/s and provides cycle-accurate AXI-Stream data alongside deterministic Subclass 1 latency. It occupies only 107 configurable logic blocks (approximately 437 LUTs), representing a 79% reduction compared to comparable commercially available IP. A modular data path featuring per-lane elastic buffers, SYSREF-locked LMFC generation, and optional LFSR descrambling facilitates scaling to high lane counts. We verified protocol compliance through simulation against the Xilinx JESD204C IP in JESD204B mode and on hardware using TI AFE58JD48 ADCs. Block stability was verified by streaming 80 MSPS, 16-bit samples over two 12.8 Gb/s links for 30 minutes with no errors.
Cardiac auscultation with a digital stethoscope is an important method for early diagnosis of cardiovascular diseases (CVDs), especially in resource-constrained settings. However, its utility is often impaired by unwanted noise (e.g., lung sounds, ambient speech) obscuring critical heart signals in the recordings. This paper addresses the challenge of denoising heart sounds in noisy environments and performing this processing on-device for minimal delay and improved data privacy and security. We propose a deep learning-based denoising method using U-Nets based on the Spleeter architecture, which separates heart sound signals from noise in the time-frequency domain. The models are implemented on two edge computing platforms - a Raspberry Pi 4 B and a Qualcomm RB3 Gen 2 - demonstrating the feasibility of AI-driven heart sound denoising at the edge. We further optimize the models via quantization to Int8 precision to leverage the RB3 Gen 2's neural processing unit (NPU) for acceleration. Experimental results show that the denoisers achieve significant enhancement in signal quality, with each yielding an average improvement in scale-invariant signal-to-distortion ratio (SI-SDR) exceeding 10 dB. The edge deployment meets real-time constraints, with the quantized model on the RB3 Gen 2 executing denoising substantially faster (at least 17 times faster than the Raspberry Pi implementation) and at lower power compared to the full-precision model, all without appreciable loss in denoising performance - 0.06 dB being the largest drop. These results demonstrate the potential of deploying advanced AI denoising on edge devices for immediate, on-device heart sound denoising.
