Several 100Base-TX receivers have been built with DFE and FFE functions implemented in DSP. These implementations tend to be large in area and high in power. In addition, the large amount of digital logic switching at high speeds can lead to EMI issues. Finally, since a finite resolution must be chosen in a DSP implementation, noise immunity is reduced due to quantization noise [1,2].This chip uses mixed-signal techniques to implement the DFE and FFE functions of a 100BASE-TX receiver. The 100BASE-TX receiver uses a switched-capacitor, fixed coefficient FFE to cancel precursor ISI and create the timing function, a mixed-signal current-summing fully adaptive DFE to cancel post-cursor ISI, a fast offset cancellation tap to cancel baseline wander, and a coarse comparator to make decisions. Timing information, extracted from the received signal, is input to a digital signal processing (DSP) engine which emulates a second-order phase-locked loop (PLL) function. The function includes both proportional and integral representations of the timing information. A phase-interpolating PLL generates the recovered clock from the output of the DSP engine.
Extracting accurate heart rate estimations from wrist-worn photoplethysmography (PPG) devices is challenging due to the signal containing artifacts from several sources. Deep Learning approaches have shown very promising results outperforming classical methods with improvements of 21% and 31% on two state-of-the-art datasets. This paper provides an analysis of several data-driven methods for creating deep neural network architectures with hopes of further improvements.
The treatment of acid rock drainage (ARD) places extraordinary financial burdens on governments and companies worldwide, and an improved efficiency in treatment by as little as 1% can save many millions of dollars in rehabilitation. We investigated a system for treating Fe-rich ARD using a three-stage reactor design. In the first reaction cell, Fe-rich ARD was partially neutralised using rapid periodic carbonate resuspension with a rotating axial mixer. This was followed by an air-sparged oxidation chamber and then a second reaction cell, with more carbonate periodically resuspended until a pH of 6.3 was reached, which was followed by a settlement chamber. This reactor design has a high capacity for neutralisation, with an efficiency of &70% of acidity neutralised by the acid neutralising capacity (g of CaCO 3 equivalent) added to the reactor. Axial mixers were tested because of their low-energy requirements and their high reliability. The intermediate chamber effectively removes Fe by oxidising Fe(II) to Fe(III). Given the amount of acidity neutralised, the sludge volume produced was low compared to other technologies, providing further potential savings in sludge handling. Waste carbonate rock proved to be an effective neutralising agent, even though it was about 60% dolomite and 40% magnesite, with minor calcite, and despite the fact that magnesite has substantially slower dissolution kinetics compared to the more dominant dolomite. The mixed waste carbonates were capable of raising the pH sufficiently to reduce the heavy metal loadings in Fe-rich ARD by more than two orders of magnitude. The final settlement stage of the process was shown to be essential for metal precipitation, for the carry-over of fine carbonates, and CO 2 loss. This was associated with a rise in pH, from 6.3 to 7.5. In addition, residual slow-reacting magnesite from the mixed carbonate remains in the sludge from the first reactor and provides acid buffering capacity within the sludge, which is commonly lacking in the ARD neutralisation sludge of other systems.
Wearable Photoplethysmography (PPG) has gained prominence as a low cost, unobtrusive and continuous method for physiological monitoring. The quality of the collected PPG signals is affected by several sources of interference, predominantly due to physical motion. Many methods for estimating heart rate (HR) from PPG signals have been proposed with Deep Neural Networks (DNNs) gaining popularity in recent years. However, the "black-box" and complex nature of DNNs has caused a lack of trust in the predicted values. This paper contributes DeepPulse, an uncertainty-aware DNN method for estimating HR from PPG and accelerometer signals, with aims of increasing the reliability, usability and interpretability of the predicted HR values. To the best of the authors' knowledge no PPG HR estimation method has considered aleatoric and epistemic uncertainty metrics. The results show DeepPulse is the most accurate method for DNNs with less than 1 million network parameters. Finally, recommendations are given to reduce epistemic uncertainty, validate uncertainty estimates, improve the accuracy of DeepPulse as well as reduce the model size for resource-constrained edge devices.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.