Challenges and Trade-offs in Real-Time Implementation of DSP for Coherent Transmission

We present an end-to-end system for real-time environment capture, 3D reconstruction, and stereoscopic view synthesis on a mobile VR headset. Our solution allows the user to use the cameras on their VR headset as their eyes to see and interact with the real world while still wearing their headset, a feature often referred to as Passthrough. The central challenge when building such a system is the choice and implementation of algorithms under the strict compute, power, and performance constraints imposed by the target user experience and mobile platform. A key contribution of this paper is a complete description of a corresponding system that performs temporally stable passthrough rendering at 72 Hz with only 200 mW power consumption on a mobile Snapdragon 835 platform. Our algorithmic contributions for enabling this performance include the computation of a coarse 3D scene proxy on the embedded video encoding hardware, followed by a depth densification and filtering step, and finally stereoscopic texturing and spatio-temporal up-sampling. We provide a detailed discussion and evaluation of the challenges we encountered, as well as algorithm and performance trade-offs in terms of compute and resulting passthrough quality.;AB@The described system is available to users as the Passthrough+ feature on Oculus Quest. We believe that by publishing the underlying system and methods, we provide valuable insights to the community on how to design and implement real-time environment sensing and rendering on heavily resource constrained hardware.

Download Full-text

Technical and Engineering Decisions That Drive Choices for Predictive Analytics of Plant Data

ASME 2011 Power Conference, Volume 2 ◽

10.1115/power2011-55329 ◽

2011 ◽

Author(s):

William Nieman

Keyword(s):

Real Time ◽

Predictive Analytics ◽

Current Data ◽

Operating Conditions ◽

Maintenance Cost ◽

Real Time Systems ◽

Efficient Technology ◽

Time Model ◽

Trade Offs ◽

Time Systems

Power generation has the goal of maximizing power output while minimizing operations and maintenance cost. The challenge for plant manager is to move closer to reliability limits while being confident the risks of any decision are understood. To attain their goals and meet this challenge they are coming to realize that they must have frequent, accurate assessment of equipment operating conditions, and a path to continued innovation-. At a typical plant, making this assessment involves the collection and effective analysis of reams of complex, interrelated production system data, including demand requirements, load, ambient temperature, as well as the dependent equipment data. Wind turbine health and performance data is available from periodic and real-time systems. To obtain the timeliest understanding of equipment health for all the key resources in a large plant or fleet, engineers increasingly turn to real-time, model-based solutions. Real-time systems are capable of creating actionable intelligence from large amounts and diverse sources of current data. They can automatically detect problems and provide the basis for diagnosis and prioritization effectively for many problems, and they can make periodic inspection methods much more efficient. Technology exists to facilitate prediction of when assets will fail, allowing engineers to target maintenance costs more effectively. But, it is critical to select the best predictive analytics for your plant. How do you make that choice correctly? Real-time condition monitoring and analysis tools need to be matched to engineering process capability. Tools are employed at the plant in lean, hectic environments; others are deployed from central monitoring centers charged with concentrating scarce resources to efficiently support plants. Applications must be flexible and simple to implement and use. Choices made in selection of new tools can be very important to future success of plant operations. So, these choices require solid understanding of the problems to be solved and the advantages and trade-offs of potential solutions. This choice of the best Predictive Analytic solution will be discussed in terms of key technology elements and key engineering elements.

Download Full-text

Real-time 80-Channel 75-GHz-Spacing Coherent Transmission over Extended C-Band for Long-haul Network beyond 100G

10.1364/oecc.2021.t3b.5 ◽

2021 ◽

Author(s):

Dong Wang ◽

Jiang Sun ◽

Yunbo Li ◽

Dawei Ge ◽

Jianghua Gu ◽

...

Keyword(s):

Real Time ◽

Coherent Transmission

Download Full-text

Probabilistic Jacobian-Based Saliency Maps Attacks

Machine Learning and Knowledge Extraction ◽

10.3390/make2040030 ◽

2020 ◽

Vol 2 (4) ◽

pp. 558-578

Author(s):

Théo Combey ◽

António Loison ◽

Maxime Faucher ◽

Hatem Hajri

Keyword(s):

Neural Network ◽

Real Time ◽

Saliency Map ◽

Black Box ◽

Saliency Maps ◽

Trade Offs ◽

Good Trade ◽

Neural Network Classifiers

Neural network classifiers (NNCs) are known to be vulnerable to malicious adversarial perturbations of inputs including those modifying a small fraction of the input features named sparse or L0 attacks. Effective and fast L0 attacks, such as the widely used Jacobian-based Saliency Map Attack (JSMA) are practical to fool NNCs but also to improve their robustness. In this paper, we show that penalising saliency maps of JSMA by the output probabilities and the input features of the NNC leads to more powerful attack algorithms that better take into account each input’s characteristics. This leads us to introduce improved versions of JSMA, named Weighted JSMA (WJSMA) and Taylor JSMA (TJSMA), and demonstrate through a variety of white-box and black-box experiments on three different datasets (MNIST, CIFAR-10 and GTSRB), that they are both significantly faster and more efficient than the original targeted and non-targeted versions of JSMA. Experiments also demonstrate, in some cases, very competitive results of our attacks in comparison with the Carlini-Wagner (CW) L0 attack, while remaining, like JSMA, significantly faster (WJSMA and TJSMA are more than 50 times faster than CW L0 on CIFAR-10). Therefore, our new attacks provide good trade-offs between JSMA and CW for L0 real-time adversarial testing on datasets such as the ones previously cited.

Download Full-text

Quasi-Operational Testing of Real-Time Storm-Longevity Prediction via Machine Learning

Weather and Forecasting ◽

10.1175/waf-d-18-0141.1 ◽

2019 ◽

Vol 34 (5) ◽

pp. 1437-1451 ◽

Cited By ~ 2

Author(s):

Amy McGovern ◽

Christopher D. Karstens ◽

Travis Smith ◽

Ryan Lagerquist

Keyword(s):

Machine Learning ◽

Real Time ◽

Weather Prediction ◽

Warning System ◽

Objective Evaluation ◽

Boosted Regression Trees ◽

Time Prediction ◽

Operational Testing ◽

Trade Offs ◽

Naturalistic Environment

Abstract Real-time prediction of storm longevity is a critical challenge for National Weather Service (NWS) forecasters. These predictions can guide forecasters when they issue warnings and implicitly inform them about the potential severity of a storm. This paper presents a machine-learning (ML) system that was used for real-time prediction of storm longevity in the Probabilistic Hazard Information (PHI) tool, making it a Research-to-Operations (R2O) project. Currently, PHI provides forecasters with real-time storm variables and severity predictions from the ProbSevere system, but these predictions do not include storm longevity. We specifically designed our system to be tested in PHI during the 2016 and 2017 Hazardous Weather Testbed (HWT) experiments, which are a quasi-operational naturalistic environment. We considered three ML methods that have proven in prior work to be strong predictors for many weather prediction tasks: elastic nets, random forests, and gradient-boosted regression trees. We present experiments comparing the three ML methods with different types of input data, discuss trade-offs between forecast quality and requirements for real-time deployment, and present both subjective (human-based) and objective evaluation of real-time deployment in the HWT. Results demonstrate that the ML system has lower error than human forecasters, which suggests that it could be used to guide future storm-based warnings, enabling forecasters to focus on other aspects of the warning system.

Download Full-text