SEARCH · Engineering Papers

Results for “Low latency”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 415 records · Page 23

All sources NASA NTRS DOE OSTI

NASA NTRS2017

Evaluation of a Head-Worn Display System as an Equivalent Head-Up Display for Low Visibility Commercial Operations

Research, development, test, and evaluation of fight deck interface technologies is being conducted by the National Aeronautics and Space Administration (NASA) to proactively identify, develop, and mature tools, methods, and technologies for improving overall aircraft safety of new and legacy vehicles operating in the Next Generation Air Transportation System (NextGen). One specific area of research was the use of small Head-Worn Displays (HWDs) to serve as a possible equivalent to a Head-Up Display (HUD). A simulation experiment and a fight test were conducted to evaluate if the HWD can provide an equivalent level of performance to a HUD. For the simulation experiment, airline crews conducted simulated approach and landing, taxi, and departure operations during low visibility operations. In a follow-on fight test, highly experienced test pilots evaluated the same HWD during approach and surface operations. The results for both the simulation and fight tests showed that there were no statistical differences in the crews' performance in terms of approach, touchdown and takeoff; but, there are still technical hurdles to be overcome for complete display equivalence including, most notably, the end-to-end latency of the HWD system.

Arthur, Jarvis (Trey) J., III↗

DOE OSTI2023

In-pixel AI for lossy data compression at source for X-ray detectors

Integrating neural networks for data compression directly in the Read-Out Integrated Circuits (ROICs), i.e. the pixelated front-end, would result in a significant reduction in off-chip data transfer, overcoming the I/O bottleneck. Our ROIC test chip (AI-In-Pixel-65) is designed in a 65 nm Low Power CMOS process for the readout of pixelated X-ray detectors. Each pixel consists of an analog front-end for signal processing and a 10b analog-to-digital converter operating at 100KSPS. Here, we compare two non-reconfigurable techniques, Principal Component Analysis (PCA) and an AutoEncoder (AE) as lossy data compression engines implemented within the pixelated area. The PCA algorithm achieves 50$×$ compression, adds one clock cycle latency, and results in a 21% increase in the pixel area. The AE achieves 70$×$ compression, adds 30 clock cycle latency, and results in a similar area increase.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2025

ExtremeMETA: High-speed Lightweight Image Segmentation Model by Remodeling Multi-channel Metamaterial Imagers

Deep neural networks (DNNs) have heavily relied on traditional computational units, such as CPUs and GPUs. However, this conventional approach brings significant computational burden, latency issues, and high power consumption, limiting their effectiveness. This has sparked the need for lightweight networks such as ExtremeC3Net. Meanwhile, there have been notable advancements in optical computational units, particularly with metamaterials, offering the exciting prospect of energy-efficient neural networks operating at the speed of light. Yet, the digital design of metamaterial neural networks (MNNs) faces precision, noise, and bandwidth challenges, limiting their application to intuitive tasks and low-resolution images. In this study, we proposed a large kernel lightweight segmentation model, ExtremeMETA. Based on ExtremeC3Net, our proposed model, ExtremeMETA maximized the ability of the first convolution layer by exploring a larger convolution kernel and multiple processing paths. With the large kernel convolution model, we extended the optic neural network application boundary to the segmentation task. To further lighten the computation burden of the digital processing part, a set of model compression methods was applied to improve model efficiency in the inference stage. The experimental results on three publicly available datasets demonstrated that the optimized efficient design improved segmentation performance from 92.45 to 95.97 on mIoU while reducing computational FLOPs from 461.07 MMacs to 166.03 MMacs. The large kernel lightweight model ExtremeMETA showcased the hybrid design’s ability on complex tasks.

large convolution kernel↗

NASA NTRS1981

Sensation of rotation about a vertical axis with a fixed visual field in different illuminations and in the dark

The effects of the oculogyro illusion of the relative motion of a spot fixed with respect to the subject during subject rotation and of a fixed striped peripheral visual field under different levels of illumination on perceptions of rotation about a vertical axis are investigated. Subjects were seated in a rotatable flight trainer cockpit with visual fields consisting of darkness, a dim peripheral field, and a bright peripheral field, all fixed with respect to the subject, and subject perceptual thresholds, frequency responses and sensations of displacement and velocity were measured during trainer rotation at constant and varying angular accelerations. The perception of angular acceleration is found to exhibit a significantly lower threshold and a reduced latency time in the illuminated visual fields which was independent of the level of illumination. Subjective frequency responses showed a higher gain in the illuminated presentations, while subjective displacements during triangular velocity stimuli exhibited no difference for the different visual cues. Finally, magnitude estimations of the after-rotation associated with deceleration from a constant velocity showed a greater rising speed, larger velocity and longer duration under illumination. Results show that, for low accelerations, the visual input enhances sensitivity to self-motion, an effect explained by the oculogyral illusion.

Huang, J.↗

NASA NTRS1997

Predicting Cost/Performance Trade-Offs for Whitney: A Commodity Computing Cluster

Recent advances in low-end processor and network technology have made it possible to build a "supercomputer" out of commodity components. We develop simple models of the NAS Parallel Benchmarks version 2 (NPB 2) to explore the cost/performance trade-offs involved in building a balanced parallel computer supporting a scientific workload. We develop closed form expressions detailing the number and size of messages sent by each benchmark. Coupling these with measured single processor performance, network latency, and network bandwidth, our models predict benchmark performance to within 30%. A comparison based on total system cost reveals that current commodity technology (200 MHz Pentium Pros with 100baseT Ethernet) is well balanced for the NPBs up to a total system cost of around $1,000,000.

Becker, Jeffrey C.↗

DOE OSTI2024

User-Centric Communication With Aerial Network for 6G: A Reinforcement Learning Approach

Meeting the diverse needs of user verticals requires innovative cellular architectures that can offer additional degrees of freedom to provide on-demand services. The terrestrial user-centric radio access network (UC-RAN) stands out as an excellent choice for this purpose. However, a drawback of UC-RAN is its tendency to prioritize high-priority verticals, often resulting in a subpar quality of experience for low-priority verticals. This issue is particularly exacerbated in hotspot areas. Here, to address this problem, we introduce an aerial network integrated with terrestrial UC-RAN to provide coverage to users which are not served by the terrestrial network. Furthermore, we analyze the impact of key configuration and optimization parameters (COPs), such as location, transmit power, altitude, and beamwidth of aerial base stations (ABSs) on system key performance indicators (KPIs), such as coverage, latency satisfaction, average spectral efficiency, and energy efficiency. We formulate a robust multiobjective function to maximize these KPIs without biasing toward any specific KPI(s). Finally, we propose a deep reinforcement learning optimization framework based on the state-of-the-art soft actor-critic algorithm to control ABS COPs and optimize system KPIs. Experimental evaluations demonstrate that the proposed optimization framework can converge to near-optimal solutions derived from the pseudo brute force in a few thousand epochs.

6G↗

DOE OSTI2026

Testing and Validation of Wireless Communication Architecture for Heliostat Fields: SIPS Final Report

This work focuses on the development and testing of a low-cost wireless communication system for heliostat fields, enabling significant capital costs reductions for concentrating solar thermal systems. Outputs of this work include a working demonstration of a multi-node communication system, clear reporting of system performance, and technical documentation of system development and architecture for reproducibility. Through this process, an open-source repository was created for manufacturing hardware at ~$30/heliostat. The system includes software for cybersecurity, achieving sub-second communication latencies and derisking of hardware for eventual scale-up to tens of thousands of heliostats. While the system is not currently off-the-shelf ready, there is now a clearly defined pathway for scaling up and completing the commercial development process.

14 SOLAR ENERGY↗

DOE OSTI2024

Ultrafast jet classification at the HL-LHC

Abstract Three machine learning models are used to perform jet origin classification. These models are optimized for deployment on a field-programmable gate array device. In this context, we demonstrate how latency and resource consumption scale with the input size and choice of algorithm. Moreover, the models proposed here are designed to work on the type of data and under the foreseen conditions at the CERN large hadron collider during its high-luminosity phase. Through quantization-aware training and efficient synthetization for a specific field programmable gate array, we show that O ( 100 ) ns inference of complex architectures such as Deep Sets and Interaction Networks is feasible at a relatively low computational resource cost.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

NASA NTRS2019

An Automatic Light Rain Detection Algorithm on NASA MPLNET Lidar Observations in the Frame of WMO GALION Project

The water cycle strongly influences life on Earth. In particular, the precipitation modifies the atmospheric column thermodynamics through the process of evaporation and serves as a proxy for latent heat modulation. For this reason, a correct precipitation parameterization (especially low-intensity precipitation) at global scale, bedsides improving our understanding of the hydrological cycle, it is crucial to reduce the associated uncertainty of the global climate models to correctly forecast future scenarios, i.e. to apply fast mitigation strategies. In this study we developed an algorithm to automatically detect precipitation from lidar measurements obtained by the National and Aeronautics Space Administration (NASA) Micropulse lidar network (MPLNET) permanent observational site in Goddard. The algorithm, once full operational, will deliver in Near Real Time (latency 1.5h) a new rain mask product that will be publicly available on MPLNET website as part of the new Version 3 Level 1.5 data. The methodology, based on an image processing technique, can detect only light precipitation events (defined by intensity and duration) as the morphological filters used through the detection process are applied on the lidar volume depolarization ratio range corrected composite images, i.e. heavy rain events are unusable as the lidar signal is completely extinguished after few meters in the precipitation or no signal detected because of the water accumulated on the receiver optics. Results from the algorithm, besides filling a gap in precipitation and virga detection by radars, are of particular interest for the scientific community because will help to better understand long-term aerosol-cloud interactions and aerosol atmospheric removal (scavenging effect) by rain as multi-year database being available for several MPLNET permanent observational sites across the globe. Moreover, we developed the automatic algorithm at Universitat Politecnica de Catalunya (UPC) Barcelona, the unique permanent observation station member of MPLNET and the European Aerosol Lidar Network (EARLINET) In the future the algorithm can be then easily applied to any other lidar and/or ceilometer network infrastructure in the frame of World Meteorological Organization (WMO) Global Aerosol Watch (GAW) aerosol lidar observation network (GALION)

Simone Lolli↗

NASA NTRS2021

Features and Characteristics of the New NASA MicroPuLse NETwork (MPLNET) Automatic Rain Detection Algorithm

S. Lolli↗

DOE OSTI2023

Runtime extension for neural network training with heterogeneous memory

Systems, apparatuses, and methods for managing buffers in a neural network implementation with heterogeneous memory are disclosed. A system includes a neural network coupled to a first memory and a second memory. The first memory is a relatively low-capacity, high-bandwidth memory while the second memory is a relatively high-capacity, low-bandwidth memory. During a forward propagation pass of the neural network, a run-time manager monitors the usage of the buffers for the various layers of the neural network. During a backward propagation pass of the neural network, the run-time manager determines how to move the buffers between the first and second memories based on the monitored buffer usage during the forward propagation pass. As a result, the run-time manager is able to reduce memory access latency for the layers of the neural network during the backward propagation pass.

Mappouras, Georgios↗

DOE OSTI2026

ClassNMSW- a real-time classification approach for non-recycled municipal solid waste using hyperspectral imaging

Real-time classification of non-recycled municipal solid waste (NMSW) is essential for efficient valorization. This study introduces ClassNMSW, a comprehensive framework for classifying 22 NMSW subclasses under industrial constraints by using hyperspectral imaging (HSI). A primary innovation of this work is the development of a variance-controlled spectral extraction algorithm. Unlike traditional methods that rely on simple averaging, this approach systematically investigates the extent of pixel extraction to minimize the loss of critical chemical information while maximizing data reduction thus ensuring high spectral fidelity with low computational cost. The approach developed in this work integrates automated, computer-vision-based background removal, eliminating the need for the manual thresholding common in current literature. To resolve ambiguities among chemically similar subclasses, a tiered classification and multi-camera fusion strategy (NIR17 and NIR22) is implemented. Results demonstrate that ClassNMSW achieves an object-wise weighted accuracy of 98.70% for single-sensor configurations and 100% under sensor fusion. A novel rolling-window strategy satisfies desired end-to-end latency of <2 s, satisfying the strict deterministic requirements of high-speed industrial sorting environments. The ClassNMSW framework provides a scalable foundation for advancing circularity and resource recovery in large-scale waste valorization operations.

99 - GENERAL AND MISCELLANEOUS↗

DOE OSTI2021

Timely Reporting of Heavy Hitters Using External Memory

Given an input stream S of size N, a Φ-heavy hitter is an item that occurs at least ΦN times in S. The problem of finding heavy-hitters is extensively studied in the database literature. In this work, we study a real-time heavy-hitters variant in which an element must be reported shortly after we see its T = Φ N-th occurrence (and hence it becomes a heavy hitter). We call this the Timely Event Detection (TED) Problem. The TED problem models the needs of many real-world monitoring systems, which demand accurate (i.e., no false negatives) and timely reporting of all events from large, high-speed streams with a low reporting threshold (high sensitivity). Like the classic heavy-hitters problem, solving the TED problem without false-positives requires large space (Ω (N) words). Thus in-RAM heavy-hitters algorithms typically sacrifice accuracy (i.e., allow false positives), sensitivity, or timeliness (i.e., use multiple passes). We show how to adapt heavy-hitters algorithms to external memory to solve the TED problem on large high-speed streams while guaranteeing accuracy, sensitivity, and timeliness. Our data structures are limited only by I/O-bandwidth (not latency) and support a tunable tradeoff between reporting delay and I/O overhead. With a small bounded reporting delay, our algorithms incur only a logarithmic I/O overhead. We implement and validate our data structures empirically using the Firehose streaming benchmark. Multi-threaded versions of our structures can scale to process 11M observations per second before becoming CPU bound. In comparison, a naive adaptation of the standard heavy-hitters algorithm to external memory would be limited by the storage device’s random I/O throughput, i.e., ≈100K observations per second.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2014

Design Concepts for a Small Space-Based GEO Relay Satellite for Missions Between Low Earth and near Earth Orbits

The main purpose of the Small Space-Based Geosynchronous Earth orbiting (GEO) satellite is to provide a space link to the user mission spacecraft for relaying data through ground networks to user Mission Control Centers. The Small Space Based Satellite (SSBS) will provide services comparable to those of a NASA Tracking Data Relay Satellite (TDRS) for the same type of links. The SSBS services will keep the user burden the same or lower than for TDRS and will support the same or higher data rates than those currently supported by TDRS. At present, TDRSS provides links and coverage below GEO; however, SSBS links and coverage capability to above GEO missions are being considered for the future, especially for Human Space Flight Missions (HSF). There is also a rising need for the capability to support high data rate links (exceeding 1 Gbps) for imaging applications. The communication payload on the SSBS will provide S/Ka-band single access links to the mission and a Ku-band link to the ground, with an optical communication payload as an option. To design the communication payload, various link budgets were analyzed and many possible operational scenarios examined. To reduce user burden, using a larger-sized antenna than is currently in use by TDRS was considered. Because of the SSBS design size, it was found that a SpaceX Falcon 9 rocket could deliver three SSBSs to GEO. This will greatly reduce the launch costs per satellite. Using electric propulsion was also evaluated versus using chemical propulsion; the power system size and time to orbit for various power systems were also considered. This paper will describe how the SSBS will meet future service requirements, concept of operations, and the design to meet NASA users' needs for below and above GEO missions. These users' needs not only address the observational mission requirements but also possible HSF missions to the year 2030. We will provide the trade-off analysis of the communication payload design in terms of the number of links looking above and below GEO; the detailed design of a GEO SSBS spacecraft bus and its accommodation of the communication payload, and a summary of the trade study that resulted in the selection of the Falcon 9 launch vehicle to deploy the SSBS and its impact on cost reductions per satellite. ======================================================================== Several initiatives have taken place within NASA1 and international space agencies2 to create a human exploration strategy for expanding human presence into the solar system; these initiatives have been driven by multiple factors to benefit Earth. Of the many elements in the strategy one stands out: to send robotic and human missions to destinations beyond Low Earth Orbit (LEO), including cis-lunar space, Near-Earth Asteroids (NEAs), the Moon, and Mars and its moons.3, 4 The time frame for human exploration to various destinations, based on the public information available,1,4 is shown in Figure 1. Advance planning is needed to define how future space communications services will be provided in the new budget environment to meet future space communications needs. The spacecraft for these missions can be dispersed anywhere from below LEO to beyond GEO, and to various destinations within the solar system. NASA's Space Communications and Navigation (SCaN) program office provides communication and tracking services to space missions during launch, in-orbit testing, and operation phases. Currently, SCaN's space networking relay satellites mainly provide services to users below GEO, at Near Earth Orbit (NEO), below LEO, and in deep space. The potential exists for using a space-based relay satellite, located in the vicinity of various solar system destinations, to provide communication space links to missions both below and above its orbit. Such relays can meet the needs of human exploration missions for maximum connectivity to Earth locations and for reduced latency. In the past, several studies assessed the ability of satellite-based relays working above GEO in conjunction with Earth ground stations. Many of these focused on the trade between space relay and direct-to-Earth station links5,6,7. Several others focused on top-level architecture based on relays at various destinations8,9,10,11,12. Much has changed in terms of microwave and optical technology since the publication of the referenced papers; Ka-band communication systems are being deployed, optical communication is being demonstrated, and spacecraft buses are becoming increasingly more functional and operational. A design concept study was undertaken to access the potential for deploying a Small Space-Based Satellite (SSBS) relay capable of serving missions between LEO and NEO. The needs of future human exploration missions were analyzed, and a notional relay-based architecture concept was generated as shown in Fig. 1. Relay satellites in Earth through cis-Lunar orbits are normally located in stable orbits requiring low fuel consumption. Relay satellites for Mars orbit are normally selected based on the mission requirement and projected fuel consumption. Relay satellites have extreme commonalities of functions between them, differing only in the redundancy and frequencies used; therefore, the relay satellite in GEO was selected for further analysis since it will be the first step in achieving a relay-based architecture for human exploration missions (see Fig.Figure 2). The mission design methodology developed by the Collaborative Modeling for Parametric Assessment of Space Systems (COMPASS) team13 was used to produce the satellite relay design and to perform various design trades. At the start of the activity, the team was provided with the detailed concept of the notional architecture and the system and communication payload drivers.

Relay Satellites↗

NASA NTRS1989

On the stability of robotic systems with random communication rates

Control problems of sampled data systems which are subject to random sample rate variations and delays are studied. Due to the rapid growth of the use of computers more and more systems are controlled digitally. Complex systems such as space telerobotic systems require the integration of a number of subsystems at different hierarchical levels. While many subsystems may run on a single processor, some subsystems require their own processor or processors. The subsystems are integrated into functioning systems through communications. Communications between processes sharing a single processor are also subject to random delays due to memory management and interrupt latency. Communications between processors involve random delays due to network access and to data collisions. Furthermore, all control processes involve delays due to casual factors in measuring devices and to signal processing. Traditionally, sampling rates are chosen to meet the worst case communication delay. Such a strategy is wasteful as the processors are then idle a great proportion of the time; sample rates are not as high as possible resulting in poor performance or in the over specification of control processors; there is the possibility of missing data no matter how low the sample rate is picked. Asymptotical stability with probability one for randomly sampled multi-dimensional linear systems is studied. A sufficient condition for the stability is obtained. This condition is so simple that it can be applied to practical systems. A design procedure is also shown.

Kobayashi, H.↗

NASA NTRS1988

Competency for graviresponse in the leaf-sheath pulvinus of Avena sativa: onset to loss

The development of the leaf-sheath pulvinus of oat (Avena sativa L. cv. Victory) was studied in terms of its competency to respond to gravistimulation. Stages of onset of competency, maximum competency and loss of competency were identified, using the length of the supertending internode as a developmental marker. During the early phases in the onset of competency, the latency period between stimulus and graviresponse decreased and the steady state response rate increased significantly. When fully competent, the latency period remained constant as the plant continued to develop, suggesting that the latency period is relatively insensitive to quantitative changes (e.g., in carbohydrate or nutrient availability) at the cell level within the plant. In contrast, the response rate was found to increase with plant development, indicating that graviresponse rate is more strongly influenced by quantitative cellular changes. The total possible graviresponse of a single oat pulvinus was confirmed to be significantly less than the original presentation angle. This was shown to not result from a loss of competency, since the graviresponse could be reinitiated by increasing the presentation angle. As a result of the low overall graviresponse of individual pulvini, two or more pulvini are required to bring the plant apex to the vertical. This was determined to occur though the sequential, rather than simultaneous, action of successive pulvini, since a given pulvinus lost competency to gravirespond shortly after the next pulvinus became fully competent.

Non-NASA Center↗

NASA NTRS1993

Evaluation Metrics for the Paragon XP/S-15

On February 17th 1993, the Numerical Aerodynamic Simulation (NAS) facility located at the NASA Ames Research Center installed a 224 node Intel Paragon XP/S-15 system. After its installation, the Paragon was found to be in a very immature state and was unable to support a NAS users' workload, composed of a wide range of development and production activities. As a first step towards addressing this problem, we implemented a set of metrics to objectively monitor the system as operating system and hardware upgrades were installed. The metrics were designed to measure four aspects of the system that we consider essential to support our workload: availability, utilization, functionality, and performance. This report presents the metrics collected from February 1993 to August 1993. Since its installation, the Paragon availability has improved from a low of 15% uptime to a high of 80%, while its utilization has remained low. Functionality and performance have improved from merely running one of the NAS Parallel Benchmarks to running all of them faster (between 1 and 2 times) than on the iPSC/860. In spite of the progress accomplished, fundamental limitations of the Paragon operating system are restricting the Paragon from supporting the NAS workload. The maximum operating system message passing (NORMA IPC) bandwidth was measured at 11 Mbytes/s, well below the peak hardware bandwidth (175 Mbytes/s), limiting overall virtual memory and Unix services (i.e. Disk and HiPPI I/O) performance. The high NX application message passing latency (184 microns), three times than on the iPSC/860, was found to significantly degrade performance of applications relying on small message sizes. The amount of memory available for an application was found to be approximately 10 Mbytes per node, indicating that the OS is taking more space than anticipated (6 Mbytes per node).

Traversat, Bernard↗

NASA NTRS2019

Overview of the New Version 3 NASA Micro-Pulse Lidar Network (MPLNET) Automatic Precipitation Detection Algorithm

Precipitation modifies atmospheric column thermodynamics through the process of evaporation and serves as a proxy for latent heat modulation. For this reason, a correct precipitation parameterization (especially for low-intensity precipitation) within global scale models is crucial. In addition to improving our modeling of the hydrological cycle, this will reduce the associated uncertainty of global climate models in correctly forecasting future scenarios, and will enable the application of mitigation strategies. In this manuscript we present a proof of concept algorithm to automatically detect precipitation from lidar measurements obtained from the National Aeronautics and Space Administration Micropulse lidar network (MPLNET). The algorithm, once tested and validated against other remote sensing instruments, will be operationally implemented into the network to deliver a near real time (latency <1.5 h) rain masking variable that will be publicly available on MPLNET website as part of the new Version 3 data products. The methodology, based on an image processing technique, detects only light precipitation events (defined by intensity and duration) such as light rain, drizzle, and virga. During heavy rain events, the lidar signal is completely extinguished after a few meters in the precipitation or it is unusable because of water accumulated on the receiver optics. Results from the algorithm, in addition to filling a gap in light rain, drizzle, and virga detection by radars, are of particular interest for the scientific community as they help to fully characterize the aerosol cycle, from emission to deposition, as precipitation is a crucial meteorological phenomenon accelerating atmospheric aerosol removal through the scavenging effect. Algorithm results will also help the understanding of long term aerosol–cloud interactions, exploiting the multi-year database from several MPLNET permanent observational sites across the globe. The algorithm is also applicable to other lidar and/or ceilometer network infrastructures in the framework of the Global Aerosol Watch (GAW) aerosol lidar observation network (GALION).

Simone Lolli↗