Optimization of a heterogeneous computing system architecture based on its coordination with a sensors group of a mobile unmanned platform

This article analyses the existing variety of sensors used in robotics and related fields and also proposes the architecture of a heterogeneous computing system designed to analyze data obtained from sensors of a mobile unmanned platform (MUP). A feature of unmanned platforms is the presence of tasks that require a significantly different level of performance of the on-board computing system for processing data from sensors of the corresponding type. Therefore, the adaptation of existing universal computing systems seems to be impractical, compared to the development of a specialized computing system with a heterogeneous architecture. The computing system is designed to solve problems of local navigation, stabilize the position of the MUP and control its movement, as well as control special equipment installed on the MUP. Often, if the goal is to ensure maximum efficiency, expressed in speed, accuracy and reliability, it is necessary to develop specialized devices. The article provides information on sensors of the main types used in robotics and indicates the requirements for the performance of a computing system necessary for processing data from sensors of this type. This, in turn, made it possible to propose a heterogeneous architecture containing processor subsystems focused on processing data from sensors requiring low, medium and high performance according to the considered classification.

Multi-Year ENSO Forecasts Using Parallel Convolutional Neural Networks With Heterogeneous Architecture

The El Niño-Southern Oscillation (ENSO) is one of the main drivers of the interannual climate variability of Earth and can cause a wide range of climate anomalies, so multi year ENSO forecasts are a paramount scientific issue. However, most existing works rely on the conventional iterative mechanism and, thus, fail to provide reliable long-term predictions due to error accumulation. Although methods based on deep learning (DL) apply the parallel modeling scheme for different lead times instead of a single iteration model, they leverage the same DL model for prediction, which can not fully mine the variability of different lead times, resulting in a decrease of prediction accuracy. To solve this problem, we propose a novel parallel deep convolutional neural network (CNN) with a heterogeneous architecture. In this study, by adaptively selecting network architectures for different lead times, we realize variability modeling of different tasks (lead times) and thereby improve the reliability of long-term predictions. Furthermore, we propose a relationship between different prediction lead times and neural network architecture from a unique perspective, namely, the receptive field originally proposed in computer vision. According to the spatio-temporal correlated area and sampling scale of lead times, the size of the convolution kernel and the mesh size of sampling are adjusted as the lead time increases. The Coupled Model Intercomparison Project phase 5 (CMIP5) from 1861 to 2004 and the Simple Ocean Data Assimilation (SODA) from 1871 to 1973 were used for model training, and the GODAS from 1982 to 2017 were used for testing the forecast skill of the model. Experimental results demonstrate that the proposed method outperforms the other well-known methods, especially for long-term predictions.

Six-photon upconverted excitation energy lock-in for ultraviolet-C enhancement

Photon upconversion of near-infrared (NIR) irradiation into ultraviolet-C (UVC) emission offers many exciting opportunities for drug release in deep tissues, photodynamic therapy, solid-state lasing, energy storage, and photocatalysis. However, NIR-to-UVC upconversion remains a daunting challenge due to low quantum efficiency. Here, we report an unusual six-photon upconversion process in Gd3+/Tm3+-codoped nanoparticles following a heterogeneous core-multishell architecture. This design efficiently suppresses energy consumption induced by interior energy traps, maximizes cascade sensitizations of the NIR excitation, and promotes upconverted UVC emission from high-lying excited states. We realized the intense six-photon-upconverted UV emissions at 253 nm under 808 nm excitation. This work provides insight into mechanistic understanding of the upconversion process within the heterogeneous architecture, while offering exciting opportunities for developing nanoscale UVC emitters that can be remotely controlled through deep tissues upon NIR illumination.

Gem5-X

The increasing adoption of smart systems in our daily life has led to the development of new applications with varying performance and energy constraints, and suitable computing architectures need to be developed for these new applications. In this article, we present gem5-X, a system-level simulation framework, based on gem-5, for architectural exploration of heterogeneous many-core systems. To demonstrate the capabilities of gem5-X, real-time video analytics is used as a case-study. It is composed of two kernels, namely, video encoding and image classification using convolutional neural networks (CNNs). First, we explore through gem5-X the benefits of latest 3D high bandwidth memory (HBM2) in different architectural configurations. Then, using a two-step exploration methodology, we develop a new optimized clustered-heterogeneous architecture with HBM2 in gem5-X for video analytics application. In this proposed clustered-heterogeneous architecture, ARMv8 in-order cluster with in-cache computing engine executes the video encoding kernel, giving 20% performance and 54% energy benefits compared to baseline ARM in-order and Out-of-Order systems, respectively. Furthermore, thanks to gem5-X, we conclude that ARM Out-of-Order clusters with HBM2 are the best choice to run visual recognition using CNNs, as they outperform DDR4-based system by up to 30% both in terms of performance and energy savings.

From Monolithic Satellites to the Internet of Satellites Paradigm: When Space, Air, and Ground Networks Become Interconnected

From the first satellite launched in 1957, these systems always have drawn the attention of telecommunications operators. Thanks to their natural orbit, satellites can provide coverage to the entire globe or serve a vast region. Is this feature that makes them potential systems to extend current ground networks over the space. The first satellites were conceived as a single backhaul system to broadcast television or phone calls. Over the years, this concept evolved to a group of satellites that compose a constellation to interconnect any user around the globe. Nowadays, these constellations are still evolving to massive architectures with thousands of satellites that are interconnected between them composing satellite networks. Additionally, with the emergence of 5G, the community has started to discuss how to integrate satellites in this infrastructure. A review of the evolution of the satellites for broadband communications is presented in this chapter, discussing the novel and future proposed architectures. The presented work concludes with the potential of these satellite systems to compose a hybrid and heterogeneous architecture in which space, air, and ground networks become interconnected.

The Adult Neck Mass: Predictors of Malignancy

Objective Distinguishing benign from malignant adult neck masses can be challenging because data to guide risk assessment are lacking. We examined patients with neck masses from an integrated health system to identify patient and mass factors associated with malignancy. Study Design Retrospective cohort. Setting Kaiser Permanente Northern California. Methods The medical records of adults referred to otolaryngology in 2017 for a neck mass were evaluated. Bivariate and multivariable logistic regression analyses were performed. Results Malignancy was found in 205 (5.0%) of the cohort’s 4103 patients. Patient factors associated with malignancy included sex, age, and race/ethnicity. Males had more than twice the odds of malignancy compared with females (adjusted odds ratio [aOR] = 2.38). Malignancy rates increased with age, ranging from 2.1% for patients younger than 40 years to 8.4% for patients 70 years or older. White non-Hispanic patients had 1.75 times the risk of malignancy compared with patients of other race/ethnicities. The percentage of patients with malignancy increased with increasing minimum mass dimension, from 3.0% in patients with mass size <1 cm to over 31% in patients with mass sizes 2 cm or larger ( P < .0001). Imaging-based mass factors most highly predictive of malignancy included larger minimum mass dimension (≥1.5 cm vs <1.5 cm: aOR = 3.87), multiple masses (2 or more vs 1: aOR = 5.07), and heterogeneous/ill-defined quality (aOR = 2.57). Conclusion Most neck masses referred to otolaryngology were not malignant. Increasing age, male sex, white non-Hispanic ethnicity, increasing minimum mass dimension, multiple neck masses, or heterogeneous architecture/ill-defined borders were associated with malignancy.