Search for: All records

Laboratory-derived optical constants are essential for identifying ices and measuring their relative abundances on solar system objects. Almost all optical constants of ices important to planetary science come from experiments with transmission geometries. Here we describe our new experimental setup and the modification of an iterative algorithm in the literature to measure the optical constants of ices from experiments with reflectance geometries. We apply our techniques to CH₄ice and H₂O ice samples and find good agreement between our values and those in the literature, except for one CH₄band in the literature that likely suffers from saturation. The work we present here demonstrates that labs with reflectance geometries can generate optical constants essential for the proper analysis of near- and mid-infrared spectra of outer solar system objects such as those obtained with the James Webb Space Telescope.

 

In this work, we propose a novel approach for the real-time estimation of chip-level spatial power maps for commercial Google Coral M.2 TPU chips based on a machine-learning technique for the first time. The new method can enable the development of more robust runtime power and thermal control schemes to take advantage of spatial power information such as hot spots that are otherwise not available. Different from the existing commercial multi-core processors in which real-time performance-related utilization information is available, the TPU from Google does not have such information. To mitigate this problem, we propose to use features that are related to the workloads of running different deep neural networks (DNN) such as the hyperparameters of DNN and TPU resource information generated by the TPU compiler. The new approach involves the offline acquisition of accurate spatial and temporal temperature maps captured from an external infrared thermal imaging camera under nominal working conditions of a chip. To build the dynamic power density map model, we apply generative adversarial networks (GAN) based on the workload-related features. Our study shows that the estimated total powers match the manufacturer's total power measurements extremely well. Experimental results further show that the predictions of power maps are quite accurate, with the RMSE of only 4.98\rm mW/mm^2, or 2.6\% of the full-scale error. The speed of deploying the proposed approach on an Intel Core i7-10710U is as fast as 6.9ms, which is suitable for real-time estimation. 

Stochastic computing (SC) can lead area-efficient implementation of logic designs. Existing SC multiplication, however, suffers a long-standing problem: large multiplication error with small inputs due to its intrinsic nature of bit-stream based computing. In this article, we propose a new scaled counting-based SC multiplication approach, called {\it Scaled-CBSC}, to mitigate this issue by introducing scaling bits to ensure the bit `1' density of the stochastic number is sufficiently large. The idea is to convert the ``small'' inputs to ``large'' inputs, thus improve the accuracy of SC multiplication. But different from an existing stream-bit based approach, the new method uses the binary format and does not require stochastic addition as the SC multiplication always starts with binary numbers. Furthermore, Scaled-CBSC only requires all the numbers to be larger than 0.5 instead of arbitrary defined threshold, which leads to integer numbers only for the scaling term. The experimental results show that the 8-bit Scaled-CBSC multiplication with 3 scaling bits can achieve up to 46.6\% and 30.4\% improvements in mean error and standard deviation, respectively; reduce the peak relative error from 100\% to 1.8\%; and improve 12.6\%, 51.5\%, 57.6\%, 58.4\% in delay, area, area-delay product, energy consumption, respectively, over the state of art work. 

Over the past decade, a series of airborne experiments in the Arctic and Antarctica explored microwave emission from sea ice and ice sheets at frequencies from 0.5 to 2 GHz. The experiments were motivated by the fact that lower frequencies penetrate deeper into a frozen surface, thus offering the possibility to measure physical temperatures at great depths in ice sheets and, subsequently, other unique geophysical observables including sea ice salinity. These experiments were made feasible by recent engineering advances in electronics, antenna design, and noise removal algorithms when operating outside of protected bands in the electromagnetic spectrum. These technical advances permit a new type of radiometer that not only operates at low frequency, but also obtains continuous spectral information over the band from 0.5 to 2 GHz. Spectral measurements facilitate an understanding of the physical processes controlling emission and also support the interpretation of results from single frequency instruments. This paper reviews the development of low-frequency, wide band radiometry and its application to cryosphere science over the past 10 years. The paper summarizes the engineering design of an airborne instrument and the associated algorithms to mitigate radio frequency interference. Theoretical models of emission built around the morphologic and electrical properties of cryospheric components are also described that identify the dominant physical processes contributing to emission spectra. New inversion techniques for geophysical parameter retrieval are summarized for both Arctic and Antarctic scenarios. Examples that illustrate how the measurements are used to inform on glaciological problems are presented. The paper concludes with a description of new instrument concepts that are foreseen to extend the technology into operation from space. 

When a fluid stream in a conduit splits in order to pass around an obstruction, it is possible that one branch will be critically controlled while the other remains not so. This is apparently the situation in Pacific Ocean abyssal circulation, where most of the northward flow of Antarctic bottom water passes through the Samoan Passage, where it is hydraulically controlled, while the remainder is diverted around the Manihiki Plateau and is not controlled. These observations raise a number of questions concerning the dynamics necessary to support such a regime in the steady state, the nature of upstream influence and the usefulness of rotating hydraulic theory to predict the partitioning of volume transport between the two paths, which assumes the controlled branch is inviscid. Through the use of a theory for constant potential vorticity flow and accompanying numerical model, we show that a steady-state regime similar to what is observed is dynamically possible provided that sufficient bottom friction is present in the uncontrolled branch. In this case, the upstream influence that typically exists for rotating channel flow is transformed into influence into how the flow is partitioned. As a result, the partitioning of volume flux can still be reasonably well predicted with an inviscid theory that exploits the lack of upstream influence. 

2.5D chiplet-based technology promises an efficient integration technique for advanced designs with more functionality and higher performance. Temperature and related thermal optimization, heat removal are of critical importance for temperature-aware physical synthesis for chiplets. This paper presents a novel graph convolutional networks (GCN) architecture to estimate the thermal map of the 2.5D chiplet-based systems with the thermal resistance networks built by the compact thermal model (CTM). First, we take the total power of all chiplets as an input feature, which is a global feature. This additional global information can overcome the limitation that the GCN can only extract local information via neighborhood aggregation. Second, inspired by convolutional neural networks (CNN), we add skip connection into the GCN to pass the global feature directly across the hidden layers with the concatenation operation. Third, to consider the edge embedding feature, we propose an edge-based attention mechanism based on the graph attention networks (GAT). Last, with the multiple aggregators and scalers of principle neighborhood aggregation (PNA) networks, we can further improve the modeling capacity of the novel GCN. The experimental results show that the proposed GCN model can achieve an average RMSE of 0.31 K and deliver a 2.6$\times$ speedup over the fast steady-state solver of open-source {\it HotSpot} based on SuperLU. More importantly, the GCN model demonstrates more useful generalization or transferable capability. Our results show that the trained GCN can be directly applied to predict thermal maps of six unseen datasets with acceptable mean RMSEs of less than 0.67 K without retraining via inductive learning. 

In this paper, we propose a new spatial temperature aware transient EM induced stress analysis method. The new method consists of two new contributions: First, we propose a new TM-aware void saturation volume estimation method for fast immortality check in the post-voiding phase for the first time. We derive the analytic formula to estimate the void saturation in the presence of spatial temperature gradients due to Joule heating. Second, we developed a fast numerical solution for EM-induced stress analysis for multi-segment interconnect trees considering TM effect. The new method first transforms the coupled EM-TM partial differential equations into linear time-invariant ordinary differential equations (ODEs). Then extended Krylov subspace-based reduction technique is employed to reduce the size of the original system matrices so that they can be efficiently simulated in the time domain. The proposed method can perform the simulation process for both void nucleation and void growth phases under time-varying input currents and position-dependent temperatures. The numerical results show that, compared to the recently proposed semi-analytic EM-TM method, the proposed method can lead to about 28x speedup on average for the interconnect with up to 1000 branches for both void nucleation and growth phases with negligible errors.