Morphological Sampling Scheme for Analysis and Optimization of Wireless Indoor Environments

In this study, the combined use of 2 new techniques, nonuniform sampling (NUS) and carbon direct-detection (CDD), for the measurement of multidimensional NMR has been demonstrated. NUS is a method for sampling time domain data of indirect dimensions at random sparse points, whereas canonical multidimensional NMR spectra require all the regular grids as sampling points. Of several sparse sampling methods, NUS is advantageous, because it can reduce the errors that originate from the lack of sampling points by randomizing the sampling points. NUS can have denser sampling points at the beginning, where signals decay less than those in the end of sampling, and in turn enhance the signal-to-noise ratio (SNR) in the frequency dimensions per unit time. NUS is applicable to proteins that have been difficult to study using conventional NMR approaches. CDD is another area where applications have advanced considerably over the last several years. CDD is an effective technique for detecting NMR signals from proteins that exist under unfavorable conditions to observe or discriminate proton signals, which include paramagnetic proteins, highly deuterated samples, proteins of high salts, and disordered proteins. Despite the advances in NMR hardware, however, CDD has not been easily employed for practical applications, because the SNR of its result is insufficient. In this paper, NUS and CDD have been used in a combination in a 3D CBCACON experiment, expecting the supporting role of NUS in improving the SNR of CDD. The 3D CBCACON experiment provides the chemical shifts information of Cα/β(i-1)-N(i)-C'(i-1). The Bruker’s default pulse sequence of “c_cbcacon_ia3d” was modified to incorporate the NUS scheme. The sample used in the present study was ubiquitin (approximately 2.0 mM) and the reference and NUS experiments were performed using a Bruker AV 600 MHz NMR machine equipped with a TCI cryoprobe. The time domains of reference 3D CBCACON data consisted of 256* (F3: C') × 28* (F2: N) × 24* (F1: Cα/β) × 2 (IPAP) (*means complex point), and the number of transients was 8, leading to the total measurement time of 26 hours. The NUS schedule comprised 270 points, and it corresponded to about 40% of the intact points and required an 11-hour measuring time. As the chemical shifts of the Cα/β atoms evolved under constant time, the points for the Cα/β dimension were designed to have equal distribution, whereas the points for N dimension distributed to a Gaussian shape with higher density at the beginning (Fig. S1). To process the NUS data, multidimensional Fourier transformation (MFT) and the maximum entropy (MaxEnt) algorithms by using in-house written MFT processing program (J-G. Jee, unpublished data) and the Rowland NMR Toolkit (version 3.0), respectively, were employed. For conventional fast discrete Fourier transformation (FFT) processing and analyses of spectra, NMRPipe software was used. The overlaid 2D spectra of [C', N] and [C, Cα/β] in FFT, MFT, and MaxEnt results are shown in Figure 1. Because the number of sampling points in the current NUS decreased more than a half, a lower SNR was expected in the results. Although some noise peaks were observed in both the MFT and MaxEnt data due to the lower sensitivity caused by smaller points (Fig. 1), they are tolerable for the backbone chemical shifts assignment. Generally, the spectra reproduced by MaxEnt display a better resolution than those reproduced by MFT. However, insufficient quality or quantity of the input data may alter the chemical shifts. MFT is expected to generate fewer erroneous results despite its

Summary form only given, as follows. ICEPIC, developed by the air force research lab, is a 3-D particle-in-cell (PIC) code capable of being operated on parallel architectures using either the MPI or PVM message passing standards. ICEPIC utilizes a Cartesian coordinate-based grid system to simulate collisionless plasma physics phenomena, such as those that occur in high power microwave devices. Previous versions of ICEPIC represented curved boundaries by using 'stair-step' approximations. These Cartesian grid 'stair-steps' may cause inaccurate results in certain simulations, particularly in regions of extreme curvature. In order to increase the accuracy of the code, while retaining the computational ease of the Cartesian grid, various methods of embedded and conformal boundaries are considered. Preliminary results of tests of the accuracy and stability of these models are presented.

In this work, an analytical model describing the noise in the retrieved three contrast channels, transmission, refraction, and ultra small-angle scattering, obtained with edge illumination X-ray phase-based imaging system is presented and compared to experimentaldata. In EI, images acquired at different displacements of the presample mask (i.e., different illumination levels referred to as points on the "illumination curve"), followed by pixel-wise curve fitting, are exploited to quantitatively retrieve the three contrast channels. Therefore, the noise in the final image will depend on the error associated with the fitting process. We use a model based on the derivation of the standard error on fitted parameters, which relies on the calculation of the covariance matrix, to estimate the noise and the cross-channel correlation as a function of the position of the sampling points. In particular, we investigated the most common cases of 3 and 5 sampling points. In addition, simulations have been used to better understand the role of the integration time for each sampling point. Finally, the model is validated by comparison with the experimental data acquired with an edge illumination setup based on a tungsten rotating anode X-ray source and a photon countingdetector. We found a good match between the predictions of the model and the experimental data. In particular, for the investigated cases, an arrangement of the sampling points leading to minimum noise and cross-channel correlation can be found. Simulations revealed that, given a fixed overall scanning time, its distribution into the smallest possible number of sampling points needed for phase retrieval leads to minimum noise thanks to higher statistics perpoint. This work presents an analytical model describing the noise in the various contrast channels retrieved in edge illumination as a function of the illumination curve sampling. In particular, an optimal sampling scheme leading to minimum noise has been determined for the case where 3 or 5 sampling points are used, which represent two of the most common acquisition schemes. In addition, the correlation between noise in the different channels and the role of the number of points and exposure time have been also investigated. In general, our results suggest a series of procedures that should be followed in order to optimize the experimentalacquisitions.

Autonomous exploration requires robots to generate informative trajectories iteratively. Although sampling-based methods are highly efficient in unmanned aerial vehicle exploration, many of these methods do not effectively utilize the sampled information from the previous planning iterations, leading to redundant computation and longer exploration time. Also, few have explicitly shown their exploration ability in dynamic environments even though they can run real-time. To overcome these limitations, we propose a novel dynamic exploration planner (DEP) for exploring unknown environments using incremental sampling and Probabilistic Roadmap (PRM). In our sampling strategy, nodes are added incrementally and distributed evenly in the explored region, yielding the best viewpoints. To further shortening exploration time and ensuring safety, our planner optimizes paths locally and refine them based on the Euclidean Signed Distance Function (ESDF) map. Meanwhile, as the multi-query planner, PRM allows the proposed planner to quickly search alternative paths to avoid dynamic obstacles for safe exploration. Simulation experiments show that our method safely explores dynamic environments and outperforms the benchmark planners in terms of exploration time, path length, and computational time.

Investigating adaptive sampling strategies for optimal building energy performance using artificial neural networks and kriging surrogate models

The distributed representation‐type three‐layered perceptron with backpropagation has such problems as the local minimum, long learning time, and ambiguity in internal representation. As a method to cope with those problems, this paper proposes the four‐layered perceptron, together with the learning algorithm, where a hidden layer is added, so that each discrete sample point can perfectly be represented by the corresponding output of the upper hidden layer.First, the learning algorithm of the perceptron is applied successively to the sample points, and the learning is executed so that the input sample points are separated perfectly by the piecewise sets of hyperplanes. In this mechanism, the output matrix of the lower bidder layer output is nonsingular. Consequently, the following four‐layered perceptron can be constructed, where the output matrix of the upper hidden layer is an identity matrix, and any discrete value can be produced as the output from the output layer by adjusting the network coefficients. Computational experiments are made for the realization of the three‐valued logic function, which is a learning problem on the two‐dimensional plane, as well as the pattern recognition problem by the representative sample points. As a result, it is shown that the learning converges in less than 1/100 computation time, compared to the three‐layered perceptron with the backpropagation.

This paper focuses on a fast ray path searching method applied to the ray tracing propagation prediction in order to reduce considerable computation time and on keeping the accuracy in the propagation analysis of the communication service environment. As results, the acceleration performance of a proposed ray path searching method is compared with one of a general method, and be enhanced on the computational prediction time.

In this paper we present two numerical methods for the calculation of Gaussian field incident on a pinhole. The first method is based on Kirchhoff's Huygens aperture field method and the other on the spectrum of plane wave approach. Numerous calculations proved that Kirchhoff's Huygens method is faster and gives accurate results in the Fraunchofer region. The plane wave spectrum is the only one giving valid results ranging from the radiating aperture to, theoretically any distance, but the computational time is rapidly increasing for distances where the other method gives accurate and fast results.

MEMS light detection and ranging (LiDAR) is becoming an indispensable sensor in vehicle environment sensing systems due to its low cost and high performance. The beam scanning trajectory, sampling scheme and gridding are the key technologies of MEMS LiDAR imaging. In Lissajous scanning mode, this paper improves the sampling scheme, through which a denser Cartesian grid of point cloud data at the same scanning frequency can be obtained. By summarizing the rules of the Cartesian grid, a general sampling scheme independent of the beam scanning trajectory patterns is proposed. Simulation and experiment results show that compared with the existing sampling scheme, the resolution and the number of points per frame are both increased by 2 times with the same hardware configuration and scanning frequencies for a MEMS scanning mirror (MEMS-SM). This is beneficial for improving the point cloud imaging performance of MEMS LiDAR.

This paper presents a Sequential Approximate Optimization (SAO) procedure that uses the Radial Basis Function (RBF) network. If the objective and constraints are not known explicitly but can be evaluated through a computationally intensive numerical simulation, the response surface, which is often called meta-modeling, is an attractive method for finding an approximate global minimum with a small number of function evaluations. An RBF network is used to construct the response surface. The Gaussian function is employed as the basis function in this paper. In order to obtain the response surface with good approximation, the width of this Gaussian function should be adjusted. Therefore, we first examine the width. Through this examination, some sufficient conditions are introduced. Then, a simple method to determine the width of the Gaussian function is proposed. In addition, a new technique called the adaptive scaling technique is also proposed. The sufficient conditions for the width are satisfied by introducing this scaling technique. Second, the SAO algorithm is developed. The optimum of the response surface is taken as a new sampling point for local approximation. In addition, it is necessary to add new sampling points in the sparse region for global approximation. Thus, an important issue for SAO is to determine the sparse region among the sampling points. To achieve this, a new function called the density function is constructed using the RBF network. The global minimum of the density function is taken as the new sampling point. Through the sampling strategy proposed in this paper, the approximate global minimum can be found with a small number of function evaluations. Through numerical examples, the validities of the width and sampling strategy are examined in this paper.

Sampling-based motion planning (SBMP) is a major trajectory planning approach in autonomous driving given its high efficiency in practice. As the core of SBMP schemes, sampling strategy holds the key to whether a smooth and collision-free trajectory can be found in real-time. Although some bias sampling strategies have been explored in the literature to accelerate SBMP, the trajectory generated under existing bias sampling strategies may lead to sharp lane changing. To address this issue, we propose a new learning framework for SBMP. Specifically, we develop a novel automatic labeling scheme and a 2-Stage prediction model to improve the accuracy in predicting the intention of surrounding vehicles. We then develop an imitation learning scheme to generate sample points based on the experience of human drivers. Using the prediction results, we design a new bias sampling strategy to accelerate the SBMP algorithm by strategically selecting necessary sample points that can generate a smooth and collision-free trajectory and avoid sharp lane changing. Data-driven experiments show that the proposed sampling strategy outperforms existing sampling strategies, in terms of the computing time, traveling time, and smoothness of the trajectory. The results also show that our scheme is even better than human drivers.

Juvenile dermatomyositis (JDM) is a rare systemic autoimmune disease specifically affecting children. Mycophenolate mofetil (MMF) is an immunosuppressant used to treat JDM. Mycophenolic acid (MPA) is an active metabolite of MMF. This study aimed to develop a population pharmacokinetic (PPK) model of MPA in children with JDM and optimize the limited sampling strategy (LSS). Fifteen JDM patients treated with MMF, at a median age of 7.35 (range, 3.09–16.1) years, were included. Blood samples were collected at 30 minutes pre-dose, 20 minutes, 60 minutes and 180 minutes post-dose to measure the MPA concentrations. Data were retrospectively collected from the electronic medical records. A two-compartment model with first-order absorption, lag time in absorption, and first-order elimination was developed. Height and co-administered cotrimoxazole were added as the covariates to the model. Concentrations at different time points were simulated and area under the concentration–time curve (AUC0–12 h) was calculated. By removing one sampling point at a time, AUC0–12 h from three-point sampling strategy was re-calculated via Bayesian approach. AUC0–12 h from the three-point sampling strategy (by removing the point at 20 minutes post-dose) had the strongest correlation with AUC0–12 h from the four-point sampling strategy (Pearson’s r = 0.971).

Exploratory landscape analysis (ELA) supports supervised learning approaches for automated algorithm selection and configuration by providing sets of features that quantify the most relevant characteristics of the optimization problem at hand. In black-box optimization, where an explicit problem representation is not available, the feature values need to be approximated from a small number of sample points. In practice, uniformly sampled random point sets and Latin hypercube constructions are commonly used sampling strategies. In this work, we analyze how the sampling method and the sample size influence the quality of the feature value approximations and how this quality impacts the accuracy of a standard classification task. While, not unexpectedly, increasing the number of sample points gives more robust estimates for the feature values, to our surprise we find that the feature value approximations for different sampling strategies do not converge to the same value. This implies that approximated feature values cannot be interpreted independently of the underlying sampling strategy. As our classification experiments show, this also implies that the feature approximations used for training a classifier must stem from the same sampling strategy as those used for the actual classification tasks. As a side result we show that classifiers trained with feature values approximated by Sobol' sequences achieve higher accuracy than any of the standard sampling techniques. This may indicate improvement potential for ELA-trained machine learning models.

Soil loss evaluation by means of radioisotopic content measurements represents a promising technique, half-way between field surveys and theoretical models, which still suffers from its practical limits when applied at basin scale. A main limit is represented by the costs of field investigations and laboratory investigations on the high number of soil samples required by a traditional sampling scheme, such as regular grid or transects. A non-conventional soil sampling scheme was tested with the aim to consider a relative scanty number of soil samples and to verify the feasibility of this technique on large areas. This scheme was based on the hypothesis that land analysis and classification could point out areas characterised by homogeneous behaviour with respect to the 137Cs deposition and transmission model. A Geographic Information System (GIS)-aided procedure allowed to classify the selected basin area in pedo-morphological units, representative of the different pedologic, morphologic and land-use conditions, to locate few sampling points for each unit. Outcomes pointed out a low correlation between 137Cs contents and soil physical and compositional characteristics. Nevertheless, the isotopic methodology allowed to estimate a total soil loss value at basin scale almost consistent with both observed data, given by reservoir sedimentation measurements and estimates from the application of the RUSLE model. Thus, the results can be considered encouraging and they allow to deem that the isotopic methodology can be refined in order to account for erosion and deposition processes even at river basin scale and with a limited number of soil samples.

How many samples do we need to be representative? Grid sampling in Danish waters for assessing the distribution of microplastics and tire wear particles in seabed sediments.