This paper addresses the optimization problem of symbol-level precoding (SLP) in the downlink of a multiuser multiple-input multiple-output (MU-MIMO) wireless system while the precoder's output is subject to partially-known distortions. In particular, we assume a linear distortion model with bounded additive noise. The original signal-to-interference-plus-noise ratio (SINR) -constrained SLP problem minimizing the total transmit power is first reformulated as a penalized unconstrained problem, which is referred to as the relaxed robust formulation. We then adopt a worst-case design approach to protect the users' intended symbols and the targeted constructive interference with a desired level of confidence. Due to the non-convexity of the relaxed robust formulation, we propose an iterative algorithm based on the block coordinate ascent-descent method. We show through simulation results that the proposed robust design is flexible in the sense that the CI constraints can be relaxed so as to keep a desirable balance between achievable rate and power consumption. Remarkably, the new formulation yields more energy-efficient solutions for appropriate choices of the penalty parameter, compared to the original problem.
The robust adaptive beamforming design problem based on estimation of the signal of interest steering vector is considered in the paper. In this case, the optimal beamformer is obtained by computing the sample matrix inverse and an optimal estimate of the signal of interest steering vector. The common criteria to find the best estimate of the steering vector are the beamformer output SINR and output power, while the constraints assume as little as possible prior inaccurate knowledge about the signal of interest, the propagation media, and the antenna array. Herein, a new beamformer output power maximization problem is formulated and solved subject to a double-sided norm perturbation constraint, a similarity constraint, and a quadratic constraint that guarantees that the direction-of-arrival (DOA) of the signal of interest is away from the DOA region of all linear combinations of the interference steering vectors. In the new robust design, the prior information required consists of some allowable error norm bounds, the approximate knowledge of the antenna array geometry, and the angular sector of the signal of interest. It turns out that the array output power maximization problem is a non-convex QCQP problem with inhomogeneous constraints. However, we show that the problem is still solvable, and develop efficient algorithms for finding globally optimal estimate of the signal of interest steering vector. The results are generalized to the case where an ellipsoidal constraint is considered, and sufficient conditions for the global optimality are derived. In addition, a new quadratic constraint on the actual signal steering vector is proposed in order to improve the array performance. To validate our results, simulation examples are presented, and they demonstrate the improved performance of the new robust beamformers in terms of the output SINR as well as the output power.
In this paper, we investigate the downlink transmission of a multiuser multiple-input single-output (MISO) channel under a symbol-level precoding (SLP) scheme, having imperfect channel knowledge at the transmitter. In defining the SLP problem, a general category of constructive interference regions (CIR) called distance preserving CIR (DPCIR) is adopted. In particular, we are interested in the robust SLP design minimizing the total transmit power while satisfying the users' quality-of-service (QoS) requirements. We consider two common models for the channel uncertainty region, namely, norm-bounded spherical and stochastic. For the spherical uncertainty model, a worst-case robust precoder is proposed, while for the stochastic uncertainties, we define a convex optimization problem with probabilistic constraints. We simulate the performance of the proposed robust approaches, and compare them with the existing methods. Through the simulation results, we also show that there is an essential trade-off between the two robust approaches.
This paper investigates the use of Reinforcement Learning for the robust design of low-thrust interplanetary trajectories in presence of severe disturbances, modeled alternatively as Gaussian additive process noise, observation noise, control actuation errors on thrust magnitude and direction, and possibly multiple missed thrust events. The optimal control problem is recast as a time-discrete Markov Decision Process to comply with the standard formulation of reinforcement learning. An open-source implementation of the state-of-the-art algorithm Proximal Policy Optimization is adopted to carry out the training process of a deep neural network, used to map the spacecraft (observed) states to the optimal control policy. The resulting Guidance and Control Network provides both a robust nominal trajectory and the associated closed-loop guidance law. Numerical results are presented for a typical Earth-Mars mission. First, in order to validate the proposed approach, the solution found in a (deterministic) unperturbed scenario is compared with the optimal one provided by an indirect technique. Then, the robustness and optimality of the obtained closed-loop guidance laws is assessed by means of Monte Carlo campaigns performed in the considered uncertain scenarios. These preliminary results open up new horizons for the use of reinforcement learning in the robust design of interplanetary missions.
Purpose. Needle placement is a challenging problem for applications such as biopsy or brachytherapy. Tip force sensing can provide valuable feedback for needle navigation inside the tissue. For this purpose, fiber-optical sensors can be directly integrated into the needle tip. Optical coherence tomography (OCT) can be used to image tissue. Here, we study how to calibrate OCT to sense forces, e.g. during robotic needle placement. Methods. We investigate whether using raw spectral OCT data without a typical image reconstruction can improve a deep learning-based calibration between optical signal and forces. For this purpose, we consider three different needles with a new, more robust design which are calibrated using convolutional neural networks (CNNs). We compare training the CNNs with the raw OCT signal and the reconstructed depth profiles. Results. We find that using raw data as an input for the largest CNN model outperforms the use of reconstructed data with a mean absolute error of 5.81 mN compared to 8.04 mN. Conclusions. We find that deep learning with raw spectral OCT data can improve learning for the task of force estimation. Our needle design and calibration approach constitute a very accurate fiber-optical sensor for measuring forces at the needle tip.
Recent works propose neural network- (NN-) inspired analog-to-digital converters (NNADCs) and demonstrate their great potentials in many emerging applications. These NNADCs often rely on resistive random-access memory (RRAM) devices to realize the NN operations and require high-precision RRAM cells (6~12-bit) to achieve a moderate quantization resolution (4~8-bit). Such optimistic assumption of RRAM resolution, however, is not supported by fabrication data of RRAM arrays in large-scale production process. In this paper, we propose an NN-inspired super-resolution ADC based on low-precision RRAM devices by taking the advantage of a co-design methodology that combines a pipelined hardware architecture with a custom NN training framework. Results obtained from SPICE simulations demonstrate that our method leads to robust design of a 14-bit super-resolution ADC using 3-bit RRAM devices with improved power and speed performance and competitive figure-of-merits (FoMs). In addition to the linear uniform quantization, the proposed ADC can also support configurable high-resolution nonlinear quantization with high conversion speed and low conversion energy, enabling future intelligent analog-to-information interfaces for near-sensor analytics and processing.
Deep neural networks (DNNs) are vulnerable to subtle adversarial perturbations applied to the input. These adversarial perturbations, though imperceptible, can easily mislead the DNN. In this work, we take a control theoretic approach to the problem of robustness in DNNs. We treat each individual layer of the DNN as a nonlinear dynamical system and use Lyapunov theory to prove stability and robustness locally. We then proceed to prove stability and robustness globally for the entire DNN. We develop empirically tight bounds on the response of the output layer, or any hidden layer, to adversarial perturbations added to the input, or the input of hidden layers. Recent works have proposed spectral norm regularization as a solution for improving robustness against l2 adversarial attacks. Our results give new insights into how spectral norm regularization can mitigate the adversarial effects. Finally, we evaluate the power of our approach on a variety of data sets and network architectures and against some of the well-known adversarial attacks.
Federated learning is a communication-efficient training process that alternates between local training at the edge devices and averaging the updated local model at the central server. Nevertheless, it is impractical to achieve a perfect acquisition of the local models in wireless communication due to noise, which also brings serious effects on federated learning. To tackle this challenge, we propose a robust design for federated learning to alleviate the effects of noise in this paper. Considering noise in the two aforementioned steps, we first formulate the training problem as a parallel optimization for each node under the expectation-based model and the worst-case model. Due to the non-convexity of the problem, a regularization for the loss function approximation method is proposed to make it tractable. Regarding the worst-case model, we develop a feasible training scheme which utilizes the sampling-based successive convex approximation algorithm to tackle the unavailable maxima or minima noise condition and the non-convex issue of the objective function. Furthermore, the convergence rates of both new designs are analyzed from a theoretical point of view. Finally, the improvement of prediction accuracy and the reduction of loss function are demonstrated via simulations for the proposed designs.
We present a new method for design problems wherein the goal is to maximize or specify the value of one or more properties of interest. For example, in protein design, one may wish to find the protein sequence that maximizes fluorescence. We assume access to one or more, potentially black box, stochastic "oracle" predictive functions, each of which maps from input (e.g., protein sequences) design space to a distribution over a property of interest (e.g. protein fluorescence). At first glance, this problem can be framed as one of optimizing the oracle(s) with respect to the input. However, many state-of-the-art predictive models, such as neural networks, are known to suffer from pathologies, especially for data far from the training distribution. Thus we need to modulate the optimization of the oracle inputs with prior knowledge about what makes `realistic' inputs (e.g., proteins that stably fold). Herein, we propose a new method to solve this problem, Conditioning by Adaptive Sampling, which yields state-of-the-art results on a protein fluorescence problem, as compared to other recently published approaches. Formally, our method achieves its success by using model-based adaptive sampling to estimate the conditional distribution of the input sequences given the desired properties.
The supplier selection problem is based on electing the best supplier from a group of pre-specified candidates, is identified as a Multi Criteria Decision Making (MCDM), is proportionately significant in terms of qualitative and quantitative attributes. It is a fundamental issue to achieve a trade-off between such quantifiable and unquantifiable attributes with an aim to accomplish the best solution to the abovementioned problem. This article portrays a metaheuristic based optimization model to solve this NP-Complete problem. Initially the Analytic Hierarchy Process (AHP) is implemented to generate an initial feasible solution of the problem. Thereafter a Simulated Annealing (SA) algorithm is exploited to improve the quality of the obtained solution. The Taguchi robust design method is exploited to solve the critical issues on the subject of the parameter selection of the SA technique. In order to verify the proposed methodology the numerical results are demonstrated based on tangible industry data.
This paper presents the building heating demand prediction model with occupancy profile and operational heating power level characteristics in short time horizon (a couple of days) using artificial neural network. In addition, novel pseudo dynamic transitional model is introduced, which consider time dependent attributes of operational power level characteristics and its effect in the overall model performance is outlined. Pseudo dynamic model is applied to a case study of French Institution building and compared its results with static and other pseudo dynamic neural network models. The results show the coefficients of correlation in static and pseudo dynamic neural network model of 0.82 and 0.89 (with energy consumption error of 0.02%) during the learning phase, and 0.61 and 0.85 during the prediction phase respectively. Further, orthogonal array design is applied to the pseudo dynamic model to check the schedule of occupancy profile and operational heating power level characteristics. The results show the new schedule and provide the robust design for pseudo dynamic model. Due to prediction in short time horizon, it finds application for Energy Services Company (ESCOs) to manage the heating load for dynamic control of heat production system.
We consider the problem of optimally designing a body wireless sensor network, while taking into account the uncertainty of data generation of biosensors. Since the related min-max robustness Integer Linear Programming (ILP) problem can be difficult to solve even for state-of-the-art commercial optimization solvers, we propose an original heuristic for its solution. The heuristic combines deterministic and probabilistic variable fixing strategies, guided by the information coming from strengthened linear relaxations of the ILP robust model, and includes a very large neighborhood search for reparation and improvement of generated solutions, formulated as an ILP problem solved exactly. Computational tests on realistic instances show that our heuristic finds solutions of much higher quality than a state-of-the-art solver and than an effective benchmark heuristic.
Face detection is a well-explored problem. Many challenges on face detectors like extreme pose, illumination, low resolution and small scales are studied in the previous work. However, previous proposed models are mostly trained and tested on good-quality images which are not always the case for practical applications like surveillance systems. In this paper, we first review the current state-of-the-art face detectors and their performance on benchmark dataset FDDB, and compare the design protocols of the algorithms. Secondly, we investigate their performance degradation while testing on low-quality images with different levels of blur, noise, and contrast. Our results demonstrate that both hand-crafted and deep-learning based face detectors are not robust enough for low-quality images. It inspires researchers to produce more robust design for face detection in the wild.