Mobile manipulation is a crucial problem in various real-world applications. However, existing methods have demonstrated unsatisfactory training efficiency and sparse rewards, requiring complex coordination strategies between the mobile base and arm. In this paper, we propose RM-Planner, a planning method for mobile manipulation tasks in unknown complex environments. By adopting a two-layer hierarchical framework, we utilize a whole-body Model Predictive Control (MPC)-based low-level planner to track subgoals and generate aggressive but safe joint commands throughout the entire manipulation process, while a Reinforcement Learning (RL)based high-level policy directly uses 3D point cloud representations of the environment, guiding the robot to achieve optimal manipulation postures based on current observations and specific task objectives. We conduct extensive simulations and real-world experiments, where RM-planner significantly outperforms state-of-the-art methods. Our code will be released at https://github.com/SYSU-RoboticsLab/RM-Planner.git.

The rapid advancement of Internet of Things (IoT) networks has positioned unmanned aerial vehicles (UAV s) as critical enablers of next-generation wireless communication technologies. This paper focuses on orthogonal time frequency space (OTFS) modulation-assisted wireless control in UAV networks with finite blocklength (FBL) transmission. In particular, we in-vestigate the optimal power allocation that maximizes the fairness of control performance in terms of linear quadratic regulator (LQR) cost, subject to rate-LQR cost bounds and maximum available power budget constraints. To address the optimization problem, we first analyze the concave-convex property of the FBL rate function, followed by developing an efficient successive convex approximation (SCA)-based algorithm to obtain a sub-optimal solution. The convergence and computational complexity of the proposed algorithm are thoroughly analyzed. Simulation results validate the effectiveness of the proposed approach, offering promising insights for UAV-enabled wireless control systems.

We propose a robust beamforming design methodology for integrated sensing and communications (ISACs) beamform, where the beamforming design is investigated under the sensing optimal beamforming designed to overcome the channel uncertainty that arises from the communication system. Under the assumption that the channel state information (CSI) error is elliptically bounded, we study the robust ISAC beamforming design problem with the minimization of the Cramér-Rao bound (CRB) under the signal-to-noise ratio (SINR) threshold constraint. We consider the long-range and near-range cases separately and categorize them into point-target and extended-target for processing. In the point target scenario, we address the problem through distributed optimization using the S-procedure and solve it with the semidefinite relaxation (SDR) method. Meanwhile, in the extended target scenario, we transform the infinite constraints of the robust ISAC design problem into a finite set, employing linear matrix inequalities (LMIs) for equivalent representation. Under specific conditions, we illustrate that the SDR problem in this scenario can yield a rank-1 solution. Simulation results verify the effectiveness of the proposed CRB optimizationmin method and prove its application value in the next generation of Internet of Things devices.

Pavement distress, manifested as cracks, potholes, and rutting, significantly compromises road integrity and poses risks to drivers. Accurate prediction of pavement distress deterioration is essential for effective road management, cost reduction in maintenance, and improvement of traffic safety. However, real-world data on pavement distress is usually collected irregularly, resulting in uneven, asynchronous, and sparse spatial-temporal datasets. This hinders the application of existing spatial-temporal models, such as DCRNN, since they are only applicable to regularly and synchronously collected data. To overcome these challenges, we propose the Spatial-Temporal Graph Autoregression Network (STGAN), a novel graph neural network (GNN) model designed for accurately predicting irregular pavement distress deterioration using complex spatial-temporal data. Specifically, STGAN integrates the temporal domain into the spatial domain, creating a larger graph where nodes are represented by spatial-temporal tuples and edges are formed based on a similarity-based connection mechanism. Furthermore, based on the constructed spatiotemporal graph, we formulate pavement distress deterioration prediction as a graph autoregression task, i.e., the graph size increases incrementally and the prediction is performed sequentially. This is accomplished by a novel spatial-temporal attention mechanism deployed by the proposed STGAN model. Utilizing the ConTrack dataset, which contains pavement distress records collected from different locations in Shanghai, we demonstrate the superior performance of STGAN in capturing spatial-temporal correlations and addressing the aforementioned challenges. Experimental results further show that STGAN outperforms baseline models, and ablation studies confirm the effectiveness of its novel modules. Our findings contribute to promoting proactive road maintenance decision-making and ultimately enhancing road safety and resilience.

For tracking a rectangular vehicle, real-world automotive radar position measurements are distributed not uniformly over the vehicle extension but typically around the edges of the vehicle, i.e., the distribution of measurements is shape-dependent. To describe this phenomenon, a shape-dependent Gaussian mixture measurement model is presented, with each mixture component being used to describe a sub-rectangle region by introducing a shape scaling factor. The shape scaling factor is also shape-dependent and can characterize the measurement spread across the corresponding edge. In this model, parameters and mixture structure are highly shape-dependent, and the rectangular shape prior information is also incorporated. Based on the proposed model, a variational Bayesian approach is derived, which recursively and efficiently estimates the kinematic, shape, shape scaling factors, and orientation states of a vehicle. Additionally, the Doppler velocity measurement can also be integrated into the variational Bayesian framework by introducing a latent variable. This approach can effectively and adaptively describe the complex measurement distribution. From the simulation and real experimental results, the proposed approach has a great improvement in the tracking performance, and the superior performance of the proposed model is more significant in estimating the centroid position compared with the state-of-the-art approaches.

In the age of big data, the massive online user activity across desktop and mobile platforms generates an immense volume of web traffic logs. Analyzing and forecasting user behaviors, particularly page views, are vital for organizations aiming to enhance personalization, recommendation systems, and search engine optimization efforts. While traditional statistical methods have long been employed for web traffic forecasting, recent advancements in deep learning offer new opportunities for more accurate and insightful predictions. This paper presents a comprehensive comparative study of traditional statistical forecasting techniques and state-of-the-art deep learning methods applied to publicly available Wikipedia web traffic data, which includes hundreds of thousands of pages with over two years of historical page views. We evaluate various deep learning architectures encompassing different model structures, including recurrent neural networks (RNNs) and long short-term memory networks (LSTMs), multilayer perceptrons (MLPs), transformer-based models without pre-training, and pre-trained foundational models. Our findings reveal that deep learning approaches, particularly those leveraging cross-learning and transfer-learning capabilities, significantly outperform conventional methods in forecasting accuracy. These advanced models provide a powerful means to better understand online users’ browsing activities. The enhanced predictive performance of deep learning frameworks equips data scientists and researchers with more effective tools, ultimately improving productivity and efficiency in the analysis of web traffic patterns.

We propose a robust beamforming design methodology for integrated sensing and communications beamforms. The beamforming design aims to address channel uncertainties in the communication system by optimizing the sensing beamform. Assuming the Channel State Information (CSI) error is elliptically bounded, we investigate the robust integrated sensing and communication (ISAC) beamforming design problem, focusing on minimizing the Cramér-Rao bound (CRB) under a signal-to-noise ratio (SINR) threshold constraint. The problem is addressed through distributed optimization using the S-procedure and solved with the SDR method. Simulation results verify the effectiveness of the proposed CRB_min method.

For extended object tracking (EOT), the shape of an extended object (EO) is usually fixed (e.g., tracking vehicles), but the orientation varies. Thus, an accurate estimate of the time-varying orientation is important. The orientation and the heading are not always identical but highly dependent, which can be used as additional prior information to improve estimation accuracy, including the orientation. In view of this, this letter proposes a velocity-dependent orientation estimation approach to EOT utilizing this information. First, we model the quantity between the orientation and the heading as a Gaussian noise with zero mean and adaptive variance. Second, based on the proposed model and the integration of a pseudo-measurement, a variational Bayesian (VB) approach is proposed to estimate the kinematic, shape, and orientation states. The proposed approach can adapt to most dynamic scenarios without the need for a sophisticated mathematical model. The effectiveness of the proposed model and estimation approach is demonstrated by using simulated data.

Millimeter-Wave Massive MIMO is important for beyond 5G or 6G wireless communication networks. The goal of this paper is to establish successful communication between the cellular base stations and devices, focusing on the problem of joint user activity detection and channel estimation. Different from traditional compressed sensing (CS) methods that only use the sparsity of user activities, we develop several Approximate Message Passing (AMP) based CS algorithms by exploiting the sparsity of user activities and mmWave channels. First, a group soft-thresholding AMP is presented to utilize only the user activity sparsity. Second, a hard-thresholding AMP is proposed based on the on-grid CS approach. Third, a super-resolution AMP algorithm is proposed based on atomic norm, in which a greedy method is proposed as a super-resolution denoiser. And we smooth the denoiser based on Monte Carlo sampling to have Lipschitz continuity and present state evolution results. Extensive simulation results show that the proposed method outperforms the previous state-of-the-art methods.

Existing machine-learning research on power grids relies on online measurements without missing data. We propose a missing data reconstruction model based on generative adversarial networks to supplement existing methods. This model fits better spatio-temporal data with several improvements over previous approaches. First, the loss function considers both distribution and value differences, leveraging all available information to minimize differences between original and generated data. Then, a deep-learning architecture incorporating convolutional neural layers and nonlocal blocks is developed to extract the spatial-temporal information in electrical feature maps. The proposed method exhibits enhanced credibility by neglecting invalid consecutive data under phasor measurement unit (PMU) failures (proven by attention maps generated in nonlocal blocks), and higher accuracy than existing models for recovering data under random data missing/PMU failure conditions (proven by numerical results). Finally, the proposed data reconstruction model is effectively applied to an online framework for transient stability assessment.

Online transient stability assessment (TSA) is crucial for ensuring the security of modern grids. However, problems with limited sample sizes and data imbalance hinder the performance of data-driven TSA classifiers. Addressing this challenge, this article presents a generative adversarial network (GAN)-based model to generate instability samples. Unlike existing methods, our approach moderately alters the long-tailed distribution of instability moments within the raw dataset, producing a more diverse database for TSA tasks. A convolutional neural network-based supervised model, mapping the relationship between fault-clearance electrical characteristics and instability moments of the power system, is incorporated to drive the GAN model to generate realistic yet rare instability cases. Numerical results have proven the superiority of the proposed model over existing case generation methods in terms of realistic and diverse sample generation. In addition, the effectiveness of the proposed data augmentation scheme is demonstrated for online TSA applications.

Considering the charging needs of the Internet of Things, we introduce the simultaneous wireless information and power transfer (SWIPT) technology into the ultra-dense cloud radio access network (UD-CRAN) with wireless fronthaul. However, SWIPT can bring potential eavesdropping issues. In this paper, we study the secure communication caused by SWIPT in the UD-CRAN network. Specifically, the transmission schemes of wireless fronthaul and access links are jointly designed, while addressing the characteristics of ultra-dense networks, such as base station diversity and high probability of line-of-sight transmission. Aiming at maximizing the security energy efficiency, we jointly optimize the power allocation in the fronthaul and the resource allocation in the access link which includes beamforming for information and energy transmission, on/off of remote radio heads (RRHs), and user-RRH association. We propose an iterative algorithm based on the Dinkelbach’s transform to deal with the fractional objective function. To solve the mix-integer non-convex inner problem, we design: (1) a successive convex approximation (SCA) based method in which the problem at each iteration is a mixed-integer second-order cone program; (2) and an alternating optimization algorithm based on semidefinite relaxation (SDR) to further balance the complexity and performance. Finally, numerical results are presented to demonstrate the efficiency of the proposed schemes. Moreover, the proposed SCA method can achieve excellent performance while preserving integer variables, which inevitably increases algorithm complexity. Furthermore, the proposed SDR method can avoid the iteration process of SCA and further reducing the algorithm complexity.

Slow-time coded waveforms are commonly used in automotive multiple-input multiple-output (MIMO) radar to achieve inter-channel spatial diversity. Typical waveforms, such as Doppler division multiple access (DDMA) and code division multiple access (CDMA), are known to suffer from Doppler ambiguity and high side lobes, respectively. In this paper, we propose a Doppler-coded joint division multiple access orthogonal waveform for automotive MIMO radar. The proposed waveform reduces the sidelobes and extends the unambiguous range of velocity by using unequal intervals for Doppler division as well. The corresponding signal processing framework is derived to correctly separate orthogonal signals. Both simulations and experimental results demonstrate the effectiveness of the proposed waveform.

This paper studies the secrecy wireless information and power transfer problem in ultra-dense cloud radio access network (UD-CRAN) with wireless fronthaul, which is a promising framework for future Internet of Things (IoT). The transmission schemes of wireless fronthaul and access links are jointly designed, while addressing the characteristics of ultra-dense network such as base station diversity and high probability of line-of-sight transmission. Specifically, we employ the idea of block diagonalization to deal with the fronthaul interference, which support multi-stream fronthaul transmission for each remote radio head (RRH). We then jointly optimize the power allocation in the fronthaul and the resource allocation in the access link which includes beamforming for information and energy transmission, on/off of RRHs, and user-RRH association. In order to solve the formulated mixed integer non-convex optimization problem, we leverage the sparsity of beamforming vectors brought by the ultra-dense RRHs. We then solve the reformulated problem by employing the successive convex approximation approach. Finally, numerical results are presented to demonstrate the effectiveness of the proposed scheme.

Remote sensing image super-resolution (RSISR) plays a vital role in enhancing spatial detials and improving the quality of satellite imagery. Recently, Transformer-based models have shown competitive performance in RSISR. To mitigate the quadratic computational complexity resulting from global self-attention, various methods constrain attention to a local window, enhancing its efficiency. Consequently, the receptive fields in a single attention layer are inadequate, leading to insufficient context modeling. Furthermore, while most transform-based approaches reuse shallow features through skip connections, relying solely on these connections treats shallow and deep features equally, impeding the model’s ability to characterize them. To address these issues, we propose a novel transformer architecture called Cross-Spatial Pixel Integration and Cross-Stage Feature Fusion Based Transformer Network (SPIFFNet) for RSISR. Our proposed model effectively enhances context cognition and understanding of the entire image, facilitating efficient integration of features cross-stages. The model incorporates Cross-Spatial Pixel Integration Attention (CSPIA) to introduce contextual information into a local window, while Cross-Stage Feature Fusion Attention (CSFFA) adaptively fuses features from the previous stage to improve feature expression in line with the requirements of the current stage. We conducted comprehensive experiments on multiple benchmark datasets, demonstrating the superior performance of our proposed SPIFFNet in terms of both quantitative metrics and visual quality when compared to state-of-the-art methods. Our code is available at https://github.com/Dr-Lyt/SPIFFNet.

Most AI projects start with a Python notebook running on a single laptop; however, one usually needs to go through a mountain of pains to scale it to handle larger dataset (for both experimentation and production deployment). These usually entail many manual and error-prone steps for the data scientists to fully take advantage of the available hardware resources (e.g., SIMD instructions, multi-processing, quantization, memory allocation optimization, data partitioning, distributed computing, etc.). To address this challenge, we have open sourced BigDL 2.0 at https://github.com/intel-analytics/BigDL/ under Apache 2.0 license (combining the original BigDL [19] and Analytics Zoo [18] projects); using BigDL 2.0, users can simply build conventional Python notebooks on their laptops (with possible AutoML support), which can then be transparently accelerated on a single node (with up-to 9.6x speedup in our experiments), and seamlessly scaled out to a large cluster (across several hundreds servers in real-world use cases). BigDL 2.0 has already been adopted by many real-world users (such as Mastercard, Burger King, Inspur, etc.) in production.

Practical and accurate localization systems are important to mobile targets that enable promising services such as navigation and augmented reality. With the proliferation of WiFi, existing WiFi-based localization systems have leveraged RSSI, fingerprints, landmarks, time of arrival, or angle of arrival to locate targets, while no related work pays attention to mobile targets themselves. For wheel-driven mobile targets, such as vehicles, bikes, and wheeled robots, we design and implement WheelLoc, a novel WiFi-based localization system leveraging the rotation of wheels. The specially designed WheelLoc hardware is cost-effective and self-powered with the composition of three commercial antennas and a solar cell, which is also easy to be installed on wheels. A hybrid WheelLoc algorithm is further proposed to realize accurate localization in diverse environments, whether the wheel of targets is static or mobile, indoor or outdoor, on flat or bumpy ground. The movements of individual antennas are exploited to emulate linear, cycloid, and circular antenna arrays using a new formulation of Synthetic Aperture Radar (SAR). Extensive experiments are conducted on bikes in the real world. Performance results demonstrate that WheelLoc does not require any user interaction, yet achieves comparable accuracy with the state-of-the-art localization systems using WiFi.

In this paper, we consider an un-cooperative spectrum sharing scenario, where a radar system is to be overlaid to a pre-existing wireless communication system. Given the order of magnitude of the transmitted powers in play, we focus on the issue of interference mitigation at the communication receiver. We explicitly account for the reverberation produced by the (typically high-power) radar transmitter whose signal hits scattering centers (whether targets or clutter) producing interference onto the communication receiver, which is assumed to operate in an un-synchronized and un-coordinated scenario. We first show that the receiver design amounts to solve a joint (non-convex) interference removal and data demodulation problem. Next, we introduce two algorithms exploiting sparsity of a proper representation of the interference and the vector containing demodulation errors of the data block. The first algorithm is basically a relaxed constrained atomic norm minimization, while the latter relies on a two-stage processing structure and is based on alternating minimization. The merits of these algorithms are demonstrated through extensive simulations; interestingly, the two-stage alternating minimization algorithm turns out to achieve satisfactory performance with moderate computational complexity.

In this letter, we propose new sub-Nyquist sampling schemes for multiple sinusoids, which require fewer number of samples than previous works. Since it is impossible to resolve the frequency ambiguity using a single sub-Nyquist sample sequence, an additional sampling channel is used to determine the correct frequencies. First, a time-staggered sampling system, with the staggered time less than or equal to the Nyquist sampling interval, is proposed. This approach requires only 3K samples to estimate the K frequency components in the signal. However, aliasing can occur when the differences between some frequencies are integer multiples of the sampling rate. Then, another sampling strategy that makes use of feedback is proposed to prevent aliasing. We demonstrate that using two sampling channels and with feedback, 4K samples suffice to resolve both frequency ambiguity and image frequency aliasing. Simulation results are provided to demonstrate the effectiveness of the proposed systems.

The millimeter-wave (mmWave) full-dimensional (FD) MIMO system employs planar arrays at both the base station and the user equipment and can simultaneously support both azimuth and elevation beamforming. In this paper, we propose atomic-norm-based methods for mm-wave FD-MIMO channel estimation under both uniform planar arrays (UPA) and non-uniform planar arrays (NUPA). Unlike existing algorithms, such as compressive sensing (CS) or subspace methods, the atomic-norm-based algorithms do not require to discretize the angle spaces of the angle of arrival and angle of departure into grids, thus provide much better accuracy in estimation. In the UPA case, to reduce the computational complexity, the original large-scale atomic norm minimization problem is approximately reformulated as a semi-definite program (SDP) containing two decoupled two-level Toeplitz matrices. The SDP is then solved via the alternating direction method of multipliers where each iteration involves only closed-form computations. In the NUPA case, the atomic-norm-based formulation for channel estimation becomes nonconvex and a gradient-decent-based algorithm is proposed to solve the problem. Simulation results show that the proposed algorithms achieve better performance than the CS-based and subspace-based algorithms.

Wireless power transfer (WPT) provides a novel solution to the painstaking power-charging issue in wireless sensor networks. However, due to the propagation loss, the fast attenuation in energy transfer efficiency over the transmission distance is the main impediment to the WPT application. In this paper, we apply the simultaneous wireless information and power transfer (SWIPT) to a wirelessly powered sensor network, where each node has two circuits, which operate on energy harvesting mode and information decoding mode separately. We propose a novel cooperative SWIPT scheme (CSS) for this system. First, we present a conflict-free schedule initialization algorithm for CSS. For a given conflict-free schedule, we formulate a resource allocation problem to maximize the network energy efficiency, which is then transformed to an equivalent convex optimization problem and resolved via dual decomposition. Finally, a heuristic algorithm is presented to achieve the transmission schedule with the maximum energy efficiency and the corresponding resource assignment policy. Simulation results indicate that the CSS can significantly improve the energy efficiency of the wirelessly powered sensor network.

In many application areas ranging from bioinformatics to imaging, we are faced with the following question: can we recover a sparse vector xo∈ ℝNfrom its undersampled set of noisy observations y ∈ ℝn, y = Axo+w. The last decade has witnessed a surge of algorithms and theoretical results to address this question. One of the most popular schemes is the ℓp-regularized least squares given by the following formulation:x̂(y, p) ∈ arg minx(1/2)∥y - Ax∥22+ γ∥x∥pp, where p ∈ [0, 1]. Among these optimization problems, the case p = 1, also known as LASSO, is the best accepted in practice, for the following two reasons. First, thanks to the extensive studies performed in the fields of high-dimensional statistics and compressed sensing, we have a clear picture of LASSO's performance. Second, it is convex and efficient algorithms exist for finding its global minima. Unfortunately, neither of the above two properties hold for 0 ≤ pothan x̂(γ, 1). Second, if we employ iterative methods that aim to converge to a local minima of arg minx(1/2)∥y - Ax∥22+ γ∥x∥pp, then under good initialization, these algorithms converge to a solution that is still closer to xothan x̂(γ, 1). In spite of the existence of plenty of empirical results that support these folklore theorems, the theoretical progress to establish them has been very limited. This paper aims to study the above-mentioned folklore theorems and establish their scope of validity. Starting with approximate message passing (AMP) algorithm as a heuristic method for solving ℓp-regularized least squares, we study the following questions. First, what is the impact of initialization on the performance of the algorithm? Second, when does the algorithm recover the sparse signal xounder a “good” initialization? Third, when does the algorithm converge to the sparse signal regardless of the initialization? Studying these questions will not only shed light on the second folklore theorem, but also lead us to the answer the first one, i.e., the performance of the global optima x̂(γ, p). For that purpose, we employ the replica analysis1to show the connection between the solution of AMP and x̂(γ, p) in the asymptotic settings. This enables us to compare the accuracy of x̂(γ, p) and x̂(γ, 1). In particular, we will present an accurate characterization of the phase transition and noise sensitivity of ℓp-regularized least squares for every 0 ≤ pp-regularized least squares (if γ is tuned optimally) exhibits the same phase transition for every 0 ≤ pp-regularized least squares with different values of p. For instance, we will show that for very small and very large measurement noises, p = 0 and p = 1 outperform the other values of p, respectively.

Content Addressable Memory (CAM) is beneficial to applications that require high-speed pattern searching as it provides fast associative lookup operations. As the amount of data to search continues to grow, reducing power consumption while minimizing the costs for speed and area is the main thread of research in designing large capacity CAMs. In this work, we are presenting an active memory architecture incorporating a searchable resistive memory. The proposed architecture incorporates processing logic in close proximity to the RRAM and the RRAM-based TCAM, where the unit TCAM cell is comprised of five transistors and two memristors. Analyzed and simulated performance (e.g., latency, energy consumption, and storage density) of the RRAM-based TCAM at various technology nodes are presented and compared to those of prior CAM/TCAM designs.

Recovering a sparse vector x0∈ ℝNfrom its noisy linear observations, y ∈ ℝnwith y = Ax0+ w, has been the central problem of compressed sensing. One of the classes of recovery algorithms that has attracted attention is the class of ℓp-regularized least squares (LPLS) that seeks the minimum of 1/2 ∥y - Ax∥22+ λ∥x∥ppfor p ∈ [0, 1]. In this paper we employ the Replica method1from statistical physics to analyze the global minima of LPLS. Our paper reveals several surprising asymptotic properties of LPLS: (i) The phase transition curve of LPLS is the same for every 0 ≤ p0. (iii) Despite the equality of the phase transition curves, different values of p show different performances once a small amount of measurement noise, w, is added.

The replica method originated from statistical mechanics has been successfully applied to analyzing performance of the global maximum-likelihood (GML) MIMO detector in the large-system limit. In this paper, the analysis is extended to the local maximum-likelihood (LML) detectors. A bit error rate (BER) formula for the LML detectors with a fixed neighborhood size is obtained by the replica method and interestingly by the method of Gaussian approximation as well. It is shown that the LML BER is always one of the solutions to the GML BER in any system configuration. Furthermore, the LML BER is the only solution of the GML BER in a broad range of system parameters of practical interest. In the high signal-to-noise ratio regime, both LML and GML detectors achieve the AWGN channel performance when the channel load is up to 1.51 bits/dimension with an equal-energy distribution, and the load can be higher with an unequal-energy distribution. This analytical result is verified by simulation that the sequential likelihood ascent search detector, which is a linear-complexity LML detector, can approach the BER of the NP-hard GML detector predicted by the analysis. This result might be practically useful in large MIMO systems.

Predicting the future risks of stroke for patients is in high demands. In this paper, we proposed a model predictive of risks of stroke in future 1 year's period for patients across all age, all payor, and all disease groups in Maine, using demographics and clinical histories extracted from Electronic Medical Record (EMR) and clinical notes provided by Health Information Exchange (HIE). A retrospective cohort of 180,196 patients and a prospective cohort of 347,504 patients were constructed for model development and validation, respectively. A logistic regression model based on multivariate analysis was built for risk prediction. The model had a c-statistic of 0.887 in prospective testing, resulting in a sensitivity of 0.410 at a positive predictive value (PPV) of 0.262. Integration of this early-warning system into online patient monitoring platforms enables better management of population with chronic conditions.

A memristor-based architecture for neuromorphic computing is proposed. With memristors mimicking key characteristics of synapses and neurons, such nanoscale neural networks exhibit learning and memory effects with high integration density and scalability. Simulations demonstrate important features including adjustable spike generation, spike-timing and spike-rate dependent plasticity.

A memristors-based Ternary Content Addressable Memory (mTCAM) is presented. A unit mTCAM cell consists of 5T2R, five transistors and two memristors to store the ternary information, having higher storage density than conventional CMOS TCAMs together with the memristors' unique non-volatility. In the write mode, each memristor in the cell is programmed individually such that high impedance is always present between searchlines to reduce the direct current. A two-step write scheme is proposed to reduce the write voltage compliance, and the search voltage used to drive the search content was chosen to optimize the sensing margin. Simulation results for a 2×4 mTCAM array demonstrate the functionality and feasibility of the proposed mTCAM structure, in both write and search modes.

Unknown Noise-MUSIC (UN-MUSIC) is a promising method of direction of arrival (DOA) estimation in unknown spatially correlated noise using sparse arrays composed of two widely separated sub-arrays. The conventional UN-MUSIC estimator only utilizes information from one calibrated sub-array. If two sub-arrays are calibrated, a joint estimator that equally weights two sub-arrays’ conventional estimators has been found in literature. But no theoretical study has been reported. To compare and improve performance of different UN-MUSIC estimators, this paper proposes a unified framework of sub-array weighting to investigate the UN-MUSIC estimators, including the conventional one and the joint ones. The closed-form expression of the sub-array weighting estimator’s variance is derived which has not been done before. With the asymptotic results, different weighting strategies are compared and optimal weighting strategy to minimize estimation variance is proposed. Numerical simulations demonstrate the theoretical analysis and the validity of optimal weighting estimator.

A unified modeling approach is proposed to cover a broad range of memristive devices. The modular structure of the model enables it to represent behaviors of different types of devices. Resulted from theoretical analyses, the window function is uniquely controlled by the memristive flux. This not only solves the stability problem at boundaries present in previous models, but also reveals that an equivalent charge-flux constitutive relationship can be obtained from various types of memristive devices. Simulations on three device examples show that our model exhibits the device properties such as the frequency-dependent hysteresis, the limited memductance switching range with boundary assurances, the linear/nonlinear dopant drift, and the threshold voltages for read/write mode.

A CMOS current-reuse transmitter for ultra-low power (ULP) applications is presented. It can provide up to 10.2dBm of output power with a total efficiency of 30% at 2.4GHz. By utilizing the stacking technique, the average current of a class-E power amplifier is reused by the accompanying VCO and an optional RX block. The breakdown issue associated with the class-E PA is mitigated. A detailed analysis of the current-reuse structure is demonstrated. Practical design issues are discussed and appropriate design guidelines are provided.

Abstract Interactive generation of falling motions for virtual character with realistic responses to unexpected push, hit or collision with the environment is interesting work to many applications, such as computer games, film production, and virtual training environments. In this paper, we propose a new method to simulate protective behaviors in response to the ways a human may fall to the ground as well as incorporate the reactive motions into motion capture animation. It is based on simulated trajectory prediction and biomechanics inspired adjustment. According to the external perturbations, our system predicts a motion trajectory and uses it to select a desired transition‐to sequence. At the same time, physically generated falling motions will fill in the gap between the two‐motion capture sequences before and after the transition. Utilizing a parallel simulation, our method is able to predict a character's motion trajectory real‐time under dynamics, which ensures that the character moves towards the target sequence and makes the character's behavior more life‐like. Our controller is designed to generate physically plausible motion following an upcoming motion with adjustment from biomechanics rules, which is key to avoid an unconscious look for a character during the transition. Based on a relatively small motion database, our system is effective in generating various interactive falling behaviors. Copyright © 2006 John Wiley & Sons, Ltd.