This paper presents the design of a novel low-profile ultra-wideband textile antenna for wearable microwave medical imaging systems. The proposed antenna is based on a monopole structure, where two triangles and a few parallel slots are cut at the bottom corners and top edge of the radiation patch, respectively, to achieve an optimized ultra-wide bandwidth and a reduced size of the antenna. Polyester fabrics and conductive copper taffeta, which are universally available on the market, are selected to construct the antenna. Thus, the fabricated prototypes well suit the needs of wearable applications, where flexible components are required to conform to the curved human bodies. The measured results show that the antenna prototype can realize a bandwidth of 109% from 1.198–4.055 GHz with a realized gain of 2.9 dBi. Furthermore, the on-phantom measurements show that there is limited effect on the operating bandwidth and realized gain of the antenna when it is working in the proximity to human bodies. The antenna is used to monitor the recovery process of a bone fracture that is emulated by a body-mimicking phantom with a size-varying blood strip. The time-domain reflection coefficient of the antenna varies significantly with the size of the fracture introduced, which demonstrates the applicability of the antenna for such use scenarios in microwave medical imaging.

A glide-symmetry Goubau line leaky-wave antenna for achieving both high scanning rate and wide scanning angle range is introduced. The novelty of the design is its combination of glide symmetry and a single-layer Goubau line. This combination leads to performance enhancement in both scanning rate and scanning angle range. Utilizing a corrugated Goubau line in a glide-symmetry fashion closes its stopband at the cutoff frequency while slowing down its group velocity. As the dispersion around the closed stopband is shifted to the fast-wave region by periodic profile modulation, the space harmonics exhibit high-scanning-rate radiation. In conjunction with this the single-layer nature of the Goubau line allows the beam to scan from nearly endfire to backfire directions, leading to a wide scanning angle range. Numerical examples and experimental validation are provided to validate the proposed design. These show that the proposed antenna exhibits dual-band operation, where one band scans 152 degree within 2.2 GHz (corresponding to scanning rate of 69 degree/GHz) with 67% efficiency while the second band scans 145 degree within 0.7 GHz (corresponding to scanning rate of 207 degree/GHz) with 26% efficiency. A tradeoff between scanning rate and efficiency always exists in high-scanning-rate leaky-wave antennas. One is able to freely choose the preferred scanning rate in this single antenna. A comparison with other related works shows that the proposed antenna exhibits the best performance.

Stroke is a serious manifestation of various cerebrovascular diseases and one of the most dangerous diseases in the world today. Volume quantification and location detection of chronic stroke lesions provide vital biomarkers for stroke rehabilitation. Recently, deep learning has seen a rapid growth, with a great potential in segmenting medical images. In this work, unlike most deep learning-based segmentation methods utilizing only magnetic resonance (MR) images as the input, we propose and validate a novel stroke lesion segmentation approach named multi-inputs UNet (MI-UNet) that incorporates brain parcellation information, including gray matter (GM), white matter (WM) and lateral ventricle (LV). The brain parcellation is obtained from 3D diffeomorphic registration and is concatenated with the original MR image to form two-channel inputs to the subsequent MI-UNet. Effectiveness of the proposed pipeline is validated using a dataset consisting of 229 T1-weighted MR images. Experiments are conducted via a five-fold cross-validation. The proposed MI-UNet performed significantly better than UNet in both 2D and 3D settings. Our best results obtained by 3D MI-UNet has superior segmentation performance, as measured by the Dice score, Hausdorff distance, average symmetric surface distance, as well as precision, over other state-of-the-art methods.

A leaky-wave antenna capable of frequency-dependent beam steering and featuring a high scanning rate is introduced. The potential usefulness of the antenna resides in its ability to act as a time-space transformer in which temporal frequency-domain information can be transformed into the spatial frequency domain. Possible applications for the antenna include analog spectrum analyzers based on time-to-space transformation and also microwave imaging systems based on space-to-time conversion. One desirable property for the antenna in these types of applications is a high scanning rate where a small change in frequency can lead to significant changes in beam angle. To meet this requirement, we propose a novel high scanning-rate leaky-wave antenna using complementary microstrip-slot stubs. The design features a simple implementation and easy fabrication. It is verified by numerical examples and an experimental example. The resultant antenna design can provide a scanning rate of 135°/GHz.

Efficient virtual machine (VM) management can dramatically reduce energy consumption in data centers. Existing VM management algorithms fall into two categories based on whether the VMs' resource demands are assumed to be static or dynamic. The former category fails to maximize the resource utilization as they cannot adapt to the dynamic nature of VMs' resource demands. Most approaches in the latter category are heuristical and lack theoretical performance guarantees. In this work, we formulate dynamic VM management as a large-scale Markov Decision Process (MDP) problem and derive an optimal solution. Our analysis of real-world data traces supports our choice of the modeling approach. However, solving the large-scale MDP problem suffers from the curse of dimensionality. Therefore, we further exploit the special structure of the problem and propose an approximate MDP-based dynamic VM management method, called MadVM. We prove the convergence of MadVM and analyze the bound of its approximation error. Moreover, MadVM can be implemented in a distributed system, which should suit the needs of real data centers. Extensive simulations based on two real-world workload traces show that MadVM achieves significant performance gains over two existing baseline approaches in power consumption, resource shortage and the number of VM migrations. Specifically, the more intensely the resource demands fluctuate, the more MadVM outperforms.

This article proposes a modified flux observer for sensorless direct torque control (DTC) of the dual three-phase permanent-magnet synchronous motor. The proposed method can improve torque performance without a position sensor. The flux observer is proposed to obtain the stator fluxes and rotor fluxes. The stator fluxes can be utilized to calculate the amplitude of torque and flux for DTC system. Meanwhile, the position information can be detected from the rotor fluxes by using a normalized phase-locked loop. This new flux observer can overcome the disadvantages due to the dc bias, initial value, and harmonic components. The stability of the proposed observer is also analyzed in this article. In addition, a fault-tolerant DTC scheme for open-circuit fault is introduced. The performances of torque control and position estimation are evaluated under healthy, open-circuit fault, and fault-tolerant conditions. Simulation and experimental evaluation are carried out to confirm the effectiveness of the proposed sensorless DTC scheme.