Summary We propose a novel sparse tensor decomposition method, namely the tensor truncated power method, that incorporates variable selection in the estimation of decomposition components. The sparsity is achieved via an efficient truncation step embedded in the tensor power iteration. Our method applies to a broad family of high dimensional latent variable models, including high dimensional Gaussian mixtures and mixtures of sparse regressions. A thorough theoretical investigation is further conducted. In particular, we show that the final decomposition estimator is guaranteed to achieve a local statistical rate, and we further strengthen it to the global statistical rate by introducing a proper initialization procedure. In high dimensional regimes, the statistical rate obtained significantly improves those shown in the existing non‐sparse decomposition methods. The empirical advantages of tensor truncated power are confirmed in extensive simulation results and two real applications of click‐through rate prediction and high dimensional gene clustering.
The historical development and current state-of-the-art of various joint wireless communication and radar sensing systems are reviewed and discussed in this study. Different kinds of systems are categorised according to their modulation waveforms and duplex schemes. Pros and cons of each category are highlighted. To showcase the current research advances, several demonstration systems are introduced with emphasis on proposed research contributions in this emerging area, and their performances are compared with respect to both communication and radar modes. Also, a number of challenges are identified for the near future system developments and applications.
A compact filter for ultra-wideband (UWB) applications is proposed and developed through quasi-equal allocation of the first three resonant frequencies of a stub-loaded resonator together with strong input/ output excitation. Measured results show low insertion loss and flat group delay in the entire passband as well as a wide stopband. The size for this single-layer UWB filter without feeding lines is only 9.5 Â 5 mm (0.53 l g Â0.28 l g in which l g is the guided wavelength of 50 V microstrip at the centre frequency).Introduction: Ultra-wideband (UWB) technology has been recognised as a promising solution for high-resolution radar, high data rate communication and power-efficient RF tracking and positioning systems. Since the Federal Communications Commission (FCC) authorised the usage of the unlicensed operation band of 3.1-10.6 GHz for commercial applications [1], academic researchers and industry engineers have been passionate in the design of UWB filters. The technical requirements for UWB filters can be summarised as follows: low insertion loss, flat group delay and high out-of-band selectivity. To meet these requirements, different techniques have been proposed. In [2], a microstrip UWB filter was studied and it is composed of five-stage cascaded dissimilar ring filters, which generate two stopbands at both low and high edges of the UWB passband. However, this kind of realisation has poor spurious performance because it is basically a dual-band bandstop filter. Another intuitive idea for designing UWB filters is to combine lowpass (LPF) and highpass filters (HPF) [3]. To compensate for the mutual perturbation between them, an iterative optimisation is required. The third way to design UWB filters is to use a so-called multiple-mode resonator (MMR). In this technique, the reflection zeros of the filtering response are brought by the first three [4] or four resonant modes [5] of a stepped impedance resonator (SIR) and extremely strong and frequencydependent input/output coupling. Recently, a UWB filter was realised using multilayer liquid crystal polymer technology [6]. Considering the availability of the mature PCB fabrication process, a single-layer realisation of a UWB filter is more attractive and cost-effective.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.