Hyperspectral (HS) images are characterized by approximately contiguous spectral information, enabling the fine identification of materials by capturing subtle spectral discrepancies. Owing to their excellent locally contextual modeling ability, conv
olutional neural networks (CNNs) have been proven to be a powerful feature extractor in HS image classification. However, CNNs fail to mine and represent the sequence attributes of spectral signatures well due to the limitations of their inherent network backbone. To solve this issue, we rethink HS image classification from a sequential perspective with transformers, and propose a novel backbone network called ul{SpectralFormer}. Beyond band-wise representations in classic transformers, SpectralFormer is capable of learning spectrally local sequence information from neighboring bands of HS images, yielding group-wise spectral embeddings. More significantly, to reduce the possibility of losing valuable information in the layer-wise propagation process, we devise a cross-layer skip connection to convey memory-like components from shallow to deep layers by adaptively learning to fuse soft residuals across layers. It is worth noting that the proposed SpectralFormer is a highly flexible backbone network, which can be applicable to both pixel- and patch-wise inputs. We evaluate the classification performance of the proposed SpectralFormer on three HS datasets by conducting extensive experiments, showing the superiority over classic transformers and achieving a significant improvement in comparison with state-of-the-art backbone networks. The codes of this work will be available at url{https://sites.google.com/view/danfeng-hong} for the sake of reproducibility.
In this work, we study the impact of the multiplicative phase noise in an IRS-assisted system. We consider an IRS-assisted system with multiplicative phase noise both at the BS and user. A novel channel estimation algorithm is proposed considering th
e phase noise. By utilizing the proposed channel estimates we investigate the system performance in the downlink, more specifically, we derive the ergodic capacity in closed form. Simulation results verify the correctness of the closed-form expression. We observe that the system becomes more robust against the phase noise as the number of reflective elements increases. Moreover, the influence of the additive channel noise in uplink vanishes as the number of reflecting elements grows asymptotically large.
The recent trend towards the high-speed transportation system has spurred the development of high-speed trains (HSTs). However, enabling HST users with seamless wireless connectivity using the roadside units (RSUs) is extremely challenging, mostly du
e to the lack of line of sight link. To address this issue, we propose a novel framework that uses intelligent reflecting surfaces (IRS)-enabled unmanned aerial vehicles (UAVs) to provide line of sight communication to HST users. First, we formulate the optimization problem where the objective is to maximize the minimum achievable rate of HSTs by jointly optimizing the trajectory of UAV and the phase-shift of IRS. Due to the non-convex nature of the formulated problem, it is decomposed into two subproblems: IRS phase-shift problem and UAV trajectory optimization problem. Next, a Binary Integer Linear Programming (BILP) and a Soft Actor-Critic (SAC) are constructed in order to solve our decomposed problems. Finally, comprehensive numerical results are provided in order to show the effectiveness of our proposed framework.
Providing coverage for flash crowds is an important application for drone base stations (DBSs). However, any arbitrary crowd is likely to be distributed at a high density. Under the condition for each DBS to serve the same number of ground users, mul
tiple DBSs may be placed at the same horizontal location but different altitudes and will cause severe co-channel interference, to which we refer as the coverage overlapping problem. To solve this problem, we then proposed the data-driven 3D placement (DDP) and the enhanced DDP (eDDP) algorithms. The proposed DDP and eDDP can effectively find the appropriate number, altitude, location, and coverage of DBSs in the serving area in polynomial time to maximize the system sum rate and guarantee the minimum data rate requirement of the user equipment. The simulation results show that, compared with the balanced k-means approach, the proposed eDDP can increase the system sum rate by 200% and reduce the computation time by 50%. In particular, eDDP can effectively reduce the occurrence of the coverage overlapping problem and then outperform DDP by about 100% in terms of system sum rate.
This paper is devoted to the lifespan of solutions to a damped fourth-order wave equation with logarithmic nonlinearity $$u_{tt}+Delta^2u-Delta u-omegaDelta u_t+alpha(t)u_t=|u|^{p-2}uln|u|.$$ Finite time blow-up criteria for solutions at both lower a
nd high initial energy levels are established, and an upper bound for the blow-up time is given for each case. Moreover, by constructing a new auxiliary functional and making full use of the strong damping term, a lower bound for the blow-up time is also derived.
Narrowband internet-of-things (NB-IoT) is a competitive 5G technology for massive machine-type communication scenarios, but meanwhile introduces narrowband interference (NBI) to existing broadband transmission such as the long term evolution (LTE) sy
stems in enhanced mobile broadband (eMBB) scenarios. In order to facilitate the harmonic and fair coexistence in wireless heterogeneous networks, it is important to eliminate NB-IoT interference to LTE systems. In this paper, a novel sparse machine learning based framework and a sparse combinatorial optimization problem is formulated for accurate NBI recovery, which can be efficiently solved using the proposed iterative sparse learning algorithm called sparse cross-entropy minimization (SCEM). To further improve the recovery accuracy and convergence rate, regularization is introduced to the loss function in the enhanced algorithm called regularized SCEM. Moreover, exploiting the spatial correlation of NBI, the framework is extended to multiple-input multiple-output systems. Simulation results demonstrate that the proposed methods are effective in eliminating NB-IoT interference to LTE systems, and significantly outperform the state-of-the-art methods.
In this paper, we consider an Unmanned Aerial Vehicle (UAV)-assisted cellular system which consists of multiple UAV base stations (BSs) cooperating the terrestrial BSs. In such a heterogeneous network, for cellular operators, the problem is how to de
termine the appropriate number, locations, and altitudes of UAV-BSs to improve the system sumrate as well as satisfy the demands of arbitrarily flash crowds on data rates. We propose a data-driven 3D placement of UAV-BSs for providing an effective placement result with a feasible computational cost. The proposed algorithm searches for the appropriate number, location, coverage, and altitude of each UAV-BS in the serving area with the maximized system sumrate in polynomial time so as to guarantee the minimum data rate requirement of UE. The simulation results show that the proposed approach can improve system sumrate in comparison with the case without UAV-BSs.
Measurements on the weak decay asymmetry parameters of charmed baryon, say $Xi_c$, provide more information on the $W$-emission and $W$-exchange mechanisms controlled by the strong and weak interactions. Taking advantage of the spin polarization in t
he charmed baryon decays, we investigate the possibility to measure the weak decay asymmetry parameters in the $e^{+}e^{-}to Xi_c^0barXi_c^0$ process. We analyze the transverse polarization spontaneously produced in this process and spin transfer in the subsequent $Xi_c$ decays. The sensitivity to measure the asymmetry parameters are estimated for the decay $Xi_ctoXi pi$.
In this article, we will prove Riemann Hypothesis by using the mean value theorem of integrals. The function $ xi(s) $ is introduced by Riemann, which zeros are identical equal to non-trivial zeros of zeta function.The function $ xi(s) $ is an entire
function, and its real part and imaginary part can be represented as infinite integral form. In the special condition, the mean value theorem of integrals is established for infinite integral. Using the mean value theorem of integrals and the isolation of zeros of analytic function, we determined that all zeros of the function $ xi(s) $ have real part equal to$frac{1}{2}$, namely, all non-trivial zeros of zeta function lies on the critical line. Riemann Hypothesis is true.
In this paper, the finite time extinction of solutions to the fast diffusion system $u_t=mathrm{div}(| abla u|^{p-2} abla u)+v^m$, $v_t=mathrm{div}(| abla v|^{q-2} abla v)+u^n$ is investigated, where $1<p,q<2$, $m,n>0$ and $Omegasubset mathbb{R}^N (N
geq1)$ is a bounded smooth domain. After establishing the local existence of weak solutions, the authors show that if $mn>(p-1)(q-1)$, then any solution vanishes in finite time provided that the initial data are ``comparable; if $mn=(p-1)(q-1)$ and $Omega$ is suitably small, then the existence of extinction solutions for small initial data is proved by using the De Giorgi iteration process and comparison method. On the other hand, for $1<p=q<2$ and $mn<(p-1)^2$, the existence of at least one non-extinction solution for any positive smooth initial data is proved.
