Millimeter Wave Wireless Communication Assisted Three-Dimensional Simultaneous Localization and Mapping

In this paper, we study the three-dimensional (3D) simultaneous localization and mapping (SLAM) problem in complex outdoor and indoor environments based only on millimeter-wave (mmWave) wireless communication signals. Firstly, we propose a deep-learning based mapping (DLM) algorithm that can leverage the reflections point on the first-order none line-of-sight (NLOS) communications links (CLs) to build the 3D point cloud map of the environment. Specifically, we design a classification neural network to identify the first-order NLOS CL and theoretically calculate the geometric coordinates of the reflection points on it. Secondly, we take the advantage of both the inertial measurement unit and the beam-squint assisted localization method to realize real-time and precise localizations. Then, combining the DLM and the adopted localization algorithm, we develop the communication-based SLAM (C-SLAM) framework that can carry out SLAM without any prior knowledge of the environment. Moreover, extensive simulations of both complex outdoor and indoor environments validate the effectiveness of our approach.

Pay Less But Get More: A Dual-Attention-based Channel Estimation Network for Massive MIMO Systems with Low-Density Pilots

To reap the promising benefits of massive multiple-input multiple-output (MIMO) systems, accurate channel state information (CSI) is required through channel estimation. However, due to the complicated wireless propagation environment and large-scale antenna arrays, precise channel estimation for massive MIMO systems is significantly challenging and costs an enormous training overhead. Considerable time-frequency resources are consumed to acquire sufficient accuracy of CSI, which thus severely degrades systems' spectral and energy efficiencies. In this paper, we propose a dual-attention-based channel estimation network (DACEN) to realize accurate channel estimation via low-density pilots, by decoupling the spatial-temporal domain features of massive MIMO channels with the temporal attention module and the spatial attention module. To further improve the estimation accuracy, we propose a parameter-instance transfer learning approach based on the DACEN to transfer the channel knowledge learned from the high-density pilots pre-acquired during the training dataset collection period. Experimental results on a publicly available dataset reveal that the proposed DACEN-based method with low-density pilots ($\rho_L=6/52$) achieves better channel estimation performance than the existing methods even with higher-density pilots ($\rho_H=26/52$). Additionally, with the proposed transfer learning approach, the DACEN-based method with ultra-low-density pilots ($\rho_L^\prime=2/52$) achieves higher estimation accuracy than the existing methods with low-density pilots, thereby demonstrating the effectiveness and the superiority of the proposed method.

Deep learning numerical methods for high-dimensional fully nonlinear PIDEs and coupled FBSDEs with jumps

We propose a deep learning algorithm for solving high-dimensional parabolic integro-differential equations (PIDEs) and high-dimensional forward-backward stochastic differential equations with jumps (FBSDEJs), where the jump-diffusion process are derived by a Brownian motion and an independent compensated Poisson random measure. In this novel algorithm, a pair of deep neural networks for the approximations of the gradient and the integral kernel is introduced in a crucial way based on deep FBSDE method. To derive the error estimates for this deep learning algorithm, the convergence of Markovian iteration, the error bound of Euler time discretization, and the simulation error of deep learning algorithm are investigated. Two numerical examples are provided to show the efficiency of this proposed algorithm.

Radar Sensing via OTFS Signaling: A Delay Doppler Signal Processing Perspective

The recently proposed orthogonal time frequency space (OTFS) modulation multiplexes data symbols in the delay-Doppler (DD) domain. Since the range and velocity, which can be derived from the delay and Doppler shifts, are the parameters of interest for radar sensing, it is natural to consider implementing DD signal processing for radar sensing. In this paper, we investigate the potential connections between the OTFS and DD domain radar signal processing. Our analysis shows that the range-Doppler matrix computing process in radar sensing is exactly the demodulation of OTFS with a rectangular pulse shaping filter. Furthermore, we propose a two-dimensional (2D) correlation-based algorithm to estimate the fractional delay and Doppler parameters for radar sensing. Simulation results show that the proposed algorithm can efficiently obtain the delay and Doppler shifts associated with multiple targets.

Vision Aided Environment Semantics Extraction and Its Application in mmWave Beam Selection

In this letter, we propose a novel mmWave beam selection method based on the environment semantics that are extracted from camera images taken at the user side. Specifically, we first define the environment semantics as the spatial distribution of the scatterers that affect the wireless propagation channels and utilize the keypoint detection technique to extract them from the input images. Then, we design a deep neural network with environment semantics as the input that can output the optimal beam pairs at UE and BS. Compared with the existing beam selection approaches that directly use images as the input, the proposed semantic-based method can explicitly obtain the environmental features that account for the propagation of wireless signals, and thus reduce the burden of storage and computation. Simulation results show that the proposed method can precisely estimate the location of the scatterers and outperform the existing image or LIDAR based works.

Environment Semantics Aided Wireless Communications: A Case Study of mmWave Beam Prediction and Blockage Prediction

In this paper, we propose an environment semantics aided wireless communication framework to reduce the transmission latency and improve the transmission reliability, where semantic information is extracted from environment image data, selectively encoded based on its task-relevance, and then fused to make decisions for channel related tasks. As a case study, we develop an environment semantics aided network architecture for mmWave communication systems, which is composed of a semantic feature extraction network, a feature selection algorithm, a task-oriented encoder, and a decision network. With images taken from street cameras and user's identification information as the inputs, the environment semantics aided network architecture is trained to predict the optimal beam index and the blockage state for the base station. It is seen that without pilot training or the costly beam scans, the environment semantics aided network architecture can realize extremely efficient beam prediction and timely blockage prediction, thus meeting requirements for ultra-reliable and low-latency communications (URLLCs). Simulation results demonstrate that compared with existing works, the proposed environment semantics aided network architecture can reduce system overheads such as storage space and computational cost while achieving satisfactory prediction accuracy and protecting user privacy.

A Generalized Semantic Communication System: from Sources to Channels

Semantic communication is regarded as the breakthrough beyond the Shannon paradigm, which transmits only semantic information to significantly improve communication efficiency. This article introduces a framework for generalized semantic communication system, which exploits the semantic information in both the multimodal source and the wireless channel environment. Subsequently, the developed deep learning enabled end-to-end semantic communication and environment semantics aided wireless communication techniques are demonstrated through two examples. The article concludes with several research challenges to boost the development of such a generalized semantic communication system.

Model-Driven Deep Learning for Non-Coherent Massive Machine-Type Communications

In this paper, we investigate the joint device activity and data detection in massive machine-type communications (mMTC) with a one-phase non-coherent scheme, where data bits are embedded in the pilot sequences and the base station simultaneously detects active devices and their embedded data bits without explicit channel estimation. Due to the correlated sparsity pattern introduced by the non-coherent transmission scheme, the traditional approximate message passing (AMP) algorithm cannot achieve satisfactory performance. Therefore, we propose a deep learning (DL) modified AMP network (DL-mAMPnet) that enhances the detection performance by effectively exploiting the pilot activity correlation. The DL-mAMPnet is constructed by unfolding the AMP algorithm into a feedforward neural network, which combines the principled mathematical model of the AMP algorithm with the powerful learning capability, thereby benefiting from the advantages of both techniques. Trainable parameters are introduced in the DL-mAMPnet to approximate the correlated sparsity pattern and the large-scale fading coefficient. Moreover, a refinement module is designed to further advance the performance by utilizing the spatial feature caused by the correlated sparsity pattern. Simulation results demonstrate that the proposed DL-mAMPnet can significantly outperform traditional algorithms in terms of the symbol error rate performance.

Hierarchical Codebook Design for Near-Field MmWave MIMO Communications Systems

Communications system with analog or hybrid analog/digital architectures usually relies on a pre-defined codebook to perform beamforming. With the increase in the size of the antenna array, the characteristics of the spherical wavefront in the near-field situation are not negligible. Therefore, it is necessary to design a codebook that is adaptive to near-field scenarios. In this letter, we investigate the hierarchical codebook design method in the near-field situation. We develop a steering beam gain calculation method and design the lower-layer codebook to satisfy the coverage of the Fresnel region. For the upper-layer codebook, we propose beam rotation and beam relocation methods to place an arbitrary beam pattern at target locations. The simulation results show the superiority of the proposed near-field hierarchical codebook design.

RIS Design to Optimize the CRB for Source Localization

Reconfigurable Intelligent Surface (RIS) plays a pivotal role in enhancing source localization accuracy. Based on the information inequality of Fisher information analyses, the Cram\'{e}r-Rao Bound (CRB) of the localization error can be used to evaluate the localization accuracy for a given set of RIS coefficients. However, there is a lack of research in optimizing these RIS coefficients to decrease the CRB under the constraint imposed by the RIS hardware. In this paper, we adopt the manifold optimization method to derive the locally optimal CRB of the localization error, where the RIS coefficients are restricted to lie on the complex circle manifold. Specifically, the Wirtinger derivatives are calculated in the gradient descent part, and the Riemannian nonlinear acceleration technique is employed to speed up the convergence rate. Simulation results show that the proposed method can yield the locally optimal RIS coefficients and can significantly decrease the CRB of localization error. Moreover, the iteration number can be reduced by the acceleration technique.