1,721,019 research outputs found
Optimal Restricted Isometry Condition for Exact Sparse Recovery with Orthogonal Least Squares
No full textOrthogonal least squares (OLS) is a classic algorithm for sparse recovery, function approximation, and subset selection. In this paper, we analyze the performance guarantee of the OLS algorithm. Specifically, we show that OLS guarantees the exact reconstruction of any K-sparse vector in K iterations, provided that a sensing matrix has unit l(2)-norm columns and satisfies the restricted isometry property (RIP) of order K + 1 with delta(K+1) < C-K = {1/root K, K = 1, 1/root K+1/4, K = 2, 1/root K+1/16, K = 3, 1/root K, K >= 4, Furthermore, we show that the proposed guarantee is optimal in the sense that if delta(K+1) >= C-K, then there exists a counterexample for which OLS fails the recovery.N
Low-Rank Matrix Completion Using Graph Neural Network
No full textIn this paper, we propose the graph neural network (GNN)-based matrix completion technique to reconstruct the desired low-rank matrix by exploiting the underlying graph structure of the matrix. The proposed approach, referred to as GNN-based low-rank matrix completion (GNN-LRMC), combines the GNN and the neural-network weight update mechanism. The GNN is used to extract the node vectors of the graph using a modified convolution operation. Empirical simulations validate the reconstruction performance of GNN-LRMC in synthetic and Netflix datasets.N
Exact recovery of sparse signals using orthogonal matching pursuit: how many iterations do we need?
No full textOrthogonal matching pursuit (OMP) is a greedy algorithm widely used for the recovery of sparse signals from compressed measurements. In this paper, we analyze the number of iterations required for the OMP algorithm to perform exact recovery of sparse signals. Our analysis shows that OMP can accurately recover all K-sparse signals within [2.8 K] iterations when the measurement matrix satisfies a restricted isometry property (RIP). Our result improves upon the recent result of Zhang and also bridges the gap between Zhang's result and the fundamental limit of OMP at which exact recovery of K-sparse signals cannot be uniformly guaranteed.OAIID:RECH_ACHV_DSTSH_NO:T201633760RECH_ACHV_FG:RR00200001ADJUST_YN:EMP_ID:A080348CITE_RATE:4.3DEPT_NM:전기·정보공학부EMAIL:[email protected]_YN:YN
Deep Neural Network-based Joint Active User Detection and Channel Estimation for mMTC
No full textAs a means to support the access of massive machine-type communication devices, grant-free access and non-orthogonal multiple access (NOMA) have received a lot of attention recently. In the grant-free environment, each device transmits information without scheduling. Hence, the device identification process called active user detection (AUD) is indispensable at the base station. After the AUD process, the channel estimation for active devices is performed in the base station before detecting the data. These processes are challenging problems in the NOMA-based systems since it is difficult to detect the active devices and estimate the channel of those devices from the superimposed received signal. In this paper, we propose a deep neural network (DNN)-based joint AUD and CE scheme for the practical mMTC systems. Specifically, the proposed scheme consists of long short term memory (LSTM)-based AUD (L-AUD) and DNN-based CE (D-CE). In L-AUD, by feeding the training data in the designed network, the proposed LSTM network is trained to exploit the extracted features when mapping the received NOMA signal to the indices of active devices. After the AUD process, by using the deeply stacked hidden layers, D-CE extracts the channel features and the codebook features of the active devices to map the received NOMA signal to the corresponding channel. As a result, the trained DNN can jointly handle the whole AUD and CE processes, achieving an accurate detection of the active devices and the small channel estimation error.N
Active User Detection and Channel Estimation for Massive Machine-Type Communication: Deep Learning Approach
No full textRecently, massive machine-type communications (mMTCs) have become one of key use cases for 5G. In order to support massive users transmitting small data packets at low rates, grant-free (GF) access and nonorthogonal multiple access (NOMA) have been suggested. Since each device transmits information without scheduling in the GF-NOMA systems, the device identification process, called active user detection (AUD), is required at the base station (BS). For the NOMA-based systems, the channel estimation (CE), an operation after the AUD, is a challenging task since multiple devices' transmit signals and channels are superimposed in the same wireless resources. In this article, we propose a deep learning (DL)-based AUD and CE in the GF-NOMA systems. In our work, DL figures out the direct mapping between the received NOMA signal and the indices of active devices and associated channels using the long short-term memory (LSTM). From numerical experiments, we show that the proposed scheme is effective in handling the AUD and CE in the mMTC environments.N
Channel Aware Sparse Signaling for Ultra Low-latency Communication in TDD systems
No full textFifth generation (5G) wireless networks are currently being developed to handle wide variety of use cases. In order to support these cases, new types of requirements other than throughput enhancement have been introduced. One such requirement is to reduce the latency down to a millisecond (ms) level in ultra reliable and low latency communications (URLL-C). In case of uplink transmission, supporting this stringent latency requirement is quite challenging and problematic since the scheduling procedure is a time-consuming and complicated handshaking process. In time division duplexing (TDD) systems, satisfying the latency requirement is far more difficult since the mobile device cannot transmit the data when the subframe is assigned for the downlink. In this paper, we propose a new type of uplink transmission scheme for TDD-based URLLC. Key idea of the proposed scheme is to transmit the latency sensitive data immediately after performing the ultra-short one-way signaling from the basestation to the mobile device. To reduce the processing time of grant signal, we present a fast signaling mechanism, referred to as channel-aware sparse signaling (CASS). Numerical results confirm that the proposed uplink transmission scheme is very effective in TDD-based URLLC systems.N
Hybrid Active User Detection for Massive Machine-type Communications in IoT
No full textMassive machine-type communication (mMTC) concerns the access of massive machine-type communication (MTC) devices to the basestation. To support the massive connectivity, non-orthogonal multiple access (NOMA) and grant-free access have been recently introduced. In grant-free access, each device transmits information without scheduling so that the basestation needs to identify the active devices among all potential devices in a cell. This process, called an active user detection (AUD), is a challenging problem for the NOMA-based systems since it is difficult to find out the active devices from the superimposed received signal. To address this problem, compressed sensing (CS) based active user detection (All)) technique exploiting the low activity of devices in mMTC has been introduced. In this paper, we propose an All) scheme that exploits both pilot and data measurements to improve the All) performance in grant-free NOMA systems. The key idea is to use the common support information of pilot and data signals in a packet. Numerical results demonstrate that the proposed ALT) scheme outperforms the conventional approaches in both AUD and throughput performance.N
AoD-Based Statistical Beamforming for Cell-Free Massive MIMO Systems
No full textCell-free massive MIMO system is one of a promising technology of 5G wireless communications that can provide high throughput from the basestation cooperation. To capitalize on the gain obtained by the basestation cooperation, the downlink channel state information (CSI) should be available at the basestations. In the popularly used frequency division duplexing (FDD) system, the downlink CSI must be fed back from the users. However, due to a large number of antennas and basestations, the feedback overhead is a serious concern in the cell-free systems. Recent studies have shown that the uplink and downlink channels have similar angle-of-departures (AoDs), so-called angle reciprocity. In this paper, we present an AoD-based statistical beamforming scheme for the cell-free massive MIMO systems that does not rely on the CSI feedback. Also, we provide an efficient solution for the power allocation problem that minimizes the total power consumption of the basestations. Simulation results demonstrate that the proposed scheme saves approximately 12% transmit power and has a 22% higher coverage probability compare to the conventional cellular systems.N
Power Minimization of Intelligent Reflecting Surface-Aided Uplink IoT Networks
No full textEmploying intelligent reflecting surfaces (IRSs) is emerging as a green alternative to massive antenna systems for improving signal quality and suppressing interference. Specifically, IRS is a planar surface consisting of a large number of low-cost and passive elements each being able to reflect the incident signal independently with an adjustable phase shift, thus the three-dimension (3D) passive beamforming can be collaboratively achieved without the need of any transmit radio-frequency (RF) chains. In this paper, we study the uplink power control of an IRS-aided Internet of Things (IoT) network under the quality of service (QoS) constraints at each user. Our goal is to minimize the total user power by jointly optimizing the phase shifts of IRS reflecting elements and the receiving beamforming at the BS, subject to each user's individual signal-to-interference-plus-noise ratio (SINR) constraint which characterizes its QoS. To solve the formulated non-convex optimization problem, we develop an efficient scheme, called the Riemannian manifold-based alternating optimization (RM-AO). Simulation results demonstrate that the proposed RM-AO algorithm saves the uplink transmit power significantly.N
Localization of Internet of Things Network via Deep Neural Network Based Matrix Completion
No full textIn this paper, we propose a technique to acquire the sensor map of Internet of Things (IoT) network. Our approach consists of two main steps to reconstruct the Euclidean distance matrix. First, we recast Euclidean distance matrix completion problem into the alternating minimization problem. We next employ a cascade of multiple deep neural networks to recover the location map of sensors (and the original distance matrix) from the noisy observed matrix. From the numerical experiments, we demonstrate that the proposed method can achieve an accurate reconstruction performance of the distance matrix with much smaller measurement required by conventional approaches and also outperforms state-of-the-art matrix completion algorithms both in noisy and noiseless scenarios.N
- …
