arXiv subjects

Taejoon Kim

Publications and source records attributed to Taejoon Kim.

At least 55 records · Page 3Linked to original sources

arXiv2022

Deep Reinforcement Learning-Based Adaptive IRS Control with Limited Feedback Codebooks

Intelligent reflecting surfaces (IRS) consist of configurable meta-atoms, which can alter the wireless propagation environment through design of their reflection coefficients. We consider adaptive IRS control in the practical setting where (i) the IRS reflection coefficients are attained by adjusting tunable elements embedded in the meta-atoms, (ii) the IRS reflection coefficients are affected by the incident angles of the incoming signals, (iii) the IRS is deployed in multi-path, time-varying channels, and (iv) the feedback link from the base station (BS) to the IRS has a low data rate. Conventional optimization-based IRS control protocols, which rely on channel estimation and conveying the optimized variables to the IRS, are not practical in this setting due to the difficulty of channel estimation and the low data rate of the feedback channel. To address these challenges, we develop a novel adaptive codebook-based limited feedback protocol to control the IRS. We propose two solutions for adaptive IRS codebook design: (i) random adjacency (RA), which utilizes correlations across the channel realizations, and (ii) deep neural network policy-based IRS control (DPIC), which is based on a deep reinforcement learning. Numerical evaluations show that the data rate and average data rate over one coherence time are improved substantially by the proposed schemes.

eess.SY↗

arXiv2022

Multiple Classifiers Based Maximum Classifier Discrepancy for Unsupervised Domain Adaptation

Adversarial training based on the maximum classifier discrepancy between two classifier structures has achieved great success in unsupervised domain adaptation tasks for image classification. The approach adopts the structure of two classifiers, though simple and intuitive, the learned classification boundary may not well represent the data property in the new domain. In this paper, we propose to extend the structure to multiple classifiers to further boost its performance. To this end, we develop a very straightforward approach to adding more classifiers. We employ the principle that the classifiers are different from each other to construct a discrepancy loss function for multiple classifiers. The proposed construction method of loss function makes it possible to add any number of classifiers to the original framework. The proposed approach is validated through extensive experimental evaluations. We demonstrate that, on average, adopting the structure of three classifiers normally yields the best performance as a trade-off between accuracy and efficiency. With minimum extra computational costs, the proposed approach can significantly improve the performance of the original algorithm. The source code of the proposed approach can be downloaded from \url{https://github.com/rucv/MMCD\_DA}.

cs.CV↗

arXiv2022

MMV-Based Sequential AoA and AoD Estimation for Millimeter Wave MIMO Channels

The fact that the millimeter-wave (mmWave) multiple-input multiple-output (MIMO) channel has sparse support in the spatial domain has motivated recent compressed sensing (CS)-based mmWave channel estimation methods, where the angles of arrivals (AoAs) and angles of departures (AoDs) are quantized using angle dictionary matrices. However, the existing CS-based methods usually obtain the estimation result through one-stage channel sounding that have two limitations: (i) the requirement of large-dimensional dictionary and (ii) unresolvable quantization error. These two drawbacks are irreconcilable; improvement of the one implies deterioration of the other. To address these challenges, we propose, in this paper, a two-stage method to estimate the AoAs and AoDs of mmWave channels. In the proposed method, the channel estimation task is divided into two stages, Stage I and Stage II. Specifically, in Stage I, the AoAs are estimated by solving a multiple measurement vectors (MMV) problem. In Stage II, based on the estimated AoAs, the receive sounders are designed to estimate AoDs. The dimension of the angle dictionary in each stage can be reduced, which in turn reduces the computational complexity substantially. We then analyze the successful recovery probability (SRP) of the proposed method, revealing the superiority of the proposed framework over the existing one-stage CS-based methods. We further enhance the reconstruction performance by performing resource allocation between the two stages. We also overcome the unresolvable quantization error issue present in the prior techniques by applying the atomic norm minimization method to each stage of the proposed two-stage approach. The simulation results illustrate the substantially improved performance with low complexity of the proposed two-stage method.

cs.IT↗

arXiv2021

An Unsupervised Domain Adaptation Model based on Dual-module Adversarial Training

In this paper, we propose a dual-module network architecture that employs a domain discriminative feature module to encourage the domain invariant feature module to learn more domain invariant features. The proposed architecture can be applied to any model that utilizes domain invariant features for unsupervised domain adaptation to improve its ability to extract domain invariant features. We conduct experiments with the Domain-Adversarial Training of Neural Networks (DANN) model as a representative algorithm. In the training process, we supply the same input to the two modules and then extract their feature distribution and prediction results respectively. We propose a discrepancy loss to find the discrepancy of the prediction results and the feature distribution between the two modules. Through the adversarial training by maximizing the loss of their feature distribution and minimizing the discrepancy of their prediction results, the two modules are encouraged to learn more domain discriminative and domain invariant features respectively. Extensive comparative evaluations are conducted and the proposed approach outperforms the state-of-the-art in most unsupervised domain adaptation tasks.

cs.LG↗

arXiv2021

Learning-Based Adaptive IRS Control with Limited Feedback Codebooks

Intelligent reflecting surfaces (IRS) consist of configurable meta-atoms, which can change the wireless propagation environment through design of their reflection coefficients. We consider a practical setting where (i) the IRS reflection coefficients are achieved by adjusting tunable elements embedded in the meta-atoms, (ii) the IRS reflection coefficients are affected by the incident angles of the incoming signals, (iii) the IRS is deployed in multi-path, time-varying channels, and (iv) the feedback link from the base station to the IRS has a low data rate. Conventional optimization-based IRS control protocols, which rely on channel estimation and conveying the optimized variables to the IRS, are not applicable in this setting due to the difficulty of channel estimation and the low feedback rate. Therefore, we develop a novel adaptive codebook-based limited feedback protocol where only a codeword index is transferred to the IRS. We propose two solutions for adaptive codebook design: random adjacency (RA) and deep neural network policy-based IRS control (DPIC), both of which only require the end-to-end compound channels. We further develop several augmented schemes based on the RA and DPIC. Numerical evaluations show that the data rate and average data rate over one coherence time are improved substantially by our schemes.

eess.SP↗

arXiv2021

Audio-Visual Transformer Based Crowd Counting

Crowd estimation is a very challenging problem. The most recent study tries to exploit auditory information to aid the visual models, however, the performance is limited due to the lack of an effective approach for feature extraction and integration. The paper proposes a new audiovisual multi-task network to address the critical challenges in crowd counting by effectively utilizing both visual and audio inputs for better modalities association and productive feature extraction. The proposed network introduces the notion of auxiliary and explicit image patch-importance ranking (PIR) and patch-wise crowd estimate (PCE) information to produce a third (run-time) modality. These modalities (audio, visual, run-time) undergo a transformer-inspired cross-modality co-attention mechanism to finally output the crowd estimate. To acquire rich visual features, we propose a multi-branch structure with transformer-style fusion in-between. Extensive experimental evaluations show that the proposed scheme outperforms the state-of-the-art networks under all evaluation settings with up to 33.8% improvement. We also analyze and compare the vision-only variant of our network and empirically demonstrate its superiority over previous approaches.

cs.CV↗

arXiv2021

Link Quality-Guaranteed Minimum-Cost Millimeter-Wave Base Station Deployment

Today's growth in the volume of wireless devices coupled with the promise of supporting data-intensive 5G-&-beyond use cases is driving the industry to deploy more millimeter-wave (mmWave) base stations (BSs). Although mmWave cellular systems can carry a larger volume of traffic, dense deployment, in turn, increases the BS installation and maintenance cost, which has been largely ignored in their utilization. In this paper, we present an approach to the problem of mmWave BS deployment in urban environments by minimizing BS deployment cost subject to BS association and user equipment (UE) outage constraints. By exploiting the macro diversity, which enables each UE to be associated with multiple BSs, we derive an expression for UE outage that integrates physical blockage, UE access-limited blockage, and signal-to-interference-plus-noise-ratio (SINR) outage into its expression. The minimum-cost BS deployment problem is then formulated as integer non-linear programming (INP). The combinatorial nature of the problem motivates the pursuit of the optimal solution by decomposing the original problem into the two separable subproblems, i.e., cell coverage optimization and minimum subset selection subproblems. We provide the optimal solution and theoretical justifications for each subproblem. The simulation results demonstrating UE outage guarantees of the proposed method are presented. Interestingly, the proposed method produces a unique distribution of the macro-diversity orders over the network that is distinct from other benchmarks.

cs.NI↗

arXiv2021

Parallel Scale-wise Attention Network for Effective Scene Text Recognition

The paper proposes a new text recognition network for scene-text images. Many state-of-the-art methods employ the attention mechanism either in the text encoder or decoder for the text alignment. Although the encoder-based attention yields promising results, these schemes inherit noticeable limitations. They perform the feature extraction (FE) and visual attention (VA) sequentially, which bounds the attention mechanism to rely only on the FE final single-scale output. Moreover, the utilization of the attention process is limited by only applying it directly to the single scale feature-maps. To address these issues, we propose a new multi-scale and encoder-based attention network for text recognition that performs the multi-scale FE and VA in parallel. The multi-scale channels also undergo regular fusion with each other to develop the coordinated knowledge together. Quantitative evaluation and robustness analysis on the standard benchmarks demonstrate that the proposed network outperforms the state-of-the-art in most cases.

cs.CV↗

arXiv2021

Multi-IRS-assisted Multi-Cell Uplink MIMO Communications under Imperfect CSI: A Deep Reinforcement Learning Approach

Applications of intelligent reflecting surfaces (IRSs) in wireless networks have attracted significant attention recently. Most of the relevant literature is focused on the single cell setting where a single IRS is deployed and perfect channel state information (CSI) is assumed. In this work, we develop a novel methodology for multi-IRS-assisted multi-cell networks in the uplink. We consider the scenario in which (i) channels are dynamic and (ii) only partial CSI is available at each base station (BS); specifically, scalar effective channel powers from only a subset of user equipments (UE). We formulate the sum-rate maximization problem aiming to jointly optimize the IRS reflect beamformers, BS combiners, and UE transmit powers. In casting this as a sequential decision making problem, we propose a multi-agent deep reinforcement learning algorithm to solve it, where each BS acts as an independent agent in charge of tuning the local UE transmit powers, the local IRS reflect beamformer, and its combiners. We introduce an efficient information-sharing scheme that requires limited information exchange among neighboring BSs to cope with the non-stationarity caused by the coupling of actions taken by multiple BSs. Our numerical results show that our method obtains substantial improvement in average data rate compared to baseline approaches, e.g., fixed UE transmit power and maximum ratio combining.

eess.SP↗

arXiv2020

Delay-Efficient and Reliable Data Relaying in Ultra Dense Networks using Rateless Codes

We investigate the problem of delay-efficient and reliable data delivery in ultra-dense networks (UDNs) that constitute macro base stations (MBSs), small base stations (SBSs), and mobile users. Considering a two-hop data delivery system, we propose a partial decode-and-forward (PDF) relaying strategy together with a simple and intuitive amicable encoding scheme for rateless codes to significantly improve user experience in terms of end-to-end delay. Simulation results verify that our amicable encoding scheme is efficient in improving the intermediate performance of rateless codes. It also verifies that our proposed PDF significantly improves the performance of the decode-and-forward (DF) strategy, and that PDF is much more robust against channel degradation. Overall, the proposed strategy and encoding scheme are efficient towards delay-sensitive data delivery in the UDN scenarios.

cs.IT↗

arXiv2020

Enhanced Beam Alignment for Millimeter Wave MIMO Systems: A Kolmogorov Model

We present an enhancement to the problem of beam alignment in millimeter wave (mmWave) multiple-input multiple-output (MIMO) systems, based on a modification of the machine learning-based criterion, called Kolmogorov model (KM), previously applied to the beam alignment problem. Unlike the previous KM, whose computational complexity is not scalable with the size of the problem, a new approach, centered on discrete monotonic optimization (DMO), is proposed, leading to significantly reduced complexity. We also present a Kolmogorov-Smirnov (KS) criterion for the advanced hypothesis testing, which does not require any subjective threshold setting compared to the frequency estimation (FE) method developed for the conventional KM. Simulation results that demonstrate the efficacy of the proposed KM learning for mmWave beam alignment are presented.

eess.SP↗

arXiv2020

Leveraging Subspace Information for Low-Rank Matrix Reconstruction

The problem of low-rank matrix reconstruction arises in various applications in communications and signal processing. The state of the art research largely focuses on the recovery techniques that utilize affine maps satisfying the restricted isometry property (RIP). However, the affine map design and reconstruction under a priori information, i.e., column or row subspace information, has not been thoroughly investigated. To this end, we present designs of affine maps and reconstruction algorithms that fully exploit the low-rank matrix subspace information. Compared to the randomly generated affine map, the proposed affine map design permits an enhanced reconstruction. In addition, we derive an optimal representation of low-rank matrices, which is exploited to optimize the rank and subspace of the estimate by adapting them to the noise level in order to achieve the minimum mean square error (MSE). Moreover, in the case when the subspace information is not a priori available, we propose a two-step algorithm, where, in the first step, it estimates the column subspace of a low-rank matrix, and in the second step, it exploits the estimated information to complete the reconstruction. The simulation results show that the proposed algorithm achieves robust performance with much lower complexity than existing reconstruction algorithms.

eess.SP↗

arXiv2020

Leveraging the Restricted Isometry Property: Improved Low-Rank Subspace Decomposition for Hybrid Millimeter-Wave Systems

Communication at millimeter wave frequencies will be one of the essential new technologies in 5G. Acquiring an accurate channel estimate is the key to facilitate advanced millimeter wave hybrid multiple-input multiple-output (MIMO) precoding techniques. Millimeter wave MIMO channel estimation, however, suffers from a considerably increased channel use overhead. This happens due to the limited number of radio frequency (RF) chains that prevent the digital baseband from directly accessing the signal at each antenna. To address this issue, recent research has focused on adaptive closed-loop and two-way channel estimation techniques. In this paper, unlike the prior approaches, we study a non-adaptive, hence rather simple, open-loop millimeter wave MIMO channel estimation technique. We present a simple random design of channel subspace sampling signals and show that they obey the restricted isometry property (RIP) with high probability. We then formulate the channel estimation as a low-rank subspace decomposition problem and, based on the RIP, show that the proposed framework reveals resilience to a low signal-to-noise ratio. It is revealed that the required number of channel uses ensuring a bounded estimation error is linearly proportional to the degrees of freedom of the channel, whereas it converges to a constant value if the number of RF chains can grow proportionally to the channel dimension while keeping the channel rank fixed. In particular, we show that the tighter the RIP characterization the lower the channel estimation error is. We also devise an iterative technique that effectively finds a suboptimal but stationary solution to the formulated problem. The proposed technique is shown to have improved channel estimation accuracy with a low channel use overhead as compared to that of previous closed-loop and two-way adaptation techniques.

eess.SP↗

arXiv2020

A Sequential Subspace Method for Millimeter Wave MIMO Channel Estimation

Data transmission over the mmWave in fifth-generation wireless networks aims to support very high speed wireless communications. A substantial increase in spectrum efficiency for mmWave transmission can be achieved by using advanced hybrid precoding, for which accurate channel state information is the key. Rather than estimating the entire channel matrix, directly estimating subspace information, which contains fewer parameters, does have enough information to design transceivers. However, the large channel use overhead and associated computational complexity in the existing channel subspace estimation techniques are major obstacles to deploy the subspace approach for channel estimation. In this paper, we propose a sequential two-stage subspace estimation method that can resolve the overhead issues and provide accurate subspace information. Utilizing a sequential method enables us to avoid manipulating the entire high-dimensional training signal, which greatly reduces the complexity. Specifically, in the first stage, the proposed method samples the columns of channel matrix to estimate its column subspace. Then, based on the obtained column subspace, it optimizes the training signals to estimate the row subspace. For a channel with $N_r$ receive antennas and $N_t$ transmit antennas, our analysis shows that the proposed technique only requires $O(N_t)$ channel uses, while providing a guarantee of subspace estimation accuracy. By theoretical analysis, it is shown that the similarity between the estimated subspace and the true subspace is linearly related to the signal-to-noise ratio (SNR), i.e., $O(\text{SNR})$, at high SNR, while quadratically related to the SNR, i.e., $O(\text{SNR}^2)$, at low SNR. Simulation results show that the proposed sequential subspace method can provide improved subspace accuracy, normalized mean squared error, and spectrum efficiency over existing methods.

eess.SP↗

arXiv2020

Joint Optimization of Signal Design and Resource Allocation in Wireless D2D Edge Computing

In this paper, we study the distributed computational capabilities of device-to-device (D2D) networks. A key characteristic of D2D networks is that their topologies are reconfigurable to cope with network demands. For distributed computing, resource management is challenging due to limited network and communication resources, leading to inter-channel interference. To overcome this, recent research has addressed the problems of wireless scheduling, subchannel allocation, power allocation, and multiple-input multiple-output (MIMO) signal design, but has not considered them jointly. In this paper, unlike previous mobile edge computing (MEC) approaches, we propose a joint optimization of wireless MIMO signal design and network resource allocation to maximize energy efficiency. Given that the resulting problem is a non-convex mixed integer program (MIP) which is prohibitive to solve at scale, we decompose its solution into two parts: (i) a resource allocation subproblem, which optimizes the link selection and subchannel allocations, and (ii) MIMO signal design subproblem, which optimizes the transmit beamformer, transmit power, and receive combiner. Simulation results using wireless edge topologies show that our method yields substantial improvements in energy efficiency compared with cases of no offloading and partially optimized methods and that the efficiency scales well with the size of the network.

cs.NI↗

arXiv2019

Coherence Statistics of Structured Random Ensembles and Support Detection Bounds for OMP

A structured random matrix ensemble that maintains constant modulus entries and unit-norm columns, often called a random phase-rotated (RPR) matrix, is considered in this paper. We analyze the coherence statistics of RPR measurement matrices and apply them to acquire probabilistic performance guarantees of orthogonal matching pursuit (OMP) for support detection (SD). It is revealed via numerical simulations that the SD performance guarantee provides a tight characterization, especially when the signal is sparse.

eess.SP↗

arXiv2019

Decorrelation Deep Learning for Fingerprint-based Indoor Localization

Indoor localization is of particular interest due to its immense practical applications. However, the rich multipath and high penetration loss of indoor wireless signal propagation make this task arduous. Though recently studied fingerprint-based techniques can handle the multipath effects, the sensitivity of the localization performance to channel fluctuation is a drawback. To address the latter challenge, we adopt an artificial multi-layer neural network (MNN) to learn the complex channel impulse responses (CIRs) as fingerprint measurements. However, the performance of the location classification using MNN critically depends on the correlation among the training data. Therefore, we design two different decorrelation filters that preprocess the training data for discriminative learning. The first one is a linear whitening filter combined with the principal component analysis (PCA), which forces the covariance matrix of different feature dimensions to be identity. The other filter is a nonlinear quantizer that is optimized to minimize the distortion incurred by the quantization. Numerical results using indoor channel models illustrate the significant improvement of the proposed decorrelation MNN (DMNN) compared to other benchmarks.

eess.SP↗

arXiv2019

A Hybrid Cache Architecture for Meeting Per-Tenant Performance Goals in a Private Cloud

The in-memory cache system is an important component in a cloud for the data access performance. As the tenants may have different performance goals for data access depending on the nature of their tasks, effectively managing the memory cache is a crucial concern in such a shared computing environment. Two extreme methods for managing the memory cache are unlimited sharing and complete isolation, both of which would be inefficient with the expensive storage complexity to meet the per-tenant performance requirement. In this paper, we present a new cache model that incorporates global caching (based on unlimited sharing) and static caching (offering complete isolation) for a private cloud, in which it is critical to offer the guaranteed performance while minimizing the operating cost. This paper also presents a cache insertion algorithm tailored to the proposed cache model. From an extensive set of experiments conducted on the simulation and emulation settings, the results confirm the validity of the presented cache architecture and insertion algorithm showing the optimized use of the cache space for meeting the per-tenant performance requirement.

cs.DC↗