arXiv subjects

Ahmet M. Elbir

Publications and source records attributed to Ahmet M. Elbir.

48 records · Page 3Linked to original sources

arXiv2021

Hybrid Federated and Centralized Learning

Many of the machine learning (ML) tasks are focused on centralized learning (CL), which requires the transmission of local datasets from the clients to a parameter server (PS) leading to a huge communication overhead. Federated learning (FL) overcomes this issue by allowing the clients to send only the model updates to the PS instead of the whole dataset. In this way, FL brings the learning to edge level, wherein powerful computational resources are required on the client side. This requirement may not always be satisfied because of diverse computational capabilities of edge devices. We address this through a novel hybrid federated and centralized learning (HFCL) framework to effectively train a learning model by exploiting the computational capability of the clients. In HFCL, only the clients who have sufficient resources employ FL; the remaining clients resort to CL by transmitting their local dataset to PS. This allows all the clients to collaborate on the learning process regardless of their computational resources. We also propose a sequential data transmission approach with HFCL (HFCL-SDT) to reduce the training duration. The proposed HFCL frameworks outperform previously proposed non-hybrid FL (CL) based schemes in terms of learning accuracy (communication overhead) since all the clients collaborate on the learning process with their datasets regardless of their computational resources.

cs.LG↗

arXiv2021

Federated Dropout Learning for Hybrid Beamforming With Spatial Path Index Modulation In Multi-User mmWave-MIMO Systems

Millimeter wave multiple-input multiple-output (mmWave-MIMO) systems with small number of radio-frequency (RF) chains have limited multiplexing gain. Spatial path index modulation (SPIM) is helpful in improving this gain by utilizing additional signal bits modulated by the indices of spatial paths. In this paper, we introduce model-based and model-free frameworks for beamformer design in multi-user SPIM-MIMO systems. We first design the beamformers via model-based manifold optimization algorithm. Then, we leverage federated learning (FL) with dropout learning (DL) to train a learning model on the local dataset of users, who estimate the beamformers by feeding the model with their channel data. The DL randomly selects different set of model parameters during training, thereby further reducing the transmission overhead compared to conventional FL. Numerical experiments show that the proposed framework exhibits higher spectral efficiency than the state-of-the-art SPIM-MIMO methods and mmWave-MIMO, which relies on the strongest propagation path. Furthermore, the proposed FL approach provides at least 10 times lower transmission overhead than the centralized learning techniques.

eess.SP↗

arXiv2020

Federated Learning for Hybrid Beamforming in mm-Wave Massive MIMO

Machine learning for hybrid beamforming has been extensively studied by using centralized machine learning (CML) techniques, which require the training of a global model with a large dataset collected from the users. However, the transmission of the whole dataset between the users and the base station (BS) is computationally prohibitive due to limited communication bandwidth and privacy concerns. In this work, we introduce a federated learning (FL) based framework for hybrid beamforming, where the model training is performed at the BS by collecting only the gradients from the users. We design a convolutional neural network, in which the input is the channel data, yielding the analog beamformers at the output. Via numerical simulations, FL is demonstrated to be more tolerant to the imperfections and corruptions in the channel data as well as having less transmission overhead than CML.

eess.SP↗

arXiv2020

A Deep Learning Framework for Hybrid Beamforming Without Instantaneous CSI Feedback

Hybrid beamformer design plays very crucial role in the next generation millimeter-wave (mm-Wave) massive MIMO (multiple-input multiple-output) systems. Previous works assume the perfect channel state information (CSI) which results heavy feedback overhead. To lower complexity, channel statistics can be utilized such that only infrequent update of the channel information is needed. To reduce the complexity and provide robustness, in this work, we propose a deep learning (DL) framework to deal with both hybrid beamforming and channel estimation. For this purpose, we introduce three deep convolutional neural network (CNN) architectures. We assume that the base station (BS) has the channel statistics only and feeds the channel covariance matrix into a CNN to obtain the hybrid precoders. At the receiver, two CNNs are employed. The first one is used for channel estimation purposes and the another is employed to design the hybrid combiners. The proposed DL framework does not require the instantaneous feedback of the CSI at the BS. We have shown that the proposed approach has higher spectral efficiency with comparison to the conventional techniques. The trained CNN structures do not need to be re-trained due to the changes in the propagation environment such as the deviations in the number of received paths and the fluctuations in the received path angles up to 4 degrees. Also, the proposed DL framework exhibits at least 10 times lower computational complexity as compared to the conventional optimization-based approaches.

eess.SP↗

arXiv2020

Sparse Array Selection Across Arbitrary Sensor Geometries with Deep Transfer Learning

Sparse sensor array selection arises in many engineering applications, where it is imperative to obtain maximum spatial resolution from a limited number of array elements. Recent research shows that computational complexity of array selection is reduced by replacing the conventional optimization and greedy search methods with a deep learning network. However, in practice, sufficient and well-calibrated labeled training data are unavailable and, more so, for arbitrary array configurations. To address this, we adopt a deep transfer learning (TL) approach, wherein we train a deep convolutional neural network (CNN) with data of a source sensor array for which calibrated data are readily available and reuse this pre-trained CNN for a different, data-insufficient target array geometry to perform sparse array selection. Numerical experiments with uniform rectangular and circular arrays demonstrate enhanced performance of TL-CNN on the target model than the CNN trained with insufficient data from the same model. In particular, our TL framework provides approximately 20% higher sensor selection accuracy and 10% improvement in the direction-of-arrival estimation error.

eess.SP↗

arXiv2020

Deep Channel Learning For Large Intelligent Surfaces Aided mm-Wave Massive MIMO Systems

This letter presents the first work introducing a deep learning (DL) framework for channel estimation in large intelligent surface (LIS) assisted massive MIMO (multiple-input multiple-output) systems. A twin convolutional neural network (CNN) architecture is designed and it is fed with the received pilot signals to estimate both direct and cascaded channels. In a multi-user scenario, each user has access to the CNN to estimate its own channel. The performance of the proposed DL approach is evaluated and compared with state-of-the-art DL-based techniques and its superior performance is demonstrated.

eess.SP↗

arXiv2020

DeepMUSIC: Multiple Signal Classification via Deep Learning

This letter introduces a deep learning (DL) framework for direction-of-arrival (DOA) estimation. Previous works in DL context mostly consider a single or two target scenario which is a strong limitation in practice. Hence, in this work, we propose a DL framework for multiple signal classification (DeepMUSIC). We design multiple deep convolutional neural networks (CNNs), each of which is dedicated to a subregion of the angular spectrum. In particular, each CNN is fed with the array covariance matrix and it learns the MUSIC spectra of the corresponding angular subregion. We have shown, through simulations, that the proposed DeepMUSIC framework has superior estimation accuracy and exhibits less computational complexity in comparison with both DL and non-DL based techniques.

eess.SP↗

arXiv2019

V-Shaped Sparse Arrays For 2-D DOA Estimation

This paper proposes a new sparse array geometry for 2-D (azimuth and elevation) DOA (direction-of-arrival) estimation. The proposed array geometry is V-shaped sparse array and it is composed of two linear portions which are crossing each other. The degrees of freedom of the sparse array is enhanced by sparse sampling property. In this respect, V-shaped coprime (VCA) and V-shaped nested array (VNA) structures are developed. VCA can resolve both azimuth and elevation angles up to MN sources with 2M + N -1 sensors in each portion and the total number of sensors is 4M+2N-3. VNA can resolve O(N^2) sources with 2N sensors. Instead of 2-D grid search, the proposed method computes 1-D search for azimuth and elevation angle estimation in a computational efficient way. In order to solve the pairing problem in 2-D scenario, the cross-covariance matrix of two portion is utilized and 2-D paired DOA estimation is performed. The performance of the proposed method is evaluated with numerical simulations and it is shown that the proposed array geometries VCA and VNA can provide much less sensors as compared to the conventional coprime planar arrays.

eess.SP↗

arXiv2019

Joint Antenna Selection and Hybrid Beamformer Design using Unquantized and Quantized Deep Learning Networks

In millimeter-wave communications, multiple-input-multiple-output (MIMO) systems use large antenna arrays to achieve high gain and spectral efficiency. These massive MIMO systems employ hybrid beamformers to reduce power consumption associated with fully digital beamforming in large arrays. Further savings in cost and power are possible through the use of subarrays. Unlike prior works that resort to large latency methods such as optimization and greedy search for subarray selection, we propose a deep-learning-based approach in order to overcome the complexity issue without causing significant performance loss. We formulate antenna selection and hybrid beamformer design as a classification/prediction problem for convolutional neural networks (CNNs). For antenna selection, the CNN accepts the channel matrix as input and outputs a subarray with optimal spectral efficiency. The resultant subarray channel matrix is then again fed to a CNN to obtain analog and baseband beamformers. We train the CNNs with several noisy channel matrices that have different channel statistics in order to achieve a robust performance at the network output. Numerical experiments show that our CNN framework provides an order better spectral efficiency and is 10 times faster than the conventional techniques. Further investigations with quantized-CNNs show that the proposed network, saved in no more than 5 bits, is also suited for digital mobile devices.

eess.SP↗

arXiv2019

Hybrid Precoding for Multi-User Millimeter Wave Massive MIMO Systems: A Deep Learning Approach

In multi-user millimeter wave (mmWave) multiple-input-multiple-output (MIMO) systems, hybrid precoding is a crucial task to lower the complexity and cost while achieving a sufficient sum-rate. Previous works on hybrid precoding were usually based on optimization or greedy approaches. These methods either provide higher complexity or have sub-optimum performance. Moreover, the performance of these methods mostly relies on the quality of the channel data. In this work, we propose a deep learning (DL) framework to improve the performance and provide less computation time as compared to conventional techniques. In fact, we design a convolutional neural network for MIMO (CNN-MIMO) that accepts as input an imperfect channel matrix and gives the analog precoder and combiners at the output. The procedure includes two main stages. First, we develop an exhaustive search algorithm to select the analog precoder and combiners from a predefined codebook maximizing the achievable sum-rate. Then, the selected precoder and combiners are used as output labels in the training stage of CNN-MIMO where the input-output pairs are obtained. We evaluate the performance of the proposed method through numerous and extensive simulations and show that the proposed DL framework outperforms conventional techniques. Overall, CNN-MIMO provides a robust hybrid precoding scheme in the presence of imperfections regarding the channel matrix. On top of this, the proposed approach exhibits less computation time with comparison to the optimization and codebook based approaches.

eess.SP↗

arXiv2019

Low-Complexity Limited-Feedback Deep Hybrid Beamforming for Broadband Massive MIMO Communications

In broadband millimeter-wave (mm-Wave) systems, it is desirable to design hybrid beamformers with common analog beamformer for the entire band while employing different baseband beamformers in different frequency sub-bands. Furthermore, the performance mostly relies on the perfectness of the channel information. In this paper, we propose a deep learning (DL) framework for hybrid beamformer design in broadband mm-Wave massive MIMO systems. We design a convolutional neural network (CNN) that accepts the channel matrix of all subcarriers as input and the output of CNN is the hybrid beamformers. The proposed CNN architecture is trained with imperfect channel matrices in order to provide robust performance against the deviations in the channel data. Hence, the proposed precoding scheme can handle the imperfect or limited feedback scenario where the full and exact knowledge of the channel is not available. We show that the proposed DL framework is more robust and computationally less complex than the conventional optimization and phase-extraction-based approaches.

eess.SP↗

arXiv2019

Cognitive Radar Antenna Selection via Deep Learning

Direction of arrival (DoA) estimation of targets improves with the number of elements employed by a phased array radar antenna. Since larger arrays have high associated cost, area and computational load, there is recent interest in thinning the antenna arrays without loss of far-field DoA accuracy. In this context, a cognitive radar may deploy a full array and then select an optimal subarray to transmit and receive the signals in response to changes in the target environment. Prior works have used optimization and greedy search methods to pick the best subarrays cognitively. In this paper, we leverage deep learning to address the antenna selection problem. Specifically, we construct a convolutional neural network (CNN) as a multi-class classification framework where each class designates a different subarray. The proposed network determines a new array every time data is received by the radar, thereby making antenna selection a cognitive operation. Our numerical experiments show that {the proposed CNN structure provides 22% better classification performance than a Support Vector Machine and the resulting subarrays yield 72% more accurate DoA estimates than random array selections.

eess.SP↗