Deep Learning Based Pilot Allocation Scheme (DL-PAS) for 5G Massive MIMO System

Kim, Kwihoon; Lee, Joohyung; Choi, JunKyun

doi:10.1109/lcomm.2018.2803054

Cited by 99 publications

(82 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…whereT th k T th k − ω k f k andη k η k + K j=1, j =k p j α k,j . 1) Solution for the problem P 4 : P 4 is still non-convex and difficult to solve due to its constraint C 1 , which can be expressed as 4 The conditions in (28) can be enforced by a proper admission control strategy [9], [22], or an appropriate choice of the fronthaul capacity or the BBU computational capability [23]…”

Section: A Hsf Design At the Xl-mimo Rrhmentioning

confidence: 99%

“…Under the assumption that ∃ p such that (28) is satisfied, f k , ∀k given by (50) converges to a KKT point of P 6 .…”

Section: Lemmamentioning

confidence: 99%

“…3) Deploying the mini-batch gradient descent based on Adam optimizer to train the DNN [28], [28], [32], [33], [43] as shown in Fig. 2 (Training phase), which effectively minimizes the mean square error (MSE) given by…”

Section: Low-complexity Implementation For the Joint Optimization mentioning

confidence: 99%

“…6(a) and 6(b), we show the pdf for the optimal transmit power p k and normalized number of CPU cycles (19), as an input parameter to train the proposed DNN along with b k and T th k , ∀k. 2) DNN training and testing: We implemented the proposed DNN scheme with the Keras machine learning toolkit where Adam optimizer was used for minimizing the MSE during the training phase [28], [33], [43]. Accordingly, we consider three hidden layers with 128, 64 and 32 neurons for l = 1, 2 and 3, respectively.…”

Section: Deep Neural Network Evaluationmentioning

confidence: 99%

See 3 more Smart Citations

Computation Offloading for IoT in C-RAN: Optimization and Deep Learning

Pradhan

She

et al. 2020

IEEE Trans. Commun.

View full text Add to dashboard Cite

We consider computation offloading for Internet-of-things (IoT) applications in multiple-inputmultiple-output (MIMO) cloud-radio-access-network (C-RAN). Due to the limited battery life and computational capability in the IoT devices (IoTDs), the computational tasks of the IoTDs are offloaded to a MIMO C-RAN, where a MIMO radio resource head (RRH) is connected to a baseband unit (BBU) through a capacity-limited fronthaul link, facilitated by the spatial filtering and uniform scalar quantization. We formulate a computation offloading optimization problem to minimize the total transmit power of the IoTDs while satisfying the latency requirement of the computational tasks, and find that the problem is non-convex. To obtain a feasible solution, firstly the spatial filtering matrix is locally optimized at the MIMO RRH. Subsequently, we leverage the alternating optimization framework for joint optimization on the residual variables at the BBU, where the baseband combiner is obtained in a closed-form, the resource allocation sub-problem is solved through successive inner convexification, and the number of quantization bits is obtained by a line-search method. As a low-complexity approach, we deploy a supervised deep learning method, which is trained with the solutions to our optimization algorithm. Numerical results validate the effectiveness of the proposed algorithm and the deep learning method.

show abstract

Section: A Hsf Design At the Xl-mimo Rrhmentioning

confidence: 99%

“…Under the assumption that ∃ p such that (28) is satisfied, f k , ∀k given by (50) converges to a KKT point of P 6 .…”

Section: Lemmamentioning

confidence: 99%

Section: Low-complexity Implementation For the Joint Optimization mentioning

confidence: 99%

Section: Deep Neural Network Evaluationmentioning

confidence: 99%

See 2 more Smart Citations

Computation Offloading for IoT in C-RAN: Optimization and Deep Learning

Pradhan

She

et al. 2020

IEEE Trans. Commun.

View full text Add to dashboard Cite

show abstract

“…With the rapid development of deep learning, it has gradually become a promising tool in solving difficult wireless communication problems due to its excellent performance and low complexity, such as resource allocation [5], channel decoding [6] and channel estimation [7], [8]. [7] designed a supervised learning based pilot assignment scheme for a massive MIMO system, in which the network is trained with the optimal pilot assignment results of exhaustive search serving as the ground truth. However, this method is applicable only when the search space is small because of the exponential complexity of exhaustive search.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning-Based Pilot Design for Multi-User Distributed Massive MIMO Systems

Zhu

et al. 2019

IEEE Wireless Commun. Lett.

View full text Add to dashboard Cite

This letter proposes a deep learning based pilot design scheme to minimize the sum mean square error (MSE) of channel estimation for multi-user distributed massive multipleinput multiple-output (MIMO) systems. The pilot signal of each user is expressed as a weighted superposition of orthonormal pilot sequence basis, where the power assigned to each pilot sequence is the corresponding weight. A multi-layer fully connected deep neural network (DNN) is designed to optimize the power allocated to each pilot sequence to minimize the sum MSE, which takes the channel large-scale fading coefficients as input and outputs the pilot power allocation vector. The loss function of the DNN is defined as the sum MSE, and we leverage the unsupervised learning strategy to train the DNN. Simulation results show that the proposed scheme achieves better sum MSE performance than other methods with low complexity.Index Terms-Pilot design, deep learning, deep neural network (DNN), multi-user distributed massive MIMO.

show abstract

Smart Antenna Designs

Tian

Luo

et al. 2020

Wiley 5G Ref

View full text Add to dashboard Cite

Smart antenna technologies are vital candidates to advance fifth generation (5G) wireless communication system owing to their merits of large capacity, big signal range, strong anti‐interference ability, flexible beam‐scanning control, and easy accessibility of point‐to‐point communications. By adopting smart signal processing algorithms, smart antennas are advantageous over others with electronically alterable radiation patterns and convenient beam‐scanning as required. Generally speaking, a smart antenna system consists of three components: the antenna, the beamforming network, and the signal processing unit with related beamforming algorithms. The purpose of this article is to review and summarize the existing works in these three parts. In the first part, the basic principles, the types, and the advantages of smart antennas are introduced. The recent development of smart antenna systems and related intelligent algorithms are presented in the second part. The third part includes the application of artificial intelligence in the design of smart antennas, the metamaterial‐based smart antennas, and the future of smart antennas. Judging from the current level in this field, it is believed that smart antennas will play a crucial role in driving the realization of the smart home and city, promoting the future communication systems, the Internet of things, the autonomous vehicles, etc.

show abstract

Deep Learning Based Pilot Allocation Scheme (DL-PAS) for 5G Massive MIMO System

Cited by 99 publications

References 10 publications

Computation Offloading for IoT in C-RAN: Optimization and Deep Learning

Computation Offloading for IoT in C-RAN: Optimization and Deep Learning

Deep Learning-Based Pilot Design for Multi-User Distributed Massive MIMO Systems

Smart Antenna Designs

Contact Info

Product

Resources

About