Performance Optimization of Federated Learning over Wireless Networks

Chen, Mingzhe; Yang, Zhaohui; Saad, Walid; Yin, Changchuan; Poor, H. Vincent; Cui, Shuguang

doi:10.1109/globecom38437.2019.9013160

Cited by 87 publications

(45 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed integrated GRU and CNN predictive model build the relationship between output I T +1 and the input time series historical illumination distribution I 1 , I 2 , · · · , I t , · · · , I T using the weight parameters. To build this relationship, a batch gradient descent approach is used to train the weight matrices which are initially generated randomly via a uniform distribution [29].…”

Section: Integrated Gru and Cnn Predictive Model Trainingmentioning

confidence: 99%

“…Theorem 1: For problem (29), the optimal user association u ij,T +1 and transmit power P i,T +1 can be respectively expressed as:…”

Section: B User Association and Power Efficiency With Fixed Uav Deplmentioning

confidence: 99%

“…Note that the UAV deployment optimization algorithm is distributed according to (18). For the user association problem, the overhead of obtaining u T +1 from (30) is O(LDU ), where L is the average iteration number until (29) converges. Note that the user association optimization algorithm is centralized according to (29).…”

mentioning

confidence: 99%

See 2 more Smart Citations

Gated Recurrent Units Learning for Optimal Deployment of Visible Light Communications Enabled UAVs

Wang

Chen

Yang

et al. 2019

2019 IEEE Global Communications Conference (GLOBECOM)

Self Cite

View full text Add to dashboard Cite

In this paper, the problem of optimizing the deployment of unmanned aerial vehicles (UAVs) equipped with visible light communication (VLC) capabilities is studied. In the studied model, the UAVs can predict the illumination distribution of a given service area and determine the user association with the UAVs to simultaneously provide communications and illumination. However, ambient illumination increases the interference over VLC links while reducing the illumination threshold of the UAVs.Therefore, it is necessary to consider the illumination distribution of the target area for UAV deployment optimization. This problem is formulated as an optimization problem, which jointly optimizes UAV deployment, user association, and power efficiency while meeting the illumination and communication requirements of users. To solve this problem, an algorithm that combines the machine learning framework of gated recurrent units (GRUs) with convolutional neural networks (CNNs) is proposed. Using GRUs and CNNs, the UAVs can model the long-term historical illumination distribution and predict the future illumination distribution. Based on the prediction of illumination distribution, the optimization problem becomes nonconvex and is then solved using a low-complexity, iterative physical relaxation algorithm.The proposed algorithm can find the optimal UAV deployment and user association to minimize the total transmit power. Simulation results using real data from the Earth observations group (EOG) at NOAA/NCEI show that the proposed approach can achieve up to 64.6% reduction in total transmit power compared to a conventional optimal UAV deployment that does not consider the illumination Y. Wang and T. Luo are with the Beijing 2 distribution and user association. The results also show that UAVs must hover at areas having strong illumination, thus providing useful guidelines on the deployment of VLC-enabled UAVs. Index TermsVisible light communication, unmanned aerial vehicles, drones, machine learning, gated recurrent units, convolutional neural networks, energy efficiency. I. INTRODUCTIONDeploying unmanned aerial vehicles (UAVs) as flying base stations (BSs) for wireless networking is a flexible and cost-effective approach to providing on-demand communications [2]-[6]. However, for tomorrow's ultra dense wireless networks that encompass a large number of ground BSs, UAVs may not have enough radio frequency (RF) resources to service ground users. Moreover, UAVs deployed as aerial BSs using RF will interfere with ground devices, hence significantly affecting the performance of the ground network. In addition, the limited energy will restrict the applicability of UAVs using RF resource to provide high-speed communication services for ground users. These challenges can be addressed by equipping UAVs with visible light communication (VLC) capabilities [7]. Indeed, VLC has recently attracted attention due to its large license-free bandwidth and high energy efficiency. For instance, a VLC system that uses light-emitting diodes (LEDs) to t...

show abstract

Section: Integrated Gru and Cnn Predictive Model Trainingmentioning

confidence: 99%

“…Theorem 1: For problem (29), the optimal user association u ij,T +1 and transmit power P i,T +1 can be respectively expressed as:…”

Section: B User Association and Power Efficiency With Fixed Uav Deplmentioning

confidence: 99%

See 1 more Smart Citation

Gated Recurrent Units Learning for Optimal Deployment of Visible Light Communications Enabled UAVs

Wang

Chen

Yang

et al. 2019

2019 IEEE Global Communications Conference (GLOBECOM)

Self Cite

View full text Add to dashboard Cite

show abstract

“…In stark contrast to the conventional machine learning methods that run in a data center, FL usually operates at the network edge and brings the models directly to the devices for training, where only the resultant parameters shall be sent to the edge servers that reside in an access point (AP). This salient feature of on-device training brings along great advantages of eliminating the large communication overheads as well as preserving data privacy, and hence making FL particularly relevant for mobile applications [6][7][8][9][10]. However, as the AP needs to link a large number of user equipments (UEs) over a limited spectrum, only a portion of the UEs can be selected to access the radio channel and send their trained updates in each global aggregation [5,11].…”

Section: Introductionmentioning

confidence: 99%

Age-Based Scheduling Policy for Federated Learning in Mobile Edge Networks

Yang

Arafa

Quek

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

162

View full text Add to dashboard Cite

Federated learning (FL) is a machine learning model that preserves data privacy in the training process. Specifically, FL brings the model directly to the user equipments (UEs) for local training, where an edge server periodically collects the trained parameters to produce an improved model and sends it back to the UEs. However, since communication usually occurs through a limited spectrum, only a portion of the UEs can update their parameters upon each global aggregation. As such, new scheduling algorithms have to be engineered to facilitate the full implementation of FL. In this paper, based on a metric termed the age of update (AoU), we propose a scheduling policy by jointly accounting for the staleness of the received parameters and the instantaneous channel qualities to improve the running efficiency of FL. The proposed algorithm has low complexity and its effectiveness is demonstrated by Monte Carol simulations.

show abstract

“…Test performance of the cases which have different subchannels and users: (M, N )= (4, 2),(8,4),(10,5).…”

mentioning

confidence: 99%

Channel Assignment in Uplink Wireless Communication Using Machine Learning Approach

et al. 2020

Self Cite

View full text Add to dashboard Cite

This letter investigates a channel assignment problem in uplink wireless communication systems. Our goal is to maximize the sum rate of all users subject to integer channel assignment constraints. A convex optimization based algorithm is provided to obtain the optimal channel assignment, where the closed-form solution is obtained in each step. Due to high computational complexity in the convex optimization based algorithm, machine learning approaches are employed to obtain computational efficient solutions. More specifically, the data are generated by using convex optimization based algorithm and the original problem is converted to a regression problem which is addressed by the integration of convolutional neural networks (CNNs), feed-forward neural networks (FNNs), random forest and gated recurrent unit networks (GRUs). The results demonstrate that the machine learning method largely reduces the computation time with slightly compromising of prediction accuracy.

show abstract

Performance Optimization of Federated Learning over Wireless Networks

Cited by 87 publications

References 40 publications

Gated Recurrent Units Learning for Optimal Deployment of Visible Light Communications Enabled UAVs

Gated Recurrent Units Learning for Optimal Deployment of Visible Light Communications Enabled UAVs

Age-Based Scheduling Policy for Federated Learning in Mobile Edge Networks

Channel Assignment in Uplink Wireless Communication Using Machine Learning Approach

Contact Info

Product

Resources

About