Energy-Efficient Resource Allocation for Federated Learning in NOMA-Enabled and Relay-Assisted Internet of Things Networks

Al-Abiad, Mohammed S.; Hassan, Md. Zoheb; Hossain, Md. Jahangir

doi:10.1109/jiot.2022.3194546

Cited by 34 publications

(12 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, the global model w is broadcasted from UAV w broadcasting (10) The size of a typical model is s = 9.098 Kb [13], [36], [37], thus T diss = 0.0059 sec. Thanks to the efficient model dissemination proposed method that disseminates models from transmitting UAVs to the closest receiving UAVs with good connectivity, the dissemination delay is negligible.…”

Section: Illustration Of the Proposed Model Dissemination Methodsmentioning

confidence: 99%

“…Several aspects of RRM, such as client scheduling, RRB allocation, and transmit power control, were extensively studied to minimize both communication and computation energy of FL frameworks [16], [17]. An energy-efficient FL framework based on relay-assisted two-hop transmission and non-orthogonal multiple access scheme was recently proposed for both energy and resource constrained Internet of Things (IoT) networks [18]. In the aforesaid studies, conventional star-based FL frameworks were studied.…”

Section: A Summary Of the Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Decentralized Aggregation for Energy-Efficient Federated Learning via D2D Communications

Al-Abiad

Obeed

Hossain

et al. 2023

IEEE Trans. Commun.

View full text Add to dashboard Cite

It is anticipated that aerial-terrestrial integrated networks incorporating unmanned aerial vehicles (UAVs) mounted relays will offer improved coverage and connectivity in the beyond 5G era. Meanwhile, federated learning (FL) is a promising distributed machine learning technique for building inference models over wireless networks due to its ability to maintain user privacy and reduce communication overhead. However, off-the-shelf FL models aggregate global parameters at a central parameter server (CPS), increasing energy consumption and latency, as well as inefficiently utilizing radio resource blocks (RRBs) for distributed user devices (UDs). This paper presents a resource-efficient FL framework, called FedMoD (federated learning with model dissemination), for millimeterwave (mmWave) aerial-terrestrial integrated networks with the following two unique characteristics. Firstly, FedMoD presents a novel decentralized model dissemination algorithm that makes use of UAVs as local model aggregators through UAV-to-UAV and device-to-device (D2D) communications. As a result, FedMoD (i) increases the number of participant UDs in developing FL model and (ii) achieves global model aggregation without involving CPS. Secondly, FedMoD reduces the energy consumption of FL using radio resource management (RRM) under the constraints of over-the-air learning latency. In order to achieve this, by leveraging graph theory, FedMoD optimizes the scheduling of line-of-sight (LOS) UDs to suitable UAVs/RRBs over mmWave links and non-LOS UDs to available LOS UDs via overlay D2D communications. Extensive simulations reveal that decentralized FedMoD offers same convergence rate performance as compared to conventional FL frameworks.

show abstract

Section: Illustration Of the Proposed Model Dissemination Methodsmentioning

confidence: 99%

Section: A Summary Of the Related Workmentioning

confidence: 99%

Decentralized Aggregation for Energy-Efficient Federated Learning via D2D Communications

Al-Abiad

Obeed

Hossain

et al. 2023

IEEE Trans. Commun.

View full text Add to dashboard Cite

show abstract

“…FL Async. FL Flexible Aggregation [10], [24] [11], [12] [16], [25], [26] [22] [23] [17]- [20] This paper Accuracy Convergence speed Energy efficiency the under-use of the computing powers of the users. An alternative to Async-FL is incomplete aggregation [21], where all users upload their local models synchronously, and some of the models are trained incompletely with fewer iterations than others.…”

Section: Kpi Referencementioning

confidence: 99%

“…In [26], the selection of IoT devices and relays, and the transmit powers and CPU frequencies of the selected IoT devices were optimized to minimize the energy consumption, subject to the delay constraint of the FL. A graph-theoretic approach was taken to design a low-complexity, suboptimal solution by applying a greedy maximum-weight-independent-set algorithm.…”

Section: Kpi Referencementioning

confidence: 99%

“…Moreover, the concurrent transmissions of the users under MC-NOMA also prevent the awkward situation (i.e., under TDMA) that the users scheduled to upload their local models earlier have to shorten their local training time. Despite NOMA being considered for local model transmissions in [23], [25], [26], it was not jointly considered with the learning parameter selection and their impact on the convergence of WFL was overlooked.…”

Section: Kpi Referencementioning

confidence: 99%

See 1 more Smart Citation

Multi-Carrier NOMA-Empowered Wireless Federated Learning with Optimal Power and Bandwidth Allocation

Li¹,

Lv²,

Cao³

et al. 2023

Preprint

View full text Add to dashboard Cite

Wireless federated learning (WFL) undergoes a communication bottleneck in uplink, limiting the number of users that can upload their local models in each global aggregation round. This paper presents a new multi-carrier non-orthogonal multiple-access (MC-NOMA)-empowered WFL system under an adaptive learning setting of Flexible Aggregation. Since a WFL round accommodates both local model training and uploading for each user, the use of Flexible Aggregation allows the users to train different numbers of iterations per round, adapting to their channel conditions and computing resources. The key idea is to use MC-NOMA to concurrently upload the local models of the users, thereby extending the local model training times of the users and increasing participating users. A new metric, namely, Weighted Global Proportion of Trained Mini-batches (WGPTM), is analytically established to measurethe convergence of the new system. Another important aspect is that we maximize the WGPTM to harness the convergence of the new system by jointly optimizing the transmit powers and subchannel bandwidths. This nonconvex problem is converted equivalently to a tractable convex problem and solved efficiently using variable substitution and Cauchy's inequality. As corroborated experimentally using a convolutional neural network and an 18-layer residential network, the proposed MC-NOMA WFL can efficiently reduce communication delay, increase local model training times, and accelerate the convergence by over 40%, compared to its existing alternative.

show abstract

Deadline‐cost aware task scheduling algorithm in fog computing networks

Sultan Hajam

2024

Int J Communication

View full text Add to dashboard Cite

SummaryThe Internet of Things (IoT) has provided us with various applications that need processing in real‐time. Fog computing has arisen as a highly effective computing paradigm that provides services closer to the IoT devices and hence fulfills the need for rapid response time and delay requirements of IoT applications. However, the heterogeneity of IoT tasks and the limited resources of fog nodes pose considerable challenges to task scheduling. Therefore, the effective utilization of fog resources is highly significant and necessitates an optimal and intelligent approach to task scheduling. This paper presents the “deadline‐cost aware task scheduling” (DCaTS) algorithm, which is designed for fog‐based IoT applications. The primary objective of this algorithm is to minimize the monetary cost while taking into account the deadline requirements and priority of tasks. The efficacy of DCaTS is assessed by analyzing various parameters, including completion time, monetary cost, response time, and waiting time. The experimental results show that in comparison with traditional policies, the proposed algorithm shows superiority in all the parameters. The proposed algorithm reduces the monetary cost up to 11.07% compared with the second best result. The results also indicate that DCaTS achieves a reduction of 10.05% in completion time, 10% in response time, and 10.12% in waiting time compared with the second best result.

show abstract

Energy-Efficient Resource Allocation for Federated Learning in NOMA-Enabled and Relay-Assisted Internet of Things Networks

Cited by 34 publications

References 34 publications

Decentralized Aggregation for Energy-Efficient Federated Learning via D2D Communications

Decentralized Aggregation for Energy-Efficient Federated Learning via D2D Communications

Multi-Carrier NOMA-Empowered Wireless Federated Learning with Optimal Power and Bandwidth Allocation

Deadline‐cost aware task scheduling algorithm in fog computing networks

Contact Info

Product

Resources

About