Energy efficiency is essential for federated learning (FL) over mobile devices and its potential prosperous applications. Different from existing communication efficient FL research efforts, which regard communication energy consumption as the bottleneck, we have observed that with ever increasing wireless transmission speed (e.g., Wi-Fi 5 or 5G), the energy consumption of wireless communications for model updates in FL is significantly reduced and sometimes is smaller than that of local on-device training. Motivated by such observations, in this paper, we propose a high-speed wireless communications inspired energy efficient federated learning over mobile devices (EEFL), whose goal is to reduce the overall energy consumption (computing + communication). In particular, we design a novel energy-aware adaptive local update policy for mobile devices by jointly considering FL performance and energy saving of high-speed wireless transmissions. Furthermore, given the device’s local update policy in each FL global round, we advance the dynamic voltage and frequency scaling (DVFS) strategy to minimize local training’s energy consumption by keeping GPU and CPU working at appropriate frequencies without triggering thermal throttling. Extensive experimental results with various learning models, datasets, and wireless transmission environments demonstrate the proposed EEFL’s superiority over the peer designs in terms of energy efficiency.

Federated learning (FL) over mobile devices is an emerging distributed learning paradigm for numerous delay sensitive applications. In FL, the training delay is composed of the computing and communication delay. Some of the participating mobile devices may have slow local computing or wireless communications, which results in high FL training delay. Intuitively, if fast devices help slow ones, the FL training delay can potentially be reduced. However, helping each other among devices requires frequent transmissions and may cause additional delay. Fortunately, we observe that device-to-device (D2D) transmission, a fast and direct transmission, may be applied between device pairs to mitigate the additional delay from frequent transmissions. Inspired by those observations, we develop the D2D transmission assisted FL (DAFL), a novel FL scheme to improve the training delay over mobile devices. Briefly, we first put the eligible mobile devices into pairs, assigning each pair to one of the four types of relation: 1) similar computing, large communication gap; 2) similar communication, large computing gap; 3) one with faster computing and the other with faster communication; and 4) one with both faster computing and communication. We design the process for each type of device pair to: 1) improve the transmission delay of each pair, by letting the fast device help with the model parameters transmission to the server and 2) improve the computing delay by splitting learning task between paired devices. The emulation results demonstrate that DAFL surpasses existing peer designs in terms of reducing training delay by more than 20%.

This artilce aims to develop a differential private federated learning (FL) scheme with the least artificial noises added while minimizing the energy consumption of participating mobile devices. By observing that some communication efficient FL approaches and even the nature of wireless communications contribute to the differential privacy (DP) preservation of training data on mobile devices, in this paper, we propose to jointly leverage gradient compression techniques (i.e., gradient quantization and sparsification) and additive white Gaussian noises (AWGN) in wireless channels to develop a piggyback DP approach for FL over mobile devices. Even with the piggyback DP approach, information distortion caused by gradient compression and noise perturbation may slow down FL convergence, which in turn consumes more energy of mobile devices for local computing and model update communications. Thus, we theoretically analyze FL convergence and formulate an energy efficient FL optimization under piggyback DP, transmission power, and FL convergence constraints. Furthermore, we propose an efficient iterative algorithm where closed-form solutions for artificial DP noise and power control are derived. Extensive simulation and experimental results demonstrate the effectiveness of the proposed scheme in terms of energy efficiency and privacy preservation.

Participant selection (PS) helps to accelerate federated learning (FL) convergence, which is essential for the practical deployment of FL over mobile devices. While most existing PS approaches focus on improving training accuracy and efficiency rather than residual energy of mobile devices, which fundamentally determines whether the selected devices can participate. Meanwhile, the impacts of mobile devices heterogeneous wireless transmission rates on PS and FL training efficiency are largely ignored. Moreover, PS causes the staleness issue. Prior research exploits isolated functions to force long-neglected devices to participate, which is decoupled from original PS designs. In this paper, we propose aresidualenergy andwirelessaware PS design for efficientFLtraining over mobile devices (REWAFL). REWAFL introduces a novel PS utility function that jointly considers global FL training utilities and local energy utility, which integrates energy consumption and residual battery energy of candidate mobile devices. Under the proposed PS utility function framework, REWAFL further presents a residual energy and wireless aware local computing policy. Besides, REWAFL buries the staleness solution into its utility function and local computing policy. The experimental results show that REWAFL is effective in improving training accuracy and efficiency, while avoiding flat battery of mobile devices.

Federated learning (FL) over mobile devices is an emerging distributed learning paradigm for numerous delay sensitive applications. In FL, the training delay is composed of the computing and communication delay. Some of the participating mobile devices may have slow local computing or wireless communications, which results in high FL training delay. Intuitively, if fast devices help slow ones, the FL training delay can potentially be reduced. However, helping each other among devices requires frequent transmissions and may cause additional delay. Fortunately, we observe that Device-to-Device (D2D) transmission, a fast and direct transmission, may be applied between device pairs to mitigate the additional delay from frequent transmissions. Inspired by those observations, we develop the D2D transmission assisted FL (DAFL), a novel FL scheme to improve the training delay over mobile devices. Briefly, we first put the eligible mobile devices into pairs, each pair consisting of a fast and a slow device. Then, we apply D2D transmission between each device pair to: (1) improve the transmission delay of each pair, by letting the fast device help with the model parameters transmission to the server, and (2) improve the computing delay by splitting learning task between paired devices. The emulation results demonstrate that DAFL surpasses existing peer designs in terms of reducing training delay by more than 20%.

Federated learning (FL) over mobile edge devices is a promising distributed learning paradigm for various mobile applications. However, practical deployment of FL over mobile devices is very challenging because (i) conventional FL incurs huge training latency for mobile edge devices due to interleaved local computing and communications of model updates, (ii) there are heterogeneous training data across mobile edge devices, and (iii) mobile edge devices have hardware heterogeneity in terms of computing and communication capabilities.To address aforementioned challenges, in this paper, we propose a novel "workie-talkie" FL scheme, which can accelerate FL’s training by overlapping local computing and wireless communications via contrastive regularization (FedCR). FedCR can reduce FL’s training latency and almost eliminate straggler issues since it buries/embeds the time consumption of communications into that of local training. To resolve the issue of model staleness and data heterogeneity co-existing, we introduce class-wise contrastive regularization to correct the local training in FedCR. Besides, we jointly exploit contrastive regularization and subnetworks to further extend our FedCR approach to accommodate edge devices with hardware heterogeneity. We deploy FedCR in our FL testbed and conduct extensive experiments. The results show that FedCR outperforms its status quo FL approaches on various datasets and models.

Energy efficiency is essential for federated learning (FL) over mobile devices and its potential prosperous applications. Different from existing communication efficient FL research efforts, which regard communication energy consumption as the bottleneck, we have observed that with ever increasing wireless transmission speed (e.g., Wi-Fi 5 or 5G), the energy consumption of wireless communications for model updates in FL is significantly reduced and sometimes is smaller than that of local on-device training. Motivated by such observations, in this paper, we propose a high-speed wireless communications inspired energy efficient federated learning over mobile devices (EEFL), whose goal is to reduce the overall energy consumption (computing + communication). In particular, we design a novel energy-aware adaptive local update policy for mobile devices by jointly considering FL performance and energy saving of high-speed wireless transmissions. Furthermore, given the device's local update policy in each FL global round, we advance the dynamic voltage and frequency scaling (DVFS) strategy to minimize local training's energy consumption by keeping GPU and CPU working at appropriate frequencies without triggering thermal throttling. Extensive experimental results with various learning models, datasets, and wireless transmission environments demonstrate the proposed EEFL's superiority over the peer designs in terms of energy efficiency.

Federated learning (FL) over mobile devices has fostered numerous intriguing applications/services, many of which are delay-sensitive. In this paper, we propose a service delay efficient FL (SDEFL) scheme over mobile devices. Unlike traditional communication efficient FL, which regards wireless communications as the bottleneck, we find that under many situations, the local computing delay is comparable to the communication delay during the FL training process, given the development of high-speed wireless transmission techniques. Thus, the service delay in FL should be computing delay + communication delay over training rounds. To minimize the service delay of FL, simply reducing local computing/communication delay independently is not enough. The delay trade-off between local computing and wireless communications must be considered. Besides, we empirically study the impacts of local computing control and compression strategies (i.e., the number of local updates, weight quantization, and gradient quantization) on computing, communication and service delays. Based on those trade-off observation and empirical studies, we develop an optimization scheme to minimize the service delay of FL over heterogeneous devices. We establish testbeds and conduct extensive emulations/experiments to verify our theoretical analysis. The results show that SDEFL reduces notable service delay with a small accuracy drop compared to peer designs.

Federated learning (FL) is a popular collaborative distributed machine learning paradigm across mobile devices. However, practical FL over resource constrained mobile devices confronts multiple challenges, e.g., the local on-device training and model updates in FL are power hungry and radio resource intensive for mobile devices. To address these challenges, in this paper, we attempt to take FL into the design of future wireless networks and develop a novel joint design of wireless transmission and weight quantization for energy efficient FL over mobile devices. Specifically, we develop flexible weight quantization schemes to facilitate on-device local training over heterogeneous mobile devices. Based on the observation that the energy consumption of local computing is comparable to that of model updates, we formulate the energy efficient FL problem into a mixed-integer programming problem where the quantization and spectrum resource allocation strategies are jointly determined for heterogeneous mobile devices to minimize the overall FL energy consumption (computation + transmissions) while guaranteeing model performance and training latency. Since the optimization variables of the problem are strongly coupled, an efficient iterative algorithm is proposed, where the bandwidth allocation and weight quantization levels are derived. Extensive simulations are conducted to verify the effectiveness of the proposed scheme.

Federated learning (FL) is a promising distributed learning paradigm, which can effectively avoid the privacy leakage and communication issues compared with the centralized learning. Specifically, in each training iteration, FL nodes only upload the local training results to the centralized server without disclosure of their raw training dataset and the centralized server will aggregate the local results of all FL nodes and update the global model. To this end, the performance of the global model is highly dependent on the nodes' cooperation. However, it is challenging to motivate mobile edge devices to involve themselves in the FL process without a desired incentive. Another significant concern of the mobile edge devices is the communication and computational energy cost of participation. Therefore, considering the high cost and weak communication channel with the centralized server specially for the distant nodes, in this paper, we propose a relay-assisted energy efficient scheme for federated learning, where each FL computational node is not only motivated by monetary awards based on their local dataset, but also further motivated to function as a relay node to assist distant nodes on local results uploading due to its locality advantage. To achieve a stable pairing solution between FL computational nodes and assisted relays in a distributive fashion, a many-to-one matching algorithm is applied, where each the computational node and relay is unable to deviate with current pairing unilaterally for higher revenue. Extensive simulations are conducted to illustrate the correctness and effectiveness of our proposed scheme.

Preventing COVID-19 disease from spreading in communities will require proactive and effective healthcare resource allocations, such as vaccinations. A fine-grained COVID-19 vulnerability map will be essential to detect the high-risk communities and guild the effective vaccine policy. A mobile-crowdsourcing-based self-reporting approach is a promising solution. However, an accurate mobile-crowdsourcing-based map construction requests participants to report their actual locations, raising serious privacy concerns. To address this issue, we propose a novel approach to effectively construct a reliable community-level COVID-19 vulnerability map based on mobile crowdsourced COVID-19 self-reports without compromising participants’ location privacy. We design a geo-perturbation scheme where participants can locally obfuscate their locations with the geo-indistinguishability guarantee to protect their location privacy against any adversaries’ prior knowledge. To minimize the data utility loss caused by location perturbation, we first design an unbiased vulnerability estimator and formulate the location perturbation probability generation into a convex optimization. Its objective is to minimize the estimation error of the direct vulnerability estimator under the constraints of geo-indistinguishability. Given the perturbed locations, we integrate the perturbation probabilities with the spatial smoothing method to obtain reliable community-level vulnerability estimations that are robust to a small-sampling-size problem incurred by location perturbation. Considering the fast-spreading nature of coronavirus, we integrate the vulnerability estimates into the modified susceptible-infected-removed (SIR) model with vaccination for building a future trend map. It helps to provide a guideline for vaccine allocation when supply is limited. Extensive simulations based on real-world data demonstrate the proposed scheme superiority over the peer designs satisfying geo-indistinguishability in terms of estimation accuracy and reliability.

Federated learning (FL) through its novel applications and services has enhanced its presence as a promising tool in the Internet of Things (IoT) domain. Specifically, in a multiaccess edge computing setup with a host of IoT devices, FL is most suitable since it leverages distributed client data to train high-performance deep learning (DL) models while keeping the data private. However, the underlying deep neural networks (DNNs) are huge, preventing its direct deployment onto resource-constrained computing and memory-limited IoT devices. Besides, frequent exchange of model updates between the central server and clients in FL could result in a communication bottleneck. To address these challenges, in this article, we introduce GWEP, a model compression-based FL method. It utilizes joint quantization and model pruning to reap the benefits of DNNs while meeting the capabilities of resource-constrained devices. Consequently, by reducing the computational, memory, and network footprint of FL, the low-end IoT devices may be able to participate in the FL process. In addition, we provide theoretical guarantees of FL convergence. Through empirical evaluations, we demonstrate that our approach significantly outperforms the baseline algorithms by being up to 10.23 times faster with 11 times lesser communication rounds, while achieving high-model compression, energy efficiency, and learning performance.

This paper presents a federated learning (FL) framework that uses multiple self-reporting crowdsourcing mobile and web apps to collaboratively construct a fine-grained COVID-19 vulnerability prediction map. The use of FL provides a reliable prediction by aggregating training results from multiple apps, while at the same time circumventing data privacy regulations that prevent user information from multiple apps to be shared with each other. Such a fine-grained vulnerability map identifies early on high-risk areas, helping to reduce the spread of the disease. To mitigate data bias from each self-reporting app, an adaptive worker selection algorithm that leverages neighbouring datasets to obtain a balanced data distribution is proposed. Further, a differential privacy scheme is adopted to protect user information. The simulation results show that the proposed framework outperforms the widely used FedAvg FL algorithm by 6% on prediction accuracy while preserving user privacy.

The pandemic of the coronavirus (COVID-19) has caused an unprecedented global public health crisis, and most countries in the world are running out of the healthcare resources. A fine-grained COVID-19 vulnerability map will be essential to track the number of people with covid-like symptoms, so that the the potential outbreak communities can be identified and the valuable healthcare resources can proactively and dynamically be allocated. Mobile crowdsourcing based symptom reporting is a promising and convenient option to construct such a map, while it may compromise the location privacy of crowdsourcing participants. In this work, we propose a novel approach to establish the COVID-19 vulnerability map based on the crowdsourced reporting without disclosing the participants' location privacy to a semi-honest crowdsourcing aggregator. Briefly, based on the differentially private geo-indistinguishability, the mobile participants are able to locally perturb their geographic data. With the masked geographic information, we employ the best linear unbiased prediction estimator with spatial smoothing to obtain the reliable vulnerability estimates in the areas of interest and construct the map. Given the fast spreading nature of coronavirus, we integrate the vulnerability estimates with a susceptible-exposed-infected-removed (SEIR) model to build up a future trend map. Extensive simulations based on real-world data verify the effectiveness of the proposed method.

The advance of edge computing pushes computing functionalities to the network edge and brings lucrative opportunities for edge operators (EOs) to cater the users with low latency requirement. Unlike in cloud computing, edge servers have limited computing capacity and require a proper resource planning. To avoid loss of potential profit, a promising way is to outsource cloud resources from a public cloud with additional cost when the edge computing capacity is insufficient to meet the real-time demands. Besides, the uncertainty of future demands also affects EOs' profits. It's essential to consider the interaction among market participants with different risk attitudes. To this end, we study multiple risk-averse EOs with one risk-neutral Cloud Provider (CP) in an edge computing market, where each EO competes to serve the users by determining the optimal resource provision strategies given the demand and the outsource price charged by the CP, and the CP sets the price based on the best responses of the EOs. We model the interaction between EOs and CP as a two stage Stackelberg game, and employ a data-driven optimization approach to characterize the uncertainty. We explore the existence and uniqueness of subgame Nash equilibrium, and find the equilibrium based on the Sample Average Approximation (SAA) method. Extensive simulations using real-world cluster data traces verify the effectiveness of the proposed method.

The development of 5G network and rapid growth of mobile traffic bring lucrative opportunities for Micro Operators (μOs), the novel local operators who own local spectrum, deploy and manage small cells (e.g., femtocells) in a specific area. Collaborating with traditional mobile network operators (MNOs), μOs gain profits from helping to offload the traffic carried by macro base stations and providing the MNOs customers with seamless service. However, due to the demand uncertainty and the sensitivity of individual user's demand profile, it is challenging for μOs to allocate the resource properly to avoid the under-and over-utilized situations. To address this issue, in this paper, we propose to employ data-driven approach to characterize the demand uncertainty, exploit differential privacy protocols to protect user's demand profile, and formulate the small cell planning problem into two-stage stochastic programming optimization with the objective of minimizing the capital and operational costs of μOs. Based on the formulated problem, we develop feasible solutions and conduct extensive simulations using real-world base station accessing real cellular data (i.e., data of 4G LTE network in Zhengzhou city, China) to verify the effectiveness of the proposed model.

Proactive caching in mobile-edge computing (MEC) networks is promising to handle the ever-increasing demand for wireless video services, and transcoding at MEC servers further improves the flexibility of video content delivery. However, how to effectively conduct caching for adaptive bitrate streaming poses great challenges due to the uncertainty of user preferences. The caching decisions also have a profound impact on the system energy efficiency since they may change the video delivery modes. In this article, by integrating caching, transcoding, and backhaul retrieving in a MEC-enabled adaptive streaming system, we propose a holistic solution to jointly determine the caching of bitrate-aware files and the scheduling of video requests in an energy-efficient manner. Specifically, we leverage a data-driven approach to characterize the uncertainty of real request arrivals. Based on the uncertainty model, we formulate a data-driven risk-averse optimization to derive a robust strategy for caching and delivery scheduling, which is a two-stage stochastic mixed-integer programming (SMIP) with the goal of minimizing the total expected energy consumption. We also develop feasible solutions and conduct extensive simulations on real-world data sets. The results validate the effectiveness of the proposed scheme in both the energy efficiency and the cache hit ratio.