Federated Learning With Multichannel ALOHA

Choi, Jinho; Pokhrel, Shiva Raj

doi:10.1109/lwc.2019.2960243

Cited by 29 publications

(22 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Choi and Pokhrel [56] propose a multichannel ALOHA scheme for improving the communication efficiency of FL in a cellular system. It is argued that an adaptation of access probability based on the significance of local updates at each user could improve the aggregation performance in FL.…”

Section: B Technologies For Enabling Fl In Wireless Iotmentioning

confidence: 99%

Federated Learning for Vehicular Internet of Things: Recent Advances and Open Issues

Yoshinaga

et al. 2020

IEEE Open J. Comput. Soc.

239

106

View full text Add to dashboard Cite

show abstract

Section: B Technologies For Enabling Fl In Wireless Iotmentioning

confidence: 99%

Federated Learning for Vehicular Internet of Things: Recent Advances and Open Issues

Yoshinaga

et al. 2020

IEEE Open J. Comput. Soc.

239

106

View full text Add to dashboard Cite

show abstract

“…where µ > 0 is the step-size and K(t) is the number of active devices at round t, which is an estimate of q p q . Note that a similar distributed approach has been used for federated learning with multichannel ALOHA in [25]. Together with the feedback of predicted values, {m q (t)}, the BS can also broadcast ψ(t) so that the sensors can decide the probability of uploading according to (39).…”

Section: B Distributed Updating Using Multichannel Alohamentioning

confidence: 99%

Data-Aided Sensing for Gaussian Process Regression in IoT Systems

Choi

2021

IEEE Internet Things J.

Self Cite

View full text Add to dashboard Cite

In this paper, for efficient data collection with limited bandwidth, data-aided sensing is applied to Gaussian process regression that is used to learn data sets collected from sensors in Internet-of-Things systems. We focus on the interpolation of sensors' measurements from a small number of measurements uploaded by a fraction of sensors using Gaussian process regression with data-aided sensing. Thanks to active sensor selection, it is shown that Gaussian process regression with data-aided sensing can provide a good estimate of a complete data set compared to that with random selection. With multichannel ALOHA, data-aided sensing is generalized for distributed selective uploading when sensors can have feedback of predictions of their measurements so that each sensor can decide whether or not it uploads by comparing its measurement with the predicted one. Numerical results show that modified multichannel ALOHA with predictions can help improve the performance of Gaussian process regression with data-aided sensing compared to conventional multichannel ALOHA with equal uploading probability.

show abstract

“…There can be a predetermined transmission order for sequential polling, where uploading is carried out regardless of local data sets. For efficient uploading, however, DAS can be employed as in [6] [7] [8]. In general, there are two types of DAS approaches.…”

Section: System Modelmentioning

confidence: 99%

“…In this paper, we introduce the key idea and approaches of DAS [6] [7] [8]. In particular, two different DAS schemes, namely centralized DAS and distributed DAS, are discussed.…”

Section: Introductionmentioning

confidence: 99%

Data-Aided Sensing Where Communication and Sensing Meet: An Introduction

Choi

2020

2020 IEEE Wireless Communications and Networking Conference Workshops (WCNCW)

Self Cite

View full text Add to dashboard Cite

Since there are a number of Internet-of-Things (IoT) applications that need to collect data sets from a large number of sensors or devices in real-time, sensing and communication need to be integrated for efficient uploading from devices. In this paper, we introduce the notion of data-aided sensing (DAS) where a base station (BS) utilizes a subset of data that is already uploaded and available to select the next device for efficient data collection or sensing. Thus, using DAS, certain tasks in IoT applications, including federated learning, can be completed by uploading from a small number of selected devices. Two different types of DAS are considered: one is centralized DAS and the other is distributed DAS. In centralized DAS, the BS decides the uploading order, while each device can decide when to upload its own local data set among multiple uploading rounds in distributed DAS. In distributed DAS, random access is employed where the access probability of each device is decided according to its local measurement for efficient uploading.

show abstract

Federated Learning With Multichannel ALOHA

Cited by 29 publications

References 12 publications

Federated Learning for Vehicular Internet of Things: Recent Advances and Open Issues

Federated Learning for Vehicular Internet of Things: Recent Advances and Open Issues

Data-Aided Sensing for Gaussian Process Regression in IoT Systems

Data-Aided Sensing Where Communication and Sensing Meet: An Introduction

Contact Info

Product

Resources

About