Data-Aided Sensing Where Communication and Sensing Meet: An Introduction
Jinho Choi
Abstract
Open-access reader
Jinho Choi
Abstract
Open-access reader
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.
OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
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.
Key concepts: Upload, Computer science, Base station, Set (abstract data type), Internet of Things, The Internet, Computer network, Embedded system