The IEEE Journal of Selected Areas in Sensors (JSAS) publishes papers in all areas of the field of interest of the IEEE Sensors Council, i.e., the theory, design, simulation, fabrication, manufacturing, and application of devices for sensing and transducing physical, chemical, and biological phenomena, with emphasis on the electronics, physics and reliability aspects of sensors and integrated sensor-actuators. The Journal is built exclusively from papers on selected topics of current interest to the Sensors community.

IEEE Data Descriptions accepts articles with a recommended length of 4—8 pages. Any extra information can be submitted as supplementary information, which will be published alongside the article on IEEE Xplore. When submitting a paper, authors must choose one of three article types: descriptor, collections, or meta. Each article type has very specific formatting instructions. The article template can be found at https://template-selector.ieee.org/.

Author Biographies (in LaTeX the IEEEbiography) are not a part of any article type. Do not include these with your submission. Any submission that does include author biographies will be returned to the authors and will not be reviewed.

ABSTRACT: For Descriptor and Collections articles the abstract is considered the Introduction section in traditional articles. Authors can introduce the dataset by briefly introducing or describing it. For Meta articles, the abstract is considered the Problem and Hypothesis section, where the problem and hypothesis central to the article are briefly described. Abstracts must be 150—250, and references/citations are not allowed. They cannot claim new scientific findings or experimental results; save this for the Conclusions section.

IEEE SOCIETY/COUNCIL: The main IEEE Section or Council that the dataset you are writing about fits in or belongs to.

DATA DOI/PID: The DOI of the dataset(s) described in the article. If more than one, then separate by using a semicolon. For example: 10.7910/DVN/FIE0S4; 10.7910/DVN/MXB7VO.

DATA TYPE/LOCATION: This information will be filled in at the time of publication, along with the selections made when submitting through the IEEE Author Portal.

INDEX TERMS: Enter keywords or phrases alphabetically, separated by commas. For more information, please follow this link.

Descriptor articles (the main type of article published) describe the published (or soon-to-be-published) dataset. In essence, this is a detailed manual on the dataset — what it is and how to use it. Each article will follow a set format for predetermined, well-defined sections. Metadata will also be collected to further classify the data that is being described, as well as funding sources, etc. The purpose of a Descriptor article is to:

• document sources and methods used to collect the data,
• to quantify the quality of the data,
• discuss any issues and caveats with the data or use of,
• discuss basic statistics about the data (e.g., collection period, location); and,
• collect enhanced metadata (define each file, columns within, and row type).

Descriptor articles must have and only have these sections (without numbering) in this order:

• Abstract (See First Page Front-Matter)
• Background
• Collection Methods and Design
• Validation and Quality
• Records and Storage
• Insights and Notes
• Source Code and Scripts
• Acknowledgements and Interests
• References

The article Title has a very specific format specified in the template file. The title must be the name and acronym of the dataset. LaTeX code for doing this is:

\title{{\color{black}\textbf{Descriptor:}} 
\textit{The Hourly Usage of Energy Dataset} (HUE)}

For example:

Background must provide an overview of the data collected and discuss how it fits with other comparable, published datasets. Authors must make clear the data's value and how it can be reused. Authors must also summarize any previous publication made using this data with a brief summary and citation for each time used. Do not include a paragraph on how your article is organized; all articles of this type are organized the same way.

Collection Methods and Design must provide details on how that data was collected. This includes details on any hardware/system designs used to collect the data (i.e., data acquisition). In addition, the steps and procedures used to collect and process the data in its final form (i.e., computational processing). It is recommended that the authors provide diagrams that show the overall system/procedure used.

Validation and Quality must provide details that support the technical quality and/or accuracy of the data collected. For example, the error rate or accuracy of any hardware sensors used for collection. Authors should provide figures and tables to support this.

Records and Storage must provide details on how the data files are structured and how the data is stored. A table that lists all files with summary details is recommended. For example, for CSV files, a description of the columns and rows of each file is needed. Additionally, details on how files relate to each other is needed, especially if there is some hierarchical structure. For example, file A is a summation of files B and C plus some error terms. The format of each data file should be sufficiently described.

Insights and Notes must use this section to provide details on any caveats and special case usages with the data. Authors can also discuss how the data can be used in other ways that they themselves cannot experiment on.  For example, authors may have a dataset that has power meter readings and may want to point out that the data can also be used for general/theoretical time-series analysis.

Source Code and Scripts must provide details on any public source code repositories that contain code used to collect, clean, or process this data. Details on any third-party software used should also be listed (including software version numbers). URLs and links to these repositories should be listed in the reference section. It is recommended that these repositories have an associated DOI to ensure their permanence.

Acknowledgements and Interests is a required and very important section to write correctly. Include the following information in this section:

• Authors may wish to acknowledge non-author contributors for their contribution, and they can do so briefly here. For example:
  

  Special thanks to John Doe for providing insight into time-series data collection and to ABC Corp. for donating the data collection hardware used.

• Authors must state the contributions each author has made. For example:
  

  S.M. was the main author. S.M. and M.T. designed the data collection system and prepared the data for publication. H.W. performed data cleaning and created diagrams. All authors reviewed the manuscript.

• Acknowledgement of funding agencies must include the grant or contribution number for each. For example:
  

  This work was funded by Natural Resources Canada (NRCan) under project XYZ123 and by the Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant DG000111222333.

  If the was no grant funding, then the authors must state so with the following statement:

  This work was done independently of any grants or awards.

• Authors must include details of any and all conflicts of interest. This may provide insight into the motivation for collecting this data. For example, the sugar industry supported the data collection in this dataset, which supports using sugar as a healthy dietary supplement. If no conflict of interest exists, authors must state so with the following statement:

  The article authors have declared no conflicts of interest.

For example what the formatting of this section must look like:

References must follow the standard IEEE format and use numbered citations. Please refer to the IEEE Reference Guide.

Collections articles follow the lead and expectations of Descriptor articles. However, they focus on groups of datasets collected during or created from a competition, hackathon, or event. These articles must have and only have the same sections in the same order as Descriptor articles. Additional information that describes these datasets as a whole must be added to any Collections article, along with statistics and metadata about the event in the Background and Collection Methods and Design sections.

The article Title has a very specific format specified in the template file. The title must be the name and acronym (optional) of the collection. LaTeX code for doing this is:

\title{{\color{black}\textbf{Collection:}} 
\textit{Datasets from IEEEXtreme 2023}}

For example:

Meta articles follow the lead and expectations of Descriptor articles. Meta articles are articles that allow for discussion about data. Authors can express their findings about data in several areas including, but not limited to:

• standardized formats for data in a given area (e.g., collecting power data from sub-stations),
• surveys of a number of datasets of a given type (e.g., GIS, time-series, etc.),
• data that are a library of models for a given AI or machine learning algorithm or system,
• data or dataset that is metadata of existing published datasets,
• comparing a number of similar datasets in an area to create meta-statistics data; or,
• any other data topics that may not have traditional scientific experimental results.

Authors are welcome to contact the Editor-in-Chief to inquire if their article fits the scope of Meta articles. EiC contact details are found on the Editorial Board page.

Meta articles have these recommended (not required) sections (without numbering) in this order; however, some articles might not fix this structure easily:

• Abstract (See First Page Front-Matter)
• Background
• Proposed Methodology
• Applied Analysis
• Discussion of Results
• Conclusions
• Source Code and Scripts
• Acknowledgements and Interests
• References

The article Title has a very specific format and should be short and to the point. The template file will give the exact format to use. LaTeX code for doing this is:

\title{{\color{black}\textbf{Meta:}} 
\textit{A Critic of Medical Image Resolutions in Publicly Available Data}}

For example:

First Page Front-Matter can have the DATA DIO/PID heading omitted if needed, as Meta articles might not have a new dataset as part of the article. Existing published datasets used can follow normal citation requirements. For the DATA TYPE/LOCATION heading, you must specify what data types the article uses all while omitting the location if it is not relevant. 

Background must provide an overview and justify why this problem is worth writing about. Authors must make clear the article’s value and how the methods described add value to the reader. Authors must also provide a literature review that summarizes any previous publication that discusses the problem/topics the respondents are writing about. If the proposed mythology/analysis was previously used in published works, then the author must have a brief summary and citation for each time used.

Proposed Methodology must describe the theory that will be used to do the analysis and collect the findings. It must also describe the design of how the hypothesis will be measured.

Applied Analysis must provide a description of how the analysis was done and how the findings were collected. Describe how the hypothesis was actually measured.

Discussion of Results must provide an objective discussion of the analysis, findings, and hypothesis. Listing how the findings can affect or solve the problem. State the limitations of the work. Describe any items that were not included and why (time, funding, equipment, etc. constraints). If this work can be extended, what might be looked at next?

Conclusions are limited to 250—350 words about the claim of new scientific findings, the significance of experimental results, or what achievements were made. References/citations are not allowed in this section.

Acknowledgements and Interests is a required and very important section to write correctly. For information about this article section, refer to the Descriptor Articles section on this page.

References must follow the standard IEEE format and use numbered citations. Please refer to the IEEE Reference Guide.