IEEE is the world's largest technical professional organization dedicated to advancing technology for the benefit of humanity encompassing electronics engineering, electrical engineering, and other related disciplines. IEEE claims to produce over 30% of the world's literature in the electrical, electronics, and computer engineering fields, publishing approximately 200 peer-reviewed journals and magazines.

This dataset contains the outline information of Research Papers published to IEEE, it was obtained by scraping the IEEE website through Selenium and Chromium. It includes essential columns such as the date, Title, Abstract, Authors, Year, and Number of citations.

This data is ideal for building a recommendation system (Traditional methods or even RAG) and can be applied in Information Retrieval (Example Posting List) or any NLP task as well

The dataset includes 39 files in both LAS and OBJ formats. The LAS format contains point cloud data, while the OBJ format provides a 3D mesh representation of the tower structures.

In collaboration with the IEEE Conference on Biomedical and Health Informatics (BHI) 2018 and the IEEE Conference on Body Sensor Networks (BSN), we are hosting a challenge to explore real clinical questions in critically ill patients using the MIMIC-III database. Participants in the challenge will be invited to present at the BHI & BSN Annual Conference in Las Vegas, USA (4-7 March 2018): https://bhi-bsn.embs.org/2018/

Dataset for the 2024 IEEE Video and Image Processing(VIP) Cup organized by Medical Image and Signal Processing Research Center, Isfahan University of Medical Sciences. The dataset was retreived from https://misp.mui.ac.ir/en/ieee-video-and-image-processing-cup on 21 February 2024. The license of the dataset published by the organizer is "Creative Commons Attribution NonCommercialShareAlike 4.0 International" as described on the official document. A mirror of the dataset has been uploaded here.

Official competition document containing the dataset description: https://2024.ieeeicip.org/wp-content/uploads/sites/579/2024-VIP-Cup-Competition-Official-Document_-Version-1.2-Jan-24-2024-FINAL-.g-link.pdf

The training dataset includes 124 volume OCT data from 124 subjects. Each volume contains about 30 to 300 B-scans of size 300x300. The OCT data have been captured from different subjects and categorized into three groups: i) Healthy (class 0), ii) Diabetic with DME (class 1), and iii) Non-diabetic patients with other ocular diseases (class 2).

The test dataset contains B-scans of 18 subjects, where each subject contains more than 70 and less than 300 B-scans. The B-scans are noisy with size 300 × 150 or 300 × 200 pixels. The ground truth classification result for the test dataset is not published by the VIP Cup organizers.

Data Set IEEE

## Overview

Data Set IEEE is a dataset for object detection tasks - it contains RETINOPATIA annotations for 470 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

IEEE Data Annotations

## Overview

IEEE Data Annotations is a dataset for object detection tasks - it contains Weeds annotations for 462 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Introduction

The 802.11 standard includes several management features and corresponding frame types. One of them are probe requests (PR). They are sent by mobile devices in the unassociated state to search the nearby area for existing wireless networks. The frame part of PRs consists of variable length fields called information elements (IE). IE fields represent the capabilities of a mobile device, such as data rates.
The dataset includes PRs collected in a controlled rural environment and in a semi-controlled indoor environment under different measurement scenarios.
It can be used for various use cases, e.g., analysing MAC randomization, determining the number of people in a given location at a given time or in different time periods, analysing trends in population movement (streets, shopping malls, etc.) in different time periods, etc.

Measurement setup

The system for collecting PRs consists of a Raspberry Pi 4 (RPi) with an additional WiFi dongle to capture Wi-Fi signal traffic in monitoring mode. Passive PR monitoring is performed by listening to 802.11 traffic and filtering out PR packets on a single WiFi channel.
The following information about each PR received is collected: MAC address, Supported data rates, extended supported rates, HT capabilities, extended capabilities, data under extended tag and vendor specific tag, interworking, VHT capabilities, RSSI, SSID and timestamp when PR was received.
The collected data was forwarded to a remote database via a secure VPN connection. A Python script was written using the Pyshark package for data collection, preprocessing and transmission.

Data preprocessing

The gateway collects PRs for each consecutive predefined scan interval (10 seconds). During this time interval, the data are preprocessed before being transmitted to the database.
For each detected PR in the scan interval, IEs fields are saved in the following JSON structure:
PR_IE_data =
{
'DATA_RTS': {'SUPP': DATA_supp , 'EXT': DATA_ext},
'HT_CAP': DATA_htcap,
'EXT_CAP': {'length': DATA_len, 'data': DATA_extcap},
'VHT_CAP': DATA_vhtcap,
'INTERWORKING': DATA_inter,
'EXT_TAG': {'ID_1': DATA_1_ext, 'ID_2': DATA_2_ext ...},
'VENDOR_SPEC': {VENDOR_1:{
'ID_1': DATA_1_vendor1,
'ID_2': DATA_2_vendor1
...},
VENDOR_2:{
'ID_1': DATA_1_vendor2,
'ID_2': DATA_2_vendor2
...}
...}
}

Supported data rates and extended supported rates are represented as arrays of values that encode information about the rates supported by a mobile device. The rest of the IEs data is represented in hexadecimal format. Vendor Specific Tag is structured differently than the other IEs. This field can contain multiple vendor IDs with multiple data IDs with corresponding data. Similarly, the extended tag can contain multiple data IDs with corresponding data.
Missing IE fields in the captured PR are not included in PR_IE_DATA.

When a new MAC address is detected in the current scan time interval, the data from PR is stored in the following structure:

{'MAC': MAC_address, 'SSIDs': [ SSID ], 'PROBE_REQs': [PR_data] },

where PR_data is structured as follows:
{
'TIME': [ DATA_time ],
'RSSI': [ DATA_rssi ],
'DATA': PR_IE_data
}.

This data structure allows storing only TOA and RSSI for all PRs originating from the same MAC address and containing the same PR_IE_data. All SSIDs from the same MAC address are also stored.
The data of the newly detected PR is compared with the already stored data of the same MAC in the current scan time interval.
If identical PR's IE data from the same MAC address is already stored, then only data for the keys TIME and RSSI are appended.
If no identical PR's IE data has yet been received from the same MAC address, then PR_data structure of the new PR for that MAC address is appended to PROBE_REQs key.
The preprocessing procedure is shown in Figure ./Figures/Preprocessing_procedure.png
At the end of each scan time interval, all processed data is sent to the database along with additional metadata about the collected data e.g. wireless gateway serial number and scan start and end timestamps. For an example of a single PR captured, see the ./Single_PR_capture_example.json file.

Environments description

We performed measurements in a controlled rural outdoor environment and in a semi-controlled indoor environment of the Jozef Stefan Institute.
See the Excel spreadsheet Measurement_informations.xlsx for a list of mobile devices tested.

Indoor environment

We used 3 RPi's for the acquisition of PRs in the Jozef Stefan Institute. They were placed indoors in the hallways as shown in the ./Figures/RPi_locations_JSI.png. Measurements were performed on weekend to minimize additional uncontrolled traffic from users' mobile devices. While there is some overlap in WiFi coverage between the devices at the location 2 and 3, the device at location 1 has no overlap with the other two devices.

Rural environment outdoors

The three RPi's used to collect PRs were placed at three different locations with non-overlapping WiFi coverage, as shown in ./Figures/RPi_locations_rural_env.png. Before starting the measurement campaign, all measured devices were turned off and the environment was checked for active WiFi devices. We did not detect any unknown active devices sending WiFi packets in the RPi's coverage area, so the deployment can be considered fully controlled.
All known WiFi enabled devices that were used to collect and send data to the database used a global MAC address, so they can be easily excluded in the preprocessing phase. MAC addresses of these devices can be found in the ./Measurement_informations.xlsx spreadsheet.
Note: The Huawei P20 device with ID 4.3 was not included in the test in this environment.

Scenarios description

We performed three different scenarios of measurements.

Individual device measurements

For each device, we collected PRs for one minute with the screen on, followed by PRs collected for one minute with the screen off. In the indoor environment the WiFi interfaces of the other devices not being tested were disabled. In rural environment other devices were turned off. Start and end timestamps of the recorded data for each device can be found in the ./Measurement_informations.xlsx spreadsheet under the Indoor environment of Jozef Stefan Institute sheet and the Rural environment sheet.

Three groups test

In this measurement scenario, the devices were divided into three groups. The first group contained devices from different manufacturers. The second group contained devices from only one manufacturer (Samsung). Half of the third group consisted of devices from the same manufacturer (Huawei), and the other half of devices from different manufacturers. The distribution of devices among the groups can be found in the ./Measurement_informations.xlsx spreadsheet.

The same data collection procedure was used for all three groups. Data for each group were collected in both environments at three different RPis locations, as shown in ./Figures/RPi_locations_JSI.png and ./Figures/RPi_locations_rural_env.png.
At each location, PRs were collected from each group for 10 minutes with the screen on. Then all three groups switched locations and the process was repeated. Thus, the dataset contains measurements from all three RPi locations of all three groups of devices in both measurement environments. The group movements and the timestamps for the start and end of the collection of PRs at each loacation can be found in spreadsheet ./Measurement_informations.xlsx.

One group test

In the last measurement scenario, all devices were grouped together. In rural evironement we first collected PRs for 10 minutes while the screen was on, and then for another 10 minutes while the screen was off. In indoor environment data were collected at first location with screens on for 10 minutes. Then all devices were moved to the location of the next RPi and PRs were collected for 5 minutes with the screen on and then for another 5 minutes with the screen off.

Folder structure

The root directory contains two files in JSON format for each of the environments where the measurements took place (Data_indoor_environment.json and Data_rural_environment.json). Both files contain collected PRs for the entire day that the measurements were taken (12:00 AM to 12:00 PM) to get a sense of the behaviour of the unknown devices in each environment. The spreadsheet ./Measurement_informations.xlsx. contains three sheets. Devices description contains general information about the tested devices, RPis, and the assigned group for each device. The sheets Indoor environment of Jozef Stefan Institute and Rural environment contain the corresponding timestamps for the start and end of each measurement scenario. For the scenario where the devices were divided into groups, additional information about the movements between locations is included. The location names are based on the RPi gateway ID and may differ from those on the figures showing the

This project involves scraping data from IEEE Xplore specifically in the field of machine learning (more than 50K data was scrapped). The collected data can be used to train a multi-label keyword prediction model. For details: https://github.com/NifulIslam/Multilabel-Keyword-Prediction

Dataset from IEEE 2014 data challenge, detailed information about the Data Challenge is included.

Ieee

## Overview

Ieee is a dataset for object detection tasks - it contains Plane annotations for 344 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

30 m

SNR levels

IEEE Dataset V3

## Overview

IEEE Dataset V3 is a dataset for object detection tasks - it contains Ducks And Pillars annotations for 5,100 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

ambient temperature

IEEE Project

## Overview

IEEE Project is a dataset for object detection tasks - it contains Objecttt annotations for 457 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

The dataset can be used to develop and test algorithms for communication and sensing in the 60GHz band. The dataset consists of synthetically generated indoor mm-wave channels between a MIMO transmitter and a MIMO receivers. Multiple targets are moving in the room. Number of targets, velocity of each target and trajectory are randomized across the dataset. The dataset contains also noisy received IEEE 802.11ay channel estimation fields. The dataset is suitable for development and testing of machine/deep learning algorithms. The dataset can be used to participate to the ITU AI/ML 5G Challenge. For information on the challenge and registration, please refer to: https://challenge.aiforgood.itu.int/match/matchitem/38. The challenge dataset relies on the open-source software available at: https://github.com/usnistgov/PS-002-WALDO.

IEEE 2

## Overview

IEEE 2 is a dataset for object detection tasks - it contains Rover annotations for 910 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

IEEE Classify

## Overview

IEEE Classify is a dataset for object detection tasks - it contains Civilian Military Helicopter annotations for 2,114 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Ieeeieee

## Overview

Ieeeieee is a dataset for object detection tasks - it contains E annotations for 700 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

%%%% 26/09/2025 - NOTE: This entry only includes the active measurements conducted in the following work.

The passive measurements can be found in https://doi.org/10.5281/zenodo.17208870

The present dataset is open-sourced with the paper "The Chronicles of 5G Non-Standalone: An Empirical Analysis of Performance and Service Evolution".

The paper is authored by Giuseppe Caso, Mohammad Rajiullah, Konstantinos Kousias, Usman Ali, Nadir Bouzar, Luca De Nardis, Anna Brunstrom, Ozgu Alay, Marco Neri, and Maria-Gabriella Di Benedetto

If you use the dataset for your own research activities and publications, please consider citing the paper as follows:

G. Caso et al., "The Chronicles of 5G Non-Standalone: An Empirical Analysis of Performance and Service Evolution," IEEE Open Journal of the Communications Society(IEEE OJ-COMS), pp. 1-21, 2024.

@article{caso2024chronicles,
title={{The Chronicles of 5G Non-Standalone: An Empirical Analysis of Performance and Service Evolution}},
author={Caso, G. and Rajiullah, M. and Kousias, K. and Ali, U. and Bouzar, N. and De Nardis, L. and Brunstrom, A. and Alay, O. and Neri, M. and Di Benedetto, M.-G.},
journal={IEEE Open Journal of the Communications Society},
pages={1--21},
year={2024},
publisher={IEEE}
}

A detailed description of the dataset is provided in the paper, and the README file provides additional details.

In the paper, an in-depth performance analysis of 5G performance is carried out, also exploiting a dataset previously collected and open-sourced at https://zenodo.org/records/8224890, and fully described in the following publication:

@article{kousias2023large,
title={{A Large-Scale Dataset of 4G, NB-IoT, and 5G Non-Standalone Network Measurements}},
author={Kousias, K. and Rajiullah, M. and Caso, G. and Ali, U. and Alay, O. and Brunstrom, A. and De Nardis, L. and Neri, M. and Di Benedetto, M.-G.},
journal={IEEE Communications Magazine},
volume = {62},
number = {5},
pages={44--49},
year={2024},
publisher={IEEE}
}

Contact Giuseppe Caso (giuseppe.caso@kau.se) for more information on the dataset(s) and potential access to additional data.