These datasets are part of Community Resource for Archiving Wireless Data (CRAWDAD). CRAWDAD began in 2004 at Dartmouth College as a place to share wireless network data with the research community. Its purpose was to enable access to data from real networks and real mobile users at a time when collecting such data was challenging and expensive. The archive has continued to grow since its inception, and starting in summer 2022 is being housed on IEEE DataPort.

Questions about CRAWDAD? See our CRAWDAD FAQ. Interested in submitting your dataset to the CRAWDAD collection? Get started, by submitting an Open Access Dataset.

The BLEBeacon dataset is a collection of Bluetooth Low Energy (BLE) advertisement packets/traces generated from BLE beacons carried by people following their daily routine inside a university building for a whole month. A network of Raspberry Pi 3 (RPi)-based edge devices were deployed inside a multi-floor facility continuously gathering BLE advertisement packets and storing them in a cloud-based environment.


The measurement results provided here are part of work on PhD thesis connected with measurement results variability reduction (main focus was on GSM/UMTS system. but other technologies were measured at the same time). All measurements were of indoor type. The duration of collecting data samples was 24h per day, with 10 seconds sampling interval. At some places it took one, two or four weeks to complete the measurements. Equipment used is the dosimeter (or exposimeter) EME Spy 140 (manufactured by Satimo).


MySignals dataset was collected by a network of approx. 10 mobile smartphone (iPhones) users via the MySignals iPhone App (www.mysignals.gr) for a period of approximately 8 months. MySignal App records the received signal strength indicator (RSSI), in dBm, of the mobile serving cell, as well as their own location, through the GPS module of their smartphone and other contextual information (timestamp, deviceID etc.). Measurements and relevant information (e.g.


Probe request frames

The sniffer was stationary. Data were obtained over 40 minutes. The sniffer arbitrates thru the 13 channels of the 802.11. The acquisition was done at the night with moderate traffic. Data was gathered from sniffing on WiFi networks in a public mall.

date/time of measurement start: 2021-06-04

date/time of measurement end: 2021-06-04



This dataset contains only the probe request frames.


Time- and frequency-variant 2.4 GHz ISM band channel gain. 

The time- and frequency-variant channel gain is measured in the presence of an industrial cyclic moving robot arm obstacle for four coexisting wireless nodes for the whole license-free 2.4-GHz-ISM band with a time- and frequency-resolution of 110 microseconds and 1 MHz, respectively. Results for two links are given.

date/time of measurement start: 2015-05-11

date/time of measurement end: 2015-05-11


Cause-Specific Episodes of Active Scanning. The dataset includes packet captures collected from controlled experiments with various devices. The dataset captures active scanning behavior of the devices.

Name of each folder represents the name of the cause of active scanning.

For details please refer to our papers - Learning to Rescue WiFi Networks from Unnecessary Active Scans, WoWMoM 2019.

data collection methodology: Controlled




Fingerprinting of wireless devices exploiting information leaked due to different device hardware compositions: Inter-Arrival-Time (IAT) of packets from wireless devices.


The preview of the road surface states is essential for improving the safety and the ride comfort of autonomous vehicles. This dataset consists of 1 million (240 x 360 pixels) road surface images captured under a wide range of road and weather conditions in China. The original pictures are acquired with a vehicle-mounted camera and then the patches containing only the road surface area are cropped. The images are classified into 27 categories, containing both the friction level, material, and unevenness properties.


This dataset includes the images, and instance segmentation masks to evaluate panoptic segmentation in the food domain. 


We present below a sample dataset collected using our framework for synthetic data collection that is efficient in terms of time taken to collect and annotate data, and which makes use of free and open source software tools and 3D assets. Our approach provides a large number of systematic variations in synthetic image generation parameters. The approach is highly effective, resulting in a deep learning model with a top-1 accuracy of 72% on the ObjectNet data, which is a new state-of-the-art result.