CRAWDAD
CRAWDAD is a community-maintained archive and directory of real‑world wireless network data, traces, and related datasets (e.g., pcap files). It aggregates and catalogs datasets from multiple contributors, enabling reproducible research in wireless and mobile networking, and exemplifying an awesome-style meta collection focused on wireless data.
About this tool
CRAWDAD – Wireless Network Data Archive
Category: Themed Directories
Slug: crawdad
Brand: Dartmouth College
Source: https://ieee-dataport.org/open-access/crawdad-duepacket-delivery
CRAWDAD is a community-maintained archive and directory of real‑world wireless network data, traces, and related datasets (such as pcap files). It catalogs datasets from multiple contributors to support reproducible research in wireless and mobile networking.
Overview of This Dataset: due/packet-delivery
Wireless sensor network (WSN) link performance dataset collected indoors over 6 months, covering:
- ~50,000 parameter configurations for 7 key protocol stack parameters
- Meta‑data for nearly 200 million transmitted packets
- 4 major performance metrics: energy, throughput, delay, loss
Key metadata:
- Nickname:
packet-delivery - Institution: University of Duisburg-Essen and Norwegian University of Science and Technology
- Measurement period: 2012‑11‑01 to 2013‑11‑30
- Release date: 2015‑03‑30
- Last modified: 2015‑03‑30
- DOI: https://dx.doi.org/10.15783/C7NP4Z
Environment & Methodology
Collection Environment
- Location: Long hallway in a university building
- Dimensions: 2 m × 40 m
- Node mounting height: 0.7 m (wooden stands)
- Line-of-sight maintained between two motes
- Sender–receiver distances: 10, 15, 20, 25, 30, 35 m
- Human presence: Students and employees may walk in the hallway during experiments
Network Configuration Parameters
For each experiment, the following parameters are varied:
- Packet inter‑arrival time (ms): 10, 15, 20, 25, 30, 35, 40, 50
- Packet payload size (bytes): 20, 35, 50, 65, 80, 95, 110
- Maximum queue size (packets): 1, 30, 60
- Max_tries (max transmissions): 1, 3, 5
- Retry delay (ms): 30, 60
- Transmission power level: 3, 7, 11, 15, 19, 23, 27, 31
- Distance (m): 10, 15, 20, 25, 30, 35
Data Collection Methodology
- Metadata of each packet recorded at sender and receiver nodes
- Data forwarded via USB cables to a laptop
- Contains no sensitive or personal information → no sanitization performed
Traceset: due/packet-delivery/delay
Per‑packet delay measurements under various stack parameter configurations.
General info:
- Files:
delay_10m_12runs.rardelay_15m_11runs.rardelay_20m_10runs.rardelay_25m_9runs.rardelay_30m_11runs.rardelay_35m_9runs.rar
- Content: 6 traces (one per distance), each with delay measurements for ~a dozen runs
- Delay definition: Time between packet generation at sender and packet reception at receiver
- Measurement purpose: Network performance analysis
- Dataname:
due/packet-delivery/delay - Version:
20150401(initial version) - Last modified / Release date: 2015‑03‑30
- Nickname:
delay
Example Trace: 10meter
- Content: Delay measurements of all packets transmitted at 10 m
- Format:
- Each file has 8064 lines
- Each line corresponds to one stack parameter configuration
- Line structure:
period | packet length | queue_size | max_retries | retry_delay | TxP_level | distance | average_SNR | delay_run1_pkt1 | ... | delay_run1_pkt300 | delay_run2_pkt1 | ... | delay_run2_pkt300 | ... | delay_runN_pkt1 | ... | delay_runN_pkt300
- Special values:
9999or1111indicate packets not received or delay measurement errors
Features
- Community‑maintained directory for wireless and mobile networking datasets
- Provides real‑world wireless network traces and related data (including per‑packet metadata)
- Supports reproducible research via well‑documented experimental setups and parameter sweeps
- Includes large‑scale WSN link measurements:
- Multi‑month measurement campaigns
- Tens of thousands of parameter configurations
- Hundreds of millions of packets
- Captures key performance metrics for wireless links:
- Energy, throughput, delay, packet loss
- Offers structured trace files with detailed parameter fields for each configuration
- Includes explicit experiment context: topology, environment, distances, human presence
- Uses open research data practices (DOI, versioning, descriptive metadata)
Pricing
Not specified. The referenced dataset is listed as open access on IEEE Dataport.
Loading more......
Information
Categories
Tags
Similar Products
4 result(s)Large-scale web crawl dataset containing 3.5 billion web pages from CommonCrawl (2012), suitable for web mining, search, and network analysis research. Listed as part of an awesome-style collection of computer networks datasets.
Clickstream dataset with 53.5 billion web clicks from 100,000 anonymized users at Indiana University, useful for studying browsing behavior, recommendation, and network traffic patterns; included in an awesome curated datasets list.
An Awesome-style collection of short, easy-to-understand JavaScript code snippets you can grasp in 30 seconds.
A GitHub repository by Brad Traversy containing 50+ small, focused web development mini projects built with HTML, CSS, and JavaScript, useful as a curated collection of example projects for learning or referencing in awesome-style directories.