Networking environments can be extremely complex and the interaction of different layers and technologies may create situations that cannot be foreseen during the design and testing stages of technology development. It is therefore critical to perform comprehensive empirical studies in a wide range of production environments to uncover deficiencies and identify possible optimizations and extensions. The availability of high-quality measurement and modeling studies would make it possible to develop wireless networks that are more robust, easier to manage and scale, and able to utilize scarce resources more efficiently.

Measurement-based data collected from diverse (wireless) networking environments, such as metropolitan areas, vehicular, houses, academic environments, research labs, and conference sites, have been made available in various data repositories. One of the largest collections with publicly available wireless traces is CRAWDAD which hosts traces from many different wireless environments. Tables B.1, B.2, and B.3 summarize the type of wireless traces available in CRAWDAD.

A comprehensive list of traffic traces and archives can also be found via the MOME Database. When you are interested in an overview of other trace repositories then just open the search page and search for DataType="WebRepository".

Further archives are:

  • the UCSD wireless topology discovery trace [56, 260]
  • the MIT Roofnet [43]
  • the MobiLib [30]
  • wireless LAN traces from the ACM Sigcomm’01 [55]
  • vehicular network traces [99, 153, 156, 95]

Empirical studies focusing on metropolitan area-based wireless networks have recently taken place:

  • in Cambridge, UK with users currying iMotes [241]
  • in Toronto with Bluetooth-enabled PDA users walking in the subway and malls to test if a worm outbreak is possible in practice [335]
  • at MIT, with one hundred smart phones that use both short-range (such as Bluetooth) and long-range (GSM) networks logging users’ location, communication, and device usage behavior information [133]
  • in Cambridge, US, with users of Roofnet, an experimental ieee802.11 b/g mesh network which provides broadband Internet access, developed at MIT CSAIL [43]
  • in a grid of six nodes placed within three different houses that produced wireless measurements to characterize connectivity and  udp and tcp  throughput [291, 361]
  • in Oulu, Finland, panOULU network provides in its coverage area wireless

The Roofnet measurements focused on the link-level of ieee802.11, finding high-throughput routes in the face of lossy links, adaptive bit-rate selection, and developing new protocols which take advantage of wireless communications’ unique properties [85, 87]. Empirical measurement studies were also performed in several conferences, such as: the 2005 IETF meeting [205], 2004 ACM Sigcomm[317], and 2001 ACM Sigcomm[55], in which snmp, and syslog traces were acquired from the deployed ieee802.11 APs.

Traces from large-scale academic deployments of ieee802.11 APs include UNC [51], Dartmouth [10], USC [30], and smaller-scale ieee802.11 APs networks in research labs or institutes, such as FORTH [51], and IBM [75, 76].

Sensor-based testbeds include the one at Columbia University using TinyOS on Mica2 motes for testing a mac protocol [135, 136].  Vehicular-based networking environments have also been explored [153, 95]. For example, [153] includes traces from a short-range communications between vehicle and roadside traffic and [95], from an ieee802.11-enabled bus in a campus and the surrounding county in UMASS. Traces from a cdma 1x EV-DO network are also available [152]. Finally, a very large collection of data tailored to measuring Internet traffic and performance can be found at the CAIDA site [11].


Extensive traces from large-scale campus-wide wireless networks (e.g. syslog, snmp, TCP flow, and signal strength based data), monitoring tools, models, and synthetic traces based on various models are publicly available to the community and can be found at

Traces Tables