This is a non-IMPACT record, meaning that access to the data is not controlled by IMPACT. For access, see the directions below.

This Resource is offered and provided outside of the IMPACT mediation framework. IMPACT and the IMPACT Coordination Council/Blackfire Technology, Inc. expressly disclaim all conditions, representations and warranties including but not limited to Resource availability, quality, accuracy, non-infringement, and non-interference. All Resource information and access is controlled by entities and under terms that are external to the IMPACT legal framework.


The kyutech/interference dataset
External Dataset
External Data Source
Crawdad (Dartmouth)
56 (lowest rank is 56)

Category & Restrictions

application layer security, network data, local networks, wireless


Measurements of HTTP requests over 802.11 in dense wireless classrooms

The common wisdom tells one not to attach more than 5-10 wireless devices to an access point. This is a problem in wireless educational classes, one can get between 10 and 50 students in the same room. Such spaces are referred to as dense wireless networks. While there are several methods that can be used to deal with dense wireless environments, installing multiple APs and separating them by channel is arguably the most readily available method today. In 802.11g/n the practical advice is to assign CH1, CH5 and CH9 to the three APs. However, even in such environments, wireless channels can become congested. It was noticed that tablet terminals would often fail to complete HTTP requests or fail getting access to the wireless channel entirely. In order to deal with this reliability problem, educational webapps had to be updated in such a way that they would detect failed requests and repeat them multiple times until a valid reply comes back. This dataset contains measurements which can help model such environments and the related application logic. The dataset has slightly non-trivial metrics which described failed/repeated. The experiment was run over multiple parameters and with on-the-fly changes in the layout. The measurement itself was simple -- a web application (in browser) on user terminal would send a GET request to the web server and would download bulk of a randomly selected size. All the performance metrics were recorded at user terminals and gradually uploaded to the server -- specifically, each new request would also contain the variables containing the results of the previous request, which had negligible effect on size but facilitated efficient log-keeping.


Additional Details

dataset, kyutech, interference, the kyutech/interference dataset, 995, corporation, inferlink, external, external data source, source, inferlink corporation, wireless, dense, requests, measurements, http, 802, classrooms, request, multiple, environments, aps, application, user, fail, metrics, web, channel, access, failed, server, size, educational, deal, terminals, parameters, updated, trivial, selected, spaces, reply, channels, send, referred, arguably, performance, separating, layout, terminal, previous, practical, browser, repeat, common, times, wisdom, installing, detect, classes, advice, assign, download, repeated, reliability, 11g, methods, log, other, gmail, bulk, webapps, congested, maratishe, devices, method, negligible, model, experiment, gradually, tablet, students, networks, simple, randomly, complete, logic, ch9, fly, ch5, tells, ch1, noticed, keeping, efficient, variables, attach, uploaded, valid, measurement, facilitated