Sven Zehl

ResFi: A secure framework for self organized Radio Resource Management in residential WiFi networks

In dense deployments of residential WiFi networks individual users suffer performance degradation due to both contention and interference. While Radio Resource Management (RRM) is known to mitigate this effects its application in residential WiFi networks being by nature unplanned and individually managed creates a big challenge. We propose ResFi - a framework supporting creation of RRM functionality in legacy deployments. The radio interfaces are used for efficient discovery of adjacent APs and as a side-channel to establish a secure communication among the individual Access Point Management Applications within a neighborhood over the wired Internet backbone. We have implemented a prototype of ResFi and studied its performance in our testbed. As a showcase we have implemented various RRM applications among others a distributed channel assignment algorithm using ResFi. ResFi is provided to the community as open source.

LoWS: A complete Open Source solution for Wi-Fi beacon stuffing based Location-based Services

Omnipresent Wi-Fi access points (APs) periodically broadcast beacon frames to inform potential stations (STAs) about their existence. Beacon frames can be extended by adding additional information (so-called beacon stuffing) making it possible to deliver this information to mobile devices (smart-phones, tablets, etc. equipped with 802.11 interfaces) without the need for association with the local infrastructure. This feature can be used to support location-based information services (LBS) related for e.g. advertising local opportunities, temporary obstacles and traffic disturbances or emergency notifications. In this paper we discuss technical solutions that support LBS using 802.11 beacon stuffing. For a selected approach we present the design, implementation details and performance evaluation for Location-based Wi-Fi Services (LoWS) - a complete open source solution. LoWS supports an original, highly efficient way to encode information. The LoWS system can be integrated in an existent 802.11 infrastructure while the receiver application can be installed on commercial off-the-shelf Android devices. The LoWS system is modular with well-defined interfaces making it an attractive tool for modifications, changes and improvements.

Practical distributed channel assignment in home Wi-Fi networks

The efficient management of radio resources in todays home or residential Wi-Fi networks is still an open research question. Due to the chaotic and unplanned deployment of Wi-Fi Access Points (APs) and the fact that all APs are managed individually by their owners, home Wi-Fi networks suffer from performance degradation due to contention and interference. In this paper we present and showcase a distributed radio channel assignment scheme, implemented using the ResFi platform, in which neighboring home Wi-Fi APs cooperate with each other in order to negotiate radio channel selection by taking into account the instantaneous network load of neighboring APs (two hop neighborhood). While the information about the network load can be directly exchanged between ResFi-enabled APs using the ResFi out-of-band Internet control channel, the information about the network load of co-located non-ResFi APs first needs to be estimated passively by monitoring the data traffic on the radio channel. The presented demonstrator of the approach is implemented using commodity hardware and enables the audience to observe the distributed channel adaptation in real-time.

SENSEFUL: An SDN-based joint access and backhaul coordination for Dense Wi-Fi Small Cells

Dense Small Cell networks are considered the most effective way to cope with the exponential increase in mobile traffic demand expected for the upcoming years and are one of the foundations of the future 5G. However, novel architectures are required to enable cost-efficient deployments of very dense outdoor Small Cell networks, complementing the coverage layer provided by macro-cells. In this regard, two important challenges need to be solved to make this vision a reality: i) increased traffic dynamics, which are translated into more frequent handovers, and ii) cost-efficient deployment of large number of Small Cells. In this paper we propose and evaluate SENSEFUL, an novel architecture addressing the two problems highlighted above: Software-Defined Networking (SDN) as the key technology to promote adaptability to a varying environment and provide efficient mobility solutions in the dense access layer, and novel wireless backhauling technologies where traditional wired connectivity does not meet cost/efficiency restrictions.

Demo: Stuffing Wi-Fi Beacons for Fun and Profit

Ubiquitous Wi-Fi access points (APs) constantly send out beacon and probe response frames to inform potential stations (STAs) about their existence. Beacon and probe responses can be extended by adding additional information (so-called beacon stuffing) making it possible to deliver this information to mobile devices (e.g. smart-phones equipped with 802.11 interfaces) without the need for association with the local infrastructure. This feature can be used to support location-based information services (LBS) related for e.g. advertising local opportunities, temporary obstacles and traffic disturbances or emergency notifications. In this paper we introduce a demonstrator for Location-based Wi-Fi Services (LoWS) based on beacon stuffing. The LoWS system can be easily installed within an existent 802.11 infrastructure while the receiver application can be installed on nearly all up-to-date commercial off-the-shelf smartphones, e.g. based on Android or iOS.

NxWLAN: Towards transparent and secure usage of neighbors' access points in residential WLANs

The increased popularity of IEEE 802.11 LANs (WLANs) in residential (home) environments leads to dense, unplanned and thus chaotic deployments. Access Points (APs) are frequently deployed unfavorably in terms of radio coverage and use interfering frequency/power settings. In fact, in some parts of a ones apartment the usage of the neighbors AP might be preferred as it might provide better signal quality than the own AP. Moreover, the network performance could be dramatically improved by balancing the network load over spatially co-located neighbor APs operating on different radio channels or having asymmetric workloads. We address these problems by presenting NxWLAN (Neighborhood extensible WLAN) which enables the secure virtual extension of users home WLANs in residential environments through enabling the usage of neighboring APs operating on the same or on different radio channels. NxWLAN requires no additional software to be installed on the client STAs while providing the same level of security as the home AP, e.g. WPA2, without revealing any kind of security credentials to untrusted neighboring APs. In this paper we prove, by prototyping and simulating the NxWLAN solution, that it is operational, performant and easy to deploy using off-the-shelf hardware. Moreover, we provide the full source code of our prototype to the community as open-source.

On practical selective jamming of Bluetooth Low Energy advertising

Bluetooth Low Energy (BLE) Advertising is an important component of BLE as it is used to broadcast connection information to other devices nearby, usually smartphones. Furthermore, it is the foundation of a variety of services, such as indoor localization. We present a selective jammer against BLE beacons. The jammer selectively jams only BLE advertising beacons for which it was configured (API) on just the BLE advertising channels actually being used (narrowband). Such a jammer is hard to detect and hence requires a sophisticated Intrusion Detection System (IDS). The prototypic implementation uses off-the-shelf embedded hardware which is low-cost, small-size and very power efficient. Experiment results demonstrate the feasibility of the proposed solution and reveal the impact of the distance between jammer and receiver on the advertising success rate. At short distances the jammer is able to successfully corrupt all BLE advertisement beacon frames making the BLE beacon source undiscoverable.