TP3.5: Proximity Services

Subproject manager
Prof. Dr. Jörg Ott

Proximity services are a form of services, which are not only produced based on the user’s location (like location-based services, LBS), but also based on the relative distance between two or more users or between a user and an object. They can be understood as LBS with a relative frame of reference. The goal of this subproject is to create a technical platform for proximity services and to realize a selection of services based on this platform. We provide decentralized mechanisms (device to device, local infrastructure) for the detection of physical proximity to fulfill the following key attributes: discovers things relevant to you, senses your environment, filters things relevant to you, knows what’s around you and interacts with your surroundings. The challenges in this field are multifaceted. Always-on services like proximity detection require continuous discovery, which drains the battery. This is especially a problem for mobile clients with limited battery power. In addition, user privacy is a challenge because the automatic recognition of nearby users allows continuous location tracking. Besides that, proprietary platforms lead to mobile app silos, which is a major problem for system interoperability.

Managing IoT at the Edge: The Case for BLE Beacons
Managing IoT devices in urban areas is becoming crucial because the majority of people living in cities and the number of deployed IoT devices are steadily increasing. In this paper we present iConfig, an edge-driven platform dedicated to manage IoT devices in smart cities. The goal is to address three major issues in current IoT management: registration, configuration, and maintenance. The core of iConfig is its programmable edge module, which can be deployed across smartphones, wearables, and smart boards to configure and interact with physically proximate IoT devices. Through testbed experiments and usability studies, we reveal the hardship and hidden pitfalls in managing IoT devices, especially for low budget devices like Bluetooth Low Energy (BLE) beacons. Our system evaluation shows that iConfig can effectively address the aforementioned IoT management challenges by harnessing the mobile and edge cooperation. To inspire community contributions, we further present concrete use cases to illustrate how iConfig can reduce operational cost and facilitate future edge-centric IoT research.

The key contributions are summarized as follows:
– We analyzed and identified key properties of IoT device management that must be attained to manage large scale deployments in smart cities.
– We designed and implemented iConfig, an edge-driven platform dedicated for IoT device management. We demonstrate the efficacy of iConfig via prototype implementations, which target at BLE beacons without backend connectivity.
– Our usability study and testbed experimentsfurther uncover hidden aspects in IoT management that are important and deserve future research.

Proximity Reasoning via Multimodal Context Fusion
Another prototype aims at close proximity detection of a few meters, where the users or devices share the same content regarding Wi-Fi signals and ambient sound. Figure 1 shows the GUI of our prototype. Initially, a Wi-Fi Direct service broadcasts the device’s available resources regarding CPU and battery. Afterwards, a metric choose one mobile client as temporary server, this device requests the required data from the clients and calculates whether the devices share the same environment. Therefore, each client sends the feature vectors of the Wi-Fi signals including received signal strength. Moreover, the device records ambient sound at a specific frequency and calculates locally sound vectors before releasing them to the temporary server to preserve user privacy. Finally, the users can send messages between group members. The groups are automatically generated by reasoning based on sensor data, which reflects the current device environment.

Connected Mobility Ecosystem Explorer
Figure 1: Proximity Detector

Enabling Practical Distance-bounding IoT Services via Visible Light Communication
Enforcing distance boundary is a highly desired attribute for various Internet of Things (IoT) services due to the compelling concern on security and privacy in smart homes and buildings where IoT devices will be deployed in a pervasive manner. In theory, such distance boundary could be achieved by combining dedicated system design and communication technologies. Yet, in practice, we rarely find IoT services that can fully benefit from this safeguard feature. This is mainly due to the complexity of exploiting various wireless communication technologies to attain distance limit in different IoT environments.

We present LocalVLC, a ready-to-deploy platform that takes advantage of visible light communication (VLC) to reinforce spatial barrier control on IoT services. Different from radio-based design which entails barrier penetration and computation overhead to determine distance-bounding threshold, LocalVLC introduces a lightweight Morse data encoding adjusted for VLC to deliver fine-grained and low-cost distance boundary control. To demonstrate the practicality of LocalVLC, we have implemented a full-fledged platform prototype and evaluated it against multiple demanding scenarios in our testbed. We find thatLocalVLC can support up to 10 meters of range under a single light source, and attain reasonable throughput and error rate in practical settings. We further developed three LocalVLC-based service prototypes, including distance-bounding authorization, indoor communication hub, and localized service discovery to illustrate how we can utilize LocalVLC to enable secure and flexible distance-bounding IoT services.

To summarize, our work makes the following contributions:
– We investigate VLC for distance-bounding services by taking advantage of its unique property of distance restricted communication path. In contrast, mid-range radio-based communications are only able to fulfill the same requirement by adding overhead to estimate a distance boundary threshold.
– We present LocalVLC, a ready-to-deploy platform that takes advantage of VLC to reinforce spatial barrier control on IoT services. LocalVLC introduces a lightweight Morse data encoding to achieve a fine-grained and low-cost distance boundary control.
– We have implemented three service prototypes, including distance-bounding authorization for smart homes, indoor communication hub, and localized service discovery. Those exemplary services demonstrate how we can utilize LocalVLC to enable secure and flexible distance-bounding IoT services.

Figure 2: openVLC platform with omnidirectional LED for transmission and low power LED which can be used as transmitter or receiver. The photodiode acting as receiver is highlighted in red.

Figure 3: International Morse code, the circle refers to a dot and the rectangle represents a dash

Figure 4: Process a raw light signal to recognize letters, words, and messages

Please contact Prof. Dr. Jörg Ott or Michael Haus, if you would like to get more information about this subproject, our work in general or if you are interested in a cooperation to develop proximitx-based services. Motivated students who are looking for a thesis or a guided research in this field are always invited to contact us as well.

Own Publications
[1] Michael Haus, Aaron Yi Ding, Chenren Xu, Jörg Ott, Demo: Touchless Wireless Authentication via LocalVLC, Proceedings of the 16th ACM International Conference on Mobile Systems, Applications, and Services (MobiSys), 2018
[2] Michael Haus, Aaron Yi Ding, Pan Hui, Jörg Ott, Demo: iConfig: What I See Is What I Configure, 12th Workshop on Challenged Networks (CHANTS), MobiCom Konferenz, 2017
[3] Michael Haus, Aaron Yi Ding, Jörg Ott, Managing IoT at the Edge: The Case for BLE Beacons, 3rd Workshop on Experiences with the Design and Implementation of Smart Objects, MobiCom Konferenz, 2017
[4] Sasan Amini, Kristian Beckers, Markus Böhm, Fritz Busch, Nihan Celikkaya, Vittorio Cozzolino, Anne Faber, Michael Haus, Dominik Huth. Alfons Kemper, Andreas Kipf, Helmut Krcmar, Florian Matthes, Jörg Ott, Christian Prehofer, Alexander Pretschner, Ömer Uludağ, Wolfgang Wörndl, Informatikforschung für digitale Mobilitätsplattformen: Am Beispiel des TUM Living Lab Connected MobilityInformatik-Spektrum, 2017
[5] Michael Haus, Muhammad Waqas, Aaron Yi Ding, Yong Li, Sasu Tarkoma, Jörg Ott, Security and Privacy in Device-to-Device (D2D) Communication: A Review, IEEE Communications Surveys & Tutorials, 2017
[6] Michael Haus, Vittorio Cozzolino, Aaron Yi Ding, Jörg Ott, P2Hub Private Personal Data Hub for Mobile Devices: Poster, 17th ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc), 2016
[7] Michael Haus, System approach towards private proximity services, ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp), 2016