Modern Building Automation Systems (BAS) consist of sensors and actuators that are connected via an IP-based network and offer their functionality via RESTful APIs. Because a single device can be exploited by an attacker to perform attacks within the local network, we put devices into isolated groups. These groups are isolated MAC-layer Trust Zones to reduce the attack surface in contrast to a BAS with fully connected devices. We propose an algorithm that leverages the so far neglected potential of Building Information Modeling (BIM) to compute Trust Zones. We assure unimpaired operation of all applications while limiting the number of infrastructure devices. The proposed mechanisms are demonstrated considering sensors and actuators that are connected via wired Ethernet and the IEEE 802.11s WLAN mesh standard. At the application layer we make exemplary use of the Constrained Application Protocol (CoAP). Finally, we experimentally evaluate the device acquisition and selection based on our network partitioning algorithm.

Due to the fact that every smart building is a unique composition of devices, services, and users there is no one-fits-all security architecture. With our proposal we address this problem by using Building Information Modeling (BIM) as source of information for our security concept. Using BIM, models are created throughout the planning of a building, which do not consist only of 3D information such as typical CAD drawings, but also provide a rich set of metadata about the building elements including devices, their interrelationships and networking. Thus, BIM can be used to derive security configurations for each individual embedded device, its services and the network as a whole. We demonstrate the function of a general Security Controller to partition the network on MAC layer. Furthermore, we draw an overall picture of the proposed architecture and the range of possible applications.

CoAP (Constrained Application Protocol) enables embedded devices to offer RESTful Web Services and exchange small binary message headers. Next to the transmission of sensor data and control information between smart home devices, there is an application field of streaming data using CoAP. There are CoAP implementations available in Java, that allow an execution on heterogeneous devices and operating systems due to the Java Virtual Machine and Java Runtime Environment. Because of the timing demands of live streaming applications in terms of low latency communication at application layer, we evaluate two different CoAP Java implementations. We compare the timing fluctuations of the packet processing with a CoAP C implementation. As a conclusion we identify the Java Garbage Collector to cause large fluctuations of the packet processing, which results in bad suitability for a live streaming scenario. Furthermore, we recommend the usage of native code applications because of better timing results.

In smart building scenarios there are a lot of vulnerable devices that could be exploited to run attacks against other devices within the same LAN. Even though existing solutions mostly tackle the problem by cluster-based authentication and key management schemes, none of them leverages the potential of isolating traffic by network interface virtualization. Thus, we proposed in a previous work a concept to avoid unauthorized communication by considering separating applications with virtual MAC interfaces as the consequence. The decreased attack surface, as the main advantage, is achieved by isolating communication through virtual MAC interfaces based on application-specific demands. To demonstrate the efficiency of this concept we developed an implementation based on state-of-the-art communication protocols. We applied our interface virtualization concept to the IEEE 802.11s WLAN mesh technology, combining it with a lightweight RESTful web service for security credentials deployment. The resulting proof-of-concept implementation in a real-world multi-hop scenario shows performance of the credentials deployment and the impact of the MAC-layer parallelization. The promising results, e. g., no drop of the overall throughput using multiple virtual MAC interfaces, show that our concept can be an efficient solution for future smart buildings.

In Smart Buildings there is a large number of connected devices. Each of them is possibly vulnerable, so that an attacker could make use of a single smart device to run attacks in the Smart Building network. We propose a concept to partition the network into trust zones depending on the application layer, so that devices that are logically linked on application layer are able to communicate on MAC layer. A trusted device is used to bootstrap new devices and reconfigure existing trust zones. We restrict the communication as far as possible, so that the potential damage caused by a compromised device is limited. In principle, malicious behavior of devices could lead to an exclusion on MAC layer. The general concept is described using a typical IoT protocol stack containing IEEE 802.11s, IP, UDP/DTLS, and CoAP.

Current Smart Home systems are often based on proprietary security solutions which hide its inner functionality and possible vulnerabilities until they are exploited. Furthermore, an initial trusted instance in Smart Home networks is necessary to provide a seamless trust relationship to Smart Home appliances. Another issue is the stalling roll-out of Smart Metering systems due to its bad price-performance ratio for the customer. Therefore, we provide the lightweight security framework SafeBase, which solves both issues and provides a possible solution for a future Smart Home and Smart Metering systems. To achieve this goal, we employ state-of-the-art protocols like Constrained Application Protocol (CoAP), Object Security for Constrained RESTful Environments (OSCORE), Authentication and Authorization for Constrained Environments (ACE) and LwM2M.

Web Services can be used as a communication structure for embedded devices. Out of numerous Web Service specifications there is a subset (Profile) to implement Web Services on resource constrained devices, called Devices Profile for Web Services (DPWS). The resulting service oriented architecture enables the user to discover new devices, e.g., in the local smart home network. The open standards for Web Services further define a metadata format for service description. Besides the simple invocation of service operations, an eventing mechanism is specified. A client which subscribes to an event will receive notifications. Those features and the inherent Plug&Play capability provided by Web Services are suitable to connect smart home devices in a user-friendly manner. However, DPWS specifies no standard procedure to combine multiple devices with each other to build more complex applications. Therefore, new concepts for embedded Web Service orchestration are needed. Some commercial solutions use a central hub, which represents a Single Point of Failure (SPoF). Hence, a failure would lead to a breakdown of all smart Internet of Things (IoT) applications and decrease the user acceptance. We propose a concept where no central broker is needed by defining a Configuration Service that runs on multiple devices. Based on that service a smartphone app is used to establish trigger-action rules between devices. A prototype is implemented as a proof-of-concept for different smart home devices. Furthermore, a mobile Android application to find and orchestrate the devices is presented.

The new standard amendment IEEE 802.11s enables low-level interoperability for future WLAN mesh networks. Support of the Hybrid Wireless Mesh Protocol (HWMP) and the Airtime Link Metric (ALM) for MAC-layer routing is mandatory. Its default distance vector routing mode facilitates scalability but also results in a limited network view per mesh node. Moreover, mesh mechanisms operate transparently to higher layers which makes the management and optimization of 802.11s networks a challenging task. Available on every standard-compliant node, ALM offers the potential to derive mesh topology information. We present AKadeMesh (Adaptive Kad-enhanced Mesh), a cross-layer approach specifically designed for 802.11s networks. It is based on the P2P protocol Kad and dynamically adapts its logical overlay to the physical mesh underlay by directly considering ALM. The resulting topology-aware P2P overlay is used to realize logical clustering for the distributed management of 802.11s networks, thereby maintaining unrestrained interoperability to the mesh standard. Our solution was implemented and evaluated in a real-world test bed. Results demonstrate its practical feasibility and verify the expected clustering benefit.

In this paper, a prototype to synchronize the Peer-to-Peer network Kad is presented. The approach bases on a deterministic algorithm, which is required for hard real-time applications, to synchronize the network in a decentralized manner. Existing industrial Ethernet solutions include and support real-time capable machine-to-machine communication in automation scenarios. However, the coordination of the communication and the synchronization to achieve hard real-time are managed by a central instance, leading to deficient resilience and scalability. Additionally, a centralized or hierarchical synchronization is required to enable the communication per se. The presented decentralized synchronization instead benefits from nodes helping to synchronize the network and bases on the decentralized structured Peer-to-Peer network Kad. However, the higher the number of helping nodes the higher is the time deviation on the nodes of the network, which contrary results in a higher time error. Therefore, a trade-off between synchronization performance and time error has to be determined to meet predefined constraints depending on the application scenario. Moreover, the individual clock drift of every device needs to be considered to define necessary re-synchronization intervals of the network. Theoretical values are compared to the measured values determined in experiments with 15 nodes. As a result of these experiments, the measured synchronization performance, which is the time needed to synchronize the complete network, confirms the theoretical worst case values. The result is a running prototype system, which is able to synchronize the network with a high performance in a decentralized manner.