Mid Sweden University

This paper describes research issues and work-in-progress concerning ubiquitous sensing. We present scenarios where the current approaches are deficient in addressing the needs for ubiquitous sensing in services and applications on the Future Internet, involving the massive sharing of information from sensors via heterogeneous networks. We propose an information-centric architecture for real-time ubiquitous sensing which capitalizes on the proposed locator/identifier split, thus extending the Network of Information (NetInf) approach. From this we identify the challenges for which we present work-in-progress within the framework of the EU-funded MediaSense project. Firstly, we integrate sensors as addressable objects, exposed by means of sensor gateways and relocatable abstract interfaces. Sensor information is thus made available to applications solely based on identity. Secondly, sensor information is made available in a distributed data model towards searching and browsing. Finally, we evaluate the effectiveness of the architecture in proof-of-concept applications for intelligent commuting, environmental monitoring and seamless media transfer, utilizing two different sensor platforms.

There are more sensors connected to the Internet today and corporations predict that there will be billions of sensors within the decade. With billions of sensors, there is an opportunity for new behavior from applications in mobile devices based on sensor input. But such applications require a mechanism to provide a manageable subset of sensors, relevant to the user's current situation; since no single device is capable of maintaining a list of billion sensors. By using a context enabled distributed storage of sensors; we can provide communication with a finite set of nodes that are deemed relevant to the user's current context.We introduce the notion of a sensor socket, that share API similarities to a TCP/IP socket, which provides communication with remote devices. We identify requirements on the grouping mechanism concerning connectivity to ensure reliable communication decoupled from location. We then provide a scenario for how sensor sockets could enable device connectivity. Finally, we demonstrate a subset of the scenario, with a proof of concept, by presenting ad-hoc association of a printing resource with a mobile device connected to the Internet via mobile broadband

The Internet-of-Things will require ubiquitous informationsharing between connected things on a global scale, whichexisting systems do not offer. Most current efforts focus on solutionsfor information dissemination, which induce single pointsof failure and introduce unnecessary communication delays. Tothis end we propose the SensibleThings platform, which is afully distributed open source architecture for Internet-of-Thingsbased applications. This article describes the major problems thatInternet-of-Things platforms must address, our technical solutionto these problems, and an evaluation thereof. We also present thecurrent progress and a series of demonstrators, which show thewide range of applications enabled by the platform. Finally, wepresent how the platform will be used in our future research andpotential spin off companies.

The Internet of Things is a research topic, where billions of things, both sensing and actuating, connect to the Internet, interact, exchange information and enable intelligent services. Current resource consumption models are adapted to information distribution where the cost of information gathering is neglected. However, the emerging Internet of Things will contain billions of sensors that are capable of producing information in real-time, which both demand more resources, and is of interest to fewer users. In this paper, we present an analytical power model for both information collection and distribution in order to find the total power consumption of the Internet of Things. We apply the model to both a centralized and a distributed scenario where billions of devices produce and consume sensor information continuously. We find through analysis that there is a reduced power consumption for content traffic in distributed solutions when compared with centralized solutions for equal numbers of devices, sensors, and users. Thus, we conclude that a fully distributed solution for information gathering and dissemination is preferable to centralized solutions for retrieving the current version of information that changes often.

One of the major challenges to realize the Internet of Things is to support IP mobility for the large amount of connected entities when they move between different locations and access methods. Current solutions for mobility are host centric, requiring support from the infrastructure, or breaks backwards compatibility, which will take a long time or high economic motivation to implement. Solutions for context information exchange are created for specific, small, or localized scenarios with centralized coordination that do not scale well. There is therefore a need for a solution which both scales well, and support IP mobility, without additional demands on current or future Internet infrastructure.We propose the use of a dual-overlay network structure for both information dissemination and as an alternative to current IP mobility technologies. It separates identities from location by introducing a second overlay network where the identity-to-location association is stored. We show analytically that the proposed solution provide logarithmic latency for localization and reduces the overall workload when the number of sensors per host increases beyond seven, with a workload reduction of 15 percentage points at fifteen sensors per host.

The Internet of Things is expected to expand several magnitudes in the coming decade surpassing over 50 billion devices. In the Internet of Things there is a need to support complex queries on the massive amount of information that will be made available. The scalability of the indexes used to support the queries is therefore critical. This paper therefore investigate what type of index that could scale to the size required by the Internet of Things. We find that range query is an approach that support continuously changing information and fast updates with the lowest increase in signaling per participating device.We find that a Chord-based distributed hash table hosting a NUBL range query indexing scheme will scale to the required size while supporting multidimensional range queries.