Contribution from Crowd / Testbed Resources

The obtained experimental data from Testbeds and Participants is sent to the Experiment Data Storage within the Experiment Manager in an OML format. An investigator researcher can send the query for experiments through the Web UI via the Experiment Manager RESTful API.

Monitoring the Resources and information about their availability and location should take place either periodically or based on the query from Experiment Manager between the Account Manager and the Resource Manager.

The Resource Manager should notify the Experiment Manager on Resources periodically as well as based on the query through the RLI interface.

6.7. Conclusions

6.7 Conclusions

In this chapter are proposed the main components of the IoT Lab platform and described their identified functionalities, interaction patterns, interfaces and communication links. The IoT Lab platform development focuses on:

• Enhancement of existing IoT FIRE testbed facilities which are traditionally built from static sensor mote platforms by including the participants’ end user mobile phones and thus achieving a crowd participation in sensing and actuation operations.

• Solutions with built-in reputation and privacy mechanisms as well as procedures for dynamic selection of suitable crowd resources.

The derivation process followed an IoT-A methodology in order to support interoperability and scalability and to enable the use of a wide range of heterogeneous devices as well as the testbeds from different application domains, therefore, satisfying a high number of requirements. Integration of smartphones with existing FIRE testbed infrastructures or any general statically deployed IoT resources represents a novel approach with respect to existing crowdsourcing solutions.

The values that IoT Lab brings in comparison to other existing crowdsourcing solutions are numerous: IoT and crowd sourcing based research and development;

access to distributed testbeds; economies of scale; access to innovative service oriented architecture (SOA) technology; a possibility to explore the future now;

market insight; crowd interaction and access to wisdom of the crowd; experimental platforms; reduction of development time and time to market; capital expenditure avoidance; cost reduction; access to ‘up to date’ and evolving IoTservices/

infrastructures; new product experimentation on advanced IoT facilities.

Our future work will focus on further implementing the proposed platform in order to test selected use cases in real situations and get a feedback from participants.

Strategies for engaging as many participants as possible will be examined as well as the new components that provide reward mechanisms towards all participants and investigators.

Part IV

IoT, Crodwsensing and Incentive

Mechanisms

7 A User-enabled Testbed Architecture with Mobile Crowdsensing Support

for Smart and Green Buildings

7.1 Introduction

The Future Internet and the Future Networks are general terms coined to address the major shift in the context and the operation of the Internet that is expected to take place in the near future. This shift is realized by cross-cutting changes in several aspects of the Internet, each one driven by a different factor. For instance, technological advances during the past decade in the field of embedded systems have driven the rise of the Internet of Things (IoT) paradigm. On the other hand, the constantly increasing adoption rates of truly portable hand-held and wearable smart devices (smartphones, smart watches and glasses, etc.) have paved the way for the Mobile Crowdsensing Systems (MCS) that seek to exploit the embedded sensory capabilities of these devices and their intrinsic mobile nature.

Between the first stage of conceiving a new paradigm, by laying down its theoretical foundations, and the final stage of rolling out at full scale a new technology, there lies an intermediate phase where small scale, experimental systems are putting the vision to the test, thus providing valuable feedback. Such testing facilities bridge the gap between abstract assumptions, necessarily present when theoretically analyzing a concept, and implementation-specific limitations that emerge due to technological dependencies and real-life limitations. Various challenges regarding the design of such

testbed facilities have been widely highlighted and relevant desired properties have been identified (e.g. see [111], [112] for IoT testbeds). Among these properties, the realism of experimentation environment,device and service heterogeneityandefficient service compositionduring the experimentation life cycle are important aspects to improve upon existing testbed facilities.

Increased realism implies matching the experimentation conditions as close as possible to the typically operating situations where the final solutions are expected to be deployed. This way design flaws or imperfections can be earlier detected and evened out, thus reducing the cost of roll out and maturation time.Device and service heterogeneityoffers experimenters with more experimentation options and captures better how technological environments are expected to mature at later deployment stages. Efficientservice compositionis thus a key requirement for efficient testing facilities. Therefore, mechanisms to control and exploit realistic experimental conditions during the evaluation phase are necessary.

Our contribution. We present an IoT testbed facility for Smart Buildings whose architecture enables the seamless and scalable interaction of crowd-enabled resources provided by the end-users of the facility [113]. This integration increases the awareness of the facility both in terms of sensory capabilities as well as in terms of users’ preferences and experienced comfort. Combined with smart actuation, IoT communication and networking technologies, the experimenter is provided with an agile experimenting platform. The facility exposes its operations as services thus greatly facilitating the definition and evaluation of diverse use-case scenarios. In order to demonstrate the use of the facility we design and evaluate several incentive policies in the context of a smart luminance scenario based on Participatory Sensing.

First the end-users are incentivised to provide access to their hand-held devices from which data on the ambient environmental conditions are collected and aggregated into live luminance maps. Then, the indoor lighting units are dynamically adjusted based on the luminance maps and the feedback provided by the users to the system on their personal preferences and experienced comfort.

7.2. Related Work

7.2 Related Work

There have been numerous testbed facilities using wireless sensor networks as the main component of their backbone infrastructure. For instance in [27] the authors present Syndesi, a framework for creating personalized smart environments using wireless sensor networks. This framework, among other services provided, is able to identify people and take personalized actions (such as control of electrical devices) based on their personal preferences. As a proof of concept, authors present a real-world deployment, where two use-case scenarios are implemented in the premises of a building. Also, in [114], they first identified some of user and operator testbed requirements, then they introduced the architecture and general concepts of a testbed approach and showed how this architecture meets the requirements of both groups. The main focus of this work was to address the perspective of both users and operators on how to experiment or respectively operate a wireless sensor network testbed based on a specific technology. In [115], the authors present PhoneLab, a smartphone testbed that provides access to smartphone users incentivising them to participate in experiments while simplifying experiment data collection. Three selected results from a usage characterization experiment are presented. In [116], they proposed a comprehensive architecture for managing a cluster of both real and virtual smartphones that are either wired to a private cloud or connected over a wireless link. Also, they proposed and described a number of Android management optimizations (e.g. command pipelining, screen-capturing, file management), which can be useful to the community for building similar functionality into their systems.

Finally, they conducted extensive experiments and microbenchmarks to support our design choices providing qualitative evidence on the expected performance of each module comprising the relevant architecture. In [117], the authors developed a distributed computing infrastructure using smartphones. More specifically, they profile the charging behaviours of real phone owners to show the viability of their approach. Furthermore, they present a simple task migration model to resume interrupted task executions. They have also implemented and evaluated a prototype

that employed an underlying novel scheduling algorithm to minimize the makespan of a set of tasks. A more diverse approach is presented in [118], where the testbed deployment is focused on smart buildings, a key building block for cities of the future.

The presented system combines heterogeneous IoT devices such as a programmable experimentation substrate in a real life office environment while making feasible the experimentations with real end users. Authors present the architecture of the facility and underline the considerations that motivated its design. Using several recent experimental use cases they demonstrate the usefulness of such experimental facilities for user-centric IoT research.

Early after the first smartphones were introduced, research teams started investigating the combined usage of their embedded sensory capabilities. In [119] authors recognize the opportunity of fusing information from populations of privately-held sensors as well as the corresponding limitations due to privacy issues. In this context they describe the principles of community based sensing and they propose corresponding methods that take into consideration the uncertain availability of the sensors, the sensitive context of sensor information and sensor owners’ preferences about privacy and resource usage. Authors present efficient and well-characterized approximations of optimal sensing policies in the context of a road traffic monitoring application.

In more recent works, the authors in [101] use the notion of Participatory Sensing (PS). They consider the problem of efficient data acquisition methods for multiple PS applications while taking into consideration issues such as resource constraints, user privacy, data reliability and uncontrolled mobility. They evaluate heuristic algorithms that seek to maximize the total social welfare, via simulations that are based on mobility datasets consisted of both real-life and artificial data traces. Finally, in [102] the authors propose a utility-driven smartphone middleware for executing community-driven sensing tasks. The proposed middleware framework considers preferences of the user and resources available on the phone to tune the sensing strategy thus enabling the execution of tasks in an opportunistic and passive manner.