Distributed FPGA-based architecture to support indoor localisation and orientation services Julio Dondo n , Felix Villanueva, David Garcia, David Vallejo, Carlos Glez-Morcillo, Juan Carlos Lopez Escuela Superior de Informtica - University of Castilla-La Mancha, Ciudad Real, Spain article info Article history: Received 9 June 2013 Received in revised form 20 May 2014 Accepted 20 July 2014 Available online 1 August 2014 Keywords: Indoor localisation and orientation Dynamic reconfiguration FPGA Partial reconfiguration Distributed systems abstract This paper introduces a proposal for an indoor localisation and orientation distributed service built on a dynamically reconfigurable platform. The integration of cameras in consumer electronic devices such as mobile phones, tablets, etc. allows the adoption of new methods based on video streaming analysis by a mobile device video camera that can enhance two essential features for a successful navigation experience: localisation and orientation. These features are important in services such as life assistance, direct marketing or localisation. The proposed infrastructure is based on the integration of hetero- geneous resources under the umbrella of the distributed object paradigm. Our ultimate goal is to provide an efficient implementation of multi-user indoor localisation and orientation services. & 2014 Elsevier Ltd. All rights reserved. 1. Introduction Indoor localisation of personal consumer electronics devices (e.g., smart phones, tablets and laptops) is a key problem, which, if addressed with enough accuracy and consideration of the orienta- tion of device (e.g., the user), will lead to other advanced services. Indoor navigation, proximity marketing, security concerns, perso- nal localisation, multitude monitoring, augmented reality, etc. are examples of services that require a localisation and orientation component. This group of services establishes a set of require- ments in terms of accuracy and efficiency for a localisation service. As we will see later, one of the most promising approaches in localisation/orientation of users in indoor environments is the analysis of video streaming from the consumer electronic devices that users carry. While analysing the video stream of the environ- ment where the user is, we can extract the position and orienta- tion of the user with several degrees of accuracy as a function of the analysis algorithm. The algorithms used in this application field have two common characteristics; first, they require a previous exhaustive characterisation of the environment, which generates a large database. As it will be detailed later, this characterisation of the environment extracts a set of points of interest (corners, windows, etc.) that are used to match against those extracted from the images taken by the user. The general idea is to find those points of interest in the users video streaming to infer his/ her position/orientation. The other characteristic in common is the computer power requirements of these algorithms to perform an ef ficient video analysis and an exact comparison with the database of the environment. These two characteristics make the implementation of the algorithms in consumer electronic devices undesirable due to the inefficiency of transmitting the environmental database to all devices present in that environment. Airports, railway stations, commercial centres, public services building, etc. are examples of the environments where these types of applications can be deployed. Public security concerns supply other limitations and restrictions to sending the database to all consumer electronics devices. In this paper we introduce an approach in which the video streaming analysis is performed by a set of specialized hardware which after processing send the localisation information to users with extra information to facilitate user localisation/orientation based on augmented reality (AR). Our main motivation for this work was the Elcano project (Villanueva et al., 2011). This project was devoted to guiding handi- capped people in indoor environments. The indoor positioning Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/jnca Journal of Network and Computer Applications http://dx.doi.org/10.1016/j.jnca.2014.07.029 1084-8045/& 2014 Elsevier Ltd. All rights reserved. n Corresponding author. E-mail addresses: Juliodaniel.Dondo@uclm.es (J. Dondo), Felix.Villanueva@uclm.es (F. Villanueva), Dave.Alcarin@gmail.com (D. Garcia), David.Vallejo@uclm.es (D. Vallejo), Carlos.Gonzalez@uclm.es (C. Glez-Morcillo), Juancarlos.Lopez@uclm.es (J.C. Lopez). Journal of Network and Computer Applications 45 (2014) 181–190