Catálogo de publicaciones - libros

We present the design, prototype implementation, and evaluation of CenceMe, a personal sensing system that enables members of social networks to share their with their buddies in a secure manner. Sensing presence captures a user’s status in terms of his activity (e.g., sitting, walking, meeting friends), disposition (e.g., happy, sad, doing OK), habits (e.g., at the gym, coffee shop today, at work) and surroundings (e.g., noisy, hot, bright, high ozone). CenceMe injects sensing presence into popular social networking applications such as Facebook, MySpace, and IM (Skype, Pidgin) allowing for new levels of “connection” and implicit communication (albeit non-verbal) between friends in social networks. The CenceMe system is implemented, in part, as a thin-client on a number of standard and sensor-enabled cell phones and offers a number of services, which can be activated on a per-buddy basis to expose different degrees of a user’s sensing presence; these services include, life patterns, my presence, friend feeds, social interaction, significant places, buddy search, buddy beacon, and “above average?”

We introduce, characterize and test a vision-based dead-reckoning system for wearable computing that allows to track the user’s trajectory in an unknown and non-instrumented environment by integrating the optical flow. Only a single inexpensive camera worn on the body is required, which may be reused for other purposes such as HCI. Result show that distance estimates are accurate (6-12%) while rotation tends to be underestimated. The accumulation of errors is compensated by identifying previously visited locations and “closing the loop”; it results in greatly enhanced accuracy. Opportunistic use of wireless signatures is used to identify similar locations. No a-priori knowledge of the environment such as map is needed, therefore the system is well-suited for wearable computing. We identify the limitations of this approach and suggest future improvements.

This paper addresses the design of a large area, high resolution, networked pressure sensing floor with primary application in movement-based human-computer interaction (M-HCI). To meet the sensing needs of an M-HCI system, several design challenges need to be overcome. Firstly, high frame rate and low latency are required to ensure real-time human computer interaction, even in the presence of large sensing area (for unconstrained movement in the capture space) and high resolution (to support detailed analysis of pressure patterns). The optimization of floor system frame rate and latency is a challenge. Secondly, in many cases of M-HCI there are only a small number of subjects on the floor and a large portion of the floor is not active. Proper data compression for efficient data transmission is also a challenge. Thirdly, locations of disjoint active floor regions are useful features in many M-HCI applications. Reliable clustering and tracking of active disjoint floor regions poses as a challenge. Finally, to allow M-HCI using multiple communication channels, such as gesture, pose and pressure distributions, the pressure sensing floor needs to be integrable with other sensing modalities to create a smart multimodal environment. Fast and accurate alignment of floor sensing data in space and time with other sensing modalities is another challenge. In our research, we fully addressed the above challenges. The pressure sensing floor we developed has a sensing area of about 180 square feet, with a sensor resolution of 6.25 sensels/in. The system frame rate is up to 43 Hz with average latency of 25 ms. A simple but efficient data compression scheme is in place. We have also developed a robust clustering and tracking procedure for disjoint active floor regions using the mean-shift algorithm. The pressure sensing floor can be seamlessly integrated with a marker based motion capture system with accurate temporal and spatial alignment. Furthermore, the modular and scalable structure of the sensor floor allows for easy installation to real rooms of irregular shape. The pressure sensing floor system described in this paper forms an important stepping stone towards the creation of a smart environment with context aware data processing algorithms which finds extensive applications beyond M-HCI, e.g. diagnosing gait pathologies and evaluation of treatment.

In this paper we present how motion sensing can be obtained just by observing the WLAN radio signal strength and its fluctuations. The temporal, spectral and spatial characteristics of WLAN signal are analyzed. Our analysis confirms our claim that ’signal strength from access points appear to jump around more vigorously when the device is moving compared to when it is still and the number of detectable access points vary considerably while the user is on the move’. Using this observation, we present a novel motion detection algorithm, based on the spectral analysis of WLAN signal’s RSSI. To benchmark the proposed algorithm, we used , which is inspired by the recent work of Sohn et al.Both algorithms were evaluated by carrying out extensive measurements in a diverse set of conditions (indoors in different buildings and outdoors - city center, parking lot, university campus etc.,) and tested against the same data sets. The 94% average classification accuracy of the proposed is outperforming the accuracy of (accuracy 87%). The motion detection algorithms presented in this paper provide ubiquitous methods for deriving the state of the user. The algorithms can be implemented and run on a commodity device with WLAN capability without the need of any additional hardware support.

An increasing number of computationally enhanced objects is distributed around us in physical space, which are equipped – or at least can be provided – with sensors for measuring spatial contexts like position, direction and acceleration. We consider spatial relationships between them, which can basically be acquired by a pairwise comparison of their spatial contexts, as crucial information for a variety of applications. If such objects do have wireless communication capabilities, they will be able to build up an ad-hoc network and exchange their spatial contexts among each other. However, processing detailed sensor information and routing it through the network lowers their battery lifetime or even may exceed the capabilities of embedded systems with limited resources. Thus, we present a novel and efficient approach for inferring and distributing spatial contexts in multi-hop networks, which builds upon qualitative spatial representation and reasoning techniques. Simulation results show its behavior with respect to common network topologies.

We exploit the ability to sense and use context information to augment or replace the traditional static security measures by making them more adaptable to a given context and thereby less intrusive. We demonstrate that by fusing location information obtained from various sources that are associated to the user and are available over time, the confidence in the identity of the user can be increased considerably. In fact, the level of confidence in the identity of the user is related to the probability that the user is at a certain location. This probability is used as a measure to parameterize the authentication level of the user making it thereby much more adaptive to changing situational circumstances. In this paper we describe the theoretical background for a context-sensitive adaptation of authentication and the design and validation of the system that we have developed to adaptively authenticate a user on the basis of the location of his sensed identity tokens.

Within the framework of a French project, which aims at developing a new human presence sensor, we intend to design a sensor system simulator. During the establishment of the requirements of that new sensor we raised that the mission of a global scene survey could only be performed by a collection of several systems using very diverse technologies. This article presents the development of a method for the placement of multi-technology and multi-sensor systems. The considered environments are room or set of rooms in office buildings or individual homes. We will explain how we managed to represent the use of different sensors considering their various environments. Then, the way of exploiting these models using genetic algorithms is discussed. Those models are oriented for finding system placement and therefore for helping sensor networks deployment.

We describe and evaluate a distributed architecture for the online recognition of user activity sequences. In a lower layer, simple heterogeneous atomic activities were recognised on multiple on-body and environmental sensor-detector nodes. The atomic activities were grouped in detection events, depending on the detector location. In a second layer, the recognition of composite activities was performed by an integrator. The approach minimises network communication by local activity aggregation at the detector nodes and transforms the temporal activity sequence into a spatial representation for simplified composite recognition. Metrics for a general description of the architecture are presented.

We evaluated the architecture in a worker assembly scenario using 12 sensor-detector nodes. An overall recall and precision of 77% and 79% was achieved for 11 different composite activities. The architecture can be scaled in the number of sensor-detectors, activity events and sequences while being adequately quantified by the presented metrics.

We are developing a context-aware application for use in homes, which detects high-level user behavior, such as “leaving the home” and “going to bed”, and provides services according to the behavior proactively. To detect user behavior, a behavioral pattern is created by extracting frequent characteristics from the user’s behavior logs acquired from sensors online, using an extraction threshold based on the criterion of frequency. Most context-aware applications need to determine such a threshold. A conventional model determines a fixed common threshold value for all users. However, the common value is improper for some users because proper values vary among users. This paper proposes a detection method of high-level behavior with a model for determining the threshold value dynamically according to individual behavioral pattern.

The introduction of mood as context of a mobile user opens up many opportunities for the design of novel context-aware services and applications. This paper presents the first prototype of a mobile system platform that is able to derive the mood of a person and make it available as a contextual building block to mobile services and application. The mood is derived based on physiological signals captured by a body sensor network. As a proof-of-concept application a simple mood based messaging service has been developed on top of the platform.