|08:00 - 09:20||Registration||09:20 - 09:30||Opening|
|09:30 - 10:30||Invited talk by Shyam Gollakota|
|10:30 - 11:00||Coffee break|
|11:00 - 12:30|
This paper investigates feasibility of device-free indoor localization using single passive receiver. Instead of local transmitters sharing one frequency channel, this work leverages frequency diversity of multiple ambient FM radio stations. Experimental results on localization performance, its dependence on time and number of utilized channels, are provided and confirm feasibility of the proposed approach.
Heba Aly and Moustafa Youssef
WiFi-based device-free localization is a main indoor localization technique that has attracted much attention recently. Typically, due to the complex wireless propagation in indoor environments, WiFi-based device-free localization requires a construction of a fingerprint map that captures the signal strength characteristics when the human is standing at certain locations in the area of interest. This fingerprint requires significant overhead in construction, and thus has been one of the major drawbacks of WiFi-based device-free localization. In this paper, we leverage an automated tool for fingerprint constructions to study novel scenarios for WiFi-based device-free localization training and testing that are difficult to evaluate in a real environment. In particular, we examine the effect of changing the access points (AP) mounting location, AP technology upgrade, and outsider effect; on the accuracy of the localization system. Our analysis provides recommendations for better localization and provides insights for both researchers ad practitioners.
Shuyu Shi, Stephan Sigg and Yusheng Ji
Due to spatial diversity, RF signals derived from a FM broadcast station differ when they arrive the receivers placed in various locations. Also, the FM signals will be altered by the change of ambient environment. Previous works focuses either the FM-based localisation or activity recognition. In this study, we propose to simultaneously classify and localise activities conducted in proximity of an FM receiver. We conducted experiments and demonstrated that the location and activities of an individual can be distinguishable with a reasonable overall accuracy in a typical indoor environment from FM broadcast signals.
Benjamin Wagner and Dirk Timmermann
Context sensing is an important part of building ubiquitous smart and assistive environments. It is the major data source for intention recognition and strategy generation systems. Device-free localization systems (DFL) join the efforts of non-instrumentation of users maintaining their privacy. In recent publications we propose an innovative approach utilizing a cluster of passive Radio Frequency Identification Transponders (pRFID) for device-free radio-based positioning. Due to the point that the RFID technology is typically not designed for that purpose we have to deal with certain drawbacks. A high number of transponders typically conclude in lower measurement frame rates while generating substantially more information for accurate positioning. To fix this tradeoff this work presents a transponder clustering approach based on inherent EPC protocol based bit masking, which allows us to calculate fast coarse grained localization results and increase the precision by time, so that the user is able to adjust between localization speed and accuracy. We made simulations and conducted experiments in an indoor room DFL scenario for validation.
|12:30 - 13:30||Lunch break|
|13:30 - 14:00||Demo and poster setup|
|14:00 - 14:40|
Session (Activity recognition)
Jihoon Hong and Tomoaki Ohtsuki
In this paper we consider device-free radio based activity recognition with localization methods that can be implemented in various applications such as e-Healthcare and security. Many researches of device-free radio based systems only focus on received signal strength (RSS) due to its ease of use. As RSS is easily affected by radio wave propagation characteristics such as fading and noise, its accuracy may degrade, particularly in indoor environments. In this paper we introduce a novel device-free based activity recognition with localization method using signal subspace that is more stable than RSS. The signal subspace can be estimated by using signal eigenvectors of the covariance matrix of an array sensor which uses an antenna array at receiver side only. To classify human activities and/or positions, we apply a machine learning method with support vector machines (SVM). We evaluate the classification performance of the signal subspace features compared with that of RSS. We analyze the impacts of antenna deployment on the enhancements of classification accuracy in non-line-of-sight (NLOS) environments to prove the effectiveness of the proposed method. Also, we compare our classification method with k-Nearest Neighbor (KNN). The experimental results confirmed that signal subspace features of array sensor provide accuracy improvements over the conventional RSS-based method.
Stephan Sigg, Shuyu Shi and Yusheng Ji
We investigate the use of received RF-signals for activity recognition in scenarios with multiple receive nodes and multiple simultaneously active individuals. Our system features a short 0.5 second window for feature values and we report on experiences in the choice of the neighbourhood size of the k-nearest neighbour (k-NN) classifier utilised. In a case study with software defined radio nodes utilised in an active, device-free activity recognition (DFAR) system, we observe a good recognition accuracy for the recognition of multiple simultaneously conducted activities with two and more receive devices. This is the first study to distinguish this particular set of activities from users conducting them simultaneously. For a single individual, we repeat the report the recognition accuracy in scenarios where the recognition area per receive node is larger than 8 square meters.
|14:40 - 15:30|
Collaboration sessionOpen session with contributions from all participants -- please contact us if you would like to contribute
(Recent and ongoing work -- demo, video, poster)
|15:30 - 16:00||Coffee break|
|16:00 - 17:30||
Panel -- Is device-free recognition the future of Ubiquitous Computing?
|19:00||Join us at Ziegel Oh Lac (next to the 'Rote Fabrik' Zurich)|