Multi-sensor data fusion embraces methodologies, algorithms and technologies for combining information from multiple sources into a unified picture of the observed phenomenon. Specifically in the context of Body Sensor Networks (BSNs), the general objective of sensor fusion is the integration of information from multiple, heterogeneous, noise- and error-affected sensor data source to draw a more consistent and accurate picture of a subject’s physiological, health, emotional, and/or actvity status.
About a decade ago the research area on wireless sensor network (WSN) technologies and applications led to the introduction of BSNs: a particular type of WSN applied to human health. Since their inception, BSNs promised disruptive changes in several aspects of our daily life. At technological level, a BSN comprises wireless wearable physiological sensors applied to the human body (by means of skin electrodes, elastic straps, or even using smart fabrics) to enable, at low cost, continuous and real-time non-invasive monitoring. Very diversified BSN applications were proposed during the years, including prevention, early detection, and monitoring of cardiovascular, neuro-degenerative and other chronic diseases, elderly assistance at home (fall detection, pills reminder), fitness and wellness, motor rehabilitation assistance, physical activity and gestures detection, emotion recognition, and so on. On of the main key benefit of this technology is the possibility to continuously monitor vital and physiological signs without obstructing user/patient comfort in performing his/her daily activities. Indeed, in the last few years, its diffusion increased enormously with the introduction at mass industrial level of smart wearable devices (particularly smart watches and bracelets) that are able to capture several parameters such as body accelerations, electrocardiogram (ECG), pulse rate, and bio-impedance.
However, since many BSN applications require sophisticated signal processing techniques and algorithms, their design and implementation remain a challenging task still today. Sensed data streams are collected, processed, and transmitted remotely by means of wearable devices with limited resources in terms of energy availability, computational power, and storage capacity. In addition, BSN systems are often characterized by error-prone sensor data that significantly affect signal processing, pattern recognition, and machine learning performances. In this challenging scenario, the use of redundant or complementary data coupled with multi-sensor sensor data fusion methods represents an effective solution to infer high quality information from heavily corrupted or noisy signals, random and systematic error-affected sensor samples, data loss or inconsistency, and so on. Most commercially available networked wearables assume that a single device monitors a plethora of user information. In reality, BSN technology is transitioning to multi-device synchronous measurement environments. With the wearable network becoming more complex, fusion of the data from multiple, potentially heterogeneous, sensor sources become non-trivial tasks that directly impact performance of the activity monitoring application. In particular, we note that the complex processing chain used in BSN designs introduces various levels of data fusion with different levels of complexity and effectiveness. Only in recent years researchers have started developing technical solutions for effective fusion of BSN data.
This special issue aims to provide a forum for academic and industrial communities to report recent theoretical and application results related to Advances in Multi-Sensor Fusion for Body Sensor Networks from the perspectives of algorithms, architectures, and applications.
Manuscripts (which should be original and not previously published either in full or in part or presented even in a more or less similar form at any other forum) covering unpublished research that clearly delineate the role of information fusion in the context of body sensor networks are invited.
The manuscript will be judged solely on the basis of new contributions excluding the contributions made in earlier publications. Contributions should be described in sufficient detail to be reproducible on the basis of the material presented in the paper and the references cited therein.
Topics appropriate for this special issue include (but are not necessarily limited to):
Data-level algorithms for multi-sensor fusion in BSNs (e.g. Digital Signal Processing, Coordinate Transforms, Kalman Filtering, Independent Component Analysis)
Feature-level algorithms for multi-sensor fusion in BSNs (e.g. Decision Trees, k-Nearest Neighbor, Naive-Bayes networks, Support Vector Machines)
Decision-level algorithms for multi-sensor fusion in BSNs (e.g. Dempster-Shafer theory, Boosting, bagging, plurality and reputation-based voting, stacking, multi-sensor ensemble)
Multi-level algorithms for multi-sensor fusion in BSNs
Hardware/software architectures (autonomic, agent-oriented, etc) for collaborative multi-sensor fusion in BSNs
Multi-sensor fusion applications in BSNs for human activity recognition
Multi-sensor fusion applications in BSNs for sport monitoring
Multi-sensor fusion applications in BSNs for emotion recognition
Multi-sensor fusion applications in BSNs for health care monitoring
Manuscripts should be submitted electronically online at http://ees.elsevier.com/inffus
The corresponding author will have to create a user profile if one has not been established before at Elsevier.
To ensure that all manuscripts are correctly identified for inclusion into the special issue, it is important that authors select “SI: BSN-Fusion\”.
Dr Giancarlo Fortino, University of Calabria (Italy), email@example.com
Dr Hassan Ghasemzadeh, Washington State University (USA), firstname.lastname@example.org
Dr Raffaele Gravina, University of Calabria (Italy), email@example.com
Dr Peter X. Liu, Carleton University (Canada), firstname.lastname@example.org
Dr Carmen C.Y. Poon, The Chinese University of Hong Kong (HK), email@example.com
Dr Zhelong Wang, Dalian University of Technology (China), firstname.lastname@example.org