6th International Conference on Body Area Networks - 6-8 November 2011, Future Internet Technology building, Tsinghua University, Beijing China.

6th Novemeber - Around 3.30 pm we landed in Beijing International Airport. Beijing Capital International Airport is located in the northeast of the city, about 30 km away from the conference site. In Beijing weather was quite cool. Usally change between 7 - 15 degree of celcius. We went to hotel in the eveing. Beijing is we


ll built city.

BodyNets conference focus on the advancement in human body based sensing, embedded computing and wireless communication. BodyNets is a forum for engineers, scientists and medical professionals to exchange ideas, discuss practices, raise awareness, and to share cutting edge technologies and experiences in various Body Area Network applications such as health monitoring, human-computer interaction, education and entertainment. In this year conference 13 full papers and 7 short papers were presented.

Morning of the 7th November we went to the conference venue and registered for the conference. The venue was hard to find Nimesha and my self we had to ask help from some citizens. Since most them are not good at speaking English we has to show them the addresses printed in chinese.

In the first day keynote speech done by Professor Majid Sarrafzadeh, University of California, Los Angeles and he had presented many of his research work and commercially available health solutions developed by his team. He is a giant research person in the body sensor networking fields and more then 1500 doctors in the medical field collaborating for his research works. There is a systems which was closure to baby monitoring project of the cute center. Further, they had used digital compass , 3 axis

accelorometers , gyros , EEG to built body sensory networks. He has presented some user study results. One of the interesting finding was in US they had found out children play video games more than 40 hours per week. In the night we were busy with finalizing the presentation that we needed to present in the following day. There was no internet facility in the hotel and wireless internet connection provided in the conference venue was so slow. Some of the websites were permanently blocked and we had to experianced very slow speed when we access some Singapore web sites. 8 th November we went to the conference venue by 7.30 am and get ready for the presentation. Keynote speech has done by Professor John Stankovic, University of Virginia. We did our presentation at 10.15 a.m Nimesha has first started with giving a broad introduction to the smell and taste communication project and went to discribe taste device and results in deep. Roughly after 15 minutes I presented the magnetic smell interface device and future goals of the research. Our presentation was very interesting for the audiance and
they had asked many questions from us regarding the taste actuation device and magnetic s timulation. There were only two projects had discussed about the actuation devices in the conference. All other papers were highly focused on body sensors and communication.
After the lunch poster madness session was started. Nimesha has done the poster present ation. After the poster madness session everybody gathered the area that we had pasted the posters which was outside of the conference hall. Everybody chechk each others posters and asked questions and gave feedbacks. We also had received some comments to improve our rersearch as
well. After the poster session, we had attened some interesting presentations in the afternoon and arounf 4p.m conference was officially closed. Next year Body Nets conference will be held at Oslo. Around 5 p.m we left the conference venue and heading to the Beijing International Airport to catch the flight to singapore. We were safly landed early morning of 10th November. This conference was a very interesting conference for me. I had learned about some new body worn sensors, algorithms, pattern reconginition techniques used for sensors, human anatomy informtation from this conference. I will plan to submit next phases developments of my research to this conference in the next year.

Interested Papers :Activity Recognition for Emergency Care using RFID presented a system that recognizes human activities during trauma resuscitation, the fast-paced and team-based initial management of injured patients in the emergency department.

StepFit: A Novel Fitness Evaluation System paper presented an automated self-administered, low cost and convenient system for fitness evaluation, heart rate response analysis and prediction. The Step-Fit system consists of unobtrusive low cost HR and motion sensors,and a computational platform that is used for data processing archival and user feedback.

A Real-time Algorithm Based on Triaxial Accelerometer for the Detection of Human Activity State, This paper proposes a human state recognition algorithm based on Kalman filter, which could identify steady state and state transition in real time. In this study, an automatic state-recognition system consisting of a Bluetooth module and a smart phone is developed.

myHealthAssistant: A Phone-based Body Sensor Network that Captures the Wearer’s Exercises throughout the Day This paper presents a novel fitness and preventive health care system with a flexible and easy to deploy platform. By using embedded wearable sensors in combination with a Smartphone as an aggregator, both daily activities as well as specific gym exercises and their counts are recognized and logged.

Toward a Trust Management Model for a Configurable Body Sensor Platform : Proposed a trust management model along with a component-based architecture to maintain the BSN as trustworthy platform under such changes in applications. With this model the trustworthiness can be autonomically evaluated and monitored at component, application, and system levels with respect to properties of security, dependability, and performance. The usage of the model is shown with an example of an ECG monitoring application.

Improving Secret Key Generation Performance for On-Body Devices :In this paper, Authors undertake experiments to assess the feasibility of generating common secret keys between two body worn devices using the near-body channel. Deriving secret keys using the wireless channel is a practical and lightweight alternative to public-key key-agreement approaches. The results indicate that key generation is good for dynamic scenarios where communicating devices are placed in non-line-of-sight positions on the body.

Does Feature Matter: Anomaly Detection in Sensor Networks: In this paper, authors propose RFS, a novel three-stage ranking based feature selection approach for anomaly detection in WSNs. Due to the lack of techniques for managing information in most of diagnostic tools, diagnosing WSNs is often information-intensive and laborious work. In this work disclose a significant fact: changes of the representative features are important indicators of WSN anomalies. Based on this fact, they propose RFS to solve the above mentioned problem, using a feature selection approach. Although feature selection method is not new in the fields of statistics and data mining, it is effective in information reduction in WSNs. The experiments demonstrate that RFS is a practical and efficient approach for anomaly detection.

Characterisation of On-Body Communications at 2.45 GHz : This work focuses on the characterisation of on-body communications using patch antennas at 2.45 GHz. An investigation on the effects of the body in the communication link is done, based on measurements, for a range of body postures and on-body links. Return reflection loss results show that, on average, the antenna performance is 0.6 dB worse than in free space, also presenting a small detuning (<5 MHz). A fitting to the Normal Distribution is proven to be acceptable. Up to 33 dB of dispersion of values is verified for path links with the same length, as a consequence of antenna mismatch, which can turn a quasi-line-of-sight link into a non-line-of-sight one, increasing path loss up to 15 dB.

Design of an ultra low power MAC for a heterogeneous in-body sensor network : This paper presents an ultra low power MAC designed for in-body implant network which shows how an in-body implant network, has its own unique set of requirements such as priority, latency, and throughput which are not addressed by generic body area network protocols which are designed assuming identical sensors. By choosing a particular use case, authors demonstrated how they can exploit disparities inherent in a typical implanted body area network to enable ultra low power operation which also meets other competing requirements. They presented a new MAC scheme, which allows ultra low power operation by handling the nodes in accordance to their power and latency requirements.