Research On Energy-Efficient Communication Scheme For Dynamic Characteristics Of Wireless Body Area Network

WBAN(Wireless Body Sensor Network)is a kind of wireless sensor network deployed on(or implanted in)human body,collecting personal healthcare information.WBAN provides the necessary conditions for the development of a universal health care monitoring system and hence has attracted a lot of attentions from academic and industry circles.Because WBAN has to work interruptedly for a long term,the energy consumption has been considered as the biggest challenge.To this purpose,we adopt power control and backscatter communication technology,since they are practical and effect low power communication technology.However,unlike the traditional wireless sensor network,WBAN users may always be in motion,which can cause highly dynamic fluctuation in channel condition and ambient RF signal intensity.Therefore,the performance of power control and backscatter communication are severely affected.To remedy this problem,the following researches have been carried out.1.Research on acceleration information fusion based power control scheme.According to the periodic channel fluctuation,the association between acceleration signal reflecting the node movement and RSSI(Received Signal Strength Indication)reflecting the channel condition is analyzed.With the analysis results,we study how to apply decision level fusion between acceleration signal and channel state information to improve the performance of power control in highly dynamic channel condition.The experiment result shows that the proposed scheme can reduce energy consumption by 18% and the packet loss rate by 40% compared to the conventional TPC.2.Research on user's gait cycle information fusion based power control scheme.Our research found that the channel fluctuation and user's gait cycle are synchronize in time.With this phenomenon,we study how to track the user's gait cycle in real time and apply decision level fusion between user's gait cycle information and channel state information to to improve the performance of power control in highly dynamic channel condition.The experiment results reveal that energy consumption can be reduced by 25% and the packet loss rate can be reduced by 65% compared to conventional TPC.3.Research on an online learning based dual model WBAN backscatter transmission scheme.To deal with the issue of unstable ambient RF signal intensity caused by user's mobility,we study a transmission policy which adopt dual backscatter modes(ambient backscatter and dedicated backscatter)and take into account of data queue length,ambient RF signal intensity,user's activity and channel condition.We formulate the problem as a Markov decision process.Since the system statistic knowledge is is always unaviable in practical,we utilize an online learning approach to derive the policy.The simulation results show that the performance of proposed online algorithem is slightly worse than the optimal MDP policy,but is still far better than the dedicated back scatter mode preffered mechanism which do not use any system dynamic statisctic knowledge.