The Research On Green Data Fusion Model And Transmission Scheduling Algorithm Of Wireless Body Area Networks

The WBAN (Wireless Body Area Network) has a significance role in carrying out continuous health monitoring and improving the level of primary health services. However, limited sensor energy resources are impeding the effective application of the WBAN. Aiming at solving the problem of the WBAN energy consumption, the characteristics of the body area communication channel and the influence of the human body motion on the WBAN architecture are discussed. Then, a data fusion green model and a transmission scheduling algorithm are established. On this basis, the prototype system of WBAN health monitoring is constructed. The results of the study provide strong support to promote the development of our grassroots telemedicine technology and promote China’s rural health services. The achievements of this dissertation are listed as follows:(1) A node movement model, the BMM (Body Mobility Model), of WBAN is constructed. In this paper, the human body transmission channel characteristics, the communication scheme regarding the GHz communication frequency band and the Bluetooth technology are determined. Also, a path loss model of wearable WBAN sensors, on the surface of the body communication is established. The logical model and physical model of the WBAN architecture are designed. Based on the analysis of the different human body postures and the affects on the network topology, the BMM model is proposed, which includes attitude selection and node movement. A method for calculating the distance between the sensor nodes and the sink nodes in the position of the arm and leg, when in the sitting, walking and lying posture is established.(2) The GDFM (Data Green Fusion Model) is established. In order to reduce the redundant data and the amount of data transmission, this model takes the green technology innovation as the guiding ideology. It consists of two parts, which are the data preprocessing based on the wavelet transform and the data fusion method, based on the prediction. Firstly, using the Sym8 wavelet function, the hard threshold function is constructed to deal with the wavelet coefficients. On this basis, the WPLA (Waveform Prediction Lightweight Agorithm) is proposed. The simulation results show that the pretreatment of the GDFM has achieved a large signal to noise ratio (SNR=54.3160) and smaller mean square error (RMSE=0.0239). These two findings indicate that the distortion is small and that the effect is satisfactory. The WPLA achieved a higher prediction accuracy (mean square error of a 10"4 magnitude). The performance of the WPLA is better than the LSSVM (Least Square Support Vector Machine) algorithm. Its lightweight characteristic has reached the goal of classifying it as a green design.(3) The transmission scheduling algorithm OPT-DSA (Outage Probability Transmission based on Dynamic Slot Allocation) is designed. The proposed algorithm is based on MAC’s (Media Access Control) TDMA (Time Division Multiple Access) super frame structure, which has three steps, including slot allocation, relay judgment and relay selection. The simulation results show that the algorithm, based on Greedy Sorting, has a higher performance, compared with the Flipping, which is used in the dynamic slot allocation. Furthermore, compared with the performance of the OPT-DSA algorithm and the simple relay transmission strategy, it is found that it has a significant advantage in terms of reliability and energy efficiency. Finally, the influences of the predefined threshold, the network size and the forwarding time slot on the performance of the algorithm are investigated.(4) The HMPS-WBAN (Health Monitoring Prototype System, based on Wireless Body Area Networks) is implemented. According to the health needs of focus groups in rural areas, the application process of the system is discussed. Then, the hardware circuit, appearance structure, software function and data structure are designed. Demonstration and application results show that there was no statistically significant difference between our HMPS-WBAN and traditional instruments for the measured vital signs results (p>0.05).The findings suggest that it had a clear effect of promoting improvement in health habits, health condition monitoring, and disease prevention (p<0.05).