Year

2018

Degree Name

Doctor of Philosophy

Department

School of Electrical, Computer and Telecommunications Engineering

Abstract

As one of the most practical and promising technologies, Wireless Boby Area Networks (WBANs) play a vital role in a variety of application areas, including ubiquitous healthcare, medical, sport, and entertainment. Meanwhile, the multiple application scenarios also result in stringent and various Quality of Service (QoS) requirements in terms of data rate, reliability, energy consumption, delay, etc. Besides, due to the complicated deployment environments and peculiar channel characteristics, the design of WBAN systems is more challenging than other Wireless Sensor Networks (WSNs). Among many transmission challenges in the context of WBANs, transmission reliability and energy efficiency are the two key challenges. This thesis focuses on investigating high transmission reliability and energy-efficient transmission schemes for WBAN systems.

In order to capture the channel information variation in the real dynamic WBAN scenarios, we build a wearable wireless transceiver from easily-assembled commercial hardware modules. The measurements are conducted by test subjects wearing multiple wireless transceivers in dynamic scenarios. This thesis mainly focuses on two typical dynamic scenarios: walking and daily scenarios. Based on the channel gain datasets collected from these two scenarios, we analyze the on-body channel characteristics, including body shadowing effect, cross-correlation, autocorrelation, and transmission outage. Moreover, to improve the authenticity of our evaluation works, these channel datasets are imported into our simulation model to represent the channel variation.

Motivated by the significant cross-correlation feature of on-body channels in the walking scenarios, we propose a novel cooperation-based Network Coding (NC) scheme, namely A3NC. In A3NC, we explore the combination of the Aggregative Allocation (AA) mechanism in the MAC (Medium Access Control) layer and the Analog Network Coding (ANC) technique in the physical (PHY) layer. By comprehensive theoretical analyses and comparisons from the perspectives of data rate, energy efficiency, and throughput balance, we explore the upper bounds of A3NC in terms of data rate and energy efficiency. Simulation results confirm that A3NC achieves a better performance in terms of data rate, energy efficiency, and throughput balance, compared to the conventional approaches.

Extending from monotone walking scenarios to daily scenarios with mixed activities, we explore the Dynamic Slot Scheduling (DSS) method to improve the transmission reliability. DSS method optimizes the duration or order of time slots of different sensor nodes. This method does not require extra hardware or software overhead on the sensor side. Motivated by the significant temporal autocorrelation of on-body channels within a time span of 500 ms, we propose a new DSS method, named DSS-TA, which applies a Temporal Autocorrelation Model (TAM) to predict the channel condition for future time slots. The performance evaluation results confirm the great potential of DSS methods (up to 52% reduction of packet loss ratio over the static scheduling method) in daily WBAN scenarios. Further, compared to the classical Markov model-based DSS methods, the newly proposed method is more effective in the prediction of channel conditions and hence achieves a better performance in terms of Packet Loss Ratio (PLR).

Lastly, we jointly consider Transmission Power Control (TPC), DSS and two-hop Cooperative Communication (CC) mechanism to achieve a better trade-off between transmission reliability and energy consumption. Motivated by the significant autocorrelation characteristic of on-body channels in the daily WBAN scenarios, we propose an autocorrelation-based adaptive transmission (AAT) scheme which uses a TAM to predict channel conditions. Then, the estimated channel conditions are used to optimize the transmission power level and the transmission order of all sensor nodes for the next superframe. Simulation results demonstrate that the proposed method can effectively reduce the PLR and increase the energy efficiency. Moreover, we also discuss the effectiveness of NC technology for two-hop transmissions. Two types of cooperative mechanisms are compared, namely the non-NC mechanism we proposed for AAT and the NC cooperative mechanism.

Share

COinS