Energy Efficiency Optimization Based On Resource Allocation In Wireless Body Area Networks

The wireless body area network(WBAN)is a special case of the wireless sensor network,which is mainly applied in general medical and entertainment applications around human beings.Generally,WBAN nodes try to collect the various human related signals like physiological,multimedia,and then transmit them to a central node,such as the PDA and cell phone,etc.Since the WBAN nodes are usually attached on the human body,the volume and transmit power of the node must be strictly restricted.On the other hand,since WBAN is usually applied for long time monitoring,it is of great importance to improve the energy efficiency.Meanwhile,the network lifetime should be maximized as well.In this thesis,we mainly focus on the energy efficiency optimization problem in WBANs,by performing proper network scheduling and power allocation.We first establish a general framework to minimize the energy consumption of WBANs,which can be formulated and solved as a standard geometric programming(GP)problem.Due to the fading effects in typical WBAN channels,we then present three dynamic network scheduling strategies,based on the proposed framework.Different tradeoffs between the energy consumption and quality of service can be achieved.In addition to the general energy consumption problem,we also consider the network lifetime optimization problem.The time duration when the first node is out of energy is defined as the network lifetime.In order to maximize the network lifetime,an optimal energy allocation is required beforehand.We first show that the general network lifetime maximization problem in static channels is also a GP problem,and can be solved accordingly.However,in dynamic channels,we try to adopt a machine-learning based non-parametric method,to perform energy allocation with respect to different node properties.Furthermore,we also consider the two-hop network topology for WBANs.Another general resource allocation framework can be adopted,in which the routing selection problem can be formulated as an integer linear programming(ILP)problem.Due to the non-convexity of the ILP problem,we present a low complexity algorithm based on the branch and bound method,to solve the problem.Numeric results show that the energy efficiency of WBANs can be greatly improved by the presented frameworks.The presented frameworks,algorithms and numeric results,are of value to the related works in WBANs,and also to practical applications.