Quality Of Information Assessment And Energy Optimization In Wireless Sensor Networks

With the rapid development of wireless communication,integrated circuits and embedded computing techniques,wireless sensor networks(WSNs)are used to observe and cognize the complicated physical world.Since the cost is very low,WSNs have been widely used in many fields,such as military defense,environment monitoring,healthy care,disaster management,smart cities,etc.WSNs are centered on the data sensed by sensors.As one of the core components in WSNs,the information is very important.Almost all applications of WSNs are dependent on information acquisition and management.Quality of information(QoI)is expressed as the measure of goodness of the information.It not only determines the performance of WSN applications,but also can be used as a constraint to manage and optimize network energy.With the advent of participatory sensing and Internet of Things,the scale of the data sensed by the sensor has expanded rapidly.How to quantify the QoI difference and efficiently obtain the high QoI information from huge sensed information has become the focus of WSNs.To effectively deal with these challenges,this paper studies QoI assessment and energy optimization in WSNs.According to the features and application requirements of WSNs,along the route of "raw data acquisition ? in-network processing ? network transport ? Sink client application",the paper focused on solving three key technical problems of "QoI modeling","QoI assessment" and "energy efficient information source selection”,and have obtained the following research contributions.1.A principle for QoI attribute organization is introduced,and a common theoretical QoI model is established.With the advent of participatory sensing and Internet of Things,the sensed information will be shared among the participants.This requires a uniform framework and common model to organize and explain the QoI.Different from the static database information,the sensed information of WSNs is transmited among nodes along multi-hop path.In the process of transmission,the sensed information may be continually processed and integrated,which will change the QoI.At the same time,different applications and users are concerned about different QoI attributes.Even for the same information,it may also give inconsistent quality judgment in different applications and users.The concept and basic theory of the traditional QoI research community is no longer applicable.Besides,the QoI attribute organization and QoI modeling at present in WSNs are highly application dependent.To share the sensed information among different applications,this paper carrys out the research of quality connotation in WSNs,according to the acquisition process of the sensed information.Under a uniform logical framework,we first define the QoI in WSNs.Then,we classify and organize the QoI attributes.Lastly,we establish a weighted QoI model by three parameters,which are the QoI attributes,the change of QoI attributes in network,and the weight of QoI attributes in particular application requrement.2.A flexible framework is proposed to dynamically assess the QoI in diferent WSN applications,with focus on accuracy and timeliness.As WSNs oten provide incorrect and outdated information about the events in a monitored environment,QoI assessment is invaluable for users to manage and use the information in particular applications,and allows a more realistic understanding and informed decision about the events.The current methods of QoI assessment are highly application dependent,which are lack of wide applicability.To establish an integrated system for QoI assessment in WSNs,we focus on the accuracy and timeliness,and propose a flexible framework to solve the problems of their online assessment in diferentWSN applications.3.An optimal problem of information source selection is designed and solved.It ensures QoI requirement and optimizes energy consumption through balancing residual energy of sensor nodes for the first time.For untethered applications in harsh environments,the challenge is to satisfy the aggregate QoI while maintaining a long-running operation due to the sensor nature of limited energy.In cases when redundant sensors are available in an area of interest,a subset of sensors can satisfy the aggregate QoI requirements,while others may be deactivated to achieve energy savings and reduce unnecessary information reporting.The problem in such cases is in determining an optimal subset of sensors which to be activated for QoI guarantee and energy optimization.The traditional methods select information sources only based on QoI,which can obtain high aggregate QoI.Howeve,the disadvantage is that the energy of sensors is not balanced.The sensors with high QoI exhaust their energy too early because of the long-term information transmission,while the energy of the low QoI sensors is abundant.To overcome the problem,we provide an optimal source selection scheme by taking account of QoI and residual energy of sensors,which can balance energy consumption of each sensor while ensuring aggregated QoI satisfaction of application.This scheme can let all sensors have a similar lifetime,making the most use of their energy and maximizing energy efficiency of each sensor.