Research On Challenging Issues Of Location-based Computing In Military Applications

With the information technology developing in a direction of interoperability and comprehensive intelligence,the seamless integration of physical world and cyber space becomes significant.The flourish of mobile and pervasive computing has enlarged the scale of the network,which makes the self-organization and the mobility of the wireless ad-hoc networks and mobile devices become prominent it motivates the studies on the network management and mobility of the nodes.Determining the locations of the nodes in wireless ad-hoc networks or valued targets rapidly and preciously has become one of the most important and interesting research areas.In the military applications,is more sig-nificant,since the strict requirements on both time and robustness.Although the global navigation satellite system(GNSS)is widely used in practice for localization purpose,it is still constrained by several factors,including the coverage,obstacle,terrain,weather and electromagnetic disturbance from the opposite.On the other hand,the sensed informa-tion with location label has already been one of the key enabling elements in the modern warfare,especially in the network-centric information war.The location based comput-ing is widely utilized in a large body of key combat operations,including intelligence,surveillance,battlefield visualization,and accurate strikes.Based on the wireless localization techniques,this thesis comprehensively exploits the theory in Graph Theory,Operation Research,Statistics and Probability to push the limits of three major challenging issues as follows:(1)localization problem in the sparse wireless sensor network,(2)target tracking with insufficient coverage,(3)user localiza-tion in a indoor environment without any pre-deployed infrastructure,and(4)comprehen-sive sensing of the generalized entities without any dedicated devices.This thesis focuses on those three problems and systemically study the localization and location computation in military operations.The major contribution of this paper is summarized as follows.1.A novel approach for the localization of an innitially non-localizable WSN.This thesis focuses on utilizing the moving passive event to achieve the network localizabil-ity in an initially non-localizable network.The hybrid TDOA and distance measure-ment is firstly used in the network localizability.We propose the sufficient conditions in terms of node localizability and network localizability,respectively.Besides,a more promising TDOA estimation approach,which is approximately synchronization-free,is proposed to well support our localization approach.We design a corresponding algo-rithm in distributed manner for the implementation of our approach.Compared with the current works,our approach requires no extra hardware cost and any adjustment of the current topology of the network,and hence is more feasible in practice.We evaluate the performance of our approach via extensive simulations on a one-thousand-node network.Simulation results show that our TDOA estimation approach achieves the accuracy of 0.1 milliseconds.More than 90%nodes can be localized within 0.04r,where r denotes the ranging radius of a sensor.2.A Fine-grained distributed localization-oriented adjustment approach for WSN.This thesis focuses on the non-localizable WSN in the initial deployment,and proposes a fine-grained distributed localization-oriented adjustment approach,which is named as LODA.LODA decide the strategy of the node based on the path information to achieve the network localizability in an initially non-localizable network.Compared with the current network adjustment approaches,our approach merely adjusts less than 11%nodes in a sparse network to achieve localizability,which is about 40%better than the current best approach.Besides,we design the algorithm in a distributed manner to balance the communication load caused by the adjustment.Therefore,the scalability limitation in the previous centralized approaches is overcome.Extensive simulation results show that our approach outperforms previous work in adjustment efficiency.3.A novel approach to facilitate the target tracking with insufficient cover-age.Target tracking is one of the most important applications of the WSN.Although tar-geting tracking is heavily studied in the past two decades,most of the innovations fail with insufficient coverage,i.e.,with less then three anchors in sensor field that receive the signal form the target.However,such networks extensively exist in the military op-erations.This thesis focuses on such a problem and proposes a novel approach based on the particle filter.4.A reliable indoor localization appraoch without any pre-deployed infrastruc-ture.It is difficult for the GNSS signal to get through the complex indoor structure to achieve the deep inside of a building.Such environments are so called GNSS-denied.The previous works show that nearly all the high-accurate indoor localization approach requires carefully pre-deployed infrastructures wit high density,while the infrastructure-free approaches lack of accuracy guarantee.This thesis proposes a novel approach based on both Kalman filter for the headings and particle filter for the trajectory to improve the localization accuracy.5.The problem of comprehensive sensing of the generalized indoor entities.The entities extensively exist in the indoor spaces,and contain plentiful semantic information.This thesis first focus on the comprehensive sensing of the generalized entities without any dedicated devices.We first present a novel idea that exploits the interaction between user and environment to identify the entities.A unique entity localization approach is proposed to leverage multiple observations from independent users.We also develop a prototype system,iScan,in an urban supermarket.