The Research On Key Technologies Of Secure Coverage In Wireless Sensor Networks

Wireless sensor networks (WSNs) consist of low-cost, low-power tiny sensor nodes that can communicate with each other to perform sensing and data processing tasks co-operatively. WSN has a large range of applications including national security, military operations, environmental monitoring, traffic management, e-healthcare, and smart home, and becomes one of the hottest research areas around the world. Network coverage and net-work security are two primary problems in wireless sensor networks. Connected coverage, which reflects how well a target field is monitored under the sink node (or base station), is the most important performance metrics measuring the quality of surveillance a WSN can provide and varies with time as sensors run out of energy or are physically destroyed by natural or intended attacks. With the rapid development and proliferation of WSNs, the increasing serious security concerns have gradually restricted the development and applica-tion of WSN. Therefore, a coverage monitoring scheme with security guarantee is necessary for a mission-critical application of WSNs. However, due to the special nature of WSNs, the task of designing a practical and robust scheme to continuously monitor the connected coverage is not trivial at all. One significant challenge comes from the strict resource limi-tation of the sensor nodes (e.g., limited battery power, memory, computational ability, etc.), which highlights the requirement of energy and computational efficiency. The unattended and distributed nature of the deployment of sensor nodes without tamper-proof also exposes them to node capture attacks, which constitutes another significant challenge.The existing coverage monitoring solutions require accurate knowledge of the sensors locations (or geographical positions), which cannot be easily obtained, or they cannot pro-vide guarantees on the coverage quality. Addressing this need, we present in this disserta-tion a distributed algorithm for coverage boundary detection. Differ from existing coverage monitoring schemes, our algorithm doesn't need the position information of nodes, it re-quires only the localized information between nodes (e.g., localized distance information or localized connectivity information). In this dissertation, we first propose a distributed algorithm for individual sensor nodes to identify whether they are on the coverage boundary, by using only localized distance in-formation. Furthermore, we extend the algorithm to k-coverage boundary detection scheme. We show the correctness and efficiency of our algorithm by theoretical proofs. In general, no assumption should be made about the distribution of the sensor nodes in the environment. Our schemes are designed to work correctly under arbitrary node distributions. So far most related literatures have assumed that each node knows its accurate location. Our schemes can also be extended to tolerate bounded location errors (defined as the distance between the actual location of a node and its estimated location).Secondly, in order to handle sensor nodes with strictly limited capabilities (e.g., cannot measure relative distances between neighbors), we propose a new distributed algorithms to detect and recover coverage holes in WSNs. The proposed algorithm does not require any coordinates or location information; it requires only local connectivity information. This al-gorithm is derived and justified through graph theory, which can detect most coverage holes and recover the hole by activating necessary redundant nodes (if exist). The complexity of the algorithms doesn't depend on the overall size of the network.Thirdly, we propose secure coverage monitoring protocols for WSNs in hostile environments. Security is very important for many WSN applications, such as military target tracking and security monitoring. Due to limited capabilities of sensor nodes and salient features of WSNs, the security design for such networks is significantly challenging. In this dissertation, we propose a secure protocol to provide current network coverage information. In contrast to previous security mechanisms, our approaches feature nearly perfect resilience to node compromise, low communication and computation overhead, low memory requirements, and high network scalability.