Study On Connectivity Of Wireless Ad Hoc Networks

The architecture of wireess networks can be classified into two main groups: infrastructure networks and ad hoc networks. As an important branch of wireless communication, wireless ad hoc networks are capable of forming a communication network without relying on any central controller. In this network, each node operates not only as a host but also as a relay station for forwarding packets to its final destination. With the capability of rapid self-organization and self-configuration without fixed network infrastucture, wireless ad hoc networks can be applied in the field of civil and military communications.As a very fundamental and important property of networks, connectivity is especially vital in wireless ad hoc networks. In such decentralized networks, there is no fixed infrastructure which each node can have a wireless link to. Besides, because of severe channel enviornment and the uncertainty of network topology resulting from the mobility of nodes, the tranmitting power and network density must be controlled to achieve a highly connected network at any time. The connectivity problem of wireless ad hoc networks is studied here.Based on large scale pathloss channel model and not considering channel fading and interference, the relationship between coverage and connectivity problem, the critical transmitting range to make the networks asymptotically connected and the connectivity probability is analyzed separately in linear and two-demensional ad hoc networks. In addition, we also study the number of neighbors for connectivity and the problem of k-connectivity in two-demensioanal case. Analytical results show that the critical range for connecitivty is different from the covering problem and that increasing transmission power and network density can improve network connectivity. It's also proved that there is no "magic" number in the sense of connectivity so that each node only needs to be connected to certain number of neighbors. The number of neighbors must increase logarithmically with the number of nodes in the system to guarantee fully connected networks.The connectivity problem in a log-shadow fading environment and the impact of interference on connecitivity is also studied in this paper. It's shown that the variations around mean power value caused by shadowing have two distinct effects on the network. Firstly, they reduce the amount of correlation betweent links. Secondly, these variations increase the probability of long links, which enhances the probability of connectivity for the network. If considering the interference between nodes, network connectivity probabiltiy stops to increase when transmission power is large enough. The convergence value is determined by threshod, network density and other system parameters.It's known that cooperative diversity can combat multipath fading and increase the reliability of system. The transmitting range is analyzed based on several strategies, such as amplify-and-forward, decode-and-forward, and selection decode-and-forward. It's shown that if selecting proper partner, the cooperative communication between nodes can greatly increase the transmitting range which improves the probability of connectivity.