This dissertation characterizes the geometry of mobile wireless networks and their performance. In mobile networks, distance variations caused by node mobility generate fluctuations of the channel gains. Such fluctuations can be treated as another type of fading besides multi-path effects. Interference statistics in mobile random networks are characterized by incorporating the distance variations of mobile nodes to the channel gain fluctuations. The mean interference is calculated at the center and at the border of a finite mobile network. The network performance is evaluated in terms of the outage probability. Compared to a static network, the interference in a single snapshot does not change under uniform mobility models. However, random waypoint mobility increases (decreases) the interference at the center (at the border).
Due to the correlation of node locations (in mobile or static networks), the interference and outage are temporally and spatially correlated. We quantify the temporal correlation of the interference and outage in mobile Poisson networks in terms of the correlation coefficient of the interference and conditional outage probability, respectively. The results show that it is essential that routing, MAC scheduling, and retransmission schemes need to be smart (i.e., correlation-aware) to avoid bursts of transmission failures.
For communication between two neighboring nodes in wireless networks, the local delay, which is defined as the time it takes a node to successfully transmit a packet, is an important quantity. Previous research focuses on the local delay in static or infinitely mobile Poisson networks with ALOHA. In this dissertation, we extend the local delay results to Poisson networks with finite mobility. Bounds of the local delay in mobile Poisson networks are derived for different mobility and transmission models. Although mobility helps reduce the local delay, its impact depends on the particular mobility model. The phase transition that marks the jump of the local delay from finite to infinite is also characterized.