We define an (n, T, 1, m)-network as a wireless ad hoc network (MANET) that possesses the following properties: (1) The network consists of n nodes placed in an 1 x m rectangle area. (2) The position of each node is a random variable uniformly distributed Over the given area. (3) Each node has a transmission radius of T unit length, where T 5 min(1, m). (4) Any two nodes that are within the transmission range of each other will have a link connecting them’. We are concerned with several fundamental properties in this model. It was commonly believed that the probability of link occurrence in MANET cannot be identical. However, we found that it is not true. The expected node degree and the expected number of links in a MANET have also been obtained. Pre- vious work on degree estimate [I], [2], [3] does not take into account border effects [2], which refers to the circumstance that a node placed near the system border will cover less area (with its radio signal) than nodes placed midway. Border effects makes the conventional estimate inaccurate. In contrast, our results are not subject to border effects. The next problem to solve is the expected area jointly covered by a finite number of nodes, which is a form of so- called coverage problem. Given the expected node coverage, which can be derived from link probability, the problem at hand is still complicated by the fact that region covered by each node may overlap one another in a stochastic way. We also found that border effects are not only a major obstacle to precise calculations of many network properties, but also the reason behind the probabilistic dependency of two links. This implies that the occurrences of any two links are independent to each other if border effects disappear.