Vehicular Ad-hoc network (VANET) is one of the fundamental parts in the intelligent transport system (ITS). Researches on the connectivity probability and vehicle positioning of VANET have remarkable significance on VANET security and system optimization. This paper presents a study on the vehicle connectivity and its impact on positioning. Firstly, we investigate the research on vehicle connectivity in the past years. Secondly, based on dual-slope path loss model, we study the connectivity probability. Based on the assumption that the distribution of vehicle nodes follows a Gaussian distribution in a dense traffic network, we present the Cramer-Rao lower bound (CRLB) for positioning error in VANET. Finally, through simulation, we analyze the influence of the path loss exponents on CRLB.