Electric Vehicle (EV) becomes a mainstream transportation tool in recent years as it is environmentally friendly and energy-usage effectively. However, due to the inconvenience of EV charging, EV requires charging optimization to improve its Quality of Experience (QoE). Previous researches focus on EV's plug-in charging mode (EVs drive to charging station for energy supplement), but it has spatial limitations (fixed location and charging station operation cost). With the development of charging technology, a Vehicle-to-Vehicle (V2V) charging mode is proposed as a replacement for plug-in mode. Here, EVs are divided into Charging EVs (C-EVs) and Discharging EVs (D-EVs). D-EVs can transfer their surplus energy to C-EVs. This V2V mode not only avoids the limitation of charging location, but also provides EVs with more flexibility in the temporal domain (avoids charging congestion). In this paper, a deadline-based V2V charging scheme is proposed. Through the aggregated information of EVs (location and energy supply/demand) and Parking Lots (PLs), V2V charging-discharging pairs (V2V-Pairs) are matched by the Global Controller (GC). Furthermore, an appropriate PL is selected for the V2V charging operation. We simulate the V2V mode in Helsinki City scenario under constrained spatial parking resources, and the results show that the proposed V2V charging scheme has better charging efficiency.