Taxi is a flourishing industry recently. It plays an important role in many kinds of popular transportation systems because of the characteristic that can be hailed conveniently, and also provided location-based service. Most of the taxi drivers carried passengers by hunting on the road or waiting in a fixed taxi queuing point, but these patterns have poor performance. Furthermore, there are about eighty percent of vacant rate and many critical issues such as air pollution and foul up traffic in the past few years. This work proposed a taxi carrying system by using location-based service and on-air queuing point techniques, including on-air queuing point establishment, on-air queuing point management and path planning protocol. It can transmit information via WAVE/DSRC and demarcate taxi service area by transmission range of road side units. Combining the concept of on-air queuing point by Taiwan Taxi, taxi drivers can both hunt on road with geographic information efficiently and wait in queue when they are in the taxi service area. We emulated the module by PRISM which is a probability module analysis simulator, and proved that this pattern has better performance for carrying passengers and reducing waiting time compared with the real data which is from the investigation of the fixed waiting point in Taiwan High Speed Rail Teaching Station. On the other hand, we design a prototype about taxi on-air queuing system. Drivers can get newest information immediately through user interface, and find passenger as soon as possible.