On the Characterization of CAR Systems Based on Mobile Computing

Collaborative Augmented Reality (CAR) systems allow multiple users to share a real world environment including computer-generated images in real time. Currently, the hardware features of most mobile phones provide excellent multimedia services, and it also includes wireless network capabilities that offer a natural platform for CAR systems. However, the performance of the mobile CAR applications under different conditions, like the number and type of devices in the system, has not been studied yet. This paper presents the experimental characterization of CAR systems based on mobile phones, providing quantitative results about well-known performance metrics in distributed systems like system throughput and system response times. The characterization results show that the system saturation point depends on the overall percentage of CPU utilization in the computer platform acting as the system server, although it is not a fixed value and it is inversely related to the number of processor cores. Also, the results show that throughput of CAR systems heavily depends on the kind of client devices, but CAR systems can efficiently support some hundreds of clients in any case. Another important result is that the CAR system throughput is limited by the server I/O in some cases. Therefore, any improvement in CAR systems should be addressed to alleviate the server I/O, even though it may add computational overhead to the server.