DMND: Collecting data from mobiles using Named Data

Technology advances in both computations and wireless communications have made it economically feasible for manufacturers to collect data from all the cars in order to monitor their operations and detect any potential problems. However to make this a reality requires a new architecture that can effectively handle vehicle mobility, intermittent connectivity, and data security, as well as scale to large number of vehicles. In this paper we address these design challenges by exploring the direction of Named Date Networking (NDN) (aka CCN1). We evaluated our design, dubbed DMND, through simulations in Qualnet. Our results show that, when data publishers (vehicles) are stationary, more than 99% of collection requests can successfully pull data packets back; even when vehicles move at a high speed of 40–50 meters per second (89.48–111.8 miles/hour), DMND can still retain its high efficiency of 97% of data replies. In contrast, under the same simulation experimental setting, the request-reply ratio of MobileIP drops from 97.9% for static publishers to 9.6% when publishers are moving at a speed of 10–20 meters/second (22.37–44.74 miles/hour).