With the increasing of Internet of Things technology, mechanisms to efficiently and accurately to a large number of mobile notes are of interest. Data failover occur due to the unstable of GPRS and the timeliness of broadcast data items. How to ensure the timeliness of broadcast data, lower power consumption and reduce the access time is a fatal problem. Some major broadcast scheduling algorithms of multiple data items including QEM, improvement QEM as well as in ondemand broadcast mode their adaptive are briefly introduced. For the two subproblem in ondemand broadcast mode, the query selection problem and the broadcast scheduling problem, we also propose minimum delay time preference algorithm and maximum data access frequency preference algorithm for both them respectively. The results of theoretical analysis, simulation and spot tests show that the algorithm aroused in this article reduces the access time, improve the system performance.