Frame length control has been widely applied to traditional communication system to improve their performance. However, most of the existing research using the data transmission efficiency oriented approach can not adapt to situation of Internet of Thing (IoT), where quality of received information is more important In this paper, based on our former work where compressive sensing is applied to facilitate efficient wireless information transmission over lossy communication links, we further investigate frame length effect on signal/information transmission. Noting that correctly received packets corresponds to a deterministic projection matrix which shows inefficiency when using large packet length. To cope with this problem, we first analyze the impact of bit error rate and signal sparsity on frame length control in signal/information transmission, then adopt traditional communication technique-data interleaving to optimize its performance. Extensive simulations has been conducted and results show that signal/information transmission efficiency can be quite different compared with data transmission efficiency, which inspires us to choose different optimal frame length to optimize different communication demand.