Explore the vast range of titles available.

Sensor data pre-processing is an essential phase of crowd sensing application. Existing studies do not effectively solve the problem, and there still exist various sensor data pre-processing optimisation problems at the acquisition end in crowd-sensing process. This study presents an improved sliding average method to achieve data compression and reduce the time complexity by using a dynamic window with improved processing time. Through adopting locally sorting and gradient change of the filter window, an improved extremum median filtering method is proposed to relieve the time-consuming problem when denoising high pixel images. A transmission strategy for optimisation is also proposed, in which only the demarcation points of each group of data and the data points with large differences when comparing with the demarcation points are recorded. This strategy reduces the storage pressure and the amount of data transmission of mobile terminal and improves the efficiency of data transmission. The experimental results show that their methods have higher speed and lower cost, and thus they can run better in crowd-sensing environment.