Data Collection Method For Mobile Crowd Sensing Based On Sampling Frequency And Data Fusion

The popularity of wearable devices and smart phones provide a great convenience for large-scale data collection.Many innovative applications and services are emerging.Sensor nodes collect the valuable data adaptively in the mobile process.And it forms a Mobile Opportunistic Networks(MONs)when data is transferred between sensor nodes.The Mobile Crowd Sensing(MCS)has a broad application prospect in the field of intelligent transportation,wildlife tracking,environmental monitoring and so on.As a novel sensing paradigm in the area of Internet of Things(IoT),MCS is gradually becoming a hot research topic in recent years.Owing to the non-uniform distribution and limited communication range of mobile sensor nodes,the data quantity collected from different regions has a wide variation.In view of the above problems,we focus on the data collection method and data fusion method between sensing nodes in MSC.The major work completed and contributions include several aspects as follows:Firstly,we propose a data collection method based on sampling frequency(DC-BSF).In this method,we firstly propose the Division of Region Grade(D-RG)algorithm to divide the whole area into regions of different grades,according to intensive degree of nodes' trajectory.On this basis,we propose DC-BSF to set different sampling frequency for sensing nodes in different regions.At the same time,we propose the Circle of Time Slice(CoTS)and the Cardinal Number Timing Method(CNTM)to overcome the perceptual error caused by the cross-regional movement of the sensing nodes,which can reduce data redundancy in the premise of ensuring data coverage.Secondly,We propose a data fusion mechanism based on clustered indexes to prevent the sensing nodes from collecting and uploading the same data.In this approach,we design a data fusion method based on the forward and receive clustered index,which ensures that only one copy of the same sensing data is stored in the encountered sensing nodes.Simulation results show that this method can not only improve the delivery ratio,but also reduce the transmission delay,and ultimately reduce data redundancy.