Research On Tree Based Provenance Compression In Wireless Sensor Networks

In wireless sensor networks(WSNs),data provenance records a data packet path from the data source to the base station(BS)as well as the operations on those data.The size of provenance expands rapidly with the increasing number of the forward and the aggregation nodes.As a result,the communication bandwidth and energy of the WSNs are drained quickly.To address such issues,we propose a tree based provenance encoding scheme,in which the generating trees of a WSN's topology graph and their indices firstly make a dictionary,and then when a provenance is encoded,besides the data source node IDs,a tree's index is appended instead of appending the entire tree with the data packet.The provenance is compressed with a high compression ratio consequently.Both simulation and experimental results show that our scheme can conserve more energy and has a shorter provenance size when compared with the others' provenance schemes in the same large-scale WSNs.Furthermore,our scheme applies the same algorithm for both linear and the aggregation provenances,which makes that our scheme has a consistent form while encoding different kinds of provenances.The main work of this paper includes the following:(1)We propose a tree based provenance encoding scheme in the same large-scale WSNs.Our scheme has high average compression ratio when compared with the others' similar provenance schemes.(2)We can take the same method to address these issues in both linear and aggregated provenance.Furthermore,our algorithm has a universal and consistent form while encoding different kinds of provenances.(3)Both simulation based on TinyOS system and experimental based on ZigBee network results show the validity of the tree based provenance encoding scheme.