| Wireless sensor network(WSN)is a data centric network.Its sensor nodes are capable of collecting,processing and transmitting the target information in the monitored area and feeding back the final information to users.However,owing to the abundant nodes in the network,the information is always over-collected,which often causes overload of the communication and accelerating consumption of energy.That's why the energy issue is the top problem to be under consideration and the data fusion is one of the key technologies to solve the above problems.To reduce the data transmission and save the nodes consumption energy from the information sending and receiving,we have to compress the data as much as we can to minimize the information amount while guaranteeing the information accuracy received by the terminal at the same time.In terms of the compressed sensing technology,the measurement matrix will directly determine the compression rate and the reconstruction signal error,so the purpose of this paper is to design a reasonable measurement matrix to reduce the reconstruction error and increase the compression rate and finally achieve the goal of easing the network's burden and saving node energy consumption.In order to make the measurement matrix work effectively and is easy to access,we proposed the design as following: firstly,we designed a binary sparse measurement matrix based on the arrangement theory,proved its qualification of isometric property restriction and analyzed its properties with experiment;secondly,focusing on the problems this measurement matrix with,we constructed a superior performance binary sparse matrix by the pseudo-random sequence,and proved its excellent performance with simulation experiment;at the last,we ran the simulation experiment with the data collecting from WSN and successfully proved that the measurement matrix is workable for WSN and the matrix we designed as well as the RIP method are able to work accurately and effectively. |
PDF Full Text Request