Research On Packet Loss Compensation For WiFi-based Structural Health Monitoring Wireless Sensor Network

In order to grasp the damage condition of the structure,this paper designs a Wi-Fi-based structural health monitoring wireless sensor network to monitor the vibration of the structure.However,the data loss problem in wireless transmission process affects the reliability of wireless sensor networks seriously.In order to ensure the accuracy of subsequent analysis,this system embeds a packet loss compensation algorithm based on compressed sensing in the node to mathematically compensate the lost data packets during wireless transmission.However,the storage space required by the measurement matrix poses a great challenge to the node.In order to make compensation algorithm applicable to this system,this paper proposes an improved method: 1)use The fixed point number replaces the floating point number: the Gaussian coefficient matrix is used instead of the Gaussian random matrix as the measurement matrix of the packet loss compensation algorithm;2)the storage mode of the measurement matrix in the node is changed: the measurement matrix is stored in bits.In order to facilitate the user to control and manage the wireless sensor nodes,this paper design a set of server-side software by Qt,including features of command interaction,receive,save and display data collected by the node,and subsequent analysis of monitoring data.In order to systematically test the structural health monitoring system designed in this paper and verify the performance of the embedded packet loss compensation algorithm,the field experiments has done on the main bridge section of Xinghaiwan Bridge.Through the analysis of the data,it can be concluded that the improved packet loss compensation algorithm using the Gaussian coefficient matrix as the measurement matrix can compensate and reconstruct the lost data within a certain error range,and its reconstruction performance is roughly equivalent with compensation algorithm using the Gaussian random matrix,but the memory space occupied by the node is only 1/32 of the Gaussian random matrix.The improved packet loss compensation algorithm in this paper is of great significance to nodes,and Wi-Fibased wireless sensor network structure health monitoring system developed in this paper is practicality.