Data Loss Compensation Based On Compressive Sensing For Wireless Data Of Structure

Depended on the advantages of rapid deployment, low cost, fast inspection technology based on wireless sensor network has been developed rapidly in the field of structural health inspection. However, due to the packet loss problem, the application of fast detection technology has been limited to a certain extent. The packet loss compensation algorithm based on Compressed Sensing provides a new research direction for resolving the above problem. The packet loss compensation algorithm's measurement matrix is difficult to be embedded in the common nodes. This situation restricts the above algorithm's application in the actual project. Research on packet loss compensation algorithm based on Random Demodulator has great theoretical significance and practical value for reducing traditional measurement matrix's storage space and computing time.Firstly, a new packet loss compensation algorithm based on Random Demodulator is proposed. Through RD theory, the structure of measurement matrix being changed, the storage space fell to 1/n times, and computing time fell to k2/n (1?k??n) times. The improved algorithm improves the running efficiency and reduces the storage space greatly, which lays the foundation for the application of the embedded node. Secondly, in order to realize the rapid detection of civil engineering structures, a wireless detection system is developed. The modular design method is used to design the WiFi-based wireless sensor nodes, which embedded in packet loss compensation algorithm. Finally, wireless systems was applied to Dalian Xinghai Bay Bridge hoisting process to test and verify the above algorithms and system.The experimental results show that the wireless system can complete the normal monitor task during the hoisting process. Compared with the traditional algorithm, the improved packet loss compensation algorithm can reconstruct the lost data in a certain extent. When the reconstruction effect is equivalent, the improved algorithm has the lower space complexity and time complexity, which is more easy to be embedded in the actual wireless sensor nodes. In a word, the improved packet loss compensation algorithm based on the Random Demodulator has an important reference value to resolve the packet loss problem of wireless sensor network, and the developed fast wireless inspection system based on the above algorithms has broad application prospects in structural health inspection field.