Autonomous Decentralized System Identification by Markov Parameter Estimation Using Distributed Smart Wireless Sensor Networks

Abstract

Decentralized data processing has the benefit of improving wireless monitoring system scalability, reducing the amount of wireless communications, and reducing overall power consumption. In this study, a system identification strategy for single-input multi-output (SIMO) subspace system identification is proposed based on Markov parameters. The method is specifically customized for embedment within the decentralized computational framework of a wireless sensor network. By using the computational resources of wireless sensors, individual sensor nodes perform local data processing to identify the Markov parameters of a structural system. The data storage and wireless communication requirements of Markov parameters are significantly less than that required by the original raw data, resulting in the preservation of scarce system resources such as communication bandwidth and battery power. Then, the estimated Markov parameters are wirelessly communicated to a wireless sensor network base station where the global structural properties are assembled by execution of the eigensystem realization algorithm, an indirect subspace system identification method. The proposed strategy is evaluated using input-output and output-only data recorded during dynamic testing of a cantilevered balcony in a historic building (Hill Auditorium, Ann Arbor, MI).