Open Access Open Access  Restricted Access Subscription Access

Compressive Sensing Approach for Structural Health Monitoring of Ship Hulls



The research presented in this paper aims to use compressed sensing algorithms to reduce the size and complexity of data sets relevant to structural health monitoring applications. Compressed sensing (CS), a powerful new data processing paradigm just beginning to emerge, proposes to accurately reconstruct unknown signals which are assumed to be compressible in some basis. Samples are taken as projections of the sparse signal onto an incoherent domain. Compressed signals are generally reconstructed via convex optimization or greedy algorithms that solve for transform coefficients most consistent with the samples taken. The advantage of CS based data acquisition lies in the amount of work saved at the sensor in data acquisition, storage, and transmission payload. In the context of wireless sensors engaged in structural health monitoring, the reduction in the number of samples possible with CS directly translates to power and communication bandwidth savings, two immensely important constraints in a wireless sensor network. CS exploits the sparsity that is so often present in natural signals such as strain response of structures subject to wave loading. As a case-study, this paper explores the use of CS in the structural health monitoring of the FSF-1 Sea Fighter, a small waterplane area twin hull (SWATH) littoral combat ship.

Full Text: