The use of wireless sensors in Structural Health Monitoring (SHM) has increased significantly in the last years. Piezoelectric-based lead zirconium titanate (PZT) sensors have been on the rise in SHM due to their superior sensing abilities. They are applicable in different technologies such as electromechanical impedance (EMI)-based SHM. This work develops a flexible wireless smart sensor (WSS) framework based on the EMI method using active sensors for full-scale and autonomous SHM. In contrast to passive sensors, the self-sensing properties of the PZTs allow interrogating with or exciting a structure when desired. The system integrates the necessary software and hardware within a service-oriented architecture approach able to provide in a modular way the services suitable to satisfy the key requirements of a WSS. The framework developed in this work has been validated on different experimental applications. Initially, the reliability of the EMI method when carried out with the proposed wireless sensor system is evaluated by comparison with the wireless counterpart. Afterwards, the performance of the system is evaluated in terms of software stability and reliability of functioning.

Contemporary design codes and guides for vibration serviceability assessment include some simplifications in load modelling. The same statistical distribution of the inter-pedestrian variability of the step interval (frequency) is proposed for all applications. Moreover, walking loads are considered to be periodic. The intra-pedestrian variability of the step interval is neglected. A more realistic load modelling trying to overcome the limitations of the codes is intended in this paper. Instead of a single mean value of the inter-pedestrian distribution of walking speed, a range of possible variation, which account for the real variations that occur in practice depending on the footbridge location and usage, is considered. An enhanced model is proposed in this paper to reproduce statistically both the intra- and inter-pedestrian variability of the step interval as a function of the walking speed distribution. This innovative model is then applied to study the sensitivity of the vertical response of footbridges to the variability of the step interval and to evaluate the influence of the aforementioned simplifications on the predicted characteristic responses. For this purpose, low-frequency footbridges excited by single-pedestrian crossings are chosen. The response is statistically characterized through Monte Carlo numerical simulations including 720 different configurations and 10,000 load cases in each configuration. Results of the study provide an overview of the influence of the footbridge and load parameters on the responses, which can be useful in practical applications where human–structure interactions are negligible. As for the simplifications of the codes, it is found that either using a single distribution to model the inter-pedestrian variability of the spatiotemporal parameters or neglecting the intra-pedestrian variability can lead to a significant underestimation of the characteristic response of footbridges.