For reliable and practical application of structural health monitoring approaches in conjunction with dense sensor arrays deployed on 'smart' systems, there is a need to develop and evaluate alternate strategies for efficient problem decomposition to rapidly and accurately determine the occurrence, location and level of small changes in the underlying structural characteristics of a monitored system based on its vibrational signature. Furthermore, there is also a need to quantify the level of uncertainties in the identified system characteristics so as to have a measurable level of confidence in the parameters to be relied on for the detection of genuine changes (damage) in the monitored system. This study presents the results of two time-domain identification techniques applied to a full-scale 17-story building, based on ambient vibration measurements. The Factor building is a steel frame structure located on the UCLA campus. This building was instrumented permanently with a dense array of 72-channel accelerometers, and the acceleration data are being continuously recorded. The first identification method used in this study is the NExT/ERA, which is regarded as a global (or centralized) approach, since it deals with the global dynamic properties of the structure. The second method is a time-domain identification technique for chain-like MDOF systems. Since in this method the identification of each link of the chain is performed independently, it is regarded as a local (or decentralized) identification methodology. For the same reason, this method can be easily adopted for large-scale sensor network architectures in which the centralized approaches are not feasible due to massive storage, power, bandwidth and computational requirements. To have a statistically meaningful results, 50 days of recorded data are considered in this study. The modal parameter and chain identification procedures are performed over time windows of 2 h each and with 50% overlap. Using the NExT/ERA method, 12 dominant modes of the building were identified. It was observed that variations in the frequency estimation are relatively small; the coefficient of variation is about 1–2% for most of the estimated modal frequencies. Chain system identification was successfully implemented using the output-only data acquired from the Factor building. Probability distributions of the estimated coefficients of displacement and velocity terms in the interstory restoring functions (which are the mass-normalized local stiffness and damping values) that were found based on the chain system identification are presented. The variability of the estimated parameters due to temperature fluctuations is investigated. It is shown that there is a strong correlation between the modal frequency variations and the temperature variations in a 24 h period.