An existing joint input-state estimation algorithm is extended for applications in structural dynamics. The estimation of the input and the system states is performed in a minimum-variance unbiased way, based on a limited number of response measurements and a system model. The noise statistics are estimated, as they are essential for the joint input-state estimation and can be used to quantify the uncertainty on the estimated forces and system states. The methodology is illustrated using data from an in situ experiment on a footbridge.