With the rapid growth of population in mega cities, a large number of high-rise buildings have been built around the world. These buildings are subjected to various loading conditions including earthquake excitation. The safety and serviceability of building structures has thus gained much attention. Long-term structural health monitoring (SHM) systems have also been developed to provide real-time measurements of various external loadings and structural responses, which are expected to be used for the performance evaluation of high-rise buildings. However, the number of sensors is always limited in consideration of the huge size and complex structural system of high-rise buildings. Accurate and complete structural performance evaluation is almost impossible by just utilizing direct measurement information from installed sensors. This is particularly true for important but inaccessible structural locations. Recently, a feasible and cost-effective way of solving the problem has been proposed by integrating response reconstruction algorithm with optimal sensor placement scheme in such a way that the optimized sensor placement confi guration enables the best estimation of structural responses at all interested locations. Yi et al. (2011) proposed an optimal sensor placement method for high-rise buildings based on the generalized genetic algorithm, but they dealt with single type of sensor. An integrated optimal sensor placement and response reconstruction method has been proposed for long-span bridges equipped with multi-type sensors by Xu et al. (2016). Corresponding experiments have also been conducted on a two-pinsupported overhanding beam and a test-bed of long-span suspension bridge (Zhu et al., 2013; Xu et al., 2016). Zhang and Xu (2016) further developed this method for the structure under unknown excitation by incorporating the excitation estimation process. However, the method designed for long-span bridges may not be applicable for high-rise buildings, for the two different types of structures adopt quite different sensor systems and the corresponding responses also differ from each other. Commonly-used accelerometers for dynamic responses measurements may not be appropriate for monitoring static and quasi-static deformation of highrise buildings of long natural periods. Instead, GPS and inclinometers are more effective sensors for monitoring high-rise building deformation (Yigit et al., 2010; Li et al., 2010; Su et al., 2013) because the safety of a high-rise building is normally evaluated in terms of its inter-story drift. In this connection, the authors have proposed an integrated multi-type sensor placement and response reconstruction method for building structures under unknown excitation on basis of the Kalman fi lter algorithm (Hu et al., 2017).