Based on the panel data of 288 cities in China from 2001 to 2016, this paper adopts improved Bass model and calculate the innovation diffusion of ICT. Then the paper constructs the fixed effect spatial econometric model, the quasi-maximum likelihood estimators is used to estimate the parameters of the spatial panel fixed effect SAR model. The main conclusions are as follows: the space-time effect of ICT innovation diffusion is significant. The innovation diffusion speed is influenced not only by local key factors, but also by neighboring cities and "neighbor to neighbor". Population density, population flow, consumer expenditure and ICT Supply Capacity have significant influences on innovation diffusion, and the short-term spatial spillover effect and long-term backflow effect of key factors is significant.