In this study, we propose an improved genetic algorithm, i.e., an interference suppression subchannel allocation (IGA-ISSA) scheme, to solve the channel interference problem that can be attributed to the layout of multiple access points in heterogeneous VLC/WiFi networks. In IGA-ISSA, decision-making is designed for the user to select a VLC or WiFi network in accordance with the VLC channel quality. Further, the users are classified into three priority levels according to the user location characteristics in VLC. Subsequently, a subchannel allocation scheme is designed based on the user conflict graph with respect to different levels for suppressing the interference. Furthermore, IGA-ISSA is introduced to optimize the subchannel allocation for the VLC network. Finally, a WiFi subchannel allocation scheme is designed based on the different user-required rates with respect to the allocated WiFi subchannel in case of users accessing a WiFi network. The simulation results denote that the IGA-ISSA scheme can improve the throughput and user satisfaction with respect to a VLC/WiFi network.