In view of the limitations of current processing of altitude data, i.e., extracting simple eigenvalues such as rhythm and energy values of electroencephalogram(EEG) signals, we considered the brain as an interconnected network, combined rhythm and energy with large-scale brain network, and explored the brain damage caused by hypoxia from a new perspective. Relevant research was conducted on residents who have lived at altitudes of 400, 2 890, 3 600 and 4 500 m for two years. The scalp energy was estimated on the basis of the resting EEG signal, the network-based source localization technology was used to obtain the cortical current density under seven rhythms, and the energy values of the eight large-scale brain networks were obtained according to the area covered by each network. The one-way analysis of variance was performed on the energy of large-scale brain networks at different altitudes. The experimental results show that the large-scale brain network exhibits significant difference between different altitude and the brain network energy of deep structure in 4 500 m is significantly reduced under θ, α1, β1 and β2 rhythms. The brain network energy of visual network and limbic under the α1 and β2 rhythms, and the default network under the α1 rhythm are all characterized by a significant increase in 3 600 m and a significant decrease in 4 500 m. In addition, the energy value increases with the increase of altitude, but the energy value at the highest altitude is the lowest, so it is speculated that there exists a threshold of the altitude influence on brain function.