Abstract: The frequent node mobility of opportunistic networks poses significant challenges for link prediction. To better reflect the temporal evolution of the topology, an opportunistic network link prediction method based on temporal generative adversarial networks is proposed. A network volatility is defined for calculating the network segmentation duration so that an opportunistic network is sliced into fine-grained snapshots. Information matrices are extracted from these network slices which integrate spatial and temporal dimensional information, and network feature vector matrices are constructed by graph embedding methods. Combining gated recurrent units and generative adversarial network, a temporal generative adversarial network model is developed. It learns the evolutionary features of network topology, so as to achieve link prediction in networks for the future. The experimental results on three real datasets demonstrate that the predictive performance of the proposed method is superior to that of the baseline method.

Key words:  opportunistic network ,  link prediction ,  generative adversarial networks ,  gated recurrent