Deep unsupervised learning resource allocation method for guaranteeing long-term user rate of multi-cell cellular network

The invention discloses a deep unsupervised learning resource allocation method for guaranteeing the long-term user rate of a multi-cell cellular network, and belongs to the technical field of wireless communication. According to the method, a deep neural network is constructed, a power distribution variable, a power control variable and a channel distribution variable are output at the same time, and a channel distribution scheme and a power distribution scheme are obtained through the output power distribution variable, the output power control variable and the output channel distribution variable; and constructing a storage pool, training and testing the deep neural network at the same time, and performing unsupervised learning on the deep neural network by adopting a Lagrange loss function in training. The method has the advantages of improving the user rate, long-term stability, self-adaptability and intelligence, saving energy, being expandability and the like. By optimizing resource allocation and netw