The proliferation of the greedy heterogeneous wireless network demands the ability to handle the sharply increasing resource requirements of power and spectrum. Thus, the study of novel and effective methods to allocate power and spectrum resources in order to reduce energy consumption for heterogeneous devices has become an urgent issue. In this context the article proposes a DJRA in which data fusion, combined with NOMA, SIC and beamforming techniques, are utilized to allocate power and spectrum resources to heterogeneous devices with a reduced energy consumption, where different radio access technologies are integrated into a cellular network for communication between devices. The data correlations between the devices are computed to group the devices so as to reduce data transmission. In addition, NOMA, SIC and beamforming are used in combination to reduce network interference between BSs and the heterogeneous devices. The main contribution of this article is the design of a large-scale heterogeneous wireless network paradigm to improve the percentage of resource utilization also reducing the energy consumption.