<p>The next generation of&#xa0;heterogeneous cellular networks is intended for a wide range of applications and scenarios. Unlike past generations, where network users typically had similar requirements, the current generation consists of diverse consumers with varying degrees of services and, consequently, distinct priorities. In this article, assuming cross-layer implementation, the physical layer prioritizes each user based on crucial characteristics of other layers or the network situation. This paper examines the issue of resource allocation to users in heterogeneous network downlinks. The service priority levels of users are transformed into corresponding weight values. The problem is formulated as priority-based cell and subband allocation using a load-balancing approach, while also effectively preventing severe inter-cell interference. The optimization problem is presented which is a non-linear integer programming and NP-hard to solve. After reformulating the problem, a distributed algorithm has been developed by applying the dual decomposition method and incorporating admission control. The proposed method is designed to be distributed, meaning that each user and base station only needs access to its own local data in order to determine its connection for service, which is more suited for real heterogeneous networks. The simulation results demonstrate that the suggested algorithm significantly enhances network utility against conventional methods while dynamically adapting to user priorities for resource allocation.</p>