LAUSR.org creates dashboard-style pages of related content for over 1.5 million academic articles. Sign Up to like articles & get recommendations!

Multicell Power Control Under QoS Requirements With CNet

Photo by acfb5071 from unsplash

Multicell power control for sum rate maximization (SRM) is a widely-studied non-convex resource allocation problem in wireless communication systems. Due to the high complexity of the traditional mathematical algorithms, this… Click to show full abstract

Multicell power control for sum rate maximization (SRM) is a widely-studied non-convex resource allocation problem in wireless communication systems. Due to the high complexity of the traditional mathematical algorithms, this letter proposes a novel deep neural network (DNN) to solve the SRM problem with both QoS and per-base station power requirements. The main challenge here is how to ensure that the power control results learned through DNN always meet the QoS requirements with coupled variables. To overcome this challenge, we first transform the feasible set of the SRM problem into its V-representation, which can be used in the gradient-based DNN by polyhedron decomposition theorem. With the V-representation of the feasible set, we directly incorporate the constraints into the DNN to obtain the feasible power control solutions. Simulation results validate that the proposed DNN can always meet the requirements and achieve a sum-rate performance close to that optimized by the mathematical algorithm but with much lower computational complexity. Moreover, our proposal can also be generalized to solve non-convex problems with linear constraints.

Keywords: qos requirements; multicell power; power control; power

Journal Title: IEEE Communications Letters
Year Published: 2022

Link to full text (if available)


Share on Social Media:                               Sign Up to like & get
recommendations!

Related content

More Information              News              Social Media              Video              Recommended



                Click one of the above tabs to view related content.