Resource Allocation for Multi-Channel Underlay Cognitive Radio Network Based on Deep Neural Network

By Woongsup Lee. 2018

99 citations recorded by the source corpus.

In this letter, a resource allocation strategy based on a deep neural network (DNN) is proposed for multi-channel cognitive radio networks, where the secondary user (SU) opportunistically utilizes channels without causing excessive interference to the primary user (PU). In the proposed scheme, the allocation of transmit power in each channel for SUs is found by utilizing the newly proposed DNN model, which separately determines the overall transmit power of individual SUs and the proportion of transmit power allocated to each channel. Both the spectral efficiency (SE) of the SU and the amount of interference caused to the PU are considered in the training of the DNN model, such that the interference caused to the PUs can be properly regulated while the SE of the SU is improved. Through simulations, we show that our scheme enables a high SE of the SU to be achieved while the interference caused to the PU can be maintained at less than the threshold.

Subfields: Artificial Intelligence, Computer Networks and Communications, Electrical and Electronic Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Sang Hyun Lee
2. Tony Q. S. Quek
3. Hoon Lee

In Computer Networks and Communications

1. Sang Hyun Lee
2. Tony Q. S. Quek
3. Hoon Lee

In Electrical and Electronic Engineering

1. Tony Q. S. Quek
2. Sang Hyun Lee
3. Hoon Lee

Citations in the corpus

• Deep Learning for Distributed Optimization: Applications to Wireless Resource Management (2019, methodology): “To resolve this issue, a penalizing strategy is applied to transform the original training task to an unconstrained training one by augmenting a penalty term associated with the IT constraint into the network utility function [10], [11].”
• Constrained Deep Learning for Wireless Resource Management (2019, methodology): “As a result, unlike the conventional penalizing strategies in [5] and [6] which require additional hyper parameter identification process, the feasibility of the proposed training algorithm is guaranteed by jointly optimizing the”