Dynamic Power Allocation Algorithms Using ML in NOMA
Dynamic Power Allocation Algorithms Using ML in NOMA
Aim: To propose a dynamic power allocation algorithm to maximize sum rate and
energy efficiency using Machine Learning in NOMA system.
Scenario 1:
Users are sorted according to the distance from the base station and are divide into
two groups (far user’s group (1) and near user’s group (2)) and now each user from
group 1 will be paired with group 2 so many pairs will be formed. In each pair the
far user will be given more power than the near user by a machine learning
algorithm which predicts the power coefficient and evaluates the performance and
then based on the performance it readjusts the power coefficients until the optimal
is arrived..… A fixed power allocation strategy in the multi-user scenario is
proposed to highlight the performance advantages of the dynamic power allocation
strategy via comparison.
Inspiration or Reference
To propose a novel Power allocation algorithms for HetNets with downlink Non
Orthogonal Multiple Accesses (NOMA).
Inspiration or Reference
Scenario 3:
A novel Dynamic power allocation for SWIPT NOMA with deep learning
approach
Inspiration or Reference