ImConv_RNN: Improved Convolutional Recurrent Model Based Channel Estimation for 6G Networks
Raghavendra Kulkarni, Venkata Satya Suresh kumar Kondeti, Binu Sudhakaran Pillai, R Surendran
Abstract
This research introduces a novel deep learning for channel estimation for the 6G networks. The channel estimation for the 6G networks is employed using the proposed Improved Convolutional Recurrent (ImConv-RNN) model. In this, the loss function optimization is employed using the proposed ChSO algorithm. The chaotic mapping based on Chebyshev is included in the skill optimization assist the algorithm to enhance the convergence rate. The proposed ImConv-RNN model is evaluated based on RMSE, R2, MAE and MAPE acquired the values of 0.0136, 0.9584, 0.0114, and 5.8969 respectively.
Topics & Concepts
Recurrent neural networkComputer scienceChannel (broadcasting)Convolutional neural networkArtificial intelligenceComputer networkArtificial neural networkMillimeter-Wave Propagation and ModelingAdvanced MIMO Systems OptimizationAdvanced Wireless Communication Techniques