Estimating the Q Factor of Reverberation Chambers Using LSTM Network

Based on the long-short term memory (LSTM) network, a novel approach to obtain the quality factor ( Q factor) of reverberation chambers (RCs) is proposed and validated in this article. We compare the predicted results obtained from the LSTM network with that from the conventional coherence bandwidth method and give their corresponding relative errors. Results demonstrate that the LSTM-based Q factor estimation method can be applied for RCs with different dimensions and loadings. The proposed network shows advantages for estimating the Q factor of RCs, particularly when the number of stirrer positions is limited.