Hybrid Parameter Estimation for Multi-Carrier Signals

Accurate estimation of unknown signal parameters, such as carrier frequency offset, timing offset, and channel state information, is crucial for reliable communication in multicarrier systems. This paper proposes a novel hybrid approach that combines the strengths of traditional signal processing and deep learning techniques to enhance parameter estimation accuracy and robustness. We leverage signal processing methods to extract informative features from the received multi-carrier signal, which are then fed into a deep neural network for parameter estimation. This hybrid strategy allows us to exploit the inherent strengths of each approach. We demonstrate the effectiveness of our proposed method through simulations and over-the-air experiments, showcasing its superior performance compared to conventional signal processing and deep learningonly methods. Our findings highlight the potential of hybrid approaches to significantly improve parameter estimation accuracy and robustness in multicarrier systems, paving the way for more efficient and reliable wireless communication.