K-Means Clustering-Aided Non-Coherent Detection for Molecular Communications

In this paper, we consider non-coherent detection schemes for molecular communication systems in the presence of inter-symbol-interference. particular, study detectors based on memory-bits-based thresholds order to achieve low bit-error-ratio (BER) transmission. The main challenge realizing is obtain channel state information only received signals. We tackle issue by reformulating through intermediate variables, which can be obtained clustering multi-dimensional data from signals, and using K-means algorithm. addition estimating thresholds, show that transmitted bits retrieved clustered data. To reduce errors, propose iterative methods one-dimensional data, are shown BER. Simulation results presented verify effectiveness proposed methods.