The Denoising Method with Variational Mode Decomposition for Signal-dependent Counting Noise in Molecular Communications

Molecular communication (MC) is considered to be one of the most feasible communication paradigms for constructing nanoscale networks within the human body, where chemical signals play significant roles. Compared with the traditional wireless communication method with electromagnetic signals, the mechanism of MC is credited to its biocompatibility. However, signal-dependent counting noise, as a type of non-stationary noise, greatly affects the detection performance of MC signals. To counteract its effect in the MC systems, we deploy the variational mode decomposition (VMD) method to recover the waveform of the original MC signals. To optimize the denoising performance, the root mean squared error and signal-to-noise ratio are used as the metrics for comparison. Finally, numerical results demonstrate the effectiveness of the VMD method for MC signals under various modulation orders, achieving significant noise reduction compared with other classical methods.