Frequency domain analysis and equalization for molecular communication

Molecular Communication (MC) is a promising micro-scale technology that enables wireless connectivity in electromagnetically challenged conditions. The signal processing approaches in MC are different from conventional wireless communications as molecular signals suffer from severe inter-symbol interference (ISI) and signal-dependent counting noise due to the stochastic diffusion process of the information molecules. One of the main challenges in MC is the high computational complexity of the existing time-domain ISI mitigation schemes that display a third-order polynomial or even exponential growth with the ISI length, which is further exasperated under the high symbol rate case. For the first time, we develop a frequency-domain equalization (FDE) with lower complexity, capable of achieving independence from the ISI effects. This innovation is grounded in our characterization of the channel frequency response of diffusion signals, facilitating the design of receiver sampling strategies. However, the perfect counting noise power is unavailable in the optimal minimum mean square error (MMSE) equalizer. We address this issue by exploiting the statistical information of the transmit signal and decision feedback for noise power estimation, designing novel MMSE equalizers with low complexity. The FDE for MC is successfully developed with its immunity to ISI effects, and its signal processing cost has only a logarithmic growth with symbol length in each block.