Low-Complexity Adaptive Threshold Detection for Molecular Communication.

This paper studies detection algorithms for diffusion-based molecular communication systems, where molecules freely diffuse as information carrier from a transmitter to a receiver in a fluid medium. The main limitations are strong intersymbol interference due to the random propagation of the molecules, and the low-energy/low-complexity assumption regarding future implementations in so-called nanomachines. In this contribution, a new biologically inspired detection algorithm suitable for binary signaling, named adaptive threshold detection, is proposed, which deals with these limitations. The proposed detector is of low complexity, does not require explicit channel knowledge, and seems to be biologically reasonable. Numerical results demonstrate that the proposed detector can outperform the common low-complexity fixed threshold detector under certain conditions. As a benchmark, maximum-likelihood sequence estimation (MLSE) and reduced-state sequence estimation (RSSE) are also analyzed by means of numerical simulations. In addition, the effect of molecular denaturation on the detection performances is studied. It is shown that denaturation generally improves the detection performances, while RSSE is able to outperform MLSE in the case of no denaturation.