An optimal decoding algorithm for Molecular Communications systems with noise, memory, and pulse width

• A new optimal decoding algorithm mitigates the effect of channel memory and noise.
• Channel memory, noise, and emission pulse width are considered within the model.
• Performance with regards to the bit error rate and capacity is shown.
• Distance measuring schemes are a critical part of performance analysis.

Molecular Communications (MC) is a promising paradigm to achieve message exchange between nano-machines. Due to the specific characteristics of MC systems, the channel noise and memory significantly influence the MC system performance. Aiming to mitigate the impact of these two factors, an adaptive decoding algorithm is proposed by optimising the symbol determination threshold. In this paper, this novel decoding scheme is deployed onto a concentration-based MC system with the transmitter emission process considered. To evaluate the performance, an information theoretical approach is developed to derive the Bit Error Rate (BER) and the channel capacity. Simulations are also carried out to verify the accuracy of these formulations, to compare the performance difference against other decoding schemes, and to illustrate the performance deviation caused by different designing of relevant parameters. Furthermore, the performance of MC systems with the distance unknown is also analysed. Comparisons between distance-pre-known and distance-unknown systems are provided.