Abstract
When designing fast and reliable molecular communication (MC) systems, some important constraints must be taken into consideration, including their power consumption, space, and ability to perform complex computational operations. Accordingly, there is a need for fast and suboptimal detection schemes to identify received bits under noise and inter-symbol-interference (ISI) distortions. In this paper, we propose and investigate a novel detection scheme that achieves a low uncoded bit error rate (BER) via a multi-layered approach in which we utilize entropy-based discrimination followed by small-size (N bits) low-complexity correlation operations. As shown by our simulation evaluation results, we have been able to achieve a low level of BER as 10 −5 versus a low level of the signal to noise ratio (SNR) as low as 8.5 dB without using error-correcting coding, compared with the traditional Fixed-Threshold Detection (FTD) approach.