This paper introduces novel approaches for the design of the linear filter and the detection algorithms in unbounded advection diffusion-based molecular communication systems affected by inter-symbol interference. The received signal samples are modeled as Poisson random variables with memory where the effect of enzymatic reactions is also included. A main characteristic of the new filter design is to allow for a real-time computation of the filter's coefficients. For the detection of the transmitted symbols, we define an averaging method suitable for time-varying channels with finite memory length. In this paper the mean value of the Poisson channel is varying with time and we quantify the memory length with a finite number, from receiver point of view. The computational burden of the proposed approaches is evaluated in terms of number of required operations and their performance is evaluated in terms of bit error rate for different sets of parameters. We show that the proposed design of the filter and of the detection algorithms enables us to achieve a performance comparable to that of the state of the art approaches in relation to their simplicity.
Low complexity receiver design for time-varying Poisson molecular communication channels with memory
Vakilipoor F.;Ratti F.;Awan H.;Magarini M.
2021-01-01
Abstract
This paper introduces novel approaches for the design of the linear filter and the detection algorithms in unbounded advection diffusion-based molecular communication systems affected by inter-symbol interference. The received signal samples are modeled as Poisson random variables with memory where the effect of enzymatic reactions is also included. A main characteristic of the new filter design is to allow for a real-time computation of the filter's coefficients. For the detection of the transmitted symbols, we define an averaging method suitable for time-varying channels with finite memory length. In this paper the mean value of the Poisson channel is varying with time and we quantify the memory length with a finite number, from receiver point of view. The computational burden of the proposed approaches is evaluated in terms of number of required operations and their performance is evaluated in terms of bit error rate for different sets of parameters. We show that the proposed design of the filter and of the detection algorithms enables us to achieve a performance comparable to that of the state of the art approaches in relation to their simplicity.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.