LMS algorithm, Adaline network, Intersymbol Interference Research Papers (original) (raw)

Resumen—El presente documento presenta el desarrollo de un filtro adaptativo que tiene la función de cancelar o eliminar el ruido mediante la implementación de una neurona adaline, la misma que se adapta a datos cambiantes de una señal indeseada, presentando a su salida una aproximación muy acertada, que puede ser restada de la señal original, produciendo una onda limpia, en este caso se implemento el sistema en Matlab, las señales utilizadas fueron una canción y una onda de ruido creada con código pseudoaleatorio.

Molecular communication (MC) aims to develop
a promising bio-inspired communication paradigm for nanotechnology,
in which molecules are used to encode, transmit,
and receive information. One of the main challenges in MC
is the intersymbol interference (ISI) caused by the nature of
the diffusion channel. The most popular solution to reduce the
effects of ISI in MC is to keep the symbol duration as long
as possible and reduce the number of molecules that can be
received in subsequent symbol durations. On the other hand, a
long symbol duration leads to a very low data rate, even for very
short distances. Furthermore, due to the size of the nano-scale
machines, production of energy becomes an essential problem.
In this paper, an ISI mitigation technique for diffusion-based
molecular communication channels, titled Molecular Transition
Shift Keying (MTSK) is proposed in order to increase the
data rate via suppressing the negative impact of the ISI on
communication quality. MTSK employs multiple molecule types
and the energy efficient extended version of MTSK with power
adjustment (MTSK-PA) makes use of the residual molecules in
the channel to reduce the ISI that would otherwise contribute to
the ISI. It is shown via computer simulations that both MTSK
and MTSK-PA outperforms the standard modulation techniques
proposed in the literature.

In numerous applications in communications and signal processing, it is necessary to remove noise and distortion from the signal received. These are mainly due to time-varying physical processes which are sometime unknown. The use of adaptive filtering that diminishes and eliminates the signal distortion completely. For adaptive filtering FIR filters are preferred because they are stable and no special adjustments are needed for implementation. The adaptation is done using the Adaline network which uses the LMS learning rule. This paper presents the LMS algorithm using Adaline network.