Visible Light Communication with Input-Dependent Noise: Channel
Estimation, Optimal Receiver Design and Performance Analysis
Abstract
In this paper, single-input single-output (SISO) visible light
communication (VLC) subject to signal-dependent shot noise (SDSN) is
investigated. We discuss both channel estimation and data transmission.
In the former, we introduce least square (LS) and maximum likelihood
(ML) estimators. Moreover, we derive the Cram´er–Rao lower bound (CRLB)
of the channel estimation error. With regard to data transmission, we
present optimal and sub-optimal receiver designs and discuss their bit
error rate (BER) performance. In particular, a closed-form expression of
the BER is derived for a sub-optimal receiver using the on-off-keying
(OOK) modulation technique. An approximated expression is then derived
for the optimal receiver. Our analysis shows that the CRLB performance
does not have a linear relation with the SDSN, thermal noise, or the
fading channel. On the other hand, the SDSN has a dramatic effect on the
channel estimation error bound, and it can degrade the BER performance.
Increased performance degradation can also be noticed due to the joint
effects of the channel estimation error and SDSN.