Open AccessCyclic gradient descent optimisation for joint MAP estimation of channel and phase noise in orthogonal frequency division multiplexing
Author(s) -
Shrivastav Kamayani,
Yadav Rajendra Prasad,
Jain Kamal Chand
Publication year - 2018
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2017.0732
Subject(s) - joint (building) , computer science , gradient descent , division (mathematics) , channel (broadcasting) , orthogonal frequency division multiplexing , multiplexing , noise (video) , algorithm , telecommunications , mathematics , artificial intelligence , engineering , arithmetic , artificial neural network , architectural engineering , image (mathematics)
This study addresses the statistically optimal channel estimation for orthogonal frequency division multiplexing systems in the presence of phase noise (PHN). The maximum a posteriori (MAP) cost function for the joint estimation of channel transfer function (CTF) and PHN is presented, which utilises the prior statistical knowledge of PHN spectral components without the assumption of small PHN. The frequency‐domain estimation of unknown frequency selective fading makes the method simpler, compared with the estimation of channel impulse response in time domain. An iterative cyclic gradient descent optimisation algorithm is derived to minimise the cost function over the unknown CTF and PHN. Cramer‐Rao lower bound is analysed and compared with the simulation results to illustrate that the proposed joint MAP estimation can achieve near optimum performance.