Channel estimation in ACO-OFDM employing different transforms for VLC

• Published in: International journal of electronics and communications Vol. 84; pp. 111 - 122
• Main Authors: Vappangi, Suseela, Vakamulla, Venkata Mani
• Format: Journal Article
• Language: English
• Published: Elsevier GmbH 01.02.2018
• Subjects: Channel Estimation; CRLB; Discrete Hartley Transform (DHT); Fast-Walsh Hadamard Transform (FWHT); Hadamard Coded Modulation (HCM); Visible Light Communication (VLC); Channel Estimation; Visible Light Communication (VLC); Hadamard Coded Modulation (HCM); CRLB; Discrete Hartley Transform (DHT); Fast-Walsh Hadamard Transform (FWHT)
• ISSN: 1434-8411; 1618-0399
• DOI: 10.1016/j.aeue.2017.11.016

The pronounced eminency of assuring simultaneous illumination and communication has driven Visible Light Communication (VLC) to gain significant ubiquity in recent times. This paper proposes comb type pilot arrangement based channel estimation for Asymmetrically Clipped Optical OFDM (ACO-OFDM) and different multicarrier transmission systems like Discrete Hartley Transform (DHT)-based ACO-OFDM and Fast-Walsh Hadamard transform (FWHT)-based Hadamard Coded Modulation (HCM) over dispersive VLC channel. Various channel estimation algorithms like Least Square (LS), Minimum Mean Square Error (MMSE), and Interpolation techniques namely Linear, Spline, and Low-Pass are evaluated and compared for the aforesaid systems employing different orders of constellation. Here, an elaborate mathematical analysis is accomplished and Cramer Rao Lower Bound (CRLB) is derived for the channel estimation error. Simulated results emphasize that, ACO-OFDM and DHT-based ACO-OFDM have improved Bit Error Rate (BER) performance than HCM at lower Signal to Noise Ratio (SNR), while at higher SNRs HCM dominates the former. Furthermore, the simulated results evidences that, in all multicarrier systems MMSE algorithm has reducible probability of error than LS because, at higher SNRs LS is more susceptible to noise and Spline Interpolation outperforms both LS and MMSE. The simulated results are validated analytically demonstrating good agreement.