Performance-enhanced gigabit/s MIMO-OFDM visible light communications using CSI-free/dependent precoding techniques.

Area of Science:

• Optical Communications
• Signal Processing

Background:

• Visible Light Communication (VLC) systems offer high bandwidth potential.
• Multiple-Input-Multiple-Output Orthogonal Frequency Division Multiplexing (MIMO-OFDM) is crucial for enhancing VLC capacity.
• Reducing Bit Error Rate (BER) is essential for reliable high-speed VLC.

Purpose of the Study:

• To propose and evaluate digital signal processing (DSP) techniques for BER reduction in MIMO-OFDM VLC systems.
• To compare performance with and without Channel State Information (CSI).
• To demonstrate the effectiveness of proposed methods in gigabit-per-second VLC systems.

Main Methods:

• Implementation of the Orthogonal Circulant Matrix Transform (OCT) technique.
• Application of Singular Value Decomposition (SVD)-based adaptive loading.
• Demonstration using a 2x2 MIMO-OFDM VLC system with ~100-MHz bandwidth.
• Testing with both symmetrical and asymmetrical MIMO configurations.

Main Results:

• Both OCT and SVD-based adaptive loading techniques achieved significant BER reduction.
• SVD-based adaptive loading provided superior performance but required CSI.
• OCT technique offered substantial BER reduction without requiring CSI.
• BER reductions of over one and two orders of magnitude were achieved with OCT and SVD, respectively.

Conclusions:

• The proposed DSP techniques effectively reduce BER in MIMO-OFDM VLC systems.
• SVD-based adaptive loading is optimal when CSI is available.
• OCT technique provides a viable, CSI-free alternative for BER improvement.
• These methods are robust across different transceiver geometrical distributions.