Towards 250-m gigabits-per-second underwater wireless optical communication using a low-complexity ANN equalizer

Area of Science:

• Underwater wireless optical communication (UWOC)
• Optical engineering
• Signal processing

Background:

• Underwater wireless optical communication (UWOC) is crucial for transmitting data over long distances underwater.
• Aquatic attenuation and nonlinear impairments severely limit UWOC system performance, particularly data rates.
• Existing methods struggle to balance extended range and high-speed data transmission.

Purpose of the Study:

• To develop a robust UWOC system capable of high-speed data transmission over extended underwater distances.
• To mitigate nonlinear impairments that degrade signal quality in UWOC systems.
• To achieve 1 Gbps data transmission over 250 meters, a significant advancement in UWOC technology.

Main Methods:

• Utilized a high-power transmitter combining 8-channel cascaded laser diodes (LD).
• Employed a sensitive receiver using a silicon photomultiplier (SiPM).
• Implemented a hybrid equalizer combining a linear equalizer with a low-complexity artificial neural network (ANN) equalizer (9-input, 2-hidden layers).

Main Results:

• Achieved 1 Gbps data transmission over a 250-meter UWOC link, a record for this distance.
• Reduced bit error rate (BER) to 3.4 × 10^-3 at 1 Gbps, below the HD-FEC limit.
• The ANN-based system significantly outperformed linear (500 Mbps) and Volterra (750 Mbps) equalizers.

Conclusions:

• The proposed hybrid linear and ANN equalizer effectively overcomes nonlinear impairments in long-distance UWOC.
• This system represents a breakthrough in achieving Gbps-level data rates for underwater optical communication beyond 250 meters.
• The findings pave the way for enhanced underwater communication networks.