Optical amplification-free deep reservoir computing-assisted high-baudrate short-reach communication

Area of Science:

• Optical communications
• Signal processing
• Machine learning

Background:

• High-baudrate intensity modulation direct detection (IM/DD) systems are crucial for intra-data center communications.
• Existing systems often require optical amplification, increasing complexity and cost.
• Reservoir computing (RC) offers a promising approach for signal equalization in optical systems.

Purpose of the Study:

• To experimentally demonstrate an optical amplification-free deep reservoir computing (RC)-assisted high-baudrate IM/DD system.
• To evaluate the performance of deep RC compared to shallow RC and decision feedback equalization (DFE) in mitigating impairments.
• To assess the feasibility of the proposed system for short-reach optical communication links.

Main Methods:

• Utilized a 100G externally modulated laser in the C-band for signal generation.
• Transmitted 112 Gbaud 4-level pulse amplitude modulation (PAM4) and 100 Gbaud 6-level PAM (PAM6) signals.
• Employed DFE, shallow RC, and deep RC for signal equalization over a 200-m single-mode fiber (SMF) link without optical amplification.

Main Results:

• Achieved bit error rate (BER) below the 6.25% overhead hard-decision forward error correction (HD-FEC) threshold for both PAM4 and PAM6 transmissions.
• The BER for PAM4 signals fell below the KP4-FEC limit after 200-m SMF transmission with RC assistance.
• Deep RC demonstrated comparable performance to shallow RC with approximately 50% fewer weights due to its multi-layer structure.

Conclusions:

• The optical amplification-free deep RC-assisted high-baudrate IM/DD system is a viable solution for intra-data center communications.
• Deep RC provides an efficient method for impairment mitigation in high-baudrate optical links.
• This technology holds promise for future high-capacity, cost-effective data center networks.