We experimentally demonstrate the LWIR FSO communication with advanced modulation formats. Up to 8.4 Gbit/s PAM8 and 5.5 Gbit/s DMT transmission is achieved with 9.15-μm directly-modulated quantum cascade laser and MCT detector at room temperature.

Next-generation communication networks require the integration of traditional and high-frequency bands to support diverse applications, positioning the optical heterodyne approach as a promising candidate for remote radio head (RRH) signal generation. However, deploying point-to-point (PtP) analog links based on such an optical heterodyne approach for the radio access networks (RAN) is challenging. Particularly, widespread deployment is expected to be hindered by practical communication resource constraints. In this context, we propose an arrayed waveguide grating router (AWGR)-based RAN architecture incorporating optical heterodyne signal generation, enabling adaptive frequency adjustment and scalable resource management. We detail the architecture and establish its generalizability through mathematical derivation. For a representative 64×64 AWGR-based system, the proposed architecture could reduce the required number of local oscillator (LO) lasers to 1/8 of conventional approaches, with potential for further reduction via scheduling. Experimental validation at both 28 GHz (mmWave) and 286 GHz (THz) bands demonstrates physical feasibility, achieving bit-error rates (BER) of 3.06×10−5 and 1.53×10−4, respectively. This AWGR-based architecture offers a practical path towards resource-efficient, high-agility fiber-wireless converged access.

Integrated sensing and communication (ISAC) is a key pillar for future wireless networks, demanding solutions that simultaneously deliver high-capacity communication and highaccuracy sensing. The terahertz (THz) band, particularly when integrated with fiber-optic networks, emerges as a highly promising candidate, offering the potential for terabit-per-second communication data rates and millimeter-level sensing resolution. Crucially, the integrated waveform is paramount to efficiently and seamlessly combine these dual functionalities, enabling compact frontends design and streamlined baseband processing for photonic THz-ISAC systems. This article presents our recent system-level investigations into diverse integrated ISAC waveform designs. We further synthesize and discuss the latest global research and development progress in advancing this rapidly evolving field.

Chaotic systems exhibit complex spatiotemporal dynamics that are notoriously difficult to characterize, limiting their applications in fields such as secure communications. To address this challenge, we propose the Chaotic SpatioTemporality-Informed Neural Network (CSTI-NN), a physics-conscious artificial intelligence (AI) framework designed to directly capture the intrinsic spatiotemporality of high-dimensional chaos. By deconstructing and deciphering the Ikeda delay dynamics, our approach successfully reveals previously hidden chaotic invariants that govern correlation scaling. The inherent stability of these invariants against noise directly endows the framework with its enhanced noise-resistant capabilities. In the experiment of the optical chaotic communication, this method achieves over 90% accuracy in reconstructing chaos from noisy signals, surpassing the performance of conventional methods by ∼ 3 dB in signal-to-noise ratio (SNR) tolerance. This endogenous spatiotemporality enables high-quality synchronization, leading to a 10-fold reduction in the bit error rate (BER) at a rate of 5 Gbaud. This work establishes a new paradigm in which AI can not only predict complex behaviors but also fundamentally uncover the underlying physics of chaos, opening new avenues for the development of noise-immune information systems

We demonstrate record unamplified transmissions of 256 Gbaud OOK, 155 Gbaud PAM4 and 128 Gbaud PAM6 using a C-band differential-drive SiP TW-MZM, achieving BER performance below the 6.25% HD-FEC threshold after 100 m SMF transmission.

Terahertz (THz, 0.3–10 THz) radar systems have garnered significant attention due to their superior capabilities in high-precision and robust sensing. However, the susceptibility to jamming, along with the sensing precision loss and ranging ambiguity induced by inflexible implementation of the conventional radar signal source, presents major challenges to the practical deployment of THz radars. Herein, a flexible photonic chaotic radar system is proposed at the THz band and investigate the ranging performance in precision and ambiguity. The photonic heterodyne detection scheme facilitates the generation of optoelectronic feedback loop-based THz chaos at 300 GHz, achieving a seamless connection between THz domains and optical domains. The system is experimentally demonstrated its superior performance of sub-centimeter resolution with 0.9345 cm and ranging unambiguity simultaneously. This work bridges the THz gap in the practical deployment of chaos theory and will pave the way for a new regime of THz radar empowered by chaos.

Terahertz (THz) waves play a crucial role in numerous imaging applications, from industrial inspection to medical diagnostics. However, most current THz imaging techniques operate in the pulsed mode, remaining entrenched in the “super-resolution versus practical-level frame rate” dilemma. In this work, we present system-level rapid, super-resolution, and large-field-of-view THz compressive imaging. In the system, by implementing high-speed sampling with low phase noise through CW radiation based on an optical frequency comb, and exploring a VO<inf>2</inf>/mica film THz spatial modulator with advanced trigger threshold (0.37 mJ/cm2) and higher modulation depth (80.4%), we simultaneously break the long-standing trade-off between spatial resolution and temporal speed in THz imaging systems. We experimentally achieve rapid super-resolution THz compressed sensing (CS) imaging with a super-resolution (<λ/65) at video-frame rates (2 fps) over a large field of view of 100 mm2. This work represents a significant step toward application-oriented THz imaging, and holds tremendous promise in biomedical diagnostics, nondestructive testing, and pharmaceutical industries

We present high-speed transmission results using two of the most promising silicon photonic (SiP) modulators towards 400 Gbps per single wavelength in intensity modulation/direct detection (IM/DD) systems. A GeSi electro-absorption modulator (EAM) enables net data rates beyond 300 Gb/s, with successful transmission of 160 Gbaud PAM4 over 100 meters of single-mode fiber, achieving bit error rates (BER) below the 6.25% overhead hard-decision forward error correction (HD-FEC) threshold of 4.5×103. Similarly, a SiP differential-drive C-band traveling-wave Mach-Zehnder modulator (TW-MZM) supports 175 Gbaud and 160 Gbaud PAM4, also meeting the HD-FEC requirement after 100 meters of SMF transmission. Results on 256 Gbaud OOK are also included for comparison purposes. While further improvements are needed, particularly in reducing optical losses, managing self-heating, and improving driver integration, these modulators represent leading candidates for pushing IM/DD links toward the 400 Gbps per wavelength target in future short-reach optical interconnects.

This paper explores photonic-based analog fronthaul solutions for 6G, highlighting their effectiveness in meeting the RF requirements of standards, supporting future distributed-MIMO networks, and providing insights into prospective solutions for radios in potential 6G bands.

The radio-over-fiber (RoF) system is promising to support broadband transmission and increased flexibility. To boost channel capacity in multi-carrier RoF systems with variable-rate forward error correction, probabilistic shaping and water-filling-based entropy loading outperforms bit-power loading in terms of achievable information rate. However, its reliance on specific channel conditions limits practical use in channel-dynamic RoF systems, highlighting the need for adaptive entropy loading that requires minimal channel state information. This paper presents a deep neural network-based transfer learning model for adaptive entropy prediction in discrete multi-tone signals, addressing frequency-selective responses in RoF systems. Numerical and experimental results confirm capacity-approaching generalized mutual information (GMI) and smoother normalized GMI (NGMI) performances, consistently achieving the 0.83 NGMI threshold across subcarriers. Unlike traditional methods requiring pre-measured signal-to-noise ratios (SNR), this approach simplifies implementation by using only demodulated data and the received SNR, providing a more channel-independent entropy loading option in dynamic RoF systems.