From the perspective of wireless transmission distance in the terahertz-band (THz-band), the 10-100 m, 100-1000 m, and ≥1000 m ranges are recommended for indoor communications, medium-distance communications, and long-distance communications, respectively. However, most existing photonic terahertz communication demonstrations reported in previous studies have a distance of less than 100 m, with some being even shorter at 10 m or 1 m. In this study, by combining high-gain and high-sensitivity terahertz modules, as well as signal processing technology, we have successfully demonstrated transmission over 850 m with a net rate of 50 Gbit/s at a 320 GHz carrier frequency through outdoor field experimental verification. This particular demonstration represents the most significant distance-rate product, as well as the longest wireless distance achieved so far for photonic terahertz wireless communication. The results demonstrate that the integration of terahertz photonics, electronics, and cutting-edge DSP algorithms, has the potential to enable long-distance and high-speed wireless communication for frequencies ≥300 GHz. Our work represents another significant stride towards achieving our goal of realizing the world's first photonic terahertz-wave communication over kilometer-level distances.

The transmission of a 65,536-ary quadrature amplitude modulation (QAM) orthogonal frequency division multiplexing (OFDM) signal supported by a hybrid fiber–terahertz (THz) multiple-input multiple-output (MIMO) system at 320 GHz is experimentally demonstrated in this Letter. We adopt the polarization division multiplexing (PDM) technique to double the spectral efficiency. Based on a 23-GBaud 16-QAM link, 2-bit delta-sigma modulation (DSM) quantization enables 65,536-QAM OFDM signal transmission over a 20-km standard single-mode fiber (SSMF) and a 3-m 2 × 2 MIMO wireless delivery, and satisfies the hard-decision forward error correction (HD-FEC) threshold of 3.8 × 10⁻³, corresponding to a net rate of 60.5 Gbit/s for THz-over-fiber transport. Meanwhile, below the fronthaul error vector magnitude (EVM) threshold of 0.34%, a maximum signal-to-noise ratio (SNR) of 52.6 dB is achieved. To the best of our knowledge, this is the highest modulation order achievable for DSM applications in THz communication.

We propose a digital-delta-sigma-modulation radio-over-fiber (DDSM-RoF) scheme for wireless fronthaul and validate it experimentally in a D-band photonics-aided RoF transmission system. The 10-Gbaud DDSM-RoF signal with a common public radio interface equivalent data rate (CPRI-EDR) of 55.8 Gb/s is successfully transmitted in a 130-GHz 4.6-km wireless channel. The spectral efficiency (SE) is 5.58 bit/s/Hz and the capacity-distance product reaches 257 Gb/s·km. Up to 34.4-dB recovered signal-to-noise ratio (SNR) is observed to support the 1024-quadrature-amplitude-modulation (1024-QAM) transmission. Compared with the digital-analog-RoF (DA-RoF) scheme, the proposed DDSM-RoF achieves an SNR improvement of 5.9 dB.

We experimentally realized a 320-GHz 320-Gbps/λ terahertz (THz) radio-over-fiber (RoF) system based on a photonics-aided scheme with the help of polarization-division multiplexing (PDM) technology and multiple-input, multiple-output (MIMO) transmission. In this system, the low-complexity MIMO single-carrier frequency-domain equalizer (SCFDE) is implemented to compensate for the polarization-related impairments of the PDM signal, and the demultiplexing performances between SCFDE and the commonly used constant modulus algorithm (CMA) are also compared in this proposed system. After 20-km standard single-mode fiber (SSMF) and 3-m 2 × 2 MIMO wireless link transmission, the bit error rate (BER) of the received 46-GBaud PDM 16-ary quadrature amplitude modulation (16QAM) signal satisfies the soft-decision forward error correction (SD-FEC) threshold with 15% overhead, which corresponds to a record-breaking net bit rate of 320 Gbit/s.

We propose and experimentally verify a photonics-aided W-band millimeter wave (MMW) radio-over-fiber (RoF) polarization-multiplexed envelope detection system for high-order quadrature amplitude modulation (QAM) signals. To solve the problem of low spectral efficiency of common public radio interface (CPRI) and severe distortion of high-order QAM of envelope detection, quantization noise suppressed delta-sigma modulation (DSM) is introduced into the system. The experimental results show that the system can transmit 131072 QAM signals when meeting the error vector magnitude (EVM) requirements of 5G new radio (NR), and transmit 1048576 QAM signals when meeting the soft decision threshold (SD@20%).

THz fiber-wireless technique can overcome the bandwidth bottleneck of electrical devices and has been popularized in different application scenarios. Furthermore, the probabilistic shaping (PS) technique can optimize both the transmission capacity and the distance, and has been extensively used in the optical fiber communication field. However, the probability of the point in the PS m-ary quadrature-amplitude-modulation (m-QAM) constellation varies with the amplitude, which leads to the class imbalance and degrades the performances of all supervised neural network classification algorithms. In this paper, we propose a novel complex-valued neural network (CVNN) classifier coupled with balanced random oversampling (ROS), which can be trained to restore the phase information simultaneously and overcome the class imbalance caused by PS. Based on this scheme, the fusion of oversampled features in complex domain increases the amount of the effective information of few classes, and thus improves the recognition accuracy effectively. It also has less requirement on the sample size than NN-based classifiers and largely simplifies the neural network architecture. By using our proposed ROS-CVNN classification method, single-lane 10 Gbaud 335 GHz PS-64QAM fiber-wireless transmission over 200 m free-space distance is experimentally realized, and the experimental results show that the efficient data rate is 44 Gbit/s considering the soft-decision forward-error-correction (SD-FEC) with 25% overhead. The results show that ROS-CVNN classifier outperforms the other real-valued NN equalizers and traditional Volterra-series by average 0.5 to 1 dB in receiver sensitivity at the bit error rate (BER) of 6 × 10⁻² magnitude. Therefore, we believe that the combination of ROS and NN supervised algorithms has an application prospect for the future 6 G mobile communication.