Millimeter-wave (mm-wave) communication using the E-band, i.e., the 71-76 and 81-86 GHz spectrum, can offer high capacity in the order of Gigabits per second (Gbps), thus becomes a candidate technology for wireless backhaul over sea. Compared to submarine fiber-optic cables, E -band backhaul systems are much more flexible to deploy and easier to maintain. Before designing such systems, we need to characterize E-band long-range propagation in over-sea environments. In the paper, based on channel measurements at 74 GHz conducted on a 12-km link across the sea near Sanya Bay, China, we study pathloss and fading properties which are crucial for wireless connectivity. Meteorological data were also recorded during the measurements, allowing us to analyze over-sea propagation in different weather conditions. Our experimental results show an average of 0.2-0.7 dB/km attenuation additional to free-space propagation loss. The additional attenuation can be largely affected by meteorological conditions above the sea surface, especially the amount of water vapour in the atmosphere. We also find that vertically-polarized waves have significantly lower propagation loss than horizontally-polarized. The pathloss model suggested by the International Telecommunication Union (ITU) Radio-communication Sector, known as the ITU-R model, well predicts the horizontally-polarized wave propagation loss but overestimates the vertical. On the 12-km across-sea link, severe fading was observed, although the transmitter and receiver were both firmly fixed. The variation of propagation attenuation can reach 15 dB in a very short time. The overall results may contribute to E-band channel modeling and future system design for wireless backhaul in oversea environments.

Existing deep learning-based methods for detecting fake news are uninterpretable, and they do not use external knowledge related to the news. As a result, the authors of the paper propose a graph matching-based approach combined with external knowledge to detect fake news. The approach focuses on extracting commonsense knowledge from news texts through knowledge extraction, extracting background knowledge related to news content from a commonsense knowledge graph through entity extraction and entity disambiguation, using external knowledge as evidence for news identification, and interpreting the final identification results through such evidence. To achieve the identification of fake news containing commonsense errors, the algorithm uses random walks graph matching and compares the commonsense knowledge embedded in the news content with the relevant external knowledge in the commonsense knowledge graph. The news is then discriminated as true or false based on the results of the comparative analysis. From the experimental results, the method can achieve 91.07%, 85.00%, and 89.47% accuracy, precision, and recall rates, respectively, in the task of identifying fake news containing commonsense errors.

Filter bank multi-carrier (FBMC) using offset quadrature amplitude modulation (OQAM) has recently attracted increasing interests for the next-generation high-speed optical transmission systems, due to the salient FBMC modulation advantages in terms of high spectrum efficiency by completely removing the cyclic prefix (CP). However, high peak-to-average power ratio (PAPR) still plays a key role of constraining the performance of FBMC systems. In this paper, we investigate a single sideband (SSB) pruned discrete Fourier transform (DFT)-spread FBMC for direct-detection (DD) passive optical network systems (PONs). It is demonstrated that the PAPR of the SSB pruned DFT-spread FBMC signal with the overlapping factor of 1.56 outperforms the conventional FBMC signal and the simple DFT-spread FBMC signal by ~2 dB and 1 dB at the complementary cumulative distribution function (CCDF) of 1×10 $^-3$ , respectively. Finally, by using the pruned DFT-spread FBMC with low PAPR, the aggregate data rate of 50-Gbit/s can be achieved after transmission over 50-km single mode fiber (SMF) below the 7%hard decision forward-error correction (HD-FEC) threshold in the DD PONs.

We propose a novel zero guard band (ZGB) multi-twin-SSB bidirectional passive optical network system which combines the multiband carrier-less amplitude phase and twin single sideband (SSB) modulation techniques. The multi-twin-SSB signal is generated at the downlink transmitter side with zero guard band between the left-sideband and right-sideband. A weak local oscillator which is extracted from the uplink optical source aids the downstream signal detection. Kramers-Kronig algorithm is used at both downlink and uplink receiver sides to reconstruct the SSB signal and eliminate the signal-signal beat interference. The downstream and upstream signals are transmitted in a single fiber without any spectral overlap between them and the Rayleigh backscattering noise is removed by the optical band-pass filter. The proposed scheme is extensively investigated and analyzed with four sub-bands using 50 Gbps 16-QAM modulation for each sub-band over a bidirectional 50 km transmission. The results show that, compared with the conventional multi-twin-SSB scheme, the proposed ZGB multi-twin-SSB scheme can improve the spectral efficiency by about 11%. The required optical signal to noise ratio (OSNR) and the required received optical power (ROP) can be reduced by more than 15 dB and 7 dB respectively at the 7% hard decision forward error correction (HD-FEC) threshold (bit error rate of 3.8×10-3), respectively. For the uplink transmission, the required ROPs for the inner and outer sub-bands at 7% HD-FEC are about -10 dBm and -11 dBm, respectively. And the signal to noise ratio to the corresponding ROPs for the inner and outer sub-bands are 15.4 dB and 15.3 dB, respectively.