An all-fiber approach to generate phase-coded microwave signals, based on spectral shaping and a frequency-to-time conversion, is proposed and experimentally demonstrated. A phase modulator cascaded by a filter module that has a sinusoidal spectral response is used to create a frequency-shifted optical spectrum. Microwave pulses with the same shape as the optical spectrum are obtained through frequency-to-time conversion in a dispersive fiber. The phase of the generated radio frequency signals can be continuously shifted by tuning the applied voltage on the optical phase modulator.

The measurement accuracy of a Brillouin optical time domain analysis (BOTDA) fiber sensor is determined by the signal-to-noise ratio (SNR) of the received sensing signal. Here, a new hybrid aperiodic coding method is proposed to improve the SNR. In the proposed method, two pre-discovered short seed aperiodic codes (SA-codes) are used to construct a new hybrid aperiodic code (HA-code) in a nested way. The HA-code inherits the good denoising capabilities of the two SA-codes and features a high coding gain. In the proof-of-concept experiment, a SNR improvement up to 8 dB is obtained, which improves the measurement certainty to 1.67 MHz over a 117.46 km sensing range under a spatial resolution of 2.6 m.

All-fiber filters with multiple narrow, flat-top channels are designed and fabricated, by incorporating both the phase sampling technique and multiple π phase shifts into a fiber Bragg gratings. The phase sampling yields multiple reflection bands with high uniformity, while the combination of multiple phase shifts allows a narrow, flat-top transmission channel in each reflection band identically, leading to multiple narrow, flattop channels. Such filters with nine channels are then fabricated and characterized with excellent comprehensive specifications, including a 3-dB bandwidth less than 800 MHz, a rectangular shape factor close to 2.5, an insertion loss less than 1 dB, and an out-of-band rejection greater than 25 dB.

Firstly, Erbium-doped fiber amplifier (EDFA) served as pre-amplifier may distort Brillouin gain spectrum (BGS) in Brillouin optical time domain analysis sensor. Here, we found that the EDFA has negligible impact on the shape of Brillouin phase spectrum (BPS). Experimental results show that a ~5.42MHz error of Brillouin frequency shift caused by EDFA has been avoided by using BPS instead of BGS. Secondly, after eliminating phase fluctuation caused by optical fiber, the combination of Golay coding and coherent detection has been realized.

Traditional portfolio theory divides stocks into different categories using indicators such as industry, market value, and liquidity, and then selects representative stocks according to them. In this paper, we propose a novel portfolio learning approach based on deep learning and apply it to China’s stock market. Specifically, this method is based on the similarity of deep features extracted from candlestick charts. First, we obtained whole stock information from Tushare, a professional financial data interface. These raw time series data are then plotted into candlestick charts to make an image dataset for studying the stock market. Next, the method extracts high-dimensional features from candlestick charts through an autoencoder. After that, K-means is used to cluster these high-dimensional features. Finally, we choose one stock from each category according to the Sharpe ratio and a low-risk, high-return portfolio is obtained. Extensive experiments are conducted on stocks in the Chinese stock market for evaluation. The results demonstrate that the proposed portfolio outperforms the market’s leading funds and the Shanghai Stock Exchange Composite Index (SSE Index) in a number of metrics.

A photonic microwave frequency down-conversion scheme with large image rejection ratio (IRR) is implemented. In experiments, the RF signal at 35.5 GHz is successfully down-converted to a 500-MHz one with an IRR beyond 60 dB.