We proposed and demonstrated a novel tilted fiber Bragg grating (TFBG)-based surface plasmon resonance (SPR) label-free biosensor via a special boronic acid derivative to detect glycoprotein with high sensitivity and selectivity. TFBG, as an effective sensing element for optical sensing in near-infrared wavelengths, possess the unique capability of easily exciting the SPR effect on fiber surface which coated with a nano-scale metal layer. SPR properties can be accurately detected by measuring the variation of transmitted spectra at optical communication wavelengths. In our experiment, a 10° TFBG coated with a 50 nm gold film was manufactured to stimulate SPR on a sensor surface. To detect glycoprotein selectively, the sensor was immobilized using designed phenylboronic acid as the recognition molecule, which can covalently bond with 1,2- or 1,3-diols to form five- or six-membered cyclic complexes for attaching diol-containing biomolecules and proteins. The phenylboronic acid was synthetized with long alkyl groups offering more flexible space, which was able to improve the capability of binding glycoprotein. The proposed TFBG-SPR sensors exhibit good selectivity and repeatability with a protein concentration sensitivity up to 2.867 dB/ (mg/mL) and a limit of detection (LOD) of 15.56 nM.

Optical multipath RF self-interference cancellation (SIC) based on phase modulation for full-duplex communication is proposed and demonstrated experimentally. Phase modulation is utilized to convert the RF signal into optical domain, in which the time delay tuning, amplitude tuning and phase inversion for multipath RF SIC are completed. The comprehensive theoretical model of the optical multipath RF SIC system is established, and the factors affecting SIC performance including the time delay, amplitude and phase deviations are analyzed. The experimental results verify the feasibility of the proposed scheme for full-duplex communication with the cancellation depth of 26 dB and 28 dB over 100 MHz at central frequency of 6 GHz and 10 GHz, respectively. A figure of merit of the maximum interference to signal of interest ratio is defined to characterize the SOI recovery capability of optical RF SIC system.

The statistical properties of semiconductor laser chaos are of great importance for practical applications. By exploring the injection parameter space of an optically injected semiconductor laser, the associated probability density functions (PDFs) are numerically investigated for generation of chaotic waveforms with rogue waves (RWs) or Gaussian distributions. First, generation of RWs are demonstrated by considering chaotic waveforms with long-tailed PDFs. The dynamical behaviors for RW generations are systematically investigated by exploring the entire injection parameter space, where a pattern is identified for parameter regions of nonzero RWs. The occurrence of RWs is also investigated for different injection parameters by evaluating the effects of laser bias current, spontaneous emission noise, and detection bandwidth. Alternatively, Gaussian PDFs are generated by bandwidth-limited detection of chaotic waveforms. The bandwidth-dependent statistical properties are investigated through tuning the injection parameters for different detection bandwidths. Generation of Gaussian PDFs are demonstrated for a large region of injection parameters and for a detection bandwidth as comparable to the laser relaxation resonance frequency. This numerical study reveals the importance of injection parameters for tuning the statistics associated with chaotic lasers, and confirms the feasibility of generating RWs and Gaussian PDFs on demand for potential applications in various fields.