Fig. 1. Recent achievable transmission distance and data rate of UVLLC systems.

Fig. 2. (a) Structure of the high-speed 5-λ Tx module. (b) Chip structure of the narrow-ridge blue laser. EBL: electron-blocking layer, SCH: separate confine heterostructure. (c) SEM image of the ridge waveguide structure. (d) Confocal microscopy image of the laser bar (top view).

Fig. 3. (a) Degree of polarization and polarization extinction ratio of the 5-λ Tx. (b) Polarization states of 5-λ Tx represented using the Poincaré sphere (left) and polarization ellipse (right).

Fig. 4. Electroluminescence spectra with the variation of bias current of (a) 685 nm, (b) 638 nm, (c) 520 nm, (d) 450 nm, and (e) 405 nm lasers.

Fig. 5. (a) Output optical power versus bias current and (b) the I-V curves of the 5-λ Tx.

Fig. 6. (a) A photograph of the 5-λ Tx; five wavelengths are independently emitted in space. (b) A schematic diagram of the experimental setup for the WDM PDM UVLLC system. (c) The transmitter, (d) the overall system, and (e) the receiver photographs from the actual experiment. Due to the limited quantity of components such as AWG and electrical amplifiers, the tests were conducted separately for each of the five wavelengths, with a photograph of the experiment of 520 nm presented in the figure. (f) The principle of the proposed ResDualNet.

Fig. 7. The variation of BER with bias current and Vpp of the (a) 685 nm, (b) 638 nm, (c) 520 nm, (d) 450 nm, and (e) 405 nm lasers.

Fig. 8. Spectral envelopes of the transmitted signals and received signals under four scenarios (without pre-equalization operations and without ResDualNet, without pre-equalization operations and with ResDualNet, with pre-equalization operations and without ResDualNet, with pre-equalization operations and ResDualNet) for (a) 685 nm, (b) 638 nm, (c) 520 nm, (d) 450 nm, and (e) 405 nm lasers. (f) BER of the five wavelengths under four scenarios, with the dashed line indicating a threshold of 3.8 × 10⁻³.

Fig. 9. Communication performance testing of the 685 nm laser. (a) The variation of BER with Vpp for both horizontal and vertical polarization directions when using ResDualNet or the traditional equalization algorithm. (b) Distribution of constellation points at the working points i (400 mV), ii (600 mV), and iii (700 mV). (c) Distribution of the constellation points in the first quadrant at the optimal Vpp, with the lower graphs showing the probability density curves of the two constellation points with the minimum and maximum amplitudes, (1,1) and (11,7). (d) Comparison of the received time-domain waveform and frequency spectrum with the transmitted waveform at the optimal Vpp. (e) Probability density curves of noise distribution and comparison of noise spectrum.

Fig. 10. The BER performance of the system utilizing ResDualNet or traditional post-equalization algorithms as a function of Vpp for wavelengths at (a) 638 nm, (b) 520 nm, (c) 450 nm, and (d) 405 nm.

Fig. 11. Achievable data rates (bar chart) and BER (line graph) for the (a) 685 nm, (b) 638 nm, (c) 520 nm, (d) 450 nm, and (e) 405 nm lasers as a function of transmission bandwidth.