Characteristics improvement of quantum dot semiconductor optical amplifier for access network

Abstract

This thesis presents the characteristics improvement and performance evaluation of Quantum Dot Semiconductor Optical Amplifier (QD SOA) in an access network. There are 3 parts: 1) improvement of internal quantum efficiency, 2) increase of chip gain and 3) implementation of QD SOA in 40 Gb/s access network. The first part, Rapid Thermal Annealing (RTA) is applied to improve internal quantum efficiency to be 1.4 times higher than without RTA and low optical loss. The second part, strain compensation technique is applied to increase the chip gain of QD SOA. Considering the design of Quantum Dot Laser Diode with optimized stacked QD layers and threshold current, then the same design is applied to QD SOA having 25-stacked QD layers and 2 mm long. It can achieve the maximum chip gain of 35 dB at 400-mA bias current. The last part of thesis, the performances of two conventional SOAs and one QD SOA are evaluated in 40 Gb/s access network. Starting from the characteristics between conventional SOAs and QD SOA are compared. QD SOA gives the lowest Noise Figure of 4.59 dB because of its highest Optical Signal to Noise Ratio (OSNR). Plus, QD SOA has the fastest response time of 70 ps with the lowest data pattern effect when operating in saturation region, which is suitable for burst-mode transmission. Next, the performance of single SOA transmission is evaluated, and the Input Power Dynamic Ranges (IPDR) of 3 SOAs are measured. Finally, the two-cascaded SOA is experimented to raise power budget of a network to successfully support 128 users and 20-km distance. Consequently, installing QD-SOA as 2nd-stage SOA following a conventional SOA provides lower Bit Error Rates (BERs) than two-cascaded conventional SOAs because QD SOA has higher saturation output power and lower data pattern effect when operating at high input power. Additionally, the BERs are computed by substituting all parameters from experiments into theoretical equations. They are compared to experimental BERs to confirm the root cause of OSNR degradation and data pattern effect.