Terrestrial free-space optical (FSO) communications is an emerging low-cost, licensefree and high-bandwidth access solution for a number of applications including the “last mile” access network. However, for a transmission range from a few meter to longer than 1 km, a number of atmospheric phenomena, such as rain, haze, fog, snow, scintillation and pointing errors become a major performance limiting factors in FSO systems resulting in link deterioration and ultimately complete link failure. Relay-assisted technique is capable of mitigating the signal fading and maintain acceptable performance levels. In this thesis, a two-way relay (TWR) channel technique is adopted to increase system spectral efficiency, which in turn boosts the network throughput. This is achieved by using a physical layer network coding (PNC) technique, where network coding (NC) is applied at the physical layer. It takes advantage of the superimposition of the electromagnetic waves, and embraces the interference, which was typically deemed as harmful, by performing the exclusive-OR mapping of both users’ information at the relay. Therefore, the main contribution of this thesis is to study the design of the TWR-FSO communication system that embraces PNC technique for the full utilization of network resources based on the binary phase shift keying (BPSK) modulation. Moreover, error control coding (ECC) in conjunction with interleaving can be employed in FSO communications to combat turbulence-induced fading, which can enhance the performance of the proposed TWR-FSO PNC system. A comparative study between convolutional code (CC) and bit-interleave coded modulation with iterative decoding (BICM-ID) code are carried out. The result shows that the BICM-ID code outperforms the CC for TWR-FSO based PNC over strong turbulence regime by ~10dB of SNR to achieve a BER of 10-4 . However, the number of users that can be simultaneously transmitted to the relay is considered the main constraint in PNC system. Therefore, to overcome this challenge, a new scheme that integrates the iterative multiuser detection (I-MUD) technique with the PNC system over RF and FSO links are introduced as another achievement. The results show that the I-MUD offers improved performance about 8, and 22dB of SNR to get a BER of 10-4 over RF and FSO channels, respectively, for number of simultaneously users equal to 14 with respect to TWR-PNC system.
|Publication status||Accepted/In press - Jun 2016|