DESIGN AND PERFORMANCE ANALYSIS OF IMPROVED FREE SPACE OPTICAL COMMUNICATION SYSTEM FOR TURBULENT ATMOSPHERE

Abstract

With the proliferation of mobile devices, the demand for bandwidth has surged, while radio frequency (RF) spectrum resources are limited leading to a bottleneck in the traditional RF cellular networks. Additionally, existing backhaul network infras- tructure often struggles to support the increased data traffic. Researchers have proposed to use optical fibers as a potential solution for alleviating the backhaul load congestion. However, as the number of cells becomes very large, networks can still suffer from the limited optical fiber installations which are very costly and sometimes even restricted. To support these challenges, free space optics (FSO) technology emerges as an alter- native solution to the RF links and optical fiber, since it is more flexible, license-free, power efficient, cost effective, and most importantly it increases the capacity of net- works. However, its performance is affected by the atmospheric turbulence, adverse weather, and pointing errors. To improve reliability, dual-hop RF-FSO system and hy- brid FSO/RF system are employed. In the dual-hop RF-FSO system, relays act as inter- mediate nodes between the source and the destination to extend coverage. In the hybrid FSO/RF system, both links operate simultaneously to transmit the same signal. Since if one link is degraded, ensuring continuous communication even if one link degrades. Multiuser (MU) diversity has been widely exploited to alleviate the performance loss due to atmospheric turbulence induced fading. However, to efficiently allocate resources to multiple users, effective scheduling schemes are required to achieve sub- stantial system performance. In the event, where users experience lower average signal- to-noise ratio (SNR), accessing channel resources becomes problematic, particularly in the presence of users enjoying comparatively higher average SNR levels. Conse- quently, the system may allocates resources to users with weaker channels to ensure their requirements are satisfied. However, such allocations do not contribute favorably to achieving optimum system performance. This situation introduces a conflict between the objectives of meeting fairness in user channel access and optimizing overall system performance. Further, to enhance network capacity in next-generation systems, one way is to re- duce the cell size. This reduction in cell dimensions brings about the challenge of co-channel interference (CCI), which can adversely affect system performance within these smaller cells. Another challenging problem in the wireless networks is assum- ing the perfect channel state information (CSI) of channel between the transmitter and receiver which is not possible to get perfect CSI in real time scenario. Therefore, this motivates the work done to address these technical challenges in this thesis. The first part of this thesis analyzes the impact of turbulence on the performance of FSO system over Inverse Gaussian Gamma (IGG) distribution. The analytical formu- lations derives closed-form expressions for the outage probability (OP) and average bit error rate (ABER). The proposed work also provides valuable insights to enhance the system performance. In the second part, the performance analysis of MU dual-hop RF-FSO systems is analyzed for cumulative distribution function based scheduling (CDFS) scheme under perfect CSI at both RF and FSO links. The performance is evaluated by deriving the closed-form expression for OP considering independent and non-identically distributed (i.n.i.d) CCI at relay. Moreover, expression for optimum power allocation and channel iii access ratio (CAR) are evaluated to enhance outage performance. The CDFS scheme is used to promotes fairness and precise control over the CAR to improve the system performance. In the third part, performance analysis of MU dual-hop RF-FSO systems is ana- lyzed employing CDFS scheme under imperfect CSI at both RF and FSO links. The performance is evaluated by deriving the closed-form expression for OP. Moreover, expression for optimum power allocation and CAR are evaluated to enhance outage performance. We also compare the performance of the CDFS based system against greedy scheduling (GS), and proportional fairness scheduling (PFS) schemes. In the fourth part, performance analysis of dual-hop MU RF-hybrid FSO/RF sys- tem is presented. Closed-form expressions for OP and ABER are derived under the influence of CCI at the relay. Moreover, expressions for optimum power allocation and CAR are evaluated to enhance outage performance. In the final part of this thesis, performance of MIMO FSO and WDM FSO systems is presented in terms of BER and Quality factor. Since the involvement of complex function such as Meijer-G function and Fox-H function in the derivations, asymptotic expressions at high SNR are derived using elementary functions to provide engineer- ing insights into system performance. The frameworks proposed in this work can be efficiently utilized in various wireless standards. It will be helpful for a communica- tion engineer to design a wireless systems without performing extensive simulations or tedious experiments.