Final Report Summary - FOCUS (Free-Space Optical Systems for Next Generation Communication Networks)
Free space optical (FSO) communication is a license-free wireless access technology with high-bandwidth capacity. It is a powerful alternative and/or complementary solution to commonly deployed fiber optic and wireless radio-frequency links. FSO is appealing for a number of applications including metropolitan area network extensions, enterprise/campus network connectivity, fiber back-up, backhaul for wireless cellular networks, disaster recovery, and quantum key distribution among others. Despite the major advantages of FSO technology and variety of its application areas, its widespread use has been hampered by its rather disappointing performance for long range links. The major performance limiting factor for FSO links with ranges longer than one kilometer is the atmospheric turbulence-induced fading. This project provides a broad research framework on the concept of relay-assisted FSO communication which is envisioned as a powerful counter-measure to atmospheric turbulence fading effects with its inherent diversity gains. Towards this main objective, we have explored the fundamental performance bounds of relay-assisted FSO systems using tools from communication and information theories, identified the optimum criteria for system design/optimization in the highlight of these, and proposed efficient physical layer solutions to approach the ultimate performance limits under practical constraints.
The project has built on three workpackages. The first workpackage addresses relay-assisted FSO systems with intensity modulation/direct detection (IM/DD). Since IM/DD systems are simple and cost-effective, they are commonly employed in today’s commercial FSO links. In our research efforts within the first workpackage, we have proposed serial (multi-hop) and parallel (cooperative) FSO relaying schemes as well as their combinations in the form of mesh networks and investigated their performance through the derivations of outage probability expressions. In our derivations, we have considered different noise models (i.e. Gaussian, Poisson) based on the employed receiver type. The derived outage expressions are functions of channel and system parameters and clearly demonstrate the dependence of system performance on turbulence strength, link distance, operating wavelength, number of relays and location of relays. Based on these expressions, we have quantified diversity gains for each scheme and demonstrated significant performance improvements through relaying. Moreover, based on the minimization of derived outage expressions, we have developed optimal relay placement strategies to further improve the overall system performance. Another focus within this workpackage is all-optical relaying, where the signals are processed in optical domain. This eliminates the electrical-optical (EO) and optical-electrical domain (OE) conversions, allowing more efficient implementation in practice. In this line of research, we have investigated the outage performance of all-optical amplify-and-relay relaying taking into account practical issues such as amplified spontaneous emission noise. Our comparison between conventional (i.e. with O/E and E/O convertors) and all-optical relaying has demonstrated that the latter presents a favourable trade-off between complexity and performance and can be used as a low-complexity solution.
In the second workpackage, we have focused on coherent FSO systems. From an information theory
point of view, we have first investigated the performance of a coherent FSO communication system
with multiple receive apertures over atmospheric turbulence channels. Our study has built on a
statistical model that characterizes the combined effects of turbulence-induced wavefront distortion
and amplitude fluctuation in coherent receivers with phase compensation. We have investigated the
link reliability as quantified by “diversity gain” and the relationship between the link reliability and
the spectral efficiency as quantified by “diversity-multiplexing trade-off (DMT).” Then, we have extended our results for relay-assisted coherent FSO systems and quantified the performance gains in terms of diversity and multiplexing available through multi-hop relaying. Our results have demonstrated impressive performance improvements over the conventional direct transmission.
We have also proposed automatic repeat request (ARQ) protocols as a way to extract time diversity gains which would further boost the performance of coherent FSO communications. Within this workpackage, we have further investigated so-called mixed RF/FSO systems where RF and FSO links are cascaded in the form of a dual-hop relaying scheme. Such a cascaded system is particularly useful to address the connectivity gap between the RF access network and the backbone network. Finally, we
we have studied differential modulation/demodulation techniques for FSO systems which can be considered as partially coherent systems and provide a low-complexity, yet sub-optimal alternative to coherent systems.
In the third workpackage, we have focused on free-space quantum relaying. In this part of this project, we have considered the quantum bit error rate (QBER) and the secrecy rate as figure-of-merits and used these in our performance analysis. Different from fiber optics, free-space quantum key distribution (QKD) systems experience performance degradations due to absorption, scattering, diffraction, and turbulent-induced scintillation experienced in atmospheric channels. Carefully integrating the effects of such channel impairments in our transmission model, we have first derived closed form QBER expressions for point-to-point direct transmission. Two main disadvantages we have observed through the analysis of free space QKD systems is the lower throughput (quantified through secrecy rate) and poor QBER performance, particularly in long ranges. Motivated by the fact that turbulence-induced fading variance heavily depends on the link range, we have proposed a relay-assisted QKD scheme. The proposed scheme increases the average number of collected background photons, but it also significantly decreases the photon loss caused by diffraction and turbulence. Overall, it is able to improve the QBER performance especially at long ranges. Our research results have provided invaluable guidelines in the design of terrestrial free space QKD systems.
Overall, the project has fully achieved its objectives and exceeded expectations in many parts. We expect that our research findings will be instrumental in the building of long-range, highly reliable, and ultra-fast FSO systems with utmost security to meet the demanding requirements of future generation communication networks. The proposed relay-assisted IM/DD FSO systems bring superior link reliability, extended link range, and higher data rate over the existing systems and address the short and medium-term market needs. In comparison to IM/DD systems, coherent receivers are more difficult to implement, but they provide more flexibility and significant performance enhancements. Therefore, the proposed coherent FSO systems bring superior performance over IM/DD counterparts and address the longer term market needs. On the other hand, the proposed relay-assisted free-space QKD systems are useful for applications which require the utmost security. These overcome the current distance limitations for long-range quantum communication and provide a powerful alternative to fiber-based QKD.
The project has built on three workpackages. The first workpackage addresses relay-assisted FSO systems with intensity modulation/direct detection (IM/DD). Since IM/DD systems are simple and cost-effective, they are commonly employed in today’s commercial FSO links. In our research efforts within the first workpackage, we have proposed serial (multi-hop) and parallel (cooperative) FSO relaying schemes as well as their combinations in the form of mesh networks and investigated their performance through the derivations of outage probability expressions. In our derivations, we have considered different noise models (i.e. Gaussian, Poisson) based on the employed receiver type. The derived outage expressions are functions of channel and system parameters and clearly demonstrate the dependence of system performance on turbulence strength, link distance, operating wavelength, number of relays and location of relays. Based on these expressions, we have quantified diversity gains for each scheme and demonstrated significant performance improvements through relaying. Moreover, based on the minimization of derived outage expressions, we have developed optimal relay placement strategies to further improve the overall system performance. Another focus within this workpackage is all-optical relaying, where the signals are processed in optical domain. This eliminates the electrical-optical (EO) and optical-electrical domain (OE) conversions, allowing more efficient implementation in practice. In this line of research, we have investigated the outage performance of all-optical amplify-and-relay relaying taking into account practical issues such as amplified spontaneous emission noise. Our comparison between conventional (i.e. with O/E and E/O convertors) and all-optical relaying has demonstrated that the latter presents a favourable trade-off between complexity and performance and can be used as a low-complexity solution.
In the second workpackage, we have focused on coherent FSO systems. From an information theory
point of view, we have first investigated the performance of a coherent FSO communication system
with multiple receive apertures over atmospheric turbulence channels. Our study has built on a
statistical model that characterizes the combined effects of turbulence-induced wavefront distortion
and amplitude fluctuation in coherent receivers with phase compensation. We have investigated the
link reliability as quantified by “diversity gain” and the relationship between the link reliability and
the spectral efficiency as quantified by “diversity-multiplexing trade-off (DMT).” Then, we have extended our results for relay-assisted coherent FSO systems and quantified the performance gains in terms of diversity and multiplexing available through multi-hop relaying. Our results have demonstrated impressive performance improvements over the conventional direct transmission.
We have also proposed automatic repeat request (ARQ) protocols as a way to extract time diversity gains which would further boost the performance of coherent FSO communications. Within this workpackage, we have further investigated so-called mixed RF/FSO systems where RF and FSO links are cascaded in the form of a dual-hop relaying scheme. Such a cascaded system is particularly useful to address the connectivity gap between the RF access network and the backbone network. Finally, we
we have studied differential modulation/demodulation techniques for FSO systems which can be considered as partially coherent systems and provide a low-complexity, yet sub-optimal alternative to coherent systems.
In the third workpackage, we have focused on free-space quantum relaying. In this part of this project, we have considered the quantum bit error rate (QBER) and the secrecy rate as figure-of-merits and used these in our performance analysis. Different from fiber optics, free-space quantum key distribution (QKD) systems experience performance degradations due to absorption, scattering, diffraction, and turbulent-induced scintillation experienced in atmospheric channels. Carefully integrating the effects of such channel impairments in our transmission model, we have first derived closed form QBER expressions for point-to-point direct transmission. Two main disadvantages we have observed through the analysis of free space QKD systems is the lower throughput (quantified through secrecy rate) and poor QBER performance, particularly in long ranges. Motivated by the fact that turbulence-induced fading variance heavily depends on the link range, we have proposed a relay-assisted QKD scheme. The proposed scheme increases the average number of collected background photons, but it also significantly decreases the photon loss caused by diffraction and turbulence. Overall, it is able to improve the QBER performance especially at long ranges. Our research results have provided invaluable guidelines in the design of terrestrial free space QKD systems.
Overall, the project has fully achieved its objectives and exceeded expectations in many parts. We expect that our research findings will be instrumental in the building of long-range, highly reliable, and ultra-fast FSO systems with utmost security to meet the demanding requirements of future generation communication networks. The proposed relay-assisted IM/DD FSO systems bring superior link reliability, extended link range, and higher data rate over the existing systems and address the short and medium-term market needs. In comparison to IM/DD systems, coherent receivers are more difficult to implement, but they provide more flexibility and significant performance enhancements. Therefore, the proposed coherent FSO systems bring superior performance over IM/DD counterparts and address the longer term market needs. On the other hand, the proposed relay-assisted free-space QKD systems are useful for applications which require the utmost security. These overcome the current distance limitations for long-range quantum communication and provide a powerful alternative to fiber-based QKD.