Naval Free Space Optical communications

Abstract

Wireless Optical Communication (WOC) has recently become a viable complementary technology in addition to existing Radio Frequency Communication (RFC). One of its critical advantages over existing RFC is its significant increase in carrier frequency, since “it is a well-known fact that an increase in carrier frequency increases the information-carrying capacity of a communication system” [24]. WOC has several critical advantages over existing RFC, namely tariff-free bandwidth allocation, low on-board power requirements, and improved channel security. Due to the rise in global tensions, the last advantage can prove a crucial asset for military applications.

There are also several critical challenges of WOC, namely its high dependence on stable atmospheric conditions. Recent advances in Vertical-Cavity Surface-Emitting Lasers (VCSELs), Avalanche Photodiodes (APDs), Adaptive Optics, Real-Time Control, Active Optics, etc., have all allowed the challenges of WOC to be addressed, increasing the feasibility of implementing such technology as a practical communication system. The application of WOC for Ground-to-Low-Earth-Orbit communication is referred to as Free Space Optical Communication (FSOC). This application remains the focus of this work.

To perform FSOC, 3 main elements are required; these include the transmitter, the transmission channel, and the receiver. The fundamental working principle of this technology is to modulate the information into an optical carrier, typically performed at the transmitter in simplex mode. As a section of the transmission channel passes through Earth’s atmosphere, this results in the introduction of disturbances due to the presence of turbulence in the different layers of Earth’s atmosphere.

Therefore, a functional FSOC system must possess precise and efficient pointing capabilities at both receiver and transmitter ends, such that the communication channel has the capacity to compensate for the atmospheric disturbances. Therefore, the Pointing Acquisition and Tracking (PAT) system must comply with stringent angular and angular-rate pointing requirements.

While there is plenty of contemporary research regarding the design and testing of ground-based Optical Communications Terminals (OCTs) meant for FSOC, currently, there is a gap in knowledge regarding mobile OCTs. Therefore, the focus of this work has centered on the expansion of knowledge by investigating the achievable performance of a mobile OCT placed on board a Royal Netherlands Navy frigate-sized ship. To summarize, this thesis has centered on answering: What is the strategy to simulate and control the effects that ship-wave disturbances have on a mobile Dual-Stage Optical Communications Terminal to determine if the achievable pointing precision is sufficient for performing Free Space Optical Communications in naval environments?

To answer this, a simulation strategy has been employed to divide the work required to answer this question into 3 main steps. Besides this strategy, literature research has been performed to determine suitable system and sub-system requirements in close cooperation with the customer and research partner, the Royal Netherlands Air Force (Koninklijke Luchtmacht). The 3 key requirements are as follows:

PATR-01: The total angular (static) and angular-rate (dynamic) Root Mean Square (RMS) pointing jitter error shall not exceed 50 [μrad] / [μrad/sec]. PATR-01-01: The precise pointing actuator model shall approximate the dynamics and performance characteristics of the Optics in Motion OIM5001 Fast Steering Mirror (FSM). PATR-01-02: The coarse actuator model shall approximate the dynamics and performance characteristics of the PlaneWave Instruments L-350 direct-drive telescope mount.

The first step regarding the implementation of the simulation strategy has been the incorporation of the recordings from an inertial measurement system (sampling rate of 200 Hz) on board a Royal Netherlands Navy frigate-sized ship into the performance analysis. These measurements provide the angular and angular-rate disturbances, which the dual-stage communications terminal would require to compensate for. These disturbances are of a stochastic nature, since, as of writing, there are limited time-based simulation tools that can accept signals/disturbances of a stochastic nature. The decision was made to perform the analysis in the frequency spectrum, where the disturbance signals are represented as Power Spectral Density (PSD) functions.

The second step involved the development of a simulation that is capable of approximating the performance characteristics of a dual-stage mobile OCT. To complete this step, the focus centered on creating 3 separate simulation models, each allowing for the expansion of the capabilities of the previous models. These include the Precise Pointing Actuator System (PPAS), which centered on recreating the closed-loop system dynamics of the FSM OIM5001 as an LTI spring-damper system, paired with a PID controller and a Quadcell Photodiode (QPD) tracking sensor sampling delay. The QPD tracking sensor errors are included in the analysis in the form of an additional PSD disturbance signal.

The Coarse Pointing Actuator System (CPS) centered on modeling the L-350 gimbal telescope mount as an LTI spring-damper system, paired with a PI controller and the same QPD sampling delay. Lastly, the 3rd model (DSAS) centered on combining the previous models into a combined dual-stage chief-deputy controller configuration, where the chief stage (PPAS) is being actively off-loaded by the deputy stage (CPAS).

The last step of the simulation strategy has been the determination of the achievable performance of the 3 simulation models. To achieve this, the Dynamic Error Budgeting (DEB) design analysis has been utilized, as it allows for the determination of the performance of a system to which stochastic disturbance signals are applied.

From the performance results, it is possible to argue that individual actuator stages are unsuitable for FSOC in naval environments. While the performance results of the PPAS simulation model result in pointing angle and angle-rate errors of ≈ 24.6% and ≈ 43.6% below PATR-01, the angular range requirements go beyond those achievable by the FSM actuator, standing at ≈ ±9.06 degrees, while the actuator specification states a maximum range of ±3 degrees [38].

While the performance results of the CPAS indicated achievable actuator effort and controller torque requirements, the system is incapable of complying with the pointing requirements, remaining ≈ 234.6% above the allowable pointing angle error requirement and ≈ 234.6% above the allowable pointing angle-rate error.

With an understanding of the performance limitations of individual stages, it is now possible to motivate the need for a dual-stage optical communications system. Therefore, it is now relevant to discuss the performance results of the DSAS, which are comprised of multiple controller bandwidth configurations. Notable from these results are the highest-performance configurations with chief and deputy controller configurations of 398/500 Hz and 765/500 Hz. Both of these are capable of complying with the pointing requirements, each remaining ≈ 5.5% and ≈ 46.3% below the pointing angle rate, while both achieving to remain > 90% under the pointing angle requirement. At the highest two performance configurations, it can be seen that the deputy controller requires ≈ 12 N·m and has a power consumption of ≈ 94 W, which is an increase from the base configuration of ≈ 66.2% and ≈ 412.8%, respectively, both still falling within the actuator specifications. Another relevant outcome is the significant reduction in angular range requirements of the chief actuator, where for the highest performance configurations it is only required to provide 1.2–1.4 mdegrees of range.

With the DSAS performance results, it is of relevance to verify and back up such findings. To this end, 3 verification steps have been performed. The first step involved the demonstration of the correct functioning of the frequency analysis; this has been obtained by passing a test signal with known properties through the entire frequency analysis pipeline. It was then possible to determine that the estimated properties differ from their original properties by only an average of 2%. The second step involved verifying the control design of the DSAS. To achieve this, the transfer functions developed under this work were compared against equivalent transfer functions from research papers, which were found to be equivalent to each other. The last step involved verifying the performance results obtained from the DEB design approach. To achieve this, an alternative time-analysis simulation has been utilized to reproduce the performance results. The difference between the results obtained from the alternative simulation methods and DEB analysis remains at a margin of 1.3–2.8%.

In conclusion, it is possible to argue that a Dual-Stage optical communications terminal with a chief-deputy controller configuration of 398/500 Hz or 765/500 Hz can be capable of performing FSOC in naval environments by fulfilling PATR-01 pointing requirements and PATR-01-01, PATR-01-02 actuator specification sub-requirements, with a preference for the higher-bandwidth configuration. Regarding future work, it is possible to recommend extending the sampling frequency to 10–20 kHz for the inertial measurements, as this could allow the recreation of phenomena that could be occurring at higher frequencies. Another crucial recommendation would be the expansion of the simulation models into specific models obtained from system identification, as these could potentially improve the estimates regarding torque and power requirements of the deputy stage of the DSAS, which has been approximated via simple geometry.