A Timescale Concept in AltPNT: A Model-Based Control of Networked Systems Approach

Terrestrial and satellite communications, tactical data links, positioning, navigation, and timing (PNT), and distributed sensing will continue requiring precision timing and the ability to synchronize and disseminate time. However, as the supply of space-qualified clocks with Global Navigation Satellite Systems (GNSS)-level performance is limited and the understanding of disruptions to GNSS due to potential adversarial actions increases, the current practice of reliance on the GNSS-level timing becomes costly and outdated when compared with an on-orbit assembly of robust and stable timescale references, especially being considered as the development of diverse alternatives to GNSS. Onboard realization of clock ensembles is particularly attractive for applications such as on-demand dissemination of GPS Time (GPST) like navigation services via a proliferated Low-Earth-Orbit (pLEO) constellation. This paper investigates the model-based control of wireless networked timing systems that embraces the vision of optimally placing critical information of the implicit ensemble mean (IEM) estimation about the multi-platform clock ensemble that has better stability than any individual member of the ensemble on the network to flexibly reduce the data traffic load. By making the remote sensor of IEM estimates running onboard the anchor platform and actuators for optimal steering of remote frequency standards located at participating platforms more “intelligent” that supports onboard IEM timescale realizations across a pLEO constellation, the networked control system can predict future behaviors of local reference clocks accompanied with low-noise oscillators and then send precise information on the IEM estimation at critical times to ensure the realization of a common pLEO timescale onboard all participating platforms. Clock steering is especially essential for the realization of timescales. Performance realizations are generally dependent on the control intervals and steering techniques chosen. Towards performance robustness beyond what the existing steering technique of Linear Quadratic Gaussian control can offer, the minimal-cost-variance (MCV) steering paradigm is proposed from the perspective of minimizing the variance of the integral-quadratic-form performance measure of the Linear Quadratic Gaussian control subject to a constraint on its mean.