NIAC project report: Solar system-scale VLBI to dramatically improve cosmological distance measurements

TL;DR

This NIAC Phase I study investigates the Cosmic Positioning System (CPS), a five-spacecraft VLBI-like network distributed through the outer Solar System to measure extragalactic distances purely geometrically by detecting wavefront curvature of fast radio bursts. By leveraging tens-of-AU baselines, CPS aims to deliver sub-percent constraints on the Hubble constant $H_0$ with a modest number of FRB detections, bypassing the traditional distance ladder. Beyond cosmology, CPS offers sensitivity to dark-matter clumpiness via differential Shapiro delays, microhertz gravitational waves, and outer-Solar-System mass distribution, with a nominal design featuring 8–9 m antennas, 3–6 GHz receivers, and space-qualified clocks. The study finds phase I feasibility with attainable technologies but highlights FRB properties at several GHz as the key uncertainty; it recommends observational campaigns to characterize high-frequency repeating FRBs and continued clock/storage technology development to make CPS viable for a future mission.

Abstract

We investigate the feasibility and scientific potential of the Cosmic Positioning System (CPS), a space mission concept enabling purely geometric distance measurements to sources at hundreds of megaparsecs by directly detecting electromagnetic wavefront curvature. CPS consists of a constellation of radio antennas distributed across the outer Solar System, operating on baselines of tens of astronomical units. By precisely timing the arrival of repeating fast radio bursts (FRBs), CPS infers source distances via trilateration -- analogous to global navigation satellite systems such as GPS but on cosmological scales. We show that CPS distance measurements could result in sub-percent constraints on the Hubble constant with even a handful of detections, whereas we predict that 10-100 FRB sources are likely visible. We evaluate dominant sources of uncertainty -- wavefront timing precision, interstellar refractive delays, spacecraft positional knowledge, and onboard clock stability -- finding these controllable at required levels using near-term technologies. Our nominal design employs five spacecraft with 8 m deployable antennas, 3-6 GHz receivers with sub-30 K system temperatures, and space-qualified atomic clocks similar to those on GPS satellites, supported by a ground network for ranging calibration and FRB alerts. Beyond cosmic expansion, CPS may enable frontier measurements in astrophysics and fundamental physics, including constraints on small-scale dark matter structure, microhertz gravitational waves (bridging pulsar timing arrays and LISA), and the outer Solar System mass distribution. The most significant viability issue concerns FRB properties at several-GHz frequencies; we recommend observational campaigns to characterize repeating FRBs in this band.

Figures (18)

• Figure 1: Left: Illustration of the wavefront curvature distance measurement: The signal arrives at detector B before it arrives at detectors A and C. By comparing the arrival times at the three detectors (four detectors in 3D), the distance to the source can be inferred via the wavefront curvature. Right: Estimated fractional distance error, $\sigma_d/d$, to a single FRB as a function of the distance to the FRB, with distances of $>100$ Mpc being interesting for measuring the cosmic expansion and the median distance of supernova for the SHOES $H_0$ measurement annotated. The typical baseline length of detectors in the constellation is annotated on the left of the curves, and the highlighted band assumes detector positions errors of $\delta x = 2-6\,$cm. CPS aims for much more precise distance constraints, which in turn would result in the most precise direct constraints on the late-time cosmic expansion.
• Figure 2: Illustration of configuration that would constrain the clumpiness of the dark matter, from xiao24. Shown are two radio antennas separated by 100 AU observing the same FRB source but along different sightlines. Each sightline experiences slightly different Shapiro time delays caused by intervening dark matter clumps. As a result, the FRB voltage-field time-series detected by CPS (illustrated by the orange jagged lines) will have different arrival times at each detector (shown here, the same signal arrives at Dish A before arriving at Dish B). Such a measurement is sensitive to diffuse dark matter structures with smaller sizes than probed by other methods.
• Figure 3: Potential configurations for CPS to detect gravitational waves in the frontier $10^{-7}$--$10^{-4}$ Hz frequency band, from 2024arXiv241115072M. Three configurations are shown: (1) traditional Doppler tracking using an Earth station, (2) a single-arm configuration leveraging an onboard atomic clock, and (3) a two-arm time-delay interferometry setup requiring two radio antennas for at least one spacecraft. The latter configuration, while requiring more onboard instrumentation for at least one spacecraft, offers substantially higher sensitivity by eliminating onboard clock noise (which likely dominates the error without significant advances in space clock technology).
• Figure 4: Conceptual design of a CPS spacecraft. The primary $D\approx 8~$m dish dominates the structure, with the receiver mounted on a passively cooled cold plate separated from the transmitter to minimize thermal coupling. The spacecraft bus below the main dish houses the mission-critical payload, including the digital receiver, digital transmitter, and atomic clock. A steerable smaller antenna for receiving terrestrial alerts is positioned at the bottom of the spacecraft. Power is supplied by a radioisotope thermoelectric generator (RTG) mounted on an extended boom to minimize thermal and radiation effects on sensitive instrumentation.
• Figure 5: The square root of the dimensionless displacement power spectrum (multiplied by frequency) for the most significant stochastic sources of spacecraft displacement in the outer Solar System. This value, when evaluated at frequencies whose period corresponds to the day to possibly even month cadence of positional calibration, approximates the spacecraft displacement due to stochastic accelerations between calibrations. These calculations assume a spacecraft with mass $M_{\rm sat}=10^3$ kg and effective area $A_{\rm eff}=10~$m$^2$ at solar distances of $r=10$ AU (left panel) and $r=30$ AU (right panel). Note that displacements from solar irradiance variations and the solar wind act primarily in the radial direction, whereas dust collisions would be predominantly aligned with the interstellar dust flow.