Kenneth Smart, CSIRO; Léa Powys, CSIRO; Douglas Hayman, CSIRO; Andrew Hellicar, CSIRO; Andrew Kintz, MIT Lincoln Laboratory; Alexander Kobsa, MIT Lincoln Laboratory; Timothy Brown, MIT Lincoln Laboratory; Gregory Hogan, MIT Lincoln Laboratory
https://doi.org/10.64861/LAAK2199
Keywords: Bistatic radar, SSA, SDA, GEO satellites, Parkes, TRADEX, L-band, range-Doppler, multi-target detection
Abstract:
Bistatic radar techniques that leverage radio astronomy receivers have shown growing promise for Space Situational Awareness (SSA) in the geosynchronous belt (GEO), where sensitivity is a limiting factor for detecting weak scatterers and for resolving small relative motions. Building on prior Southern Hemisphere results obtained with the CSIRO Mopra 22-m telescope, we report the first bistatic radar observations using the 64-m Parkes (Murriyang) radio telescope within the NATO SET-340 Long Baseline Bistatic Radar (LBBR) effort. Our goal is to quantify the sensitivity, multi-target resolution, and measurement accuracy achievable with Parkes at L-band, and to compare two independent receive backends while benchmarking performance against the earlier Mopra dataset._x000D_
On 23 December 2025, the TRADEX radar at the Reagan Test Site (RTS) served as the illuminator, transmitting the L-band waveform (carrier 1.32 GHz, bandwidth 8 MHz, pulse width 565 microseconds) in three 5-minute dwells on GEO spacecraft. Parkes pointed and recorded simultaneously with two receive chains: (i) the VLBI backend and (ii) a GPU voltage-capture backend. The planned schedule targeted Sky Muster II (NORAD 41794), TJS-1 (40892), and Eutelsat 172B (42741); all three were detected in both backends. Parkes operated in normal astronomy pointing with additional voltage recording; pointing predictions were cross-checked locally using Celestrak TLEs via NASA Horizons workflows._x000D_
We performed standard pulse-compression processing on the recorded IQ to form range-Doppler maps for each dwell. For the Sky Muster II pass, we examined all resolvable peaks and computed relative bistatic range and relative bistatic range-rate with respect to the dominant return. Predictions from contemporaneous TLEs were used as a reference to assess agreement and likely error sources. We compared SNR statistics (max-median SNR) between the VLBI and GPU backends, and against the June 2025 Mopra L-band collect under similar illumination conditions, to estimate the sensitivity advantage provided by Parkes._x000D_
During the Sky Muster II dwell, Parkes resolved multiple GEO targets within the same beam. Alongside Sky Muster II, a strong secondary peak and additional candidates were present; cross-checks indicated that Express AMU-7 was contemporaneously visible to both Parkes and TRADEX, consistent with the range-Doppler structure. The relative bistatic range between the dominant and secondary peaks measured 6.197 km, versus a TLE-predicted 5.768 km. The relative bistatic range-rate measured -0.5208 m/s, compared to -0.01755 m/s predicted by TLEs. The range agreement (approximately 0.43 km difference) is within typical TLE-driven uncertainty for GEO baselines, while the larger Doppler discrepancy reflects known TLE limitations for fine radial-rate estimation. These results demonstrate simultaneous multi-target detection and characterisation in a single Parkes dwell._x000D_
Across all three dwells (Sky Muster II, TJS-1, Eutelsat 172B), both VLBI and GPU receive chains produced consistent detections with closely matched SNRs, differing only by minor, target-dependent variations. This backend consistency validates the operational flexibility of Parkes for SSA collections, enabling either backend to serve as a primary receiver or providing redundancy for campaign robustness._x000D_
Relative to the June 2025 Mopra L-band collect against GEO targets, Parkes delivered approximately 4-7 dB higher max-median SNR, as observed in the paired VLBI and GPU datasets for the same session. The improvement is consistent with expectations for a larger aperture (64 m) with low-noise receive systems and stable timing references. The sensitivity margin has practical implications: shorter dwells can achieve equivalent detectability, multi-target scenes can be disentangled more reliably, and weaker scatterers (including smaller spacecraft features or co-located objects) become accessible._x000D_
These experiments establish Parkes as a high-sensitivity Southern Hemisphere SSA receiver for bistatic radar at L-band. We demonstrate simultaneous multi-target resolution in GEO, validated dual-backend performance, and quantified sensitivity gains over Mopra. The ability to detect multiple objects in a single dwell, and to extract relative bistatic range and range-rate with kilometre- and sub-m/s-level precision against TLE predictions, directly supports characterisation of relative motion and spatial separation within busy geostationary slots. With additional baselines and absolute time-of-transmit-to-receive alignment, these measurements can be extended to recover full 3D state vectors for GEO objects. The performance demonstrated by Parkes enhances LBBR coverage in the Pacific domain and motivates continued campaign operations combining Parkes, Mopra, ATCA, and RTS radars._x000D_
We report the first Parkes-based bistatic radar detections of GEO satellites using TRADEX at L-band, with multi-target resolution and backend-to-backend consistency. Near-term work includes (i) coordinated collections to incorporate additional baselines for geometry diversity, (ii) time-transfer refinements to tighten absolute range and range-rate estimates, and (iii) validation against high-accuracy ephemerides (for example GPS-class references) to bound measurement error independently of TLEs. The sensitivity demonstrated by Parkes opens a path to routine GEO monitoring with improved detectability and characterisation fidelity across the Southern Hemisphere.
Date of Conference: September 15-18, 2026
Track: SDA Systems & Instrumentation