An engineering team at the University of California, San Diego, is trying to replace the traditional bulky, mechanically rotating parabolic antennas of satellite ground stations with distributed small planar antennas, which is expected to significantly improve satellite data transmission capabilities while reducing costs. The research system, called "ArrayLink," works together by distributing multiple laptop-sized phased array antennas on rooftops, communication towers and other buildings to provide a more flexible and scalable ground access method for increasingly crowded low-Earth orbit satellites.
Although the satellites themselves have experienced rapid development in the past decade, transforming from large communication satellites of several tons in the past to highly integrated, software-definable small low-Earth orbit satellites, ground infrastructure is still largely stuck in the old model of relying on large mechanical pointing antennas. Current satellite communications not only serve the satellite Internet, but also support key scenarios such as global positioning and navigation, financial transactions, weather forecasts, military communications, emergency response, aviation and shipping operations, telemedicine, and earth observation. Its importance far exceeds the general public awareness.
At present, the vast majority of satellite data still needs to be "landed" through ground stations to access the wide-area Internet, and these ground stations usually rely on parabolic antennas with a diameter of about 1.8 meters or larger to provide high-gain "feed links." Although this type of antenna is powerful in performance, it is extremely inflexible: each antenna can only track one satellite at a time, and must also be mechanically rotated to track low-Earth orbit satellites that pass through the sky at about 28,000 kilometers per hour. This model has increasingly become a bottleneck in the context of the proliferation of low-orbit constellations.
The research team mentioned that the mechanical rotation speed of some active ground station satellite antennas is only 2 to 5 degrees per second. Switching from one satellite to another often takes several seconds or even close to a minute. During this period, the ground station is in an "unavailable" state, further limiting the overall throughput capacity. Although electronically scanned phased arrays can theoretically replace mechanical antennas, the cost and complexity of stacking enough antenna elements on a single array to match the gain of a large parabolic antenna are currently too high to be deployed on a large scale.
ArrayLink's idea is to abandon "making one array larger" and instead use multiple off-the-shelf small phased array panels and coordinate them as a distributed system. This architecture can incorporate up to 16 phased array panels, with a distribution range of up to kilometers. Each panel has limited link capabilities. However, after unified coordination, the overall performance is like a "virtual large antenna", approaching the traditional parabolic antenna in gain.
Dinesh Bharadia, the corresponding author of the paper, pointed out that the current fundamental bottleneck in expanding satellite communication capacity is no longer in space, but on the ground. This is the core problem that ArrayLink is trying to solve. He said that this solution can help the industry expand the scale of ground stations at a lower cost and faster speed, and can even be implemented through "crowdsourcing" deployment: any owner or enterprise with rooftop resources can install the system and transmit satellite data back to the Internet.
It is worth noting that ArrayLink’s innovation lies not only in the flattening and distribution of physical forms, but also in the in-depth utilization of spatial channel characteristics. By spreading the panels apart over a larger physical range, the team found that they could take advantage of an effect called "near-field line-of-sight MIMO" to form multiple parallel data streams between the same satellite and ground stations, significantly improving throughput.
In traditional line-of-sight satellite links, each receiving antenna often "sees" almost the same signal, making it difficult to achieve spatial multiplexing. When the panel spacing is large enough, the incident electromagnetic waves observed by each panel will differ in parameters such as phase, allowing the system to separate multiple independent data streams from signals from the same satellite. This is similar in principle to the MIMO technology commonly used in Wi-Fi routers and mobile communication networks, but is amplified to the satellite scale.
According to the team's simulation results, ArrayLink can support up to 4 spatially parallel data streams at a transmission distance of hundreds of kilometers, and can still maintain 2 data streams at a distance of more than 2,000 kilometers. Researchers say that compared to traditional single-stream parabolic antenna systems, the overall throughput of this architecture is expected to be increased to about three times.
ArrayLink also demonstrates an unconventional ability: it can not only focus energy in the angular dimension, but also achieve "fixed-point delivery" of energy in the distance dimension. Traditional antennas mainly control the beam direction by changing the pointing angle, while ArrayLink can finely control the concentrated location of energy in both angular and radial dimensions, which is expected to reduce interference to other satellite systems in complex orbital environments.
This system does not remain on paper. The research team has completed outdoor hardware experiments under line-of-sight conditions using phased array antennas and software radio platforms in the 27GHz frequency band. The measured data is highly consistent with the theoretical analysis and simulation results, which to a certain extent verifies the key physical mechanism behind the scheme.
From an engineering implementation perspective, ArrayLink also emphasizes practicality and implementability. Its design is based on commercially available phased array hardware. These devices are similar to current mass-produced satellite Internet terminal antennas, avoiding the need to rely on expensive and highly customized "laboratory-specific" equipment, thus leaving a feasible path for future large-scale deployment.
The team also proposed a very practical idea: to install this type of array directly on the existing 5G base station tower, so that it can "part-time" as a satellite ground station while maintaining the original cellular communication function. Since these tower stations themselves have basic conditions such as power supply, optical fiber backhaul and site leasing, the marginal cost of superimposed deployment of ArrayLink will be much lower than that of new dedicated ground stations, which is expected to accelerate the expansion of global satellite access infrastructure.
Currently, ArrayLink is still in the experimental research stage and has not yet completed end-to-end verification on real orbiting satellites. The research team is continuing to optimize the system design and explore how to solve engineering challenges such as coordinated control, operation and maintenance management, and compatibility with existing satellite network architectures in large-scale deployments.
From an industry perspective, as the number of low-orbit constellations continues to surge and satellite capabilities continue to increase, it is almost inevitable that the ground station architecture will be forced to shift from "few but refined" large sites to a new paradigm of "multi-point distribution and flexible expansion". If a distributed phased array-based solution like ArrayLink can achieve breakthroughs in cost, reliability and standardization, it is expected to provide a feasible new infrastructure form for future global satellite communication networks.