Satellite Collision, Space Debris and PETs – How Privacy-Preserving Technologies can Advance Space Situational Awareness

In February 2009, the first ever accidental collision took place between two satellites in earth’s orbit: the U.S.-based Iridium 33 and the inactive Russian satellite Cosmos 2251. The collision produced almost 2000 pieces of space debris. 

This incident raised serious concerns about the Kessler Syndrome, which describes a phenomenon where the amount of debris in orbit around earth would reach a point that would lead to a chain reaction of collisions that in turn would produce more and more space debris. Too much debris could ultimately block access to space and cause severe problems for satellites and the global space community.

The exponential growth of space debris.

In this day and age, it is critical that we do the utmost to prevent satellite collisions. With applications including television, weather, navigation, scientific research, military reconnaissance and much more, there are over 8,000 governmental and private satellites in orbit – and collisions continue to occur. Take for example the 2021 breakup of Yunhai 1-02, a Chinese military satellite that collided with a small piece of space junk, creating at least 37 debris objects.

In this blog post, we will explore the global challenges of effectively sharing and analyzing space situational awareness (SSA) data for the purpose of satellite collision avoidance, and how Privacy-Enhancing Technologies (PETs) advance the sharing of sensitive data to help mitigate such risks.

Challenges in Sharing Sensitive Data

The global problem of space debris can only be solved through cooperation. Following the Iridium-Cosmos crash in 2009, the U.S. government began making more SSA data publicly available and notifying all entities of collision risks. The U.S. Space Force’s 18th Space Defense Squadron has long had the most extensive SSA network in the world, and in order to share higher quality data, the U.S. currently has more than 170 data sharing partnerships. 

Collected in part from ground-based sensors, highly accurate SSA data is considered among the most sensitive military information. As a result, the U.S. government must balance working with satellite operators and protecting data. The U.S. tends to take a conservative approach to data sharing, adding synthetic noise to public domain SSA data, for instance, which can reduce the accuracy of space object tracking information shared with other parties. 

This security-focused approach has led to trust issues around SSA data. For example, some South Korean government officials estimate that their country receives data on only about 40% of the objects tracked by the Department of Defense. A lack of insight into processing methods is an additional concern. A 2021 paper published by the European Space Agency (ESA) Space Debris Office presents both a literature review and stakeholder survey. One of the clearest findings is a general dissatisfaction with the current quality and timeliness of SSA data and recommends areas of improvement. 

An Evolving Data Landscape

In fact, a key driver for recent advancements in countries and regions developing independent SSA capabilities may be the lack of confidence in shared SSA data. The number of countries, regions, and commercial entities with SSA capabilities has proliferated. The EU launched its own Space Surveillance and Tracking (SST) program in 2014. Russia, South Korea, and India are among the countries that are developing or improving national SSA capabilities. Commercial SSA providers have also emerged, selling data and services to satellite operators. In fact, the commercial sector now has more sensors in the southern hemisphere than the U.S. government.

Satellite operators are increasingly integrating multiple data sources – an opportunity to improve the resilience and redundancy of collision avoidance systems. Some countries, including Canada, Japan, and Australia have defined interoperability with U.S. systems as a goal in developing their own SSA systems. Driven in part by the quality of commercial data, a current initiative for the U.S. Commerce Department to take over civil space traffic management is aimed at creating a system that draws SSA data from the Department of Defense as well as commercial and international sources. 

Machine Learning for Satellite Collision Avoidance

As the quantity of data and the complexity of the satellite environment continues to compound, space traffic management is becoming increasingly vested in leveraging artificial intelligence (AI) and machine learning (ML). ML, a branch of AI, involves developing algorithms that can learn from data, gradually improving accuracy. These algorithms can then be used to uncover insights in new datasets. ML models are especially effective and relevant in applications with large amounts of data – such as, potentially, space traffic management.

The European Space Agency (ESA) looks to ML for calculating collision risks as it could prove more effective than other methods. In 2019 the European Space Agency held the Collision Avoidance Challenge. A number of conference papers about ML and satellite collision avoidance, using a dataset of real-world historical conjunction data messages released by the ESA, appear on their website.

In the U.S. the 18th Space Defense Squadron is emphasizing AI and ML because of the need to integrate and analyze data from various sources. According to a recent article, the goal is to incorporate AI/ML in areas including parsing data, predictive maintenance, and predictive maneuvers – yet the challenge is “making the data accessible for industry partners to help us truly take advantage of what the models can inform.” In other words, how to collaborate on sensitive data remains a stumbling block.

PETs Enable Cross Industry Collaboration 

Those concerned about the need for collaboration on sensitive SSA data can look to other industries that have applied innovative privacy-preserving technologies to address such challenges.

For example, a large defense contractor needed to monitor the performance of its fleets in order to improve usage analysis. However, parties were not willing to share information to support this analysis – so overall maintenance performance suffered. Inpher’s XOR platform, which utilizes Privacy-Enhancing Technologies (PETs), including Secure Multiparty Computation (MPC), enabled the defense contractor to privately compute distributed data across fleets without sharing specific mission critical data. The improved maintenance models deliver operational value to fleet managers.

In the financial sector, a large multinational financial services firm wanted to improve their fraud detection and risk models by accessing more data. But fraud data sharing initiatives had historically been difficult to implement for reasons including jurisdictional restrictions, data confidentiality concerns, and trade secret sensitivities. Inpher’s XOR platform enabled the firm to train their fraud and risk models on distributed data from participating partner banks without sharing individual data inputs. This led to improved predictive accuracy while maintaining all parties’ security and privacy requirements. 

PETs also address security concerns, as encryption-in-use safeguards even against quantum attacks. While to date few publicly known cyber attacks have targeted space systems, multiple countries possess the cyber capabilities to conduct such attacks, and a growing number of non-state actors are discovering vulnerabilities in commercial satellite systems. 

Conclusion

In a context where advanced data analysis and collaboration on sensitive data are critical to avoiding the out-of-control Kessler Syndrome, the space sector should look to innovative technologies that are enabling privacy-preserving data collaboration in other industries.

Learn more about secure data collaboration.