Why Satellite Edge Computing?
Technology has reached unforeseeable heights in the last few decades, and we run on the advancements made because of science and technology. The world is becoming more and more data-driven. Large volumes of raw data is generated in nearly every industry and digital activity. Storage, movement and processing of this data is an important activity that needs to be facilitated efficiently to maximise speed, reduce latency and reduce cost. This is where edge computing comes into play. Edge computing refers to the form of data computation carried out precisely at the data generation site or closer to sources of data.
Satellite Edge Computing specifically refers to performing data processing on the satellite, treating satellite as the edge device. And truly so, satellites carry various sensors, optical cameras being the prominent ones. These generate vast amount of raw sensor data (raw imagery) that needs to be transported and processed. Edge computing eliminates the need to transport raw data and perform computation on board the satellite.
Here are the key benefits of Satellite Edge Computing that contribute to its growing interest in space data management across various missions.
1. Avoiding Bandwidth Issues in Data Downlink
Data generation on a satellite is the beginning of the data process lifecycle. Traditionally this data is downlinked to the computing site on the ground (e.g. cloud servers) and subsequently processed into information. Downlinking this data is infrequent due to the dependency on ground stations. A satellite pass is when a satellite passes over the ground station to downlink data and this is limited to a few times a day. Each satellite pass averages 10–15 minutes and this limits the amount of data that can be downloaded. Moreover downlinking data is expensive. As an example the AWS Ground Station bills between $10 per minute to $22 per minutes to its satellite customers for downlinking data. Satellite Edge computing can eliminate the need for raw data downlink altogether, by moving the computing process closer to the data generation site i.e. on the satellite. This avoids bottleneck when several satellites are eyeing for the same ground station connectivity.
2. Speed to Insights generation
Timely insights from satellite data holds more commercial value and enable several use cases which could not be exploited otherwise. As soon as the data is captured on board a satellite, it can be immediately processed into meaningful information. As a result, this speeds up the decision-making process, thus bridging the gap between data generation and insights generation. As an example, monitoring the oceans using satellites for illegal activities is best done if insights are available immediately — know as actionable insights. Information older than 20–30 minutes is stale and holds minimum commercial value. After all, data processing and computing can only be said to be “real time” and “fast enough” when it mimics reality.
3. Improving efficiency
Data storage and data computing are inalienable organs of any satellite related activity. Reducing the bandwidth demands and lowering the networking costs makes the entire data computing process easier to manifest. Increased efficiency also contributes to closing the gap between real-time and data computing, which is the ultimate goal of edge computing. When it comes to satellites higher up in MEO and GEO then edge computing helps in eliminating the latency issue. Efficiency can also bought by distributing processing — partially on the Edge and partially on the ground. This flexibility makes the best use of resources, both the Edge and on the ground, instead of carrying out all the computing either on a Satellite or a distant computing location. Furthermore satellites can connect with other nearby satellites (via inter-satellite relay for example) to form a distributed processing platform bringing in more flexibility on the Edge.
Author: Gaurav Bajaj is the co-founder & CTO at Little Place Labs. With vast experience working in FinTech and Blockchain in UK and silicon valley, Gaurav transitioned to building a SpaceTech startup focused on bringing intelligence to satellites and space probes using the power of AI/ML. Gaurav is an Executive MBA candidate at Said Business School, University of Oxford focus on entrepreneurship, strategy and innovation.
Also published on Medium.