Big Data, a marketing term on everyone's lips today, in fact does not refer to a new technological invention that only evolved lately. In fact, Big Data has a longtime history. Traditionally, governments were always confronted with the need for processing large amounts of data in real-time, for example when monitoring the airspace. If you are a frequent flier, you will feel safe when flying, since the airspace is being continuously monitored. But not only civil air traffic is being watched. For example, the surveillance allows radar operators determining whether non-cooperative aircrafts commit an overflight violation. The observation of airspace and the search for possible attackers is even more strictly performed, when, for example, a country hosts a major event such as the Soccer World Cup, or if an armed conflict occurs in a nearby state.

Technologically, it is the so-called Multi-Sensor Data Fusion, which allows a real-time situational awareness of an environment. This situation analysis is more than just displaying the sum of all raw input data from diverse sensors. The situational awareness arises from the pre-processing of raw sensor data, and then their meaningful composition - the Fusion -, which provides de facto new information with the aim of being accurate and complete.

For Multi-Sensor Data Fusion, many sensors of different type and design - and therefore referred to as non-commensurable data sources - are scanning a particular environment. Objects of the environment that are being observed will generate data - very heterogeneous data, such as oscillations or text or quantitative data like figures. The so-called Sensor Level Fusion pre-processes such raw data before handing them over to the very Fusion Engine. The Fusion Engine combines the data to more meaningful, relevant and new information. The Fusion Engine can detect and classify objects which are not natural to the observed environment: for example, a non-cooperative fighter aircraft, an obstacle in the way of a moving vehicle, or explosive mines.

While Multi-Sensor Data Fusion is used by governments to ensure the safety of their civil societies, at the age of Big Data it is also increasingly used by the commercial sector. (Partially) autonomous automobiles rely on Data Fusion, and so do the Internet giants. Their target, however, is no longer an object, but the Internet user himself - an individual, a person. Sensors observing persons are smartphones, their networked next-generation automobiles, their fitness wearables or their smart home technology - in short, the "Internet of Things". Fusion Engines analyze the output data of personal sensors, putting together a virtual, but complete and quite accurate picture of a person. A person is monitored, then analyzed, classified and standardized - with unpredictable consequences for a person’s future, which will no longer depend on education, social status, or trust. It will be replaced by the judgment of a machine, the Fusion Engine, which will know everything about a person and even draw conclusions for her future: Is the person married, divorced, over-debted or previously convicted, and what will be her probable future financial requirements or her probable health status in five years from now… Applying data fusion technologies to persons and, even more, selling the new information derived from data fusion technologies, e.g. to banks or insurance companies, violates European constitutions with their protection rights for the inviolation of human dignity. Such violation was never proposed by the early architects and pioneers of Data Fusion; however, it turns out to become the biggest social challenge of our decade.

Teramark Technologies strictly disapproves applying data fusion technologies to individuals. However, the benefit of increased safety remains intact, when applied by governments under their strict laws to observe the civil rights of their people. With almost twenty years of experience in implementing data fusion systems based on artificial intelligence, Teramark’s core team has pionieered Multi-Sensor Data Fusion and implementations that really work. And, we have an equally long experience operating these types of complex intelligent systems.