History
August 05, 2024
The success or failure of a military operation often depends on the ability of troops to collect, analyze, and exploit vast amounts of information. As the volume and complexity of data continues to grow, so do the challenges and opportunities associated with leveraging big data for military applications.
Today’s warfighters are inundated with data from multiple sources, a stark contrast to the relatively limited data available during the wars in Iraq and Afghanistan two decades ago. In past conflicts, soldiers collected data primarily through traditional means such as reconnaissance reports, human intelligence, and satellite imagery. While valuable, these sources provided limited information compared to the multidimensional data streams available today.
Personnel on the ground generate data from a wide range of sources, including advanced sensors, unmanned systems, real-time communications, and satellite feeds. The volume of data collected is staggering. For example, aircraft and drone sensors can generate terabytes of data during a single mission. This data includes high-resolution imagery, radar, electronic warfare (EW) information, and much more.
There are many challenges in managing all this data and making it useful to warfighters, says David Mercado, director of field engineering at Wind River (Alameda, Calif.). These include “shortening the OODA (Observe, Orient, Decision, and Act) loop to accelerate and optimize battlefield decision-making (i.e., speed and accuracy); collecting data from all relevant sources (which can be overwhelming in terms of volume and speed) such as sensors or devices; interpreting that data in situ and in the context of data from other sources; and analyzing the data to understand relevant potential courses of action.”
The visualization of this data has also evolved: previously, data was processed and analyzed by a limited number of analysts and then disseminated in reports. Today, data is often visualized in real time using advanced software platforms, which provide commanders and soldiers with immediate information and actionable intelligence.
Big Data Requires Cutting-Edge Technology
Ideally, this data should improve situational awareness, inform strategic and tactical decisions, and predict future threats.
For example, data collected by drones and surveillance systems can provide detailed images and movement patterns of enemy forces. Communications intercepts can reveal plans and intentions, while logistics data ensures that resources are deployed effectively. This holistic view of the battlefield allows commanders to make informed decisions that improve operational effectiveness and reduce risk.
The problem is that the usefulness of this data is highly dependent on technology: can all this data, however useful, be processed and transmitted to the warfighters in a timely manner? Otherwise, it is useless.
This is where advanced technologies like artificial intelligence (AI) and machine learning (ML) come in. These technologies can quickly and accurately analyze large data sets, identifying patterns and trends that would be impossible for human analysts to discern.
“By sorting and deciphering data using learned and programmed patterns, AI can fill in gaps in the data (infer), make decisions, and recommend or even take actions faster than a human operator,” Mercado explains. “Over time, as the AI engine is exposed to more data and patterns, it also has the ability to recognize trends in the data, predict outcomes, and adapt accordingly.”
For example, Wind River’s VxWorks integrates AI and ML frameworks to optimize embedded systems for real-time data processing. This allows systems to analyze and act on data with minimal latency, the company explains. (Figure 1.)
(Figure 1 | Wind River’s VxWorks is a real-time operating system (RTOS) that provides a scalable and secure environment for mission-critical computing systems.)
According to an Abaco Systems spokesperson, this challenge can be addressed with technology that can process massive amounts of edge sensor data using AI and ML, with the goal of ensuring that only the most relevant and critical data is distributed across the network. The idea is not only to conserve bandwidth, but also to enable faster decision-making across all domains for the Cross-Domain Command and Control (CJADC2) framework, they explain.
For these applications, Abaco offers the SBC3901, a 3U VPX single-board computer (SBC), to handle complex AI and ML tasks in challenging environments in real time. (Figure 2.)
(Figure 2 | Abaco Systems SBC3901 single board computer.)
Predictive analytics
Predictive analytics is another promising application of big data in the military. By analyzing historical data and identifying trends, predictive analytics can forecast future events and trends, allowing leaders to anticipate and prepare for potential threats. For example, analyzing past supply chain data can help predict future logistics needs.
By monitoring data such as fuel consumption, maintenance records and deployment history, forces can anticipate equipment failures and perform maintenance proactively, reducing downtime and extending the lifecycle of an asset.
Integrating real-time data analytics also facilitates decision-making on the battlefield. Commanders can receive up-to-date information on the status of their forces, enemy movements, and environmental conditions. This real-time data allows for rapid adjustments to strategies, troop movements, and tactics, increasing the likelihood of mission success.
Challenges in data access
The military faces bandwidth limitations. The world is saturated with sensors, communications devices, and unmanned systems, all generating massive amounts of data. Transmitting this data in real time to command centers or data processing facilities can strain existing communications networks, leading to latency issues and potential data loss. All of these operations are made even more challenging in remote or hostile environments where infrastructure may be limited, compromised, or destroyed.
“Battlefields face challenges of bandwidth limitations, communication latency, interoperability between services, and data security,” Mercado says. “Today’s data collection systems provide access to a staggering amount of information that must be correlated and evaluated.”
Another factor: the security of data transmission and storage. Secure data encryption and robust cybersecurity measures are essential, but they can add layers of complexity and slow down data access and processing speeds.
The Growing Importance of Real-Time Data Visualization
Collecting and storing data is just the beginning. Making that data useful requires overcoming several additional challenges. One of the most important is the ability to visualize it for the warfighter.
Integrating this disparate data to form a coherent picture is not an easy task, requiring sophisticated algorithms and advanced analytics tools. But it is a significant puzzle to solve, Mercado says. “The point of data in battle is to improve combat effectiveness – data at high speed and at large scale – to use data faster than adversaries,” he says.
Wind River’s VxWorks seeks to address this challenge by providing a highly reliable real-time operating system (RTOS) that collects data from various sources, interprets that data, and delivers it to the warfighter at the edge.
