Exploiting advanced imagery with a standards-based network architectureStory
May 16, 2009
Modern military forces rely heavily on sensor-based imaging systems, but extracting only the most valuable imagery information and getting it into the right hands quickly can be challenging. However, a new Converged Sensor Network (CSN) Architecture is delivering critical sensor-based information to users on the tactical edge.
New generations of sensor-based imaging systems are generating invaluable information for our military forces, using increasing numbers of both manned and unmanned platforms. Video imagery is now highly detailed, but it's not just video; Synthetic Aperture Radar (SAR) and Electro-Optic Infrared (EO/IR) sensors can create detailed images at night and through cloud cover. Soon Hyperspectral Imaging (HSI) and Laser Radar (LADAR) will additionally advance our imaging capabilities. These technologies operating from new platforms provide an unrivaled ability to gather detailed imagery of any geography, giving our armed forces both strategic and tactical advantages in current and future conflicts.
That is all good, but a significant challenge remains: How do we extract truly valuable information from all that imagery and then get that information to the people who really need it, when they need it? Imagery by itself is often of little value; for example, seeing tire tracks in the soil outside a village is not too important, but knowing they were not there the previous evening may be critical information. And the person who needs that information is probably not an analyst in the Pentagon, but a war fighter preparing to enter the village.
A flexible network or a network of networks is needed to link sensor systems to the ultimate users of the information, and the Converged Sensor Network (CSN) Architecture is delivering this capability.
Data imaging challenges
This challenge of extracting and delivering useful information actually has several dimensions. First, there is the issue of bandwidth limitations in the data links that provide the transmission backbone from a sensor platform to a ground station. Currently deployed data links cannot transmit the full data streams generated by advanced sensors. Improvements in data communications will increase available bandwidth, but data links will still be overwhelmed by ever-larger streams of raw data from new generations of sensors.
A second issue is the difficulty of communications between different types of platforms and systems. Too often, information has to flow all the way up, then down different chains of command because there is no good way to move imagery directly from point A to point B, even though both are out on the tactical edge.
And lastly, there is the issue of merging and synthesizing information from multiple sensors. For example, combining infrared and radar images may give real insight into what is happening at a particular place, but allowing that combination to happen quickly is very difficult with today's systems. Because the infrared and radar images are produced by separate systems, focused human intervention is needed to create combined coverage of a specific physical location.
Addressing these interrelated issues requires a new paradigm for processing and communicating sensor-generated data. Military operations need flexible networks of intelligent nodes that can process imagery, share that information, and deliver it quickly wherever it is needed (Figure 1).
Figure 1: The CSN Architecture vision encompasses a flexible network of networks.
Converged Sensor Network Architecture
The CSN Architecture provides a flexible, standards-based approach to creating such a network, using a novel approach for digital signal processing that conjoins the agility of cluster computing with sensors at the tactical edge. Rather than funneling all network traffic through a host processor, the CSN Architecture makes each sensor and its associated embedded computing assets look like a standard cluster on an IP-based network.
To enable the CSN Architecture, Mercury Computer Systems has introduced a set of SigmaNET software and hardware components, including:
- Interconnect software that upgrades the RapidIO fabric to support high-performance Ethernet tunneling.
- A gateway hardware module that bridges a RapidIO subsystem with interconnect software installed on an external 10 GbE network. The gateway module is responsible for forwarding Ethernet traffic into and out of a RapidIO subsystem.
- Failover software that enables a redundant RapidIO fabric to isolate faults and recover from failures, including failures in an external network, via routing changes.
Thomas Roberts is a product marketing manager at Mercury Computer Systems. He has more than 25 years of experience in systems engineering and technical marketing with IBM, Nixdorf, Data General, Digital Equipment, and Compaq. He holds a BS in Engineering from Cornell University and an MBA from the University of Kansas.