Digital video and GbE backbones keep pace with video sensor proliferationStory
September 04, 2012
The use of legacy analog and newer digital multispectral, daylight, infrared, and HD video sensors continues to proliferate at a rapid rate on military platforms such as ground vehicles.
The use of legacy analog and newer digital multispectral, daylight, infrared, and HD video sensors continues to proliferate at a rapid rate on military platforms such as ground vehicles. The demands of new multisensor video-intensive applications – such as 360° situational awareness display systems that enable uninterrupted visual display of the external environment – have increased the importance of open-architecture-based Video Management Systems (VMSs). Today’s VMS technologies are helping system designers keep pace with sensor proliferation by embracing digital video switching techniques and high-bandwidth data backbones such as GbE and faster, to provide operators with greater flexibility and optimal use of greater amounts of incoming data.
More sensors equals more video channels
Typical system requirements expand from 2 or 3 video channels to designs that now demand as many as 12 or more channels. In the meantime, high-resolution HD video displays have become common. As sensors multiply, system architects are confronting a mix of analog and digital interfaces. Some sensors are driven directly to a dedicated single-mode display while others are displayed on demand on multifunction consoles. To handle the confusion of cables, sensor standards, and multiple displays, the COTS market has worked to develop open-standards-based architectures that leverage technologies such as digital network-based switches. Sensor vendors are now starting to add native support for digital networking interfaces, making it easier to add sensors into a digital switch.
Moving to a digital backbone
The move from traditional analog video switched architectures to Ethernet is speeding and simplifying video acquisition and transmission, while providing a common infrastructure for data and control. With GbE backbones in vehicles and the increasing use of high-speed HD serial digital interfaces – such as the 1.485 Gbps High-Definition Serial Digital Interface (HD-SDI) and 2.970 Gbps 3G-SDI – digital sensor data, together with analog video after conversion, can be handled by an internal common digital video data standard.
An example of an advanced VMS is Curtiss-Wright Controls Defense Solutions’ VRD1, a turnkey technology that combines A/D video conversion, digital switching, scaling and windowing, compression, networking, and recording onto solid-state storage in a “configure and go” rugged enclosure (Figure 1).
Figure 1: The VRD1 from Curtiss-Wright Controls Defense Solutions
(Click graphic to zoom by 1.9x)
New ICs bring better video switching performance
Semiconductor vendors are increasingly adding support for high-density digital video switching. The emergence of GbE as the vehicle’s data backbone enables system designers to take advantage of this enabling technology via highly dense digital matrices that can support multiple video inputs simultaneously, which, after conversion into digital formats if originally analog, can then be sent out to multiple outputs and display stations. Another semiconductor trend that is helping to improve VMS performance is the trend of the past couple of years to use FPGAs rather than the specialty, dedicated ASICs traditionally used for advanced functions. A common digital environment enables greater availability of these desirable functions such as video scaling, windowing, quad display, and picture-in-picture. As FPGAs have become more powerful, they have provided VMS designers more flexibility. For example, rather than just rescaling a single video input, FPGAs enable multiple video inputs to be combined. Today it’s not unusual to support as many as eight video inputs on a single FPGA, which can then simultaneously perform various windowing and scaling functions.
The flexibility of digital video
As the number of video sensors increases, so does the amount of video data. This raises the risk of “information overload” as human operators try to manage the multiple displays and flood of real-time data. The move to a digital switched architecture helps minimize this problem by making it easier to move to multifunction displays, making data more useful and manageable for the user. The digital approach also makes the configuration of video stations more flexible, because the move away from dedicated one-to-one sensor/display interfaces makes it much easier and faster to add additional operator consoles or to display in a platform. The move to digital video switching also results in very scalable architectures because the VMS separates the sensors from the displays, making it effortless to add additional sensors without adding displays.