From a highly anticipated border protection program to a mess of malfunctioning sensors, solution providers and system integrators can learn some valuable lessons from the unsuccessful U.S.-Mexico border project.
From a highly anticipated border protection program to a mess of malfunctioning sensors, solution providers and system integrators can learn some valuable lessons from the unsuccessful U.S.-Mexico border project.
In March 2010, the Obama administration froze any new work on the Secure Border Initiative (SBI) project, diverting US$50 million in the planned economicstimulus budget for the project to other purposes. SBI network (SBInet), the technology and people components of the SBI, has been underway since it was announced by the US Department of Homeland Security in 2006. It was intended to secure the northern and southern land borders of the U.S., with a new, integrated system of personnel, infrastructure, technology and rapid response.
However, with $2.4 billion already spent on covering a fraction of the border, the team led by the Boeing Company has been largely criticized for, among other things, technical glitches and schedule delays.
A&S takes a closer look at the possible reasons behind the snafu and suggests how others can learn from when working with perimeters of such magnitude, as it is a From a highly anticipated border protection program to a mess of malfunctioning sensors, solution providers and system integrators can learn some valuable lessons from the unsuccessful U.S.-Mexico border project. By Monica Chang Treading Cautiously with Border Security 8262 Special Feature 22 OCT 2010 www.asmag.com 8272 huge market according to IMS Research, with the world market for perimeter security sensors projected to exceed $440 million by 2014.
Technical Challenges
Project 28, the initial deployment of the SBInet, is a 28-mile pilot project south of Tucson. The pilot virtual fence — nine mobile towers, radar systems, cameras and vehicles custom-fitted with laptops and satellite phones or handheld devices — was to be linked and then distributed to control rooms to enable agents to track targets and direct law enforcement resources. However, large amounts of sensor data could not be processed by Boeing's software, and the satellite communication system took too long to relay information to the command center. Radar systems were affected by rain and other environmental factors, and cameras that were expected to resolve images at 6 miles had trouble doing so at 3.
Radar detection should be successful in open areas when the number of moving targets is confined in a constrained area. In the case of Project 28, radar could not distinguish between people, vegetation and animals (of similar sizes) in bad weather. These circumstances render the system futile as it “lacks the information detail which is necessary to distinguish a pedestrian intruder in the presence of all this other clutter of approximate size and motion,” said John Romanowich, CEO of SightLogix. “The challenge further increases when you're trying to zoom in with a camera and have yet to distinguish the viability of a threat in the presence of so many nuisance alerts.”
Relying on a one-solution-fits-all technology can be dangerous due to the inevitable presence of defects, and perimeter protection is no exception. “Relying on analytic algorithms tends to be challenging, especially if the technology is not fully understood by those involved with the project,” said Yariv Lenchner, Product Marketing Manager of Nice Systems. The notion that sensor-based technologies and video analytics can provide 100 percent intrusion detection is simply not true. “Unrealistic expectations are unattainable.”
Targeting Terrains
The 20 sectors that divide the U.S.-Mexico border pose unique challenges based on weather, terrain, population and other factors. Tucson, for example, is 60 miles north of the border and especially difficult to work with due to its varied landscape and rugged land. When working with this kind of environment, it is best to assess each section as to ascertain the best and most cost-effective solution, said Lenchner. These factors include the number of poles, number of cameras per pole, exposure to sunlight, field of view and camera type.
Physical fences are good complements to a virtual fence as they sustain in various terrains, extreme weathers and densely populated areas. “Physical barriers are critical in any space in order to deter, detect and delay,” said Hagai Katz, VP of Marketing for Magal Security Systems. Katz states that 60 percent of Israel's borders are protected by fence, most notably, the Israeli- North border with Lebanon and Syria barrier. In critical areas, more cameras are placed, and in other areas, response forces are closer in distance than others.
As the terrain becomes more rugged and cluttered with rocks, land masses and foliage, few open areas are available to leverage the benefits of radar, Romanowich said. He suggested placing video sensors closer together than the radar system with each individual sensor covering a smaller area. “The advantage is that video surveillance would provide the necessary accuracy as a basic foundation to achieve operational control over the border.”
Jason Burger, Sales and Marketing Manager for Navtech Radar, believes the use of a shortrange, high-resolution system which employs a narrow beam can circumvent the problem of accuracy. “The radar will resolve/distinguish moving or stationary targets between obstacles and undulations.”
COTS = Mistake?
Another reason the undertaking was unsuccessful was perhaps because Boeing was tasked to use commercial off-the-shelf products (COTS) which were never intended for the varying terrain found on the border. “It's likely that there was an attempt to use a uniform solution across the border, when the application might have been better served by emphasizing different technologies for different environments,” Romanowich said.
Lenchner believes the issue lies not in using COTS, but rather the lack of understanding the solution and how to achieve it. “It's important to carefully consider the environment as well as the changes it brings, and the ability to integrate everything into a working solution. While intrusion detection technologies are mature, those dealing with them need to fully understand them,” Lenchner said. “This not only lowers the rate of false alarms but also allows the solution to deliver usable information about intrusion attempts.”
A better alternative could have been testing small rather than big as some issues are difficult to foresee. “The most difficult aspect of system integration when it comes to such a large-scale perimeter security project is trying to use a ‘cookie-cutter' solution to solve a diverse problem that constantly involves new variables,” Katz said. “As a border project involves influence on wildlife, immigration, laws and public opinion, it is best to take small steps and compose a series of tests on a part of the system before doing it big time.”
Integration and CONOPS
The challenges of wide-area security, with border security in particular, lie in system integration, process reengineering as well as technology deployment. “We think it lies in a combination of the three,” Lenchner said. “The integration challenge is not presenting massive amounts of generated information on a single screen. The real challenge is correlating the information, and setting up proper workflows so the command-and-control center is able to effectively handle events, and not be flooded with irrelevant information.”
The basics of integration should begin with concept of operations (CONOPS) before addressing further problems. “CONOPS is the name of the game,” Katz said. “Implementation is not that difficult once CONOPS has been strategically thought out.” Initial consultation is important in understanding customer needs and to define the performance criteria, Lenchner said. “This allows the company to confirm the project's validity. Then from the implementation phase and to the final phase, adjustments are fully monitored and measured, while enhanced setup iterations are done to improve overall system performance.” Katz added: “Customer needs may sometimes be overlooked by companies who only wish to use their products. For example, not everything necessarily requires IP or high-end technologies. For some areas, low-bandwidth, lean energy consumption and long-range technology is more effective.”
Initial Testing and Monitoring
Overlooking initial testing and progress monitoring is detrimental to any perimeter project as it affects the outcome and the budget. Extensive site surveys, CONOPS and risk reduction plans enable parameters to be adjusted while adhering to the project's original objectives. “This assumes a diverse interaction of system and human factors,” Romanowich said. “Basic to this functionality is accurate and actionable detection of valid targets. Since radar by itself does not have sufficient accuracy, it's unlikely that the CONOPS could be fulfilled through this technical approach for this environment.” A coverage diagram of the area and characterization of the radar's performance against different types of terrains prevent various issues that may come up further along, Burger added.
Advancements
Perimeter fencing, intrusion detection and camera and radar technologies have made leaps and bounds over the years. The following concentrates on how these advances cater to outdoor perimeter security solutions.
Perimeter Fencing and Intrusion Detection
Fusing multiple sensors from various technologies in order to correlate the information to a common picture is a challenge that continues to plague integrators today. However, standard IP interfaces and APIs that enable connections to multiple systems have made this problem easier. “Physical security information system is the concept that integrates a variety of information sources — video, sensors, geographical information systems (GIS), access control and others — to achieve a more holistic understanding of unfolding events. The introduction of IP networking and APIs has also eased the task to integrate substantially,” Lenchner said.
Cameras and Radars
Camera and radar technologies have advanced substantially for securing outdoor areas. Specifically, add-on features like edge computing abilities have enabled GPS-based video analytics and high levels of accuracy and accountability, to secure outdoor areas. “These capabilities can be scaled to large systems which are becoming easier to install, manage and maintain, thus inherently precluding gaps in systemic accountability,” Romanowich said. The end result could be better security solutions installed on time, on budget and, ultimately, yielding a stable solution for the end user.
Outdoor requirements and capabilities have also migrated to more complex algorithms that support various features — anti-passback, dual card access authorization, visitor escort, zone counting, intruder alarm setting and unsetting — as well as changes between those modes based on a time schedule, said Peter Francis, Sales Manager for Asia, Gallagher Security Management Systems
While optical and thermal technologies have had their share in technological advances, radar sensors are realistically the only technology to detect in all directions across large, open spaces, Burger said. “All other detection technologies suffer inconsistencies under poor light conditions or weather/wildlife challenges.”
Future Trends
Emerging trends include biometrics, intelligent video, unified situational awareness, GIS and GPS positioning for on-the-ground mobile forces and the use of layered approach combining multiple technologies, Lenchner said.
Biom etrics
Biometric technologies have, over the past 30 years, moved from a single application identifier with data stored in plain text to one where the data is encrypted and locked in sectors inside a chip built in a card, Francis said. Cost, convenience, identification token and invasion of privacy are important considerations.
Intelligent video
Video is the only detection source that provides the intrinsic detail to visually display the source of an alert. Thus, there is a growing trend to use intelligent video to protect large-area perimeters. “Essentially, all modalities of perimeter detection require video as a validation source today, such as fence sensors, microwave or radar,” Romanowich said. However, even though video can be used as the sole detection source for high-security applications, incorporating multiple modalities is often recommended.
At the end of the day, a no-nonsense, cautious approach that does not defy the laws of physics is the best course of action for any large-area perimeter security project. What's important is accurate security information as opposed to countless false alarms that cause nuisance and defeated attitudes. Boeing and its partners had to learn it the hard way, but others can learn and benefit from their invaluable experience.