Advances in security camera technology, data analytics, digital video storage and mobility applications have dramatically improved the performance and capabilities of electronic security systems.

These functional improvements also have changed the time-tested fundamentals of security system programming, design and implementation.

The required level of understanding goes far beyond specification of security electronics, reaching into a wide variety of facility infrastructures and data network systems. Collaboration also has grown more complex, requiring that managers and their senior technical staff be involved throughout the design and implementation process.

Even operations and maintenance strategies must be examined closely and defined to maintain alignment among all parties.

The new generation

Historically, access control and closed-circuit television (CCTV) surveillance equipment existed as proprietary systems, built on analog technology and installed independently from one another. Most CCTV systems comprised analog cameras, single-application coaxial cable connecting devices, a centralized recording device and a central termination point colocated with camera monitoring equipment. Specific to ongoing system operations and maintenance, the facility would contact its local security integrator to repair problems and maintain the system.

But the new generation of Internet Protocol (IP)-based electronic security systems rapidly have become the standard around the world, largely accepted due to better system performance, image quality, more cost-effective installations and improved functionality that address many of the gaps and limitations of analog systems.

For many years, the initial focus when designing electronic security systems was identifying and coordinating equipment locations and ensuring sufficient space in the security room for video monitoring and associated equipment. Subsequent tasks would focus on coordinating rack locations and cable management while provisioning enough cooling to allow security staff to work as comfortably as possible. Processes for designing system architecture and specifying performance generally were equivalent to the time when technology transitioned from videocassette recorders to digital video recorders (DVRs). But the transformation to IP-based electronic security systems and network-based equipment and the introduction of the network-based systems has forced an entirely new category of intersystem reliance, network design coordination and planning complexity.

Consequently, health facilities professionals must find out early in the design phase whether the organization's information technology (IT) department will support security devices and applications on a converged IP network as well as whether the existing data network can support demanding video streaming from cameras. They must also know how and where the Power over Ethernet (PoE) will be provided to end devices, where the security video storage servers will be located and how access to security equipment as well as the organization's IT department equipment will be monitored and controlled.

Back to front

With so many complex issues involving multiple departments, consultants, scopes of work and spaces within a facility, it's important to begin with a discussion of the end devices and work back to the head-end equipment.

Surveillance cameras. Original IP cameras could only transmit Moving Picture Experts Group (MPEG) and Joint Photographic Experts Group (JPEG) video streams at 640 x 480-pixel resolution with significant variances in bandwidth requirement depending upon individual camera settings. Because of the novelty of IP cameras and the variance in data network design considerations, the vast majority of IT network administrators would not allow security cameras onto their converged data network.

Manufacturers quickly adapted by releasing a new generation of cameras that delivered H.264 compression, providing a high-quality video image that could be configured for a 5-megabyte video data stream. High-definition (HD) cameras were the next wave of products that, despite better compression tools, created another bandwidth concern. Higher megapixel cameras gave security staff incredible image clarity and higher resolution playback, allowing designers to deploy fewer cameras in hallways due to the exceptional quality of video images.

In a great example of "too much of a good thing," the HD craze has led to nearly all cameras being specified for high definition. In many facility locations, this is far more resolution than is needed at a higher price and bandwidth requirement. For example, when recording one IP camera using MPEG-4 video compression at a screen resolution of 640 x 480 pixels at 7.5 frames per second, the bandwidth requirement will be 325 kilobits per second (Kbps). This means the data network must be able to provide that camera an upload speed of 650Kbps. The data network supporting the video storage device (DVR or network video recorder) must also be able to provide 650 Kbps download speed. In another instance, a 5-megapixel camera installed to monitor a door location can soak up as much as 22 megabits per second (Mbps) when a smaller, less expensive 1.3-megapixel camera performing the same function would only require 2.5 Mbps.

These complexities and resulting demands on IT infrastructure force an expanded perspective on the CCTV design process. It is very important for facilities professionals to keep total infrastructure requirements in mind when specifying camera locations and performance. Network cabling must match bandwidth requirements and protocol specifications of the cameras.

Video distribution. Each facility has its own unique space requirements. Large facilities have dozens of intermediate distribution frames (IDF), commonly mislabeled as data closets, that connect to a centralized main distribution frame which, in turn, connects to a data storage facility or data center. Each of these spaces is capable of supporting cameras and possibly video storage utilities. The first focus area for these spaces should be the network switches that will provide data connectivity and, more than likely, electrical power to the cameras. It is essential to determine the capacity of the data network to reliably support the surveillance system.

Health facilities professionals must find out how many switch ports are needed to support the number of cameras that will be serviced from the space, whether the existing or planned switches support PoE or a separate injection device will be required and whether the backplane and uplink capacity of the network switch can support the estimated bandwidth requirements of the cameras.

Some other issues include: whether there is a need to virtually isolate security camera video streams from other types of enterprise data traffic or will a physically separate network be required; whether the combination of video streams and other organizational applications serviced by the switch can be accommodated by an existing or new switch; what the physical and virtual paths are that video streams need to connect with the storage servers; and where the video storage servers will be located.

Indeed, video storage server location is critical because there may be several network devices beyond the IDF that must be evaluated for capacity, performance and configuration. In some cases, it's necessary to analyze the viability of centralizing video storage at remote geographic locations.

Recently, early designers for a large campus project assumed a remote video storage solution without fully evaluating bandwidth requirements. As the design grew more detailed, outside experts were called in to review the design. The significant amount of bandwidth required was quickly calculated as was the total cost of the service provider links that would transmit video streams to the central storage location. Ultimately, the total cost of leased Internet links was so extreme the original plan had to be redesigned around local video storage with remote access available as needed. The team was lucky to identify this issue before equipment had been purchased or service provider contracts were being adjusted. If the flaw had been discovered during deployment, the cost would have been much greater.

Video storage. Specifications for the storage media often can be difficult to determine. The challenge is not so much selecting specific manufacturers as it is calculating specifications for bus speeds, processing capacity, flash media thresholds, hard-drive speed and capacity, network card performance and levels of redundancy. Each of these factors must be measured against the total performance capabilities of the surveillance system, including the video stream and video quality settings for each camera, total number of streaming devices, frame rates per device, hours each camera is configured to record daily and the total days of video storage required.

Compiling these calculations from each camera into a complete and accurate performance model is a challenge. In many cases, senior solutions designers closely collaborate with network design engineers to make sure no data are missed or that some data have not been mistakenly exaggerated, thus leading to unnecessary cost. Fortunately, this process has been automated recently by a variety of manufacturers who are providing Web-based camera and storage calculators at low cost or free, depending on the level of sophistication provided. One particularly useful calculator is called Workbench by Iomnis Surveillance Solutions, Houston. The free tool can be accessed at Adding an online calculator to the planning tool kit is a great way to speed up the performance modeling as well as improve the accuracy of calculations.

Video monitoring. The final pieces of the surveillance system puzzle are to identify if and where camera views will be monitored, and designing the presentation of the camera feeds to meet the needs of the security staff. Common wisdom argues that because the video streams are riding the data network, camera views should be readily available anywhere a computer or laptop is located. Technically, this is true, but several issues must be considered and resolved before finalizing the design approach.

Specifically, health facilities professionals must determine the bandwidth demands that will be placed on the data network when video streams are routed from central storage back to the personal computer; How many cameras need to be viewed simultaneously by security staff and what specific cameras are required for viewing?

Once the specific camera feeds, bandwidth implications and optimal presentation of those feeds to monitoring personnel are determined, hardware can be engineered as needed for video displays and video cards. Working with users is extremely important in gauging the best number of video streams for various-sized video displays. After the desired user experience has been finalized, the computing hardware that will process the incoming signal and project the video content can be specified.

Health facilities professionals should keep in mind that the resolution of the video monitors, processing power of the central processing unit (CPU) and video card capacities can become bottlenecks for high-resolution video. If these components are underpowered, video feeds will be distorted with jitter and lag or staff will not be able to access the full resolution being produced by cameras. During assessments, it's not uncommon to encounter HD-1080p cameras being viewed by monitoring staff on HD-720p video displays, thus impairing the efforts of the security team. Video imagery that is not smoothly rendered causes oversaturation of CPU capacities and leads to poor performance.

A new era

As new technologies have brought great enhancements to functionality and flexibility, the processes for planning, designing and implementing surveillance systems have greatly evolved as well.

In many ways, this consolidation of technologies and functionality further emphasizes the efforts of health care organizations to fully align their staff and assets with customer service and employee satisfaction.

Security, in particular, is a fundamental value that reaches throughout a health care organization to protect people and assets alike.

Scott McChesney is the senior security systems designer and Nathan T. Larmore is a principal and practice leader of technology services at Sparling. They can be reached at and

Sidebar - Network limitations disrupt video
Last year, a new hospital in the Pacific Northwest completed a complicated design-build project that included a new surveillance system equipped with more than 150 Internet Protocol (IP) cameras configured at 14 frames per second, 50 percent motion and H.264 compression.

During preconstruction, the integrator used an online calculator to determine data bandwidth requirements and thresholds, data switch backplane specifications and processing capacity of the storage servers. Additional considerations included deploying multiple video storage servers to balance overall system loads.

Real-time video monitoring was assumed to take place at any personal computer in the facility without identifying specific locations. Once hooked up to a local area network dedicated to the surveillance system, the complete solution was fully tested and commissioned back to the supporting data switches and video storage devices located in the facility main distribution frame. No testing of video monitoring locations took place.

Upon opening, the hospital security teams immediately reported poor video quality at the surveillance monitoring stations throughout the facility. Video imagery was choppy with severe latency affecting usability of the images. Eventually, quality was so poor that security staff deemed it unusable and called the installer back to troubleshoot. Troubleshooting included comprehensive testing on all components and applications, including the video storage server array.

It was concluded that the IP surveillance system was operating properly and the poor video quality was a direct result of inadequate bandwidth on the hospital's production data network. Although IP cameras were not using the production network, video streams from the centralized servers to each monitoring location were. This same network supported standardized business applications, Voice over IP traffic, patient education programming and various departmental data transfers.

While planning for the surveillance system was detailed, it was never coordinated with the design of the hospital's production network.

Sidebar - Camera issue creates problems
A large psychiatric facility opened with an elaborate electronic security system, including a surveillance solution that supports more than 300 Internet Protocol (IP) cameras deployed across hundreds of thousands of square feet.

The surveillance system was equipped with multiple integration points to parallel systems, including primary logic controllers, the facility access control system and emergency call stations throughout the campus. Surveillance cameras were configured at 14 frames per second, 75 percent motion and H.264 compression.

Infrastructure and connectivity of the networked systems was similar to many modern facilities, utilizing a dedicated local area network (LAN) to transport video streams from cameras to the centralized video servers in the campus data center. The completed system was fully tested and commissioned prior to the hospital opening.

Once the new facility was occupied with patients and staff, the surveillance system immediately began to experience problems.

Oddly, the system worked as expected at night, but would fail during the day. The integration team was called back to troubleshoot the system and completed extensive testing and analysis in an attempt to pinpoint the cause. After three weeks, the root cause was determined. The motion settings on the cameras were triggering higher than planned video streaming during the day.

Motion settings on cameras were designed to minimize unnecessary video streaming and recording when there was no activity in the field of view. These settings are an excellent means of building efficiency into network planning and video storage capacity. However, in the case of this project, the motion settings were creating such excessive daytime bandwidth utilization, the supporting LAN could not sustain the load.

Ultimately, each camera was evaluated and reprogrammed and the owner had to accept reduced frame rates per second.