This article originally appeared in the August 2022 issue of Security Business magazine. When sharing, don’t forget to mention Security Business magazine on LinkedIn and @SecBusinessMag on Twitter.
We often judge performance of a product by its specifications. While valid, it overlooks the role of the network that connects these devices. Regardless of how well a camera or server performs, its specification is meaningless if its signal quality or even the signal itself cannot reliably travel from point A to point B. The role of the network is the most significant part of performance and reliability.
Analog security systems were simple – just connect the cable and go; in fact, even the most complex video components simply required connecting two ends of a coax cable.
IP/PoE security systems are far more complex, and adding to this complexity is the assumption that “standard” networking products – those designed for data files – will work for security video applications. Based on this concept, each network must be designed based on its own application requirement. There is no one-size-fits-all solution.
Initial Considerations
When approaching network design for security applications, start with the simple concept that security networks have unique requirements that are not addressed by standard networking equipment.
For security systems, the most challenged networks involve cameras powered by PoE so we will focus on cameras, but the same requirements can be applied to any type of device.
There are two components to security networks: Bandwidth and PoE. Both are moving targets. Since the introduction of IP cameras powered by PoE, their demands for increased bandwidth and PoE have grown. Many times, these requirements have increased faster than the ability of developing standards to support them. All of this adds to the confusion and complexity of network design for security applications.
Making a purchasing decision solely based on the performance of an individual component without considering all the network components and connected devices is a recipe for failure.
Following the standards for networking set by the Institute of Electrical Engineers (IEEE) the cable must be CAT 5, (later CAT 5e, 6, 6a). The distance can be no more than 328 feet (100m). PoE transmission must also comply with this and adds very specific limited ranges of power and signal signatures for the source and destination to recognize each other.
5 Steps to Designing Networks for Security Systems
Integrators must be cognizant of five key considerations for designing secure networks for security technology. It starts with the connected device itself, and progresses to Power over Ethernet compatibility, cable choice, the number of connected devices per location, and finally to collection devices. Here is a closer look at each of the steps:
Step 1: Powering the Connected Device
Camera pixel sizes can range between 2MP to as high as 33MP. The ability to process the camera resolutions must remain consistent throughout the network – even though the resulting image packet size will change, depending image activity and composition. If a network does not have the ability to handle packet sizes or bandwidth, the result is image distortion and losses.
Image problems do not mean much if you cannot power the cable. A camera specification indicates a specific amount of power. First, determine if that specification indicates the required source power or the power required at the camera. Cameras generally will surge about 20% as accessory functions such as day/night, LED turn on, and PTZ operation is activated. If these functions result in exceeding the ability of the PoE source to provide the necessary power, the source can stop providing PoE, shutting down the camera.
This is often the case when a user discovers some of their cameras stopped operating overnight when functions such as day/night and LEDs are most active.
Rule one: If the spec sheet notes PoE power as both a value and PoE class, always provide enough PoE power equal to the highest level of the class PoE.
Step 2: PoE CompatibilityNot only is the amount of PoE important but also the type. PoE standards or lack of means that even if you have the right amount, it may not work with the connected device. PoE is defined in several ways. One of them is type. 802.3af, 802.3at are standards that also defined PoE levels. Recently a third standard 802,3b was adapted as a standard raising PoE levels to 90 watts. In the need for increased power camera manufacturers did not wait for standard approvals. This led to non-standard types such as PoE++ and UPoE. Most important are not compatible with 802.3ab. A device requiring 60W but based on UPoE will not work with connected device requiring 802.3bt.
Rule two: The source must match the device requirements.
Step 3: Transmission and Cables
Start with knowledge that IP/PoE transmission was only designed for CAT cable. The requirements such as resistance and capacitance are very specific for a PoE source to successfully power a connected device. Outside a limited range, either the connected device will not power up or will not successfully be able to handle the normal PoE surges that occur resulting to the device going offline.
There are various applications and installations that require going outside the CAT cable at a distance limitation of 328. These include CAT cable at distances greater than 328 feet, and the use of coax and single pair. Regardless of the type of cable or cable distance the two critical values of cable resistance value CAT cable of 22 ohms and for Coax between 19 to 33 ohms and cable capacitance of under 10uf. All types of cable will have different characteristics which will affect performance.
An additional consideration is how much power a cable can handle. With reference to CAT 5 cable, which is 24 AWG, maximum power is limited to 37W within the range of 802.3af/at. As PoE requirements increase to greater than 37W more pairs are required. CAT cable consists of 4 pair, but coax and single pair are limited to only a one pair, which limits their PoE capacity to a maximum of 37W.
Step 4: Bandwidth
The number of cameras routed to a single location will determine the total bandwidth required. In general, that point is a network switch. Packet size must also be taken into consideration. Can your selected switch handle jumbo frames at 100Mbps? Can it pass all connected cameras operating at the highest bandwidth? Without answers to these questions video frames (packets) can either be rejected by the switch or pass through with distortions.
Step 5: The Endpoint
The endpoint has two considerations: First is how switches are configured; then the server or NVR.
General switch configurations are either IDF to MDF where one switch in the field is directly connected to a main switch usually referred as a core switch, or where several switches are connected in a daisy chain configuration. In both, the ability to maintain bandwidth and packet handling is reduced. The highest bandwidth and transmission values must be maintained throughout the total path.
Finally, there is the server or NVR. Recording devices can have three individual paths: Recording, viewing and playback; or viewing and recording can be combined with playback being separate. Can the recording/viewing device handle the bandwidth? Regardless of configuration an endpoint must be able to handle all the images transmitted to it.
Neil Heller is VP of Business Development for Vigitron, a provider of IP/PoE transmission and networking solutions. The company offers several educational resources on the unique requirements of network design for security applications. Email [email protected] for details, or request more info at www.securityinfowatch.com/10215546.
