Network Fundamentals for the Security Installer

Jan. 6, 2009

IP technology in the security market today continues to demand presence and attention from many industry professionals. With IT continuing to play an integral part of growing security technology, integration professionals are given the chance to learn networking from the IT perspective. Although networking concepts are not foreign to most security installers, as they’ve been creating networks of pan-tilt-zoom (PTZ) cameras using RS-485 technology for many years now, they face the challenge of absorbing 20-plus years of IT network specific knowledge at a rapid pace.

Know your network cabling

Be aware of the wiring and cabling that is available to you. Although today’s security market is seeing a growing presence of wireless technologies, it is vital to remember the basics, as cable serves as the medium of communication through which information is sent.

Copper

Unshielded twisted pair cabling (UTP) serves as a popular and most convenient form of cabling. The Electronic Industries and Telecommunications Industry Associations (EIA/TIA) established standards of UTP and grouped the wires into five basic categories, dependent upon the Mb/s traveled.

Copper Category 5 (CAT 5) cable serves as the long-time standard for networks and is generally capable of working at up to 10 Mb/s. CAT 5 is composed of four twisted pairs of wire wrapped by a common sheath. The CAT 5e cable, which according to EIA/TIA standards is the next generation of CAT 5, features a bandwidth of up to 100Mb/s due to increased manufacturing tolerances and techniques.

An increasing need for network speed created demand for higher-performance cable such as CAT 6. Rated at 1Gb/s data bandwidth, the larger conductors of CAT 6 have specifically designed fillers to maintain the positions of the individual pairs, which minimize crosstalk within the cable. Further development continues to drive the capability of these cables with CAT 6a cable rated at 10Gb/s.

Fiber optic

Present since the early 1970s, fiber optics is what one may be familiar with in talking about their telephone line, Internet or cable television system. Serving as another cost-effective solution, it provides two primary purposes in networks. It allows the distance between the core switches and the edge switches to exceed the 300-foot (100 meters) maximum distance of an Ethernet connection and provides transmission of signals over higher-bandwidth applications, such as data centers. Fiber optic cable is characterized by the size and type of the fiber. Ranging from single mode fiber to multimode fiber, the use that each provides depends on the applications in which it is used. Multimode fiber is the most common fiber in networks today and has a 62.5-micrometer core.

Networking 101

Cabling and electronic devices, such as hubs, switches, gateways, routers and wireless access points, make up the physical part of the network. Today, most of the traffic on an IT network uses the Transmission Control protocol/Internet protocol (TCP/IP), which can be considered the network’s “language.”

Network bandwidth is measured in bits per second (bps), which may seem strange to installers who are used to talking about bandwidth measured in Hertz or KHz. However, in both cases, the term bandwidth indicates the ability of the system to carry a signal without change or degradation.

As more and more network devices produced more data traffic, networks had to handle more bandwidth and data switches were created to address this need. A data switch acts similar to that of a video multiplexer by switching the full bandwidth from one source to another, but the switch makes the source selection quickly and automatically. The connections can be from any port to port because all of the ports are bi-directional. A data switch can handle more traffic than a hub since the connection is at the full bandwidth of the switch. Current production switches typically have 100Mb/s or even 1Gb/s bandwidth specifications.

Switches with modifiable parameters are known as managed switches, while units without configurable parameters are known as unmanaged switches. Initially, managed switches had a serial communications port that enabled configuration of the parameters via a terminal or PC. Yet, going from switch to switch to modify configurations was time consuming and as a result, switches today are generally configured over a network connection. The serial port connection is called out-of-band control, while a network connection is known as in-band control.

As the need to interconnect more equipment grew, IT couldn’t connect everything to a single switch, so they chain-linked multiple switches. This resulted in a design using a core switch and edge switch(above). The core switch, which is at the heart of a larger network, provides very high bandwidth specifications in the 10 to 100’s of Gb/s with a large number of ports to connect other switches and devices, such as servers. Edge switches are typically deployed in closets throughout the building to meet the 300-foot maximum distance of an Ethernet connection.

PoE gets the job done

A new enhancement to switches is the implementation of the 802.3af standard, known to the industry as Power over Ethernet (PoE). The standard enables the UTP to carry data and power to a device at the edge of the network. Many new IP video cameras today are powered by PoE connections. The digital data is carried on wires 1,2 and 3,6 while power is carried on wires 4,5 and 7,8.

The process for an 802.3af compliant connection is more complex than what one might think. When the standard was being developed, significant concern was expressed about non-powered devices connecting to a powered port and the potential device damage that could occur. As a result of these concerns, a power-up process was developed that protects non-powered devices from potential damage. PoE is also limited in the total amount of power that can be supplied. The 802.3af standard provides for a maximum of 350 mA at a voltage of 44VDC. This equals a maximum of 15.4 watts of power. In actual operation due to circuit constraints the usable power is only approximately 12.95 watts. There is a new IEEE standard under consideration that would increase the amount of power available via PoE to 25 watts, however the maximum power amount is a cause of concern by UTP cable manufactures due to cable heating at higher power levels.

PoE can be implemented by either installing a switch which has PoE capabilities or by placing a midspan injector into the data stream. In either case the total amount of power available to the end devices is limited to the capacity of the switch or injector, i.e., if a 24-port switch has all ports connected to devices using the full 15 watts per port then the switch would need to be able to supply a total of 360 watts of power. Many PoE switches are only capable of supplying 200 watts. With a 200-watt power supply, the switch could power a maximum of 13 ports.

Some power injectors that are not 802.3af compliant have been manufactured. Depending on the system design 802.3af compliance may not be needed. These devices serve a need but use non-standard voltages and are capable of supplying non-compliant power levels. When using these devices be sure to allow for the heating effect on cables and properly identify the device so that there is no possibility of confusing them with 802.3af compliant devices.

In a nutshell

Many diverse systems are utilizing networks to share information, both within a single application as well as between multiple specifications requiring the same information. Systems are now sharing network bandwidth with lighting controls, security cameras and access control systems and while these systems can clearly coexist on the same physical network, the bandwidth required, system responsibility, possible regulatory requirements (HIPPA, Sarbanes-Oxley, PCI) and data security concerns can be quite different. The use of parallel networks creates a physical network for a specified set of applications, allowing IT personnel to better manage the regulatory environment from both the network side and in providing the facilities department with the necessary tools.

Although some argue that networks shouldn’t be kept separate because modern switches can establish a virtual LAN (V-LAN) to control bandwidth, many manufacturers are pushing for open-platform solutions, allowing the use of many applications on a single network. With the growing presence of IP-based systems, the ability to view video from a remote location is becoming more demanding. The application of IP network fundamentals will be a part of the security landscape well into the future as physical security systems migrate to IT networks.

Paul Koebbe is the National Market Manager-Security for Graybar, St. Louis, Mo.

Lots of ‘E’s in Standards

The world of IT networking is controlled by various standards that have been established to allow multiple manufacturers to operate on a common network. The Institute of Electrical and Electronic Engineers (IEEE) or the Electronic Industries Association and Telecommunications Industry Association (EIA/TIA) established many of the network standards prevalent in IT networks.

The following IEEE Standards define Ethernet communications and are for information technology (IT), telecommunications and information exchange between systems, local and metropolitan area networks.

IEEE 802.11

IEEE Standard -Specific requirements Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications.

IEEE 802.3

Specific requirements Part 3: Carrier Sense Multiple Access with Collision Detection (CSMA/CD) Access Method and Physical Layer Specifications.

IEEE 802.3af

Clause 33 specification explains how Power over Ethernet is usually implemented.