Power over Ethernet (PoE), a technology for the delivery of power and Ethernet data on the same cabling, has been around in various forms for almost four years. In June 2003, the IEEE 802.3af Task Force ratified the technology into the Ethernet standard. The standard involves delivering power over the existing cable plant, including Category 5, 5e, patch cables, patch-panels and connecting hardware. It is endorsed by all network players that power IP-connected devices, such as IP phones, wireless LAN access points and security cameras, over the existing LAN cabling infrastructure. This not only allows for a convenient and cost-effective method of installation, but also a simple, centralized power protection of the whole installation directly from the switch or wiring closet.
PoE technology saves time and the cost of installing separate power cabling, ac outlets and wall warts, and eliminates the need for a dedicated uninterruptible power supply (UPS) for individual devices. The power delivered over the Ethernet infrastructure is automatically activated when a compatible terminal is identified and blocked to non-compliant devices, such as desktop computers and laptops. This lets users freely and safely mix non-compliant and PoE-compatible devices on their network.
There are two main implementations of PoE. The endspan is a PoE-enabled Ethernet switch where power is supplied directly from the data ports, while the midspan is a stand-alone plug-and-play device, residing between an ordinary Ethernet switch and the terminals, often referred to as a “Power Hub.”
In the past year, PoE technology has matured rapidly with millions of ports installed worldwide. As awareness of PoE grows, so does the span of devices that wish to benefit from this safe and reliable method of supplying power. Recently, the issue has been raised that some of these devices require more than the current power limit defined in the 802.3af standard (about 13 W at the power device [PD] end). This is preventing them from using PoE (Fig. 1).
Revolution of Ethernet-Based Security Systems
Security systems are only useful if they can communicate and control the environment they secure. The video streaming from the closed-circuit television (CCTV) system should be watched or recorded; otherwise, it is useless. In addition, access control devices enable access to authorized personnel, while restricting others, and must have a central database of the personnel.
Professional CCTV systems based on digital streaming of video over Ethernet are beginning to satisfy these security requirements as technological limiting factors, such as quality and rates of video passing over Ethernet, are solved.
Consequently, the use of the readily available and cost-effective LAN-based security solutions is becoming more common. However, once end devices, such as a network camera or an access control terminal, are up and connected to the LAN, they still require power to run.
New high-end surveillance cameras are typically dynamic, remotely controlled devices that can move in all directions, allowing tracking of suspicious events or intruders. These quick-responding motors turn the camera rapidly to scan a larger area. However, they add to the camera power requirements.
PoE technology enables LAN-capable devices to be powered over the network-cabling infrastructure, thus avoiding the need for separate power and data-cable infrastructure and costly ac outlets near cameras. Furthermore, it allows the removal of the electricians from the installation team.
Network cameras are traditionally installed in high open places, such as corridor ceilings, airports, lecture halls and other large venues. The installation of power infrastructure is costly and time-consuming, and requires a team of dedicated electricians for pulling power cables, changing of building plans and safety approvals. In addition, digital CCD cameras are complicated devices that tend to need resetting from time to time. A “dark” camera, once discovered to require a reset, forces the IT manager to locate the network camera, reach it and then reset it. By using PoE, it is a trivial matter of resetting via the camera's respective PoE port via SNMP management, rather than climbing up high ceiling installations.
Enter High-Power PoE
PowerDsine, a manufacturer of PoE midspan and endspan solutions, recently launched a high-power PoE solution — the high-power midspan. The high-power midspan provides up to 39.5-W power, as well as data over standard Ethernet cabling, while still supporting IEEE 802.3af detection and safety requirements. The high-power midspan is the first implementation of the PowerDsine high-power PoE concept. (Fig. 2)
In similar fashion to the evolution of Gigabit Ethernet solutions, which required backwards compatibility with fast Ethernet, it was clear that high-power PoE would need backwards compatibility with the 802.3af standard PoE. This support would allow seamless connection of all powered devices, whether high-power compatible or 802.3af compatible. Also, the high-power solution would have to maintain the same levels of safety and reliability as provided by the IEEE 802.3af PoE standard. Both the high-power PoE standard and the current IEEE 802.3af standard must perform detection of non-compliant devices and real-time fault monitoring for short circuits, overloads and other failures of compliant powered devices.
The high-power PoE midspan eliminates the need for external power supply and associated ac-dc power cabling for equipment, such as point tilt zoom (PTZ) network cameras, high-end color touchscreen IP phones, multichannel and outdoor wireless LAN access points, information kiosks, point-of-sale terminals and other high-power-consumption Ethernet devices in enterprise installations. Furthermore, a high-power midspan can support centralized power distribution by deploying PoE in conjunction with a central UPS. This approach provides a cost-effective way to distribute backup power and ensures uninterrupted operation of the network during electrical power failure.
Technical Challenges
The challenge lies in the power-sourcing equipment (switch or midpsan) as well as in the compliant high-powered device (HPD).
In the 802.3af Power Source Equipment (PSE), power is provided either on the data or spare pairs. Each pair may safely detect and pass the 802.3af standard power (limited to 13 W). In similar fashion to the Gigabit Ethernet using all pairs, the high-power PSE now uses both data and spare pairs to double the power available, without affecting safety (Fig. 3).
The HPD must be able to receive power from both power and spare pairs, and be able to use both sources simultaneously to benefit from the dual-power capabilities (Fig. 4); such a device is the PowerDsine high-power splitter. This splitter contains a small HPD-compatible dc-dc stepdown converter, which provides a simple means of powering legacy 12-V, 2.5-A devices with high-power PoE.
Naturally, a new algorithm has been developed to enable the PoE detection concept to allow the high-power switch or midspan to detect a compliant HPD, looking for power on both pairs and power it. The algorithm of detection also determines whether a regular 802.3af-compliant device, looking for power on either spare or data pairs can be powered safely. This provides a seamless single-power source for the end user, who need not bother which device complies with which standard.
In similar fashion, the algorithms now must provide real-time protection of both data and spare pairs simultaneously per port to maintain safety for any power interruption event, such as a disconnected port, shorted line or overloaded failing device.
A New Standard is Emerging
As new devices requiring the higher power prepare for the new high-power opportunity to enjoy the PoE advantages, the demand to standardize high-power PoE is rising.
Today's market legacy devices are either using the active splitter as an interim solution, until the HPD front end can be modified to comply with high power, or are already adapting to the small changes required to be a compliant HPD. With the demand also comes a call to initiate and enhance the PoE standardization world to include the high-power PoE approach. Undoubtedly, a standard will be finalized to answer this demand.
With the introduction of the IEEE PoE standard, the advantages of easy installation and robustness of the powered network were quickly adopted by wireless LAN and IP telephony as the norm for new installations.
Higher-power devices needing a higher-power “bandwidth” are demanding to enjoy the same advantages, while needing higher power than the 13-W limit of 802.3af-compliant PDs.
PowerDsine has introduced the first high-power PoE solutions, the high-power 8000 midspan series. This technology allows for higher power without compromising on safety and keeping backwards compatibility with 802.3af devices, for seamless installation to the end user. With the demand for higher power and early adaptors rushing to use this technology, a high-power PoE standard is bound to appear in the near future.
For more information on this article, CIRCLE 345 on Reader Service Card