Published In January 2002
The first Category 6 application has been approved and published as of June 2001. It is an alternate transmission scheme for 1Gigabit Ethernet taking advantage of the increased bandwidth offered by Category 6 cabling systems. ANSI/TIA/EIA-854: A Full Duplex Ethernet Specification for 1000Mbit/s (1000BASE-TX) Operating Over Category 6 Balanced Twisted-Pair Cabling is the second standard Telecommunications Industry Association (TIA) published to provide an alternate means of transmitting Ethernet. The first was a short wavelength version of Fast Ethernet for fiber reported in September. In 1999, IEEE published the original and by far predominant Gigabit Ethernet specification as IEEE 802.3ab: Physical Layer Parameters and Specifications for 1000 Mb/s Operation over 4 Pair of Category 5 Balanced Copper Cabling, Type 1000BASE-T. This specification is the basis for all products being shipped today and, for all practical purposes, requires a minimum cabling performance of Category 5e. So why have two standards? What is the benefit of a Category 6 Gigabit Ethernet standard, and what are the differences? The IEEE standard was written before Category 6 cabling was developed; therefore, a solution had to be developed to operate over the existing cabling system—Category 5e. Based on the technology at the time, this was necessary to address the large installed base on Category 5 and Category 5e cabling. No one, except maybe those in the cabling industry, wanted to require the customer to install new cabling to implement Gigabit Ethernet. However, to operate Gigabit Ethernet over a copper cabling system with a maximum bandwidth of 100 MHz proved to be very difficult and specifically required complex circuitry in the chip set and a complex transmission scheme. With the development of Category 6 cabling, which doubles the bandwidth to 200 MHz, the ability exists to use that increased bandwidth with less complex circuitry. The TIA standard for 1,000BASE-TX does exactly that. Also, the precedence for having two standards for a solution had been established with Fast Ethernet, which has a Category 5 solution (100BASE-TX) and a Category 3 solution (100BASE-T4). With Fast Ethernet, 100BASE-TX for Category 5 is the predominant solution adopted by the industry. The TIA states that the purpose of standard “1,000BASE-TX is to provide an Ethernet physical layer specification that is less complex compared to 1,000BASE-T. This standard will take advantage of the improved transmission performance provided by Category 6 cabling... and eliminate the requirement of canceling the effects of crosstalk and return loss.” The TIA accomplishes this objective by using directional transmission on the pairs versus bi-directional transmission as specified by IEEE. To keep the frequency (bandwidth) within the capabilities of Category 5e, the IEEE specifies using all four pairs in a bi-directional means with each pair transmitting 250Mbps. This bi-directional transmission requires the electronics (chip set) to cancel out a portion of the crosstalk and return loss that exists in the cabling system. The TIA, 1,000BASE-TX uses the additional bandwidth provided by Category 6 to allow each pair to carry 500 Mbps, but each pair operates only in one direction, with two pairs being used for transmitting and two pairs for receiving. This transmission scheme eliminates the need for electronics to cancel out crosstalk and return loss properties. Figure 1 provides a simple schematic of the two schemes. Both methods use the same form of encoding—Pulse Amplitude Modulation (PAM5). However, the two methods are not interoperable meaning 1,000BASE-T and 1,000BASE-TX cannot be mixed. Both methods can operate in the same enterprise (building), just not on the same channel. When the project was approved, it was projected that this alternate scheme would reduce the number of transistors required in the chip set from 1,000BASE-T by over 50 percent. This reduction is believed to result in a lower-cost chip, because of the smaller size, reduced packing requirements necessary to compensate for heat generation and, hopefully, increased competition because the reduced complexity would allow for more manufacturers. Only time will determine if the cost benefits result in price benefits. BEAM is director of systems marketing at AMP NETCONNECT Systems. He can be reached at (336) 727-5784 or firstname.lastname@example.org.