Cisco Aironet 340 Series User's Guide download pdf (Page 18)

3-18 Background

Besides faster data transmissions, 2.4 GHz radios offer an even bigger advantage—unlike 900 MHz, the 2.4 GHz

is unlicensed and available in most of the world. Because of this worldwide acceptance, the Instituteof Electrical

andElectronics Engineers(IEEE) StandardsActivity Board802.11 committeedesignated 2.4GHz asthe RFband

for emerging standards for wireless LAN speciﬁcations.

In June 1997 theIEEE Standards Activity Board approved the 802.11 wireless LAN standard. This long-awaited

standard will allow mobile wireless LAN components, from different manufacturers that are 802.11 compliant,

to be used with the same wireless infrastructure. More and more 802.11-compliant products are coming onto

the market. The entire industry is now focusing on the implementation of IEEE 802.11 wireless networks,

conformance testing, and proving interoperability between multiple vendors.

As a champion for industry standards, Cisco intends to fully participate in all pertinent conformance and

interoperability testing of wireless LAN products. In April of 2000, Cisco was in the ﬁrst group of wireless LAN

vendors to be awarded the wi-ﬁ™ certiﬁcation of inter-operability for IEEE 802.11b products by the Wireless

Ethernet Compatibility Association (WECA). WECA was founded in 1999 to certify inter-operability of IEEE

802.11b wireless LAN products.

Real-world throughput vs. idealistic data rates

When comparing the performance of communications systems, it’s important to make sure the systems are on a

level playing ﬁeld. Two terms in particular can lead to a lopsided understanding of system speed—“data rate”

and “throughput.”

Data rate in communications systems is not a measure of speed. The raw data rate actually describes the capacity

of a system—the size of the pipe.

Throughput describes how much useful data is transferred through a system in a given period of time, and is a

better measure of the actual speed of a system.

Throughput will always be less than the raw data rate. The difference between the two measures is due to

protocol overhead and transmission errors. The real data that you want to send over the network is encapsulated

in packets that contain addressing and other protocol control information. Each protocol layer adds its own

protocol functions, and therefore its own overhead.

In a world dominated by the Internet, everyone uses the TCP/IP protocol stack. Therefore, a reasonable

real-world test of the “speed” of a network is to measure a ﬁle transfer time over a network using FTP, the File

Transfer Protocol. Cisco bases most network performance comparisons on FTP ﬁle transfers. There are at least

ﬁve levels of protocol overhead involved in that transfer: the FTP protocol, the UDP protocol, the IP protocol,

the LLC protocol, and the MAC and PHY for the particular network. This is a good test for most users, but the

results are always less than the throughput claims put forward by many marketing departments.

To avoid protocol overhead and show their network in a more favorable light, companies usually conduct very

low-level tests of their products. The resulting marketing claim is usually based on transmitting frames directly

over the MAC interface without any higher level protocols. This is more impressive than the FTP test, but still

less than the raw data rate.

1 2 ... 13 14 15 16 17 18 19 20 21 22

No comments

Cisco Aironet 340 Series User's Guide Page 18