Optimizing and Testing WLANs : Proven Techniques for Maximum Performance

As WLANs (Wireless Local Area Networks) become increasingly common, it's becoming vital to be able to test them for proper performance and to optimize their operation. This book, written by a member of the IEEE committee that develops WLAN standards, is the first book addressing that need. It covers test equipment and methods for the RF (wireless) and physical layers of WLAN, protocols, the application layer, and manufacturing testing. The emphasis throughout is on underlying engineering principles along with modern metrics and methodologies, ensuring this book gives both a solid theoretical background along with field-proven techniques and applications. Much of the subject material is drawn from the author's own hands-on experience in the field, as an architect and engineer of WLAN test equipment, and a writer of standards for measuring WLAN equipment performance. A particularly engaging chapter deals with manufacturing test, describing some of the different manufacturing test setups and equipment. A concise introduction to deployment testing of "hotspots" and WLANs in enterprises is also provided. *Gives proven techniques for maximizing the coverage area and performance of wireless networks *Author's insider position on the IEEE committee developing standards for WLAN testing ensures timeliness and technical integrity of the material *Includes coverage of newer multiple input/multiple output (MIMO) wireless networks
Chapter One IEEE 802.11 WLAN Systems

In order to successfully test something, it is essential to have a good understanding of how it works and what it does. We will therefore begin with an introduction to the important technical factors behind IEEE 802.11 wireless LANs (WLANs), as well as the standards and regulatory documents that govern how WLANs are developed and operated. By necessity, only brief explanations can be provided here; the reader is encouraged to consult the actual standards documents and other references for more information.

1.1 IEEE 802.11 Wireless Local Area Networks

Contrary to popular misconception, 802.11 is not merely "wireless Ethernet." Instead, 802.11 WLANs use an entirely different network protocol and are deployed in different topologies. The purpose of a WLAN is primarily to provide LAN connectivity to portable and mobile stations (laptop computers, voice handsets, bar-code readers, etc.), though fixed-station use is becoming more popular as the technology becomes widely adopted.

Essentially, WLANs provide data communications over radio links, and are subject to all the vagaries of RF propagation and interference that any radio communications system suffers. Wired (optical or copper) LAN links are nearly error-free (normal bit error rates are on the order of 1 x 10-9), physically secure, independent of environmental influences or mutual interference, and provide extremely high bandwidth. A single optical fiber, for instance, is capable of supporting hundreds of gigabits/second of bandwidth. By contrast, radio links are subject to error rates as high as 10%, subject to both eavesdropping and denial of service, highly affected by propagation characteristics and nearby equipment, and support only 10–500Mb/s of bandwidth that must be shared between all users of the RF channel. As radio signals propagate well outside the area covered by the WLAN and could interfere with other radio services, the operation of WLANs is governed by national and international regulations rather than being exclusively limited by technical or market considerations. The following table summarizes the key differences between wired (optical or copper) and wireless LANs.

While the IEEE 802.11 protocol allows for different types of WLAN topologies to be set up, nearly all deployed WLANs comprise two types of stations: clients and access points (APs). Clients such as laptops are the endpoints in the WLAN, and run the applications that source and sink data traffic. APs, on the other hand, provide portals into the remainder of the wired LAN; it is rare to find a LAN that is exclusively comprised of wireless devices. They support wireless interfaces on the "front" and wired interfaces such as Ethernet, DSL, or DOCSIS cable at the "back", and act as bridges between the wired and wireless infrastructure. Clients associate (connect) with APs to exchange data traffic with each other or the remainder of the LAN or WAN.

A group of clients and APs is collectively referred to as a service set. The 802.11 standard defines two kinds of service sets: a basic service set (BSS), which comprises a single AP and some number of clients; and an extended service set (ESS), which joins together several APs into a common network by means of a wired infrastructure. We will be concerned principally with ESS network operations in this book.

The following figure depicts the reference model under which 802.11 WLANs operate.

It is plain from the above figure that the wireless data links of WLANs coexist with wired Ethernet links. WLANs normally replace the "last 30 feet" of a data communications network to provide mobility, but are not used in the remainder of the network, where the emphasis is on bandwidth (large servers and routers, after all, do not move about). Data traffic carried over WLAN links uses the Transmission Control Protocol (TCP)/Internet Protocol (IP).

1.2 WLAN Standards Today

In 1985, the Federal Communications Commission (FCC) decided to open up the socalled ISM (Industrial, Scientific, and Medical) bands for use by unlicensed low-power communication devices using spread-spectrum modulation methods. This spurred significant interest in the US in developing wireless networking equipment utilizing these bands for computer communications (i.e., radio LANs) to serve as a radio version of the popular Ethernet LAN technology. As a result, in 1990 the IEEE standards development organization set up a group, referred to as the IEEE 802.11 committee, to standardize WLANs in the ISM bands. However, it took 7 years (until 1997) before the first 802.11 standard was ratified and published. That first standard defined a relatively low-speed digital WLAN technology, with data rate options of 1 and 2Mb/s, and using a new Carrier Sense Multiple Access/Collision Avoidance (CSMA/CA) medium access protocol, which was roughly modeled after the Carrier Sense Multiple Access/Collision Detection (CSMA/CD) protocol used by half-duplex IEEE 802.3 (Ethernet) LANs.

In parallel with the work of the IEEE committee, the European Telecommunications Standards Institute (ETSI) started work in 1991 on a radio LAN technology called HIPERLAN (High Performance European Radio LAN). HIPERLAN was standardized somewhat earlier than IEEE 802.11 (1996) and offered considerably more performance: 10Mb/s, as compared to 2Mb/s. A subsequent enhancement called HIPERLAN/2 raised this to 54Mb/s in the year 2000. However, due to complexity and market reasons, HIPERLAN and HIPERLAN/2 have been largely superseded by IEEE 802.11 LANs, though some of the principles of the former have been subsequently incorporated by the latter.

WLAN standards are set today by the IEEE 802.11 Working Group (WG), which is a subsection of the IEEE 802 LAN/MAN Standards Committee (LMSC), which in turn is a subsection of the IEEE Standards Association and sponsored by the IEEE Computer Society. As of this writing, the 802.11 WG has about 350 voting members and several hundred observers, and meets six times a year to work on WLAN-related standards. The 802.11 committee works within the constraints set by various national and international regulatory bodies to define the actual radio functionality and protocol.

The IEEE 802.11 standard does not try to specify how a WLAN device should be constructed – it leaves the design and operation of the actual clients and APs up to the implementer. Instead, it specifies the interactions between WLAN devices, collectively referred to as the WLAN protocol. The purpose of the standard is to ensure interoperability between devices without unduly constraining the device designer or vendor.

The WLAN protocol is partitioned into a number of pieces or layers:

1. The physical or PHY layer, which deals with the transmission and reception of radio signals, and is further divided into the physical media-dependent (PMD) portion and the PHY-layer convergence protocol (PLCP).

2. The Medium Access Control or MAC layer, which deals with the exchange of suitably formatted packets.

3. The PHY management layer, which handles the interactions required to control the PHY layer.

4. The MAC management layer, which likewise deals with the interactions needed to control the MAC layer.

The 802.11 WLAN standard is thus actually a collection of related standards, specifying all of the pieces described above. To date, there are over 25 different protocols and subprotocols comprising the 802.11 protocol stack, each being created (or having been created) by a separate subgroup within IEEE 802.11. The following figure shows a rough map of this plethora of protocol elements. The reader should observe the caveat that, as with any dynamic standards body, the number of protocols grows by leaps and bounds every year.

IEEE 802.11 subgroups are known as Task Groups (TGs), and are assigned letter suffixes to distinguish one from the other. The standards documents that they create are also assigned these same letter suffixes. For example, TGg created a PHY layer standard for Orthogonal Frequency Division Multiplexing (OFDM) transmission in the 2.4GHz band, which promptly became known as 802.11g. Similarly, TGi introduced a much enhanced security system, which was enshrined in the 802.11i standards document (more commonly known as WPA2, after the Wi-Fi® Alliance nomenclature). A curious convention is used when assigning letter suffixes: lowercase letters denote standards documents that will eventually be folded into the main 802.11 standard, while uppercase letters indicate that the document will remain permanently stand-alone. Thus the output of the 802.11b group was folded into the main 802.11 document in 2003 (forming Clause 18), but the 802.11T group is creating the 802.11.2 document, which will remain as a stand-alone performance test specification.

1.2.1 PHY Standards

In the US, the PHY layer of 802.11 occupies two principal microwave frequency bands: the ISM band at 2.400–2.483GHz, and the Unlicensed National Information Infrastructure (U-NII) band at 5.150–5.825GHz. (There is a further allocation in the 4.900GHz public service band, but this is a relatively recent development.) All 802.11 WLANs share these frequency ranges with other users, most notably microwave ovens in the 2.4GHz band. In theory, as 802.11 WLANs only have a secondary allocation in these bands, a WLAN must cease operation if it causes interference to the primary users; in practice, however, this almost never happens, due to the low power used by 802.11 radios. Figure 1.2: A Zoo of Protocols

The original 802.11 standard called for a 2.4GHz time-division-duplex (TDD) radio link with data rates of 1 and 2Mb/s, using DBPSK and DQPSK modulation, respectively. Both direct-sequence spread-spectrum (DSSS) and frequency-hopping spread-spectrum (FHSS) methods were specified and deployed; TDD was used to allow the uplink and downlink signals to share the same channel, taking turns to transmit. While FHSS was generally more robust to interference, DSSS proved to be more efficient and flexible, and FHSS was gradually abandoned; no vendor sells 802.11 FHSS radios today. Subsequently, the 802.11b standard added Complementary Code Keying (CCK) at 5.5 and 11Mb/s data rates to the mix, in addition to carrying forward the 1 and 2Mb/s data rates of the original. The following figure shows the general process used in CCK modulation. See Clause 18 of IEEE 802.11 for more information.

The data exchanged between 802.11 stations, at the PHY layer, is encapsulated within a frame format known as the PLCP frame. PLCP frames are different for the various modulation schemes, but generally contain a short header that indicates the coding and length of the encapsulated MAC frame; the receiver then uses this to properly decode the frame. The PLCP frame transmitted by an 802.11b radio is shown in the figure below.

The 802.11a standard was approved after the adoption of the 802.11b standard. (Actually, work on the 802.11a standard was started prior to 802.11b, but as it used a much more complex modulation scheme – OFDM – it took longer to develop than 802.11b. Hence the puzzling inversion in the nomenclature.) The 802.11a standard operates in the 5.8 GHz band, and calls for several different modulation types to achieve a large range of PHY bit rates. The modulation types are not only the BPSK and QPSK used in the 1 Mb/s PHY, but also include 16-QAM (quadrature amplitude modulation) and 64-QAM, leading to much higher data rates: 6, 9, 12, 18, 24, 36, 48, and 54 Mb/s. These modulation types are imposed on a set of 52 subcarriers spread over a 16.6 MHz channel bandwidth. A block diagram of the OFDM modulation and transmission process is shown below; Clause 17 of IEEE 802.11 provides details.

The 802.11a PHY operates in the 5.15–5.825 GHz band, which suffers from indoor propagation limitations. Due to market demand, therefore, the 802.11 WG began work on extending these same data rates to the 2.4 GHz band shortly after 802.11a was published. The result was the 802.11g standard, which incorporated all of 802.11b for backwards compatibility, and added the OFDM modulation types from 802.11a as well, producing a plethora of data rates: 1, 2, 5.5, 6, 9, 11, 12, 18, 24, 36, 48, and 54 Mb/s. (The specific data rate to be used is selected by the transmitter according to the channel conditions, to assure the best chance of getting the data across in the shortest time.) The 802.11g standard remains today the most widely used WLAN physical layer.

(Continues...)

Excerpted from Optimizing and Testing WLANs by Tom Alexander Copyright © 2007 by Elsevier Inc.. Excerpted by permission of Newnes. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site. Optimizing and Testing WLANs explores proven techniques for maximizing the coverage area and performance of wireless networks. The author's insider position on the IEEE committee developing standards for WLAN testing ensures timeliness and technical integrity of the material. The book includes coverage of newer multiple input/multiple output (MIMO) wireless networks. The techniques provided will allow engineers to help maintain continuous wireless connectivity to laptops and other mobile devices. Optimizing and Testing WLANs is the first book to address the need to test WLANs (Wireless Local Area Networks) for proper performance and to optimize their operation as they become increasingly common. It covers test equipment and methods for the RF (wireless) and physical layers of WLAN, protocols, the application layer, and manufacturing testing. The emphasis throughout is on underlying engineering principles along with modern metrics and methodologies, ensuring this book gives both a solid theoretical background along with field-proven techniques and applications. A particularly engaging chapter deals with manufacturing test that describes some of the different manufacturing test setups and equipment. A concise introduction to deployment testing of "hotspots" and WLANs in enterprises is also provided. This text will be of interest to RF wireless engineers and designers, networking engineers, IT professionals and managers, and graduate students. *Gives proven techniques for maximizing the coverage area and performance of wireless networks *Author's insider position on the IEEE committee developing standards for WLAN testing ensures timeliness and technical integrity of the material *Includes coverage of newer multiple input/multiple output (MIMO) wireless networks As WLANs (Wireless Local Area Networks) become increasingly common, it’s becoming vital to be able to test them for proper performance and to optimize their operation. This book, written by a member of the IEEE committee that develops WLAN standards, is the first book addressing that need. It covers test equipment and methods for the RF (wireless) and physical layers of WLAN, protocols, the application layer, and manufacturing testing. The emphasis throughout is on underlying engineering principles along with modern metrics and methodologies, ensuring this book gives both a solid theoretical background along with field-proven techniques and applications.

Much of the subject material is drawn from the author's own hands-on experience in the field, as an architect and engineer of WLAN test equipment, and a writer of standards for measuring WLAN equipment performance. A particularly engaging chapter deals with manufacturing test, describing some of the different manufacturing test setups and equipment. A concise introduction to deployment testing of “hotspots” and WLANs in enterprises is also provided.

*Gives proven techniques for maximizing the coverage area and performance of wireless networks
*Author's insider position on the IEEE committee developing standards for WLAN testing ensures timeliness and technical integrity of the material
*Includes coverage of newer multiple input/multiple output (MIMO) wireless networks