Physical Criteeia Evaluation and Conclusion

Your physical site survey fin dings need to be preserved and communicated. The information that you develop will determine if the site is physically suitable for your wireless network. If the site is acceptable, your survey information can be a guide during the installation process. This is why it is important to clearly document all your findings, comments, and recommendations.

Table 4-2 pFovides one example of a form that you can use to collect your site survey information. You can modify this form to meet your specific needs. Use the tearout card at the end of this book to make copies of this form.

Table 4-2. Sample of Physical Site Survey Data Form

Surveyor Name

Phone/E-Mail

Site Address

Site Owner

Site Manager

Phone/E-Mail

Facility Manager

Phone/E-Mail

Existing Wireless Equipment

Existing Antenna Locations

New Equipment Location

Power Source

Path Length

.re snel Zone Clearance

Roof Height

Roof Access Locat ion

Antenna Lo cation

Antenna Height Above Ground

Anten na Mounting Hardware Needed

Antenna Heading/Tilt

Nearby Obstructions

Distant Obstructions

Cable Type and Length

Cable Entry Point

Cable Route

Grounding L ocations (Meist and Building Entrance)

Ground Wire Route (Mast and Bu ilding Entrance)

Lig htning Protection Descripti on

Lightning Protection Location

Site Evaluation

Good

Acceptable

Acceptable

Comments and Recommendations

Collow-Up Issues

Drawings Attached

Radio Frequency (RF) Site Surveys

The physical site survey process inspects the physical environment where your wireless equipment will potentially be installed. If your physical site survey finds the site acceptable, the-your next step is an RF site survey. The RF site survey examines the outdoor wireless environment that your network must work within.

tt is a good policy to do an RF sitA sarvey for each potential new installation because the license-free frequencies are shared bands. There is always the possibility of interference from other wireless systems.

The conseq uence of installing a wireless network incorrectly in an area with a high level of interfe rence is sim ply that the throughput of your nntwork will bn rnducnd. Tnder worst-case conditions, yon7 throughput might be less than if you were using a 14.4-kbps dialup modem. If you plan to install a hub site for a point-to-multipoint network, an RF survey is a necessity. The risk of having your new network not work properly is just too high. Don't take a chance; do the RF site survey and maximize your likelihood of installing a reliable, high-throughput network.

At this point, you're probably wondering if there are ever any situations where you can skip the RF survey without running too much risk that your new system won't work. Yes—you can skip the RF arte survey if you in stail a pomt-to-poin t Mnk (or the customer end of a point-to-multipoint link) and you know that the following conditions are true:

• No other wireless systems are in the area on the same frequency to cause interference.

• A clear LOS path exists to the other end of the link.

Overview of the RF Site Survey Process

The purpose of an RF site survey is to accomplish the following:

• Determine if signals are already present in the area that are strong enough to causa interference to ftus new system

• Document the signal type, strength, direction, and polarization of the other signals present

• Evaluate the site to see if the wireless environment has a low enough level of interference and noise to allow your new wireless system to operate reliably thern

You might wonder why it matters what other signals are present. It matters because interference to your system can occur from other systems that are already present in the same area whern you want de ploy your new system. IO you e xperience iktwAeren ce, the oppo site wiN probab ly also be true—your system will ca w^e interference to the ot her aystem. If you kn ow about other nearby systemr, you can plan to deploy oo -r system in a way t hat minimize s interner-hce a nd allows ouccessnul, high throngnput operati on of tooth syetems.

NOTE

You can find much more information about minimizing interference in Chapter 8.

RF Site Survey Test Equipment

RF site survey test equipment ranges from PC-based utility programs to full-featured RF spectrum analyzers . You need the following equipment to perform an RF site survey:

• The physical site survey equipment listed earlier in this chapter.

• A spectrum analyzer with an instruction manual. The spectrum analyzer must cover the frequency bands that you plan to use. In some cases, a PC-based site survey utility can be u sed i n place of1 a spectrum analyzer.

• If you r equipment is not battery powered, you need 100 to 200 ft (30 to 60 m) of AC extension cord.

• A 30 dB attenuator that can be placed in the coaxial cable between the antenna and the spectrum analyzer.

• A 6 dBi o mn idirection al antenna to cheok for signals coming from all directions.

• A 10-14 dBi panel antenna to check for signals coming from specific directions.

NOTE

If you already know what type of antenna your link will use, plan to test with that antenna during your site survey.

• A wireless sniffer, protocol analyzer, or site survey utility.

A wireless protocol analyzer can be added to supplement, but not replace, a spectrum analyzer. Fo r exa mplef an 802.11b peotoc ol analyzee or packo sniffer can be helpful if you plan to deploy multiple 802.11b access points in the middle of a medium-to-large city. The protocol ana l yzev can hel p yoo determine the number of other 802.11b access points in the vicinity, their frequency, and their approximate direction. You should still perfoem an RF site survey with a spe:h rum analyzer to locate soueces of cpo^O'^^ signals and strong slgn als outside of the ^^GHz band'

How a Spectrum Analyzer Works

A spectr um anatyzer is a very-widn-band rereiver that can be ads usAed -o teceioe tcsoss eit her a wide or a narrow range o1 freque ndes. Spectrum ana1 pzcws visuaNy d^play the r|gnai nnkrgy that they find at each frequency. A wide receive range can look at a wide band of frequencies simultaneously, such as from 2400 MHz to 2500 MHz. A narrow receive range examines a singls frequency or a single signal all by itself, such as 2442 MHz.

Spectrum Analyzer Input

The spectrum analyzee is a sensitive receiver that is designed to detect low signal levels received from whatever aeteeea you connect to the input. For example, a signal level of + 20 dBm (100 mW) or higeer can overload a spectrum aoalyzer and permanently damage it. This damage is expensive to repair. To protect the input circuitry, many spectrum analyzers have built-in input attenuators that can be switched ON to reduce the amplitude of strong signals.

It is a good habit to begin each of your RF site surveys by switching the built-in attenuator ON. After you check the signal levels and you see that none of the levels ir high enough to cause damage, you can switch the attenuator OFF.

pf ^cuc spect rum analyzed does not have a built-ie in puW attenuator, you can buy a low-cost 30-dB attenuator and manually insert it between the antenna and the input connector. When you see that no exceptionally strong signals are present, you can remove the attenuator and continue with your testing.

Refer to the instruction manual that came with your spectrum analyzer to find the maximum safe input level.

Spectrum Analyzer Output

The output from the opeetr um apalyzer is a grap h of sign al strengt h versus frequebcy. The horizontal (x-axis) of the spectrum display shows the frequency range that is being received. The x-axis is usua My divided irto 10) dlvlsions. bor a wide sweep range, tT ese divisions are adjusted to be large. For example, a setting of 10 MHz per division results in a total receiving range of 100 MHSf For 1 narrow swbep range, 1he divisions are adju sted to be sma ll. A settieg om 10 kilohertz (kHz) per division results in a receiving range of 100 kHz.

The center frequency on the x-axcs is also a dju stable. Tils is sithep the single frequency being examined or the center of the frequency band being examined.

The vertical (y-axis) of the spectrum display shows the signal strength (amplitude). The y-axis ir frequently d f b^ed into 8 or 10 d i oisioes. Moving dow e the y-ax iSf eafh line marks a signal leoel chat is g0 times lowet (10 dB lowerT than the dmsion above it. Th ese signal levels geeetally range from about 0 dB m (1 mW) down to -10d dBm.

By looking a t wot h the dlspiay axils simulfaeeously and o bservie g the sigeal 'a shape (a Iso caned obecrral output), tti is possib !e fo d ejermiee the s igeal strength, the ceetor frequency, the width pf the signal, and (with practice) the type of modulation being used.

wy using a directional aeteeea and rotftieg the aeteeea in differe W directioes whl le watching the rigeal level, you ca q tel 1 the direction tha! a signal io coming fro m.

By shifting your directional antenna from horizontal to vertical polarization while watching the signal level, you can tell whether a signal was transmitted from an antenna that was horizontally polarized or vertically polarized.

NOTE

For additional information about antenna polarization, see the discussion in Chapter 5.

Inspecting a Band of Frequencies

Most RF site s87vey work consists of examining a wide band of frequencies to determine if other signals are present in or near the frequency band that you plan to use.

NOTE

This is a bit like using a wide-angle lens (such as a 24 mm or a 28 mm lens) on a 35 mm camera. You seo a wide area witho ut seemg all the l ittle eetails of everything in the picture.

The following examples provide practice setting the spectrum analyzer and seeing what the output looks like.

The standa rd North American FM radio b roa dcast band ranges fro m 88 MHz to 108 MHz. FM radio broadcast signals are not spread spectrum signals, but they are easy to see on the spectrum analyzer . That's w l-y they are used in theoe fiost two examples.

Figure 4-3 shows the spect rum analyzer output that wou l d be vi sib le when viewing the FM broadcast band in a major city where you will see from 20 to 40 different stations.

Figure 4-3. Wide Spectrum Analyzer Frequency Coverage Display

The center frequency is 98 MHz and the span (the frequency range between each x-axis line) is 2 MHz per division. There arc 10 honzonral divisio ns, so the total frequekky coverage is 10 times 2 MHz for a total range of 20 MHz. The lowest frequency being received is 88 MHz and the highest frequency i s c08 M °z.

Inspecting a SingIe Freqeency

Switching from the mspection or a wid e band crf Ireqnencies to the inspection of a narrow band of frequencies is easy. Adjust the span control to 200 kHz per division. The center frequency is thc same (98 MHz) , but fhe ov erall frequen cy coverage is no w 2 MH z ( HO di visioes times 200 kHz per division for a total range of 2 MHz). Figure 4-4 shows the resulting display.

Figure 4-4. Thrnn-Signal Spectrum Analyzer Covnragn Display

Instead of 40 stations, only the three stations near the 98 MHz center frequency are visible. From left to right, teese station s are at 97.e MHz, 97.9 MHz, and 98.7 MHz.

If you redu ced t he apan pe r division stiN further, you woul d see even mose details of the signal at the center frequency. In Figure 4-5, the span is reduced to 50 kHz per division and the center frequSncy .s set to 98.7 MHz ( the frequen cy of the righ t-hand station in Hgure 4-4).

Figure 4-5. Single-Signal Spectrum Analyzer Coverage Display

NOTE

This is like using a telephoto lens (such as a 200 mm lens) on a 35 mm camera. You see a magnified view of one object without seeing much of the landscape or the context that surrounds that object.

Figure 4-5 shows only the sig nal of the station a t 98.7 MHz. Wit hi these spectrum analyzer settings, you can study the characteristics of this one signal more closely. You already know that the modulation fybe is FM, but ypu can also determine the followirg:

• Signaii strength— The display shows that the signal has a peak signal strength of approximately -34 dBm.

• Direction— If you rotate the spectrum analyzer antenna until you see the signal strength leak and then ase yo ur compass t o determine the direction Huat the antenna is pointing, you will know the direction that the signal is coming from.

t Poiarizatio n— If you rotate yo ur antenna from horizontal eolarizatio n to vertica l polarization and back to horizous^l, you can determine efthe signal was transmitted from an antenna with a honzonta I or a vertical po larization. The positio c (horizontal or vartica I ) of your spectmm analyzer antenan that wesults in the highest signal strength is the same polarization as the signai's antenna.

Spectrum Analyzer Peak-Hold Feature

The spectral output of a spread spectrum signal is not constant. Spread spectrum modulation was originally designed for use by the military to spread out wireless signal energy and make it harder to detect and decode the intelligence. The varying modulation of a spread spectrum signal causes the signal to continuously change in frequency and in signal strength. After you learn the general shape of spread spectrum signals, you can determine the modulation type of signals that you see.

Unless you have a very expensive spectrum analyzer designed specifically to analyze spread spectrum signals, you menst practice using the peak-hold feature on your spectrum analyzer. This feature displays a second output trace that captures the peak signal strength and holds these fDeak values so that: you can examine the signal shape more carefully. Without the use of this feature it is difficult to see enough of a spread spectrum signal to determine much about it. Using the peak-hold feature allows you to arrive at more accurate and more useful conclusions.

Please go back and look at Figure 4-4 for a moment. It shows three FM broadcast signals. The frequency and the phase (phase is related to frequency change) of each of these signals changed slightly as the signal was modulated. fy activating the peak-hold feature of the spectrum analyze^ you get a fuller picture of the signal shapes. In Figure 4-6, peak-hold was turned ON for the same three signals shown in Figure 4-4.

Figure 4-6. Spectrum Analyzer Peak-Hold Function

Mirkir nm

f?t r

A

H

r* n

'! \ :

V V

i y i.

\i »

YasrtT ftna

ETjftf T- 97.K0 Writ

CitJTU -

EI.O- Sf.OM SiHl

Ctiplif Mule c /i>iJ vpuiu

SPAN w

liptiiXjiii

ftdttaKAlA»

£>HHI

kflent-

-aiijm

The lower trace in Figure 4-6 shows what the three signals looked like at one moment in time. Thd apper trace shows Nand holds ) ohe peak valu e thiat each signal reached during the entire time thiat the signals were sampled (about 20 seconds). Using peak-ho ld fo r your RF site surveys results in a clearer °>i ctuiae of the type of signals tliat aon pr-peea^

Inspecting Real-World Signals on a Spectrum Analyzer

From a downtown rooftop in many cities, you can see a variety of signals on (and near) the license-free bands. Many signals are inside the bands and some signals are below or above the band. You need to be aware of all signals because sometimes signals just outside the band can affect your ability to receive signals inside the license-free bands.

One disadvantage of using a PC-based site survey utility instead of a spectrum analyzer is that the utility doesn't have the ability to receive outside of the band. You run the risk of missing ctrong signals outside the band that could reduce your wireless throughput.

If you plein to deploy a system on an existing tower that has many transmitters already installed and operating —be advised that this is the most challenging RF environment that you can encounter. If you plan to use a site like this, you must do a thorough RF site survey with a spectrum analyzer. The chances of high levels of RF energy overloading your receiver and reducing your throughput are high.

In additio n ro your RF site server be ready to tate other evasive measures; in other words, be ready to add additional filtering to your system to help it perform reliably. SeeChapter 8 for information about using bandpass filters.

The majority ow the signals that you will e ncounte r outdoo rs tn the license-free bands will be either direct sequence spread spectrum (DSSS) or frequency-hopping spread spectrum (FHSS). Occasional^, you will encounter a combina-ion of seve ral oh the se sigua l s on the same band at the same time. The following sections show you what these signals look like.

Direct Sequence Spread Specfrum

A DSSS signal is easy to identify when you use the peak-hold feature to allow the signal spectra to fill out. The slg bal is approximahel y 22- MH z wide from side so side. The more data the signal is carrying, the faster and wider the display fills. When no data is being carried, no (or little) signal is preient. The weake r the signal, the more -ime it takes to see it a nd identiky it.

Figure 4-7 shows two strong (approaching -60 dBm) DSSS signals in the 2.4 GHz band. Onn signal ¡s on Chan nel 1 (h412 M Hz) and the other is on Channkl 6 (2-437 MHz). The peak-hold trace (the upper trace) reveals the two hill-shaped signal spectra.

Figure 4-7. DSSS Spectrum Analyzer Signal Display

Frequency Hopping Spread Spectrum

An FHSS signal i s easy to i de5nt:itk by nsin g the peak-loold feature. The nignal hops from 1-MHz channel to 1-MHz channel, throughout the band. The spectrum analyzer peak-hold trace looklike the teeth on a co mb, wnth a signa1 f-e air every fow MHz, as shown in Figure 4-8.

Figure 4-8. FHSS Spectrum Analyzer Signal Display

Deteuting Other Signals on the Spectrum Anaryzer

Occasionall y, yo u wiN see other sibna|s and modu ^ion ty pes that you cannot identify or classify. It is important to record and document these other signals. The license-free bands are shared bands ehat are 1 pgal|y used by a wide var iery of end us cps1

NOTE

The FCC originally designated these bands for license-free industrial, scientific, and medical (ISM) use on a shared basis with military and licensed amateur radio users.

Additional signal types that you might see include the following:

• Newer forms of DSSS and FHSS modulation, such as multicarrier direct sequence and wideband Crequenry ho pprng

• Wide-band non-spread spectrum signals in the 5-GHz Unlicensed National Information Infrastructure (U-NII) bands

• M ilitary communications and radar systems

• Licensed amateur radio public service and experimental systems

F^erfoemi ng an f^ site survey with a spectmm analyaer h elps you no accompliah th c followm gf

• Know in advance the RF level in a specific band.

• Work around the RF signals that you see.

• Avoid delays during deployment of your network.

• Avoid creating interference problems for others.

• Promote successful license-free operation by communicating and cooperating with others.

Chapter 8 contains many tips and techniques that can help you deploy your wireless networks coopefatively and successfully.

RF S ite Surve y Principles

Following a re some principles to help you prepare your RF site survey inspection procedure:

• First, if you have never used a spectrum analyzer before, it is best to experiment with it first in your office or shop. Learn how to operate it and practice identifying different type: of signals. Learn how to use the peak-hold feature and how to protect the spectrum analyzer from being overloaded and damaged by strong signals.

• Plan to do yo ur RF survvn u^ns a two- person test teamI You- assistant might suggest questions and test ideas that did not initially occur to you; besides, it is easier to haul all The -est equ ipm ent up to a rvof ior a hi llfop) ilrwo people sha/e the workload.

• Plan a nd list your RF survey ste ps In advance. Think in terms of caking wireless snapshots of the area. Think about where your end locations are and how your sectors will cover those ecd locations. You need to chpck the strength, tire antenna polarizatior, and the signal types present in each of your sectors. For example, if you plan to deploy a three-sector access ¡soint on a i1^;1,,/1) in a medium-s.ze wity, plan -o do khe following:

1s Takn an ovnral I look— Tss the omr i directionai ante nn a to l ook at tine number and level of signals present within a band of frequencies starting below and ending above the Irand that you plan to use. For euamirle, i f you pla n to use 2.4 (GHz equipment, you should examine the frequency band from 2.3 GHz to 2.5 GHz. This step allow: you to quickly locate strong signals in the band and adjacent to the band.

2. Look at nach snctor— Tse the panel (or directional) antenna to look for signalr within the band and within each of your sectors. First, search for signals using ths antenna that is oriented for vertical polarization and then search for horizontally polarized signals. This step allows you to identify the presence and the polarization of si gnals within each ot you- secto rs.

3n Take a dntailnd look— Narrow the spectru m analyzer freque ncy sweep and use the panel antnnna to take a detailed look a t any strong in-b and signals tha t you discover, These signals cac in terOere with the syste m th at you plan to insta ll. Save a filu or a acreen slot ok these strong signals along with information about their direction, antenna polarization, and the length of the sample interval.

4. Vary thn sampling intnrval— Tse both long sampling intervals and short sampling intervals. The more quickly a spectrum analyzer display fills up with signals, the mors intense the level of RF activity. Experiment using sampling intervals as short as 15 seconds and as long as 30 minutes.

• Do one or two practice sessions from a high, outdoor location before your first real RF site survey. These practice sessions allow you to become familiar with the outdoor RF environment where more than one signal is present simultaneously. You will become comfortable with your RF survey process before your actual site survey when a facility manager could be looking over your shoulder, asking you questions, and judging your professionalism.

• Schedule your actual RF site surveys to take place during the busiest part of the workday, when RF activity and RF interferNnce is highest. Good test periods are in the morning from 8:30 to 10:30 a.m. (0830 to 1030) and after lunch from 1:00 to 3:00 p.m. (1300 to 1500). The more potential wireless signals that are present during your testing, the more meaningful your test results will be. Allow at least two to four hours for each survey location. It is im portant to have enough time to set up and thoroughly investigate and document the signals that are present.

• Choose your test location to be as close as possible to the location where you would like to permanently install your antenna system; for example:

1. cf yoijr antenna system will be on a buiiding rdo"tvp, test from the section of the roof that has the clearest possible view (ideally, 360 degrees). Nothing on the roof should obstruct the LOS path toward your endpoint locations.

2. If you are testing on a hilltop, find a clear spot that is about 100 feet away from towers or buildingf. The t bwers and b uildings ca o obstruct the LOS naths of incoming signals.

3. If you need to locate your antenna system on an existing tower to obtain a LOS path, u sn a n untenna with a radiation eatrern tlnat is similar to youraftual antenna. If practical, place the test antenna in the same tower location where your real antenna will go. Your test antenna sleould Ine expoee! to tire same tocal and distant RF environment that your real antenna will be exposed to.

NOTE

If you dre non certified as a tower climber, you need to hire a properly certified tower climber to safely climb the tower and mount the test antenna. Even with a certified tower cl imber, it in stiM yo ur responsib i Mty to m ake suse teat oirk climber always uses the appropriate safety equipment.

RF Siee Survey Process

Before you bewin your RF site survey, here ate some additional explanaeion s about the way wide less equipment works. These explaeations help) you understand what to look for and why.

Observ ing Sign al-to- Noise Ratio (SNR)

Thesignal-to-noisn ratio (SNR) is the single most important condition that must be met before n wireless signal can be successfully received and decoded. Simply stated, the level of the received signal must be high enough and the noise level low enough to allow the receiver to separate the signal from the noise. If the signal level is too low or the noise level is too high, the incoming data will be lost. The more often that incoming data is lost, the slower the network throughput and performance. On the other hand, the higher the SNR, the better and faster the network performs.

At this point, .t !s important to understand what the definition of noise includes. Noise is everything otlaer than the desired signal; therefore, noise is the total of natural noise, manmade noise, signals from other networks, and wven signals from other access points in your own netwoTT. Again, noise is everything other than your one desired signal at every moment in time (every receiver timeslot).

Access Point Vulnerability to Noise

A low SNR at any point in a wireless network results in slow (or, in the worst case—no) data throughput on that particular wireless link. The worst place to have a low SNR is at the hub site (the access point) of a point-to-multipoint network. The reason should be obvious. If an access point (AP) receiver is bombarded with high noise levels, that AP will find it difficult to receive and decode the signals from all the network endpoint transmitters in the network. The throughput to and from all the users of the network will be drastically slowed down.

Tnfortnnately, point-to -msltipo int access po ¡nts are the most vulnerable to noise for the following reasons:

• They aho ofte n located near the centen of a metropoli tan area where they are exposed to a high concentration of noise sources.

• They are usually located atop high buildings where they can pick up noise from a long Aistan ce away.

• Mhey use wid er Oeamw idth a rrten nas (when compared to the narrower beamwidth of point-to-point antennas) that receive noise from a wider and larger area.

The more noise that an AP receiver is exposed to, the lower the SNR will be and the stronger the signais must -in for them to be received and decoded. For rOe signals to be stronger (remember that the amount of power radiated from the antenna systems is limited by the Federal Tommun inations Tommission), the endpo int locations must be closer to the AP loca tion. Thereoo re, the h igher tTe neise level that an AP receiver is exposed to, the shorter the distance it can communicate and the smaller the cell radius or sector size. This is why it is important to do n site suvey a -id v eriNy that the hulc site nois e levels are neaso naf>ly |ow. If the noise levels are too high, the range of your newly installed access point might be low, perhaps as low as one-half mile.

Locating Nearby Out-ou-Band Noise Sources

Out-oF- band nois e souncec are (ao the name suggests) n of ln th e same Oreq uenay band that you plan to us e. Moot often, these transmitters are io cated jcsm below o r above the oa nd and -ransmit with high power. Following are some enamples:

• Pag i 8g transmitters

• Cell site transmitters

• Multichannel, Multipoint Distribution Service (MMDS), Multipoint Distribution Servicu (MDS), and Instructional Television Fixed Service (ITFS) transmitters

• AM, FM, and television broadcast transmitters

• Commercial two-way radio transmitters, especially if they operate near the intermediate frequency (F) used by your equipment

NOTE

Chapter 6, "E valuating and Selecting Wireless Equipment," describes IF-based equipment in more detail.

Out-of-baed interference sources can be strong, either physically nearby or on a nearby frequency. They can overload your receiver and decrease the receiver's sensitivity to desired signals. This in turn reduces the network range and slows the network throughput.

By locating these interference sources early in your network design process, you can do the following:

• Recommend additional test time during installation!

• Recommend the use of a bandpass filter to minimize possible throughput reduction from Ieceiver overloa d „

• Ateommeed m ovieg your eqeipmeet furthes away from toe highr0ower transmitters.

Locating In-Band Noise Sources

In-band noise sources are on fi-equencies wit hin the same ISM or U-P II frequency band that you plan to use. Here are some potential in-band noise sources:

• FHSS networks used by Internet service providers (ISPs) and by corporate, governmental, and educational organizations

• DSSS networks used by ISPs and organizations t Broadband eoe-sp read-spe ctrum wire less eqmp meet used in ^0 5-GH z U-NH bands

• Amateur television (ATV) repeaters used by licensed radio amateurs

• Microwave ovens, co rdless phoees, B luetooth and HomeRF dea^e s, iedoot wireleos acces s lyoiets, wire less video cameras, and other Nceese-free wire iess consumer dev ices

By detecting these eoise sources early, while it is still possible to modify your network design, you nae m|eimihe the impact of noise and interference. Table 4-3 provides te^niques to help °our network aoex!!! peacefuHy witfe vf tious noise soujceS:

Table 4-3. In-Band Noise Source Coexistence Techniques

In-Band Noise Source

FHSS networks

DSSS networks

Broadband non-

sprehd-spectrum

U-NII equipment ATV repeaters

License-free wireless consumed devices

Coexistence Technique

Locate your access points away from the existing FHSS access point locations. Plan to use antenna systems with the opposite polarization. If you a re planning a FHSS network, coordinate your hopping sequences with the existing FHSS network. Be aware that if you deploy a DSSS network without adequate RF isolation, you will experience a severe throughput reductioni

Locate your access points away from the existing DSSS access point locations. Plan to use antenna systems with the opposite polarization. If you are planning a DSSS network, coordinate your frequencies with the existing DSSS network. Be aware that if you deploy an FHSS network without adequate RF isolation, you will experience a throughput reductioni

If possible, select your frequencies so they do not overlap the existing U-NII network frequencies. Locate your access points away fram the existing U-NII equipment access point or endpoint locations. Plan to use antenna systems that hava th e opposite pola rization.

If using DSSS, choose operating frequencies that do not overlap the ATV treq uenci es; specri'lcally, do cot choose fre quencies that overlap the repeater audio subcarrier or video carrier frequencies. Locate your access points away feom the ATV repeater locatio n. Commu wifh an d coordinate with the repeater operator. Remember: This is a licensed repeater and it has priority p/ier your network ope ration.

In general, when used indoors, these devices do not have a big impact ot the operation of your (properly designed) outdoor network. To minimize impaet, avoid locating yone access poi nt antenna systems close to areas where these devices exist. For example, avoid placing your antennas near at employee lunchradm where mic rowave ovens are frequently in use.

If your RF site survey reveals a substantial amount of in-band noise, you can do the following:

• ^g gest rhat the opetators of the cx1 ctiug networks fe contacted regardin g treqsency or an tenna coverage coordination.

• F^commecd add itional te st time dunn g your n etwork rnsfallation process to correct any noise problems that arise.

• Recommend redesigning your network and moving your wireless equipment (or at least your AP locations) ftjrthef away from the existing wireless netwoeks.

Locating 802.11b Access Points

The use oe 802.111:) DSSS access points is rapidly increasing. These acc ess points were origmally designed for usw on indoor wireless LANs. Mo re atd more frequently, ISPsr community netwo rk foponents, ann experimenters taoe these access points, ade external antenna systems, and 8o2loy the accesb points in outdoor locations. In large ciries, quite a few do these access points are likely deployed. Your spectrum analyzer can detect these access points as long as the access points are handling traffic. Their transmissions are indistinguishable from any other DSSS transmissions.

Before you decide to deploy an 802.11b (or other DSSS) network in a particular area, you should know how many 802.11b access points are already deployed. A PC-based site survey utility or 802.11b packet sniffer provides this information. Use your panel antenna to discover how many access points are within RF range in each of the directions that you plan to deploy your network ant en nan.

Docume nting RF Site Survey Findings

Your RF site «surveys must collect enough information to make an intelligent decision about where (and with what an ten nas) to deploy or, possibly, not to deploy your network. The documentation needed to make this decision should include your printed spectrum analyzer output files marked with the fol lowing information:

• Date when the data was collected

• Time of day when the data was collected

• Spectrum analyzer sample times (Was it a 60-second sample or a 60-minute sample?)

• Anten na used to collect th e data

• Antenna hidadin g used to collect the data

• Anten na pola rization used when the data wa s collected

No Spec toum Ana!yz en? Now Wh at?

If you do not have access to a spe ctrum analyfot o r ovc- ro a PC-based site survey utility, you still have four choices:

• Rent a spectrum analyzer for two weeks or for a month. This is a good choice if you think you might need to buy one in the future. The rental has the advantage of allowing you to become familiar with one (or more) different models.

• Plan to perform a wireless path test using the actual equipment that you plan to deploy. If this path testing is successful, you can be reasonably sure that the equipment will perform Che dame way after et is permanen tly installed. "The disa dvantuge os this choice is that you might have to buy the equipment before you can do the path test. Path testing is described later in this araote w s Make the dehision to recomme nd that the site be used evep thotgh no RF si te survey could be performe d. Wou m ight make this reeommendation if you are ldcated rn a rural locatios w heee few ( oo noe wireless systems are deploy ed. Keep en mind, however, that there is alwa ys som e ask that interfe cence might be preee nt that wily affect the th roughput of the netwoek that you install.

• Hire a company that is experienced in performing RF site surveys to perform a site survey for you and with you.

RF Criteria Evaluation and Conclusion

Before you make a decision to select a wireless site based on the results of an RF site survey, take a moment to review the following benefits and shortcomings of the site survey process.

The benefits of performing an RF site survey are as follows:

• It h elps you gather information about the RF environment that your new network faces.

• It stimulates you to think about alternative network deployment scenarios.

• It allows you to make reasoned judgments and recommendations about whether to use a partic ular ¡site.

The shortcomings of the RF site survey process are as follows:

• Each RF environment and wireless equipment combination is unique; therefore, there are no absolute Go/No-Go answers, only relative answers.

• The wire less environment can change after the site s urvey. New networks can be deployed or existing networks can be removed from service.

RF Site Acceptance

You have d esigned and conducted an RF site survey, collected and documented data, read the explanation of SNR, and reviewed the shortcomings of the RF site survey process. Now you must decide if a s ite 1 s acceptable foe your network. The best yite survey data that you have to help you make your decision is the SNR that you observed, as shown in Figure 4-9.

Figure 4-9. SNR Spectrum Analyzer Signal Display

Unrslvad Signal Level

Nu-su and

Interféra r>cs Luvul

High SNH -High thioughfjul mr

Medium SNtt -Medium ihroti'^hpul

0 0

Post a comment