Wi-Fi

Wi-Fi 101: Multipath environments and how they affect Wi-Fi propagation

Ever wonder why moving your computer or changing the direction of your Wi-Fi antenna can make such a difference in signal strength? In this Wi-Fi 101 lesson, I explain how RF signals-especially Wi-Fi ones-react to real world conditions.

Ever wonder why moving your computer or changing the direction of your Wi-Fi antenna can make such a difference in signal strength? In this Wi-Fi 101 lesson, I explain how RF signals-especially Wi-Fi ones-react to real world conditions.

RF terminology

Before I jump into a discussion on multipath environments and RF propagation, here are a few basic terms you should be familiar with:

RF propagation: The movement of RF signals through a medium. In theory the simplest example would be a signal traveling directly from the transmitter to the receiver. Under real-world conditions RF signals do not emanate as a single wave, but as an expanding wave front similar to ripples on a pond. Multipath environment: A term used to describe real world conditions encountered by the expanding RF wave front. RF Line of Sight (LoS): A measurable parameter where anything less than 20 percent blockage of the Fresnel Clearance Zone is considered RF LoS conditions-- RF LoS and visual LoS are not one and the same. RF Non-Line of Sight (NLoS): Defining RF NLoS is slightly more complicated. Two distinctly different RF propagation models need to be considered. One model-essentially the compliment of RF LoS-is when physical obstructions block the Fresnel Zone by more than 20 percent. The second less intuitive model focuses on situations where the original RF wave front has been altered into multiple RF signal components that may or may not reach the implied destination. In theory, virtually every WLAN is affected by RF NLoS to some degree.

RF propagation basics

Initially a uniform RF wave front leaves the transmitting antenna. As the wave front traverses space, it may encounter obstacles that alter the original wave front or create new RF signals. One or more components of the original RF wave front may continue traveling straight to the receiving antenna, other components may diffract, scatter, or reflect off of obstructions.

  • Diffraction: The phenomena where radio waves are bent around sharp objects creating a new wave front.
  • Scattering: Is where RF energy is reflected off of a non-uniform surface in multiple directions.
  • Reflection: Occurs when the wave front contacts a uniformly flat surface such as sheet metal siding and is reflected at a predictable angle.

These elements can have either a positive or negative impact on WLAN performance. For example, without diffraction, scattering, and reflection, WLAN's would not work in the typical office environment, which has less than optimal RF LoS conditions. Negative impact elements or anomalies are also very important to understand. Having that insight may help explain why a WLAN is working poorly or help decipher location specific problems.

RF propagation anomalies

RF propagation anomalies are the results of the original RF wave front being altered, broken into separate RF components, and the resultant RF components reacting to each other when they meet at a given physical location-receiver's antenna. Since each RF component is following its own distinct path, there will be differences in travel times and RF wave geometry. Signal strength, timing, phase and angle of arrival are just a few of the affected parameters. It now becomes the receiver's job to decipher what is the original signal and what are distorted copies. Described below are some of the most frequently encountered propagation anomalies.

  • Multipath fading: Is a phenomena associated with wave propagation and occurs when the receiver sees the superposition of multiple copies of the transmitted signal, each traversing a different path. The results can either be constructive (amplified) or destructive (attenuated) interference at the receiver. The moving of computers and/or fiddling with antennas mentioned in the first paragraph are methods of combating destructive interference.
  • Time delay: An anomaly that goes hand in hand with multipath fading. Time delay is the amount of timing variation between different RF signals. Time delay can cause phase and polarization changes as well as multipath fading. Unlike multipath fading which affects signal amplitude, time delay adversely affects the receiver's ability to decode signals due to distortion.
  • Doppler effect: RF Doppler effects are created in two different ways, relative motion between the transmitter and receiver and RF NLoS conditions that alter the relative motion of the RF signals themselves. Both of which especially affect 802.11a/g and the use of OFDM technology.

RF NLoS should be considered a double-edged sword and any knowledge gained about how RF signals disseminate in a specific environment will be helpful when deciding what type of technology and equipment is needed. In a future post, I'll examine technologies that reduce or even use propagation anomalies to improve RF signal transmission.

About

Information is my field...Writing is my passion...Coupling the two is my mission.

25 comments
michael.brodock
michael.brodock

I wonder if a leaky cable antenna would reduce multipathing or would it be worse?

mjd420nova
mjd420nova

I've worked with these tecnologies for over thirty years and find the info good for non-technical types in understanding a little about how these very high frequecy signals are affected by their individual environments. I'm not so much concerned about boosting the range of my WIFI in the home, but eliminating outside interference and blocking those same signals from leaving the premises. This can be achieved with grounded screens and foils strategically placed around or near the antennas to block outside WIFI users from attempting to use my always on cable modem. Some newer "N" type WIFI routers will interfere with some "B" and "G" type units. I have experienced this and with directional antennas and spectrum analyzer was able to place grounded foils to block them out. Likewise with neighbors and their laptops trying to use my connection. I've not had any trouble with in home lack of signal strength but this could be usefull in boosting signals with directional elements mounted to the antennas themselves.

georgeou
georgeou

Diversity (dual) antennas help multipath issues. 802.11n MIMO systems actually utilize multipath to its advantage.

Michael Kassner
Michael Kassner

Hello Michael, I have used leaky cable antenna systems on two occasions and I suspect it eliminated much of the negative multipath effects. Both installs were in large metal warehouses that would have been difficult to cover properly in any other way. We knew we were in trouble during the site surveys (assumed it was due to multipath interference) and took the risk of testing a leaky cable antenna before getting the contract. The fact that the signal is localized and at a reduced power level may have been the key. That way the RF wave fronts may lose sufficient energy before reaching all of the reflective surfaces. As a side note, if you are considering using leaky cable for an installation, be aware of the increased chance of having hidden nodes as that tends to become a challenge.

Michael Kassner
Michael Kassner

I agree with your comments completely. I was hoping to get more granular about these very subjects in my next blog post. 802.11b/g RF pollution is becoming a real problem, especially in urban environments where many different wireless networks try to co-exist in the same physical space. The use of directional antennas and only enough power to achieve a good link has been a mantra used by "us" amateur radio ops for many years and increasing awareness of this would help a great deal in the 802.11 realm.

Neon Samurai
Neon Samurai

I've a router at home which, asside from being locked tighter than a bank safe, sits in the basement of my house. The location is closer to the rest of my network hardware and hopefully uses the building foundation to limit the horizontal distance outside my house which can recieve a signal. (I want to paint the basement and two floors up rather than walk around my backyard or offer connections to the parkinglot) Based on the primarly locations being directly above or beside where the router sits, have you any recommendations on angles for the two antenna? I've currently got them both at 45 degrees twisted out and tilted back hopefully causing the greatest surface area for both above and beside. Would it be better to have one up and down with the other 90 degrees horizontal or something similar? (I really should take a half hour and just try a bunch of angles I guess)

Michael Kassner
Michael Kassner

Hello George, You are exactly correct. My humble intent was to initially review the exact mechanism of multipath propagation. Having the next post delve into simple techniques and technologies that will help reduce the negative aspects of multipath interference that plague 802.11a/b/g networks. Finally I would like to devote an entire post to MIMO technology and how it actually fosters a multipath environment to improve RF signal propagation.

michael.brodock
michael.brodock

Thanks for the reply. It has been a long time since I worked in radar but most of the fundamentals are the same. :) Are the hidden nodes a function of the antenna creating self amplifying waves due to propagation overlay, or are you talking about just not remembering that you put this access point over here?

Michael Kassner
Michael Kassner

I think they have a place, but I am not a big fan of having all of my resources in one device. I do not see any technical information on 802.11n equipment yet either. Which seems odd.

Michael Kassner
Michael Kassner

I am curious as to how your existing setup works, especially two floors up. Aside from the obvious answer of moving the wireless device to the center of the house there are several options available to you. As usual it depends on how important it is to get complete coverage and how much money you want to spend. Mjd420nova made mention of some things you could try to shape the RF coverage area. The vertical dipoles normally used on wireless devices are not that efficient and have a 360 degree horizontal coverage area. So in your case, with the antennas tilted, 50% or more of the signal is being radiated toward the floor of the basement. There are several DIY methods to add reflectors and direct the RF wave front to cover the areas you want. This link is to a DIY page that has several small reflectors that you can make. http://binarywolf.com/249/diy-parabolic-reflector.htm You also can purchase small panel antennas that will do the same thing. A house is a very good example of a multipath environment so it will as you mentioned take some experimentation to determine what is the best attitude and location of the antennas. Another interesting option that appeals to many people who do not want to run Cat5/6 cable though out the house is to setup signal repeaters or actually make a small mesh network. The only important requirement is that the remote wireless devices have a solid RF link with the Internet connected main wireless unit. http://meraki.com/ Finally if that is not possible, you could use your in house power wiring to extend the Ethernet portion of the network with devices like this or wifi enabled devices at the location needing coverage. When using powerline devices you have to make sure that the circuits being used by both devices are not isolated by a transformer. http://reviews.cnet.com/bridges/netgear-powerline-hd-ethernet/4505-3304_7-31970278.html?tag=txt http://www.netgear.com/Products/PowerlineNetworking/PowerlineWirelessAccessPoints/WGX102.aspx

Neon Samurai
Neon Samurai

Replace one's existing 2.4 ghz cordless phone with a 900mhz or 5.? ghz cordless. I finally did that a few weeks back after playing network russian rulette with a previous cordless phone for far longer than I care to admit.

Michael Kassner
Michael Kassner

Found this explanation in my notes, not exactly sure where it is from. Hidden nodes are client devices that are all within range of the access point but are not necessarily within range of each other. Picture an access point in the middle of a circle. Client A is at 9 o'clock, 164 feet from the access point. Client B is at 3 o'clock, 164 feet from the access point. The distance between the two clients will be 328 feet. 802.11b/a/g uses a media access control mechanism called carrier sense multiple access with collision avoidance. Client nodes more than 300 feet apart are not likely to "hear" each other transmitting in order to avoid a collision. Two nodes transmitting on a common channel at once causes collisions, which results in interference and lowers throughput and response times.

Michael Kassner
Michael Kassner

I am not sure what types of devices you have but using directional antennas on one of the two if a dual antenna system and directing at the remote device could be of significant help. Also adding a device that uses WDS (repeater) might be the best option. Except when considering throughput, but throughput is probably already low due to signal strength. If you are in the mood to experiment and are not locked into the devices you have you may want to look at systems like Meraki Networks. It is a true mesh network and works very well. http://meraki.com/ I have Meraki systems setup in several upscale large homes. The coverage is more than adequate. The pricing is reasonable as well.

Neon Samurai
Neon Samurai

I could throw some shielding behind it to catch anything tryong to go out towards the parking lot. So far, I've simply stuck to keeping my configuration nice and tight. I have since upgraded from a 54gs to a 350n router too though I'm still sticking to 11g/wpa for now. I hadn't considered reflecting the signal though, maybe a reflective shield on the shelf under the router to bounce everything upward would have some effect.

Michael Kassner
Michael Kassner

I remember one client that bought a building around the corner form the original one. He wanted to use Wi-Fi to connect the two. LoS was out of the question, but there was a sign on top of a building that would allow a perfect bank shot. It was pretty amazing as it worked quite well.

JCitizen
JCitizen

I use a metal hutch to reflect energy downward(successfully); maybe a metal screen or tin insert under your AP router would help reflect upward? Just a thought!

Michael Kassner
Michael Kassner

I am glad to see that it works in your situation. I have been involved with many multi-level facilities where that was not the case. You maybe correct about your ability to receive other more distant wireless signals as RF propagation in multipath environments is very difficult to predict.

Neon Samurai
Neon Samurai

The router sits five feet up on a shelf along the wall of the basement. I figured the foundation around it would promote the signal going up but not outward. Mobiles are primarily used on the couch almost directly above on the ground floor. The signal is weaker the next floor up but I've never had the connection drop. The floors don't offer too much resistance. For reception, even when I'm in the same room as the router, I get 50~60% at most. It seems the other radios in the area are broadcasting at 70% and greater. I thought this may be due to antenna angles on the router box. My router's not new so I also expect the others in the area could be newer boxes.

Michael Kassner
Michael Kassner

The audio on Skype is amazing. PSTN and cellular need to take a hint. I know that VoIP is seriously better sounding, but I did not expect that from Skype. Kudos and it is as secure as a non-open source entity can be. As I know you are concerned about that.

JCitizen
JCitizen

because the Public Service Tele here is so reliable I can stay online after a power failure! The only problem is that I have to have a tele account to use their service. Oh Well! We own the phone company around here, it is an association, thank goodness! It doesn't hurt to keep my WIFI router on it too; of course. Can't afford an ethernet failure either. It is good to hear your comments about Skype as I am thinking of dipping my toes in that next.

Michael Kassner
Michael Kassner

I suspect that most equipment on the ISM band has consumer written all over it. Except for Wi-Fi, these products are as cheap as possible and have a minimal amount of intelligence employed in the circuitry. I don't use these types of phones, totally cell phone. But, if I did use a PSTN line I'd use VoIP and wireless handsets. I've set this up for clients already. Actually I think I even prefer Skype over the PSTN line for clarity and ease of use. The only exception and is the reason I don't get rid of the PSTN line coming into the house is that it always works. Every emergency that you hear about PSTN is the only thing that is still working. It's pretty amazing.

JCitizen
JCitizen

the static was so terrible. Now I'm back on 2.4Ghz and these are working well. I wonder if the FCC is not only working to reduce RF noise on walk abouts but also working on digital circuit/programing improvement?

Michael Kassner
Michael Kassner

The cordless phone issue is more prevalent than most people think. It sounds like you also had to deal with the fact that cordless phones are designed to hunt for the best frequency and it eventually gets to the one used by the 802.11b/g network.

Editor's Picks