WLAN System Design Fundamentals Part 1: General Concepts

Print Friendly

With much talk about small cells, a review of wireless LAN (WLAN) design is in order. This review is a basic design guide for WLAN coverage enhancement, the first of a multipart series on in-building wireless enhancement.

The Wi-Fi Alliance, the organization that owns the Wi-Fi (registered trademark) term specifically defines Wi-Fi as any WLAN product based on the Institute of Electrical and Electronics Engineers’ (IEEE) 802.11 standard. Some folks think WiFi technology is not part of a “small cell”. However, one can argue that unless a wireless technology is part of a tower’s local infrastructure, it is considered a small cell. A WLAN design includes “small cells” of wireless access points (WAP) much like a cellular DAS. Even if a carrier adds “WiFi offloading” the WLAN is detached from the tower and hence could be described as a small cell.

Typical complex Wireless LAN architecture by Cisco

Typical complex WLAN architecture by Cisco

The demands on WLANs for functionality and scalability are growing due to the rapid proliferation of new network devices and applications. The number of devices and connections per user is steadily increasing. It is common for most users today to not only have a primary computing device but also at least one other smart device. Wireless operators have worked hard to accommodate the increased demand for data services over wireless networks. They have been forced to consider alternative offload strategies, including wirelessly connecting electronic devices (Wi-Fi). Unfortunately, the majority of smartphones being introduced into the marketplace only support Wi-Fi at 2.4 Gigahertz (GHz), which is rapidly increasing pressure on Wi-Fi designers and administrators to design products for the smallest segment of bandwidth available. Many devices now include the 5 GHz band (vis-à-vis 802.11n). Administrators and IT Managers are finding themselves faced with the challenge of providing ever-increasing levels of WLAN service in areas where simple coverage is the singular design goal. While there have been great advances made in the speed and ease of implementation of Wi-Fi networks, the basic nature of radio frequency (RF) is generally unchanged. Increasing the number of users who can access the WLAN in a small physical space remains a challenge. The steps and process for a successful high user density WLAN design that can be proven, implemented, and maintained.

General Process for Wirless Local Area Network (WLAN) Design

The general concepts underlying medium-density Wi-Fi design remain true for many environments. But it is important to note that the content and solutions presented here will not fit every WLAN design scenario. Rather, the intent of the design guide is to explain the challenges in WLAN design and to offer successful strategies so that engineers and administrators understand them and are able to articulate the impact of design decisions.

Not treated here, but of potential concern in some situations and jurisdictions, is the matter of radio frequency safety. In general, power density from most WLAN installations is sufficiently low as to be acceptable under most codes and standards. However, the designer should always examine and provide for these possibilities.

WLAN design refers to any environment where client devices will be positioned in office environments where coverage expectations of a normal enterprise deployment, in this case a traditional, multi-floor, carpeted office are both ubiquitous and robust. For reference, a typical office environment has indoor propagation characteristics for signal attenuation as shown below. User density is a critical factor in the design. Aggregate available bandwidth is delivered per radio cell, and the number of users and their connection characteristics (such as speed, duty cycle, radio type, band, signal, and signal-to-noise ratio) occupying that cell determines the overall bandwidth available per user. A typical office environment may have APs deployed for 2,500 to 5,000 square feet with a signal of -65 dBm coverage and a maximum of 15 to 20 users per cell. That is a density of one user every 100 square foot (sq. ft.) and yields a minimum signal of -65 dBm.

WLAN 2.4 GHz Coverage Heat Map (WAPs are white-shaded circles; Coverage scale is received signal power)

WLAN 2.4 GHz Coverage Heat Map (WAPs are white-shaded circles;
Coverage scale is received signal power)

Before a WLAN design is completed, a heat map survey and RF spectrum measurements are performed. A WLAN survey detector as well as a spectrum analyzer can reveal both existing WAPs and any interference that may be present in a facility. Graphical heat maps help visualize anticipated wireless LAN behavior for planning and design of a new or upgraded system.   During the survey general information on the architectural characteristics of an office should be noted. A sample of existing AP signals is collected while spectrum readings are made to support the study and analysis.

Propagation of electromagnetic waves inside buildings is a very complicated issue. There are various models based on various principles with various requirements for input data complexity. In real situations, any piece of furniture, open doors and windows, moving people, reflections from outside the building and other effects influence the signal propagation. In a propagation study, several assumptions are made based on either typical WLAN equipment parameters including AP power output and average antenna gain. Predicted coverage levels on an individual multiple input multiple output (MIMO) WLAN channel may vary up to plus or minus several Decibels (dB) due to antenna gain variations in azimuth. The WLAN signal also changes due to attenuation of the walls and ceiling. Assuming a median attenuation for both frequency bands, wall losses are also estimated based on the survey and empirical data. Heavy wall losses are modeled as moderate cement member unit losses. Floor to floor penetration is modeled as the floor construction is assumed to be concrete poured on a metal pan. However, with some signals penetrating the floors, coverage will be enhanced with AP parameter optimization. Exterior wall and window losses are modeled as based on survey observations and other empirical data. It is also assumed that the 802.11n AP’s will be operated in narrow channel mode (i.e. 802.11b) so that multiple channels can be utilized for frequency division multiplexing of users. The design criteria are -65 dBm in coverage under a load of 15 users. A typical office may have APs deployed for 2,500 to 5,000 square feet with a signal of -65 dBm. However, to the end, typical AP layout goal is to assure contiguous coverage throughout all of the building.

In summary, WLAN design fundamentals include planning, analysis and design of the heat map data, RF spectrum background noise levels and propagation characteristics of the building. WLAN AP layout is only effective after these steps so that excellent coverage for 802.11n technologies will be available to all users in the building. A WLAN upgrade may be necessary at your facility to ensure business needs are met with technology- business drives technology!

 


About the author:

Chris Horne, Chief Technology OfficerChris Horne is Chief Technology Officer with the LBA Group, Inc. Chris is a Professional Engineer, and holds a Doctorate in Electrical Engineering.

He specializes in wireless and industrial communications, including DAS system design and evaluation, RF safety, and RF interference management. Chris has successfully undertaken many challenging WLAN and DAS projects. Contact Chris at chris.horne@lbagroup.com or 252-757-0279.

 

 


About LBA Group Inc.
LBA Group

LBA Group, Inc. has 50 years of experience in risk management, design, and integration of industrial and wireless telecommunications infrastructure assets, worldwide. It is comprised of the professional technology consultancy Lawrence Behr Associates, Inc. ; LBA University, Inc. providing on-site and online professional training; and LBA Technology, Inc., a leading integrator of radio frequency systems, lightning protection, and EMC equipment for broadcast, industrial, and government users. The companies are based in Greenville, N.C., USA.

 

Keep up with the latest LBA news and industry information on Facebook at:  www.facebook.com/LBAGroup.

 

19 Comments on "WLAN System Design Fundamentals Part 1: General Concepts"

  1. William B. Cheney, III April 28, 2013 at 5:17 am · Reply

    Isn’t interesting that after years of ignoring RF Engineering, it has finally become important to the EE comunity again. As a Nuclear Engineer, I spent a lot of my time resolving RF interference with Neutron counters and other Rad Monitoring Equipment. I was a Ham. No body in the Nuclear Electronics Branch, all degreed and PEI or PE had any RF Engineering training. Yet I couldn’t qualify as a PEI because I was a Physicist!
    Welcome back into the fold. It ain’t all about chips.

    On another subject, have you thought about using a real antenna with a ground plane and a capacitive Hat to reduce losses to ceiling and floor? All I see is omni-directional antennas with vertical polarization.
    I also assume in your example that all your nodes are slave routers. All set to different channels, but the same ssn and password, so people can move from one area of the office to another, seamlessly sliding from one connection to another?

    Like or Dislike: Thumb up 1 Thumb down 0

  2. Jack Hill, W4KH April 29, 2013 at 10:41 am · Reply

    The blog is interesting… Although I doubt that ANY IT Department would overlook it, when setting up a WLAN, there are some BASICS no one should overlook: 1) Change the SSID; 2) Hide the SSID; 3) Select MAXIMUM encryption available; 4) Use a complex password; 5) Be vigilant in monitoring the logs; 6) Do NOT allow any employee with a laptop, smart phone, tablet, or whatever to access *ANY* insecure WLAN!!!

    ln

    Like or Dislike: Thumb up 0 Thumb down 0

  3. Lawrence Behr April 29, 2013 at 10:53 am · Reply

    Thanks for the comment. Agreed, but the paper dealt with the infrastucture aspects. Operational matters are important, but another layer beyond the discussion.

    Like or Dislike: Thumb up 0 Thumb down 0

  4. Jack Hill, W4KH April 29, 2013 at 10:54 am · Reply

    Interesting suggestion about the antenna design… The ground plane under the antenna would tend to reduce loss to the floor, but a cap hat would have some interesting “side” effects – i.e., the distributed capacitance of the cap hat would couple with the antenna and tend to reduce antenna “Q”, and might not prevent ceiling loss so much as broaden the frequency response of the antenna… perhaps an antenna scaled down from the WSM tower design (two triangular prisms mounted base to base with points at top and bottom) would better drive the signal in the desired (omni) direction? ln

    Like or Dislike: Thumb up 0 Thumb down 0

  5. James Carlini April 30, 2013 at 7:00 am · Reply

    What the “Fundamentals” paper does not touch upon is the design reality of the network vs. the design theory of the network. How many networks have been installed and look good from a theoretical standpoint, only to be overwhelmed when they are actually used at the venues?

    How many design engineers have been brought back to re-design what was thought to be a “good network” only to find out that it did not provide adequate capacity? (Some Baseball and Football stadiums come to mind. They had to be re-engineered in one year.)

    Some other points in “Fundamentals” should be to somehow test the network once it is put in place (A Stress test where you see how much traffic it can handle). A post-implementation review is also good to add.

    How well does the network desgn perform in a real life situation? The importance of “Capacity” should take precedence over “Coverage”. Both are important,, but “capacity” is just as critical as coverage, if not more so.

    What I have also seen is that 99% of the networks in place are obsolete because they only pulled copper to the end points for all the antennae. With next-generation antenna coming out with fiber connections, all of them will have to re-cable in order to get higher capacities from the fiber-based antennae. Instead of a quick switch out at all the endpoints, it will be a major job to replace not only the antennae, but all the cable backbone as well.

    Like or Dislike: Thumb up 0 Thumb down 0

    • Chris April 30, 2013 at 10:41 am · Reply

      The paper’s design fundamentals criteria were used by the Corporation to upgrade their existing 802.11n WLAN network at their headquarters. The author worked with an infrastructure team at the Corporation in the planning, analysis and design phases. Due to some extreme aesthetic requirements by the Corporation, antenna style and placement was not standard. Then again, much of RF design is an “art”, not a science; not “engineering.” The Corporation is now implementing the new solution. The upgraded system is all SMF to the WAPs. In any wireless network (RFID, cellular, WiMax…) there is a tradeoff between coverage and capacity. As modulation schemes evolve and technology advances, the physics remains the same. Thanks for the post.

      Like or Dislike: Thumb up 0 Thumb down 0

  6. Bob Johnson April 30, 2013 at 11:40 am · Reply

    My only experience with wireless LAN systems has been that they didn’t pose any RF or environmental safety issues! ln

    Like or Dislike: Thumb up 0 Thumb down 0

  7. Lawrence Behr April 30, 2013 at 11:40 am · Reply

    @Bob Johnson

    Normally no, but there is a perception of hazard by some building owners and others. We are involved in a large project now that will require MPE compliance verification throughout. Both the GC safety manager and owner’s insurer insist on it. Several years ago, we were retained by a major carrier to verify a new system in front of reported employee ill effects. Workers comp claims were being made, and their legal staff wanted a third party report.

    Like or Dislike: Thumb up 0 Thumb down 0

  8. Nir Cohen April 30, 2013 at 2:19 pm · Reply

    Here is a great experience that was one of the highlights of this year Wi-Fi covered events – 2013 NBA All Star Weekend Games Powered by Extricom’s WLAN Channel Blanket Technology. More details on the link below:
    http://www.extricom.com/category/2013_NBA_AllStar_Weekend ln

    Like or Dislike: Thumb up 0 Thumb down 0

  9. Lawrence Behr April 30, 2013 at 2:20 pm · Reply

    @ Nir Cohen Good case. Sports venues are a challenge all to themselves!

    Like or Dislike: Thumb up 0 Thumb down 0

  10. James Carlini, MBA, IC April 30, 2013 at 2:28 pm · Reply

    What the “Fundamentals” paper does not touch on is the design reality of the network vs. the design theory of the network. How many networks have been installed and look good from a theoretical standpoint, only to be overwhelmed when they are actually used at the venues?

    How many design engineers have been brought back to re-design what was thought to be a good network only to find out that it did not provide adequate capacity?

    Some other points in “Fundamentals” should be to somehow test the network once it is put in place (A Stress test where you see how much traffic it can handle). A post-implementation review is also good to add.

    How well does the network perform in a real life situation? The importance of “Capacity” should take precedence over “Coverage”. Both are important, but “capacity” is just as critical as coverage, if not more so.
    ln

    Like or Dislike: Thumb up 0 Thumb down 0

  11. Lawrence Behr May 2, 2013 at 12:28 pm · Reply

    @ James Carlini Excellent points, James. This was Part 1. Several new Parts are being written to cover specific areas. I’ll pass your suggestions on to Dr. Horne.

    Like or Dislike: Thumb up 0 Thumb down 0

  12. William Faehse May 2, 2013 at 12:28 pm · Reply

    As a new engineer just starting in the DAS industry, I appreciate the effort put into writing this blog and I look forward to reading his next posting. ln

    Like or Dislike: Thumb up 0 Thumb down 0

  13. William A. Boyd, RCDD/OSP May 2, 2013 at 12:28 pm · Reply

    Wow, this could almost set off a firestorm of discussion about mixing WLAN and DAS on the same infrastructure. Some say Yea, some say Nay. What was that Yogi Berra said?
    ln

    Like or Dislike: Thumb up 0 Thumb down 0

  14. Randy McInnis May 2, 2013 at 12:31 pm · Reply

    For past 6-7 years or so, I’ve been providing Power Backup/Power Protection solutions for DAS systems (among other things of course). Is it typical for a non-cellular based WiFi system to use backup and/or power quality/protection equipment? ln

    Like or Dislike: Thumb up 0 Thumb down 0

  15. Chris Horne May 2, 2013 at 12:35 pm · Reply

    Hi Randy, with global DAS growth expected to generate revenue of 13+ billion by 2015, traditional power equipment firms like Emerson are smart to capitalize on it. After all they’ve been providing “BTS” back up for 10 + years. Head end back up can be straightforward like a BTS but as you may know the remote unit DAS back up could be challenge. Remote units may be powered by either AC or DC depending on the system. For public safety DAS, back up uninterruptible power supplies at the remote nodes are critical. For WiFi WAPs, the back-up I’ve seen is the building’s emergency power. As you may know, the FCC adopted backup power requirements for cell towers in 2007 following Hurricane Katrina, but a appeals court ruled against the regulations in 2008. Sounds like a technical opportunity to me but the carrier’s cost is not trivial (i.e., rising cost of batteries ) nor is the potential loss of lives due to failed circuits. That’s one reason I still have my ham radio for natural disaster communications. Thanks for the post.

    Like or Dislike: Thumb up 0 Thumb down 0

  16. David Beck May 2, 2013 at 12:37 pm · Reply

    Hello Lawrence.

    Personally I love wireless Lan’s. I’m sure you were refering to technological stories, but still….

    There is a local coffee shop near me that sells extremley good quality coffee: all you can drink for £1.75.

    I had about 8 cups of the strong fine brown stuff and also played with my phone on the wireless network whilst sitting with a friend. I showed him some great youtube video’s and we just drunk so much coffee that we could have died!

    I don’t see how that shop makes a profit! ln

    Like or Dislike: Thumb up 0 Thumb down 0

  17. James Carlini, MBA, IC May 3, 2013 at 11:15 am · Reply

    Randy – EVERYTHING should be backed up if it is a mission critical application. (and I would say at this point – anything in communications is mission critical).

    If you go into a telephone central office – EVERYTHING is backed up and when it comes to power, each office has battery back-up as well as a diesel generator besides power from the electric company. (Triple back-up) WHY? Because no one wants that money generator to go dead.

    People who design or accept “cheap” systems with no power back-up have no clue as to what network infrastructure should be.

    We have a problem in the industry today with those who think they are engineers because they built a couple of WiFi systems for the local coffee shops and those who built the large totally redundant networks of the phone company and many private companies (like the Exchanges, the Airlines, etc)

    And remember this – even those networks that are totally redundant never get to 100% reliability. You can design to 99.99, 99.999, 99.9999 even 99.99995 but you will never attain 100%. Anyone who says the phone company’s network is totally redundant is wrong – look up the Hinsdale Central Office Fire which became a huge disaster.

    Bottom line – you are selling a GOOD solution. Anyone who doesn’t put in battery back-up (or something even more substantial) on any network doesn’t understand the nature of the industry. ln

    Like or Dislike: Thumb up 0 Thumb down 0

  18. Bob Hanshaw May 3, 2013 at 2:39 pm · Reply

    One of the contributors to the LBA Group Blog, indicated “…global DAS growth [is] expected to generate revenue of 13+ billion by 2015…,” which makes me wonder about the viability of the current infrastructure. I’m not an engineer, but through my military service and Ham Radio, believe I have a pretty good grasp on technology. That said, what’s the feasibility of having individual WAP devices–in a local neighborhood, for example–all working together in a manner similar to the TCP/IP Packet concept. Could each WAP, wireless router, or other smart devices be configured to share a secondary channel–like a Hot Point–to extend coverage without the need to build redundant networks? Aside from getting ISPs working on the same page, what other concerns would need to be overcome?

    In my third life (military, educator, and now career and transition advisor) I find the prospects, for employment and jobs, interesting for those looking for a career, especially in light of the, once again, increasing reliance on RF. RF is not only being used in wireless-radio technology, but there is a growing need to comprehend analog devices to understand, mimic, and reproduce artificial human behavior, as many of you already know. Some see RF as a more direct way for circuitry to react to interface situations without all the need for D/A–>A/D conversion and coding algorithms. It seems as though, like William Cheney said in an earlier post, “…It ain’t all about chips [digital].” This would not only open the doors for the employment of more RF engineers/techs, but also increase the need for qualified educators. ln

    Like or Dislike: Thumb up 0 Thumb down 0

Leave a Comment