Defining Access Point Range Boundaries
May 27, 2003
The range boundary of an access point is very important when determining the radio frequency coverage of a WLAN. Discover several methods for defining range boundaries so you'll know which ones you should use.
A major task in designing a wireless LAN is identifying the number and location of access points. This involves performing a radio frequency (RF) site survey in order to test specific access point locations as the basis for determining the range boundaries of access points within the actual facility. The results of this testing indicates the optimum number and location of access points.
Range boundaries characterize the circumference of the radio cell that the access point produces. With the boundaries known, a designer can make certain that the radio cells overlap enough to enable roaming and an adequate degree of redundancy. This may seem straight forward, but complications arise because of the nature of radio waves and the 802.11 protocol.
Range Boundaries Based on Connectivity
The range boundary of an access point based on connectivity indicates the point at which a user device can associate with an access point at a specific data rate. For example, you might define the connectivity range boundary to be at 11 Megabits per second (Mbps), which means that the boundary is met when the associated data rate drops from 11Mbps to 5.5Mbps. Within this range boundary, all users will have 11Mbps associations. Similarly, you may define the boundary to be at 1Mbps, which is when the associated data rate drops from 1Mbps to the disassociated state. In general, you can define the connectivity range boundary to be any data rate that the access point supports.
Even though designers generally specify range boundary based on connectivity, it's not sufficient for WLAN systems needing to support higher capacities, many users, or guaranteed levels of performance. The issue is that data rate alone does not provide a true metric for user performance. The definition of range boundaries based on connectivity introduces significant risk when planning and managing the performance of the WLAN.
One user having an 11Mbps association with an access point may be relatively free from RF interference and have nearly maximum throughput (approximately 6Mbps for 802.11b). In this case, using range boundaries based on connectivity would probably be satisfactory, assuming the user conditions don't change.
A more likely scenario, however, would entail multiple users experiencing RF interference from adjacent access points. Each user may have an 11Mbps association with the access point, but the interference (and collisions among the users) could cause each user to experience 50% packet retransmissions. This could significantly reduce the total throughput for users to unacceptable levels that don't adequately support the wireless applications.
Range Boundaries Based on Performance
As WLANs begin to support more users requiring greater capacity, I'm finding that designers should utilize a more accurate method of defining access point range. For example, I've been working with Cellular Design Services (CDS) over the past few months as a consultant and designing a common WLAN infrastructure for the BAA airports, which includes London Heathrow Airport and several others located in the United Kingdom.In addition to lower performance bar code applications, the BAA WLAN will support travelers requiring broadband wireless connections throughout airport public areas. A combination of these requirements requires careful planning of access point locations to ensure bandwidth requirements are met. In this case, we've found that the use of range boundaries of access points based merely on the ability to connect with an access point at a specific data rate is simply not good enough.
A critical goal of planning a high performance WLAN is to ensure that the packet retransmissions stay down to an acceptable level. In addition to simply indicating association with an access point, the range boundary definition should also include affects of both interference and noise. This defines the useful service boundary of an access point, which is the distance at which the signal-to-interference-plus-noise-ratio (SINR) is just large enough to provide sufficient performance. Designers should utilize this definition of range boundary when specifying the location of access points, such as when performing an RF site survey.
This method will generally produce a range boundary that's closer to the access point than when using connectivity alone to define the boundary. An acceptable SINR will depend on the actual application, but 30dB SINR will generally provide optimum performance for maximum capacity WLANs. In order to make use of this range boundary definition, however, ensure that the site survey tools you select have the ability to measure SINR.
Range Boundaries Based on Interference
Of course, the range boundary of an access point is important to know in order to guarantee a specific level of performance for mobile users, but another element to consider is potential interference with other systems. An important point to realize is that the range of an access point in terms of potential interference with other systems can be much farther than the range necessary for user association.
For example, a user may lose association with the access point at 150 feet range, but the access point may still interfere with another system located 500 feet away. As a result, designers should determine the range boundary based on interference when performing site surveys to ensure the adequate spacing of access points to minimize interference with other systems and access points set to the same channel. The actual signal level that will cause interference depends on the external system. As a result, be sure to understand the toleration level of all existing and future systems that could be the recipient of RF interference, and ensure that these systems fall outside the range boundary (based on interference) of your access points.
Stay tuned! In future tutorials, we'll look at tools and RF site survey processes that take advantage of performance and interference range boundaries.
Jim Geier provides independent consulting services to companies developing and deploying wireless network solutions. He is the author of the book, Wireless LANs and offers workshops on deploying WLANs.
Join Jim for discussions as he answers questions in the 802.11 Planet Forums.
Got a question for Jim Geier and you're just too shy to post anonymously in our forums? Join us at the 802.11 Planet Conference & Expo, June 25 - 27, 2003 at the World Trade Center Boston in Boston, MA. You can meet Jim in person as he gives his talk called 2.4 GHz or 5GHz? Strategies for Choosing The Right Spectrum .