Overview
The site survey is an assessment of signal coverage per each access point within a specific building. It confirms with the client 100% coverage at a specified signal guaranteeing employees will have no issues with sending wireless packets across the network. Each building structure will have a design that determines signal characteristics. The most often issues that cause problems with decreasing wireless signals include water, metal, building contours and devices that emit electro-magnetic interference. Some access points can send a pattern of coverage extending 100 – 170 feet while some will emit a signal 50 feet with same characteristics.
The building structural design has a lot to do with that.
The following is an overview of the site survey process:
1) Discuss Signal / Noise and Data Rate with Client
2) Obtain Floor Plans
3) Survey Buildings Noting Signal and Coverage
4) Document Access Point Placement
What then is an acceptable signal? Signal to noise separation of around 20 – 25 dBm at 54 Mbps is a minimum with most designs. The site survey software measuring a continuous 75 dBm signal and 95 dBm noise (75/95) would be enough to have employees connected and sending packets with same service level as wired clients. As signal weakens it becomes more difficult to distinguish between a signal of 80 dBm and noise of 95 dBm (80/95). There is less separation between signal and noise making the signal difficult to detect. It is counter intuitive that a larger number is a weaker signal however the key with signal and noise measurements is the signal is a negative number and hence a lower value is a bigger number. Discuss with the client what specific client adapter, 802.11a/b/g wireless standard and data rate they’ll implement and survey with that. Deploying the 802.11a standard doesn’t require a separate survey if you implement dual band and surveyed with 802.11g. Some companies with higher bandwidth demands will improve cell coverage with a separate 802.11a survey. It isn’t as cost effective with additional access points deployed per cell coverage. Configure the access point with the selected 802.11a/b/g protocol, transmit power and data rate. Set the data rate and transmit power of the client adapter to maximum during the survey.
The site survey of a building involves walking around with a laptop on a powered cart configured with a Cisco wireless client adapter and the Cisco Aironet Desktop Utility (ADU). The software is a utility available with Cisco client adapters. The Cisco access point is placed at a specific spot on the ceiling or above the ceiling tile, wall, desktop or cubicle and measurements of signal coverage is noted. The access point is moved until optimal coverage occurs for that cell. Select ADU active mode to examine data rate, link status and errors while surveying. The cell coverage can be transferred to a Visio drawing floor plan with specific access placement and signal boundaries. The access point is moved around until the entire building is mapped. The minimum signal strength should be recorded per coverage cell. Be as specific as possible as you describe where access points should be implemented. Some clients will have floor plans which makes the site survey much easier to document coverage. Scan each floor plan as a Visio drawing noting specific placement of access points and assignment of channels.
Your map will have some overlapping of coverage which isn’t a concern. You should assign a channel that doesn’t overlap with neighbor cells and employees won’t have any problem with channel interference. As mentioned the 802.11b and 802.11g wireless standards use the 2.4 GHz frequency band which can cause interference however there are 3 non-overlapping channels (1, 6 and 11) that can be assigned. As an example should your building have a main floor with 17 rooms and each access point cover 3 rooms use the following channel assignments:
AP1 Channel 1 Rooms 1-3
AP2 Channel 6 Rooms 4-6
AP3 Channel 11 Rooms 7-9
AP4 Channel 1 Rooms 10-12
AP5 Channel 6 Rooms 13-15
AP6 Channel 11 Rooms 16-17
In some instances you could have signal overrun between AP2 and AP5 with specific building design shapes. Decrease the transmit power from 100 mW to 20 mW on either or both access points and confirm signal strength is still acceptable at those rooms. Should the decreased transmit power setting not allow for same coverage, change position of the access point, implement an additional access point at decreased wattage or an external antenna. Site surveys take some trial and error knowing where possible signal coverage will occur. You will know after doing some preliminary testing at a specific building what coverage to expect. Walk around with the laptop when starting and get some understanding of signal patterns. Signal overrun can often occur outside the building posing a security risk with hackers on the street. The site survey should confirm no signal overrun with minimized access point transmit power.
Multi-floor buildings will have some access points that will send signals across several floors. You can of course use fewer access points should access points cover those floors or decrease transmit power and assign non-overlapping channels as discussed. The issue with signal overrun is a concern where an access point covers some of a floor and not all of a room on another floor. You want access points spaced with proper signal coverage and minimal overlapping. Again the building could allow for an access point to cover a single floor or 2 floors with a signal strength as needed everywhere. Should a third floor exist you can deploy it on that floor. The first and second floor would have an access point on the first floor. Deploying on the second floor would cause signal overlap. That isn’t an issue with non-overlapping channels, however clients with fast roaming should associate with the nearest access point. Having access points on second and third floors could cause some clients on the third floor to select a second floor access point over utilizing it and decreasing performance.
The client sends out a beacon and selects the nearest access point with the best signal, lower number of clients and matching security settings. A negotiation between access point and client adapter occurs and data is sent at the highest supported rate available. The coverage map with access point placement should be specific as possible during the survey describing access point placement. Deploy all access points and test connectivity. With all access points transmitting the chances for interference occurs and the design can be modified before deployment. Signal overrun outside can be tested confirming the signal stays within wall boundaries and there is no access from the street. Spectrum analyzers can be deployed during the site survey or after to examine frequencies with interference and possible sources. Defining a security test plan is a good idea after deploying access points.
Make survey notes concerning mounting issues, where AC power outlets unavailable, distance between wiring closets and cabling distance between access point and switch. Cisco does have outside wireless bridges available connecting buildings with line of sight. There is a survey involved with deploying those bridges that considers path loss with transmitted signals and how environmental issues such as wind and rain will affect data transmitted.
Shaun Hummel is author of Cisco Wireless Network Site Survey available at amazon.com and CiscoDesignBooks.com featuring Networking Books, eBooks, Certifications, Articles and Design Tools.
Wireless Network Site Survey Overview
No comments:
Post a Comment