Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Do you really want to delete this prezi?
Neither you, nor the coeditors you shared it with will be able to recover it again.
Make your likes visible on Facebook?
You can change this under Settings & Account at any time.
EU - Lesson 2 - What is a good site survey
TC Leeon 28 August 2014
Transcript of EU - Lesson 2 - What is a good site survey
What is a Good Site Survey?
WiFi network is already designed before the project starts
1. Construction team designs WiFi
2. CFO designs WiFi
3. Hotel star ratings designs WiFi
Problems We Face Today
1. Enable DHCP server (on AP or ZD or network)
2. Configure 2 SSIDs: one for 2.4GHz, one for 5GHz
e.g. "Survey2.4" and "Survey5"
3. (Suggestion) Configure fix frequency:
- channel 11 for 2.4GHz
- channel 165 for 5GHz (or the highest available channel for your regulatory domain)
Site Survey Major Configuration Items
Get floor plans
Understand requirements and scope
Mark coverage areas
2. Bandwidth requirements?
3. Uplink (Internet) bandwidth?
4. Client devices?
-65 RSSI on 2.4GHz measured using a laptop
Get Rooms and Areas for Testing
Meeting rooms, ballrooms
May include back-of-house areas
1. Get adjacent rooms
2. Get rooms across the corridor
3. If possible, room one floor above
4. And/or room one floor below
Scan existing WiFi
Test signal between clients across the corridor
This will determine if you should partition your coverage area
1. Enable tethering or portable hotspot on the Android phone or iPhone
Note that most tethering runs in 2.4GHz only
2. Place it inside the guest room
3. Use another smartphone or laptop running InSSIDer or WiFi Analyzer to detect the portable hotspot in the room diagonally opposite
If you can detect the portable hotspot across the corridor, then you will be able to position access points along the corridor.
If not, you will need to plan each side of the corridor separately.
Note: this totally depends on the room design/layout and construction.
Use the laptop to run InSSIDer and scan surrounding WiFi to understand the environment. Or use spectrum analyzer.
Deploy the access point
Determine the likely locations for access point and test each location separately.
Test signal levels & throughput at various points
Test 2.4GHz and 5GHz
using a laptop or smartphone running InSSIDer (
without association i.e. passive survey
using smartphone running Speedflex or ZapPerf
using smartphone running Speedflex or ZapPerf
Record results on floor plan
location 2.4Grssi down up
.a -35 45M 35M
.b -45 40M 30M
.c -30 55M 40M
.d -39 42M 32M
.e -65 35M 27M
.f -75 20M 12M
.g -55 38M 30M
Repeat For Different AP locations
Mount AP in different locations
Repeat tests: downlink signal, downlink throughput, uplink throughput
Record results on floor plan
Site Survey Software
Ekahau Site Survey or AirMagnet Survey software
Perform walk-through to collect survey data and generate heatmap
Follow somebody else's design
1. Another vendor came with a design - follow
2. Existing WiFi in place - follow
3. Other hotels did this successfully - follow
Myths and misconception
1. Strong signal from AP is better
2. Professional tools gives best results
3. Site survey can be done during construction
Ekahau Site Survey walk-through
Airmagnet Survey walk-through
Generate Heat Map Report
1. Using Ruckus ZonePlanner
2. For the sections that are surveyed
3. For selected floors (predictive model) and areas
4. Attach heat map report together with BOM
What do we measure?
1. Downlink signal: AP-to-client signal
2. Uplink signal: client-to-AP signal
3. Downlink throughput: AP-to-client throughput
4. Uplink throughput: client-to-AP throughput
Downlink Signal Measurement
There are differences when the client is associated vs when the client is not associated.
When client is not associated, it is a
When client is associated, beamforming comes into play and signal is expected to be better.
Recommended to measure the worse-case: not associated.
Uplink Signal Measurement
No site survey tool can provide this measurement*.
Has to be measured from the AP.
Client has to be associated.
*One way is to configure the client to be a soft AP.
Or use tethering/portable hotspot on the client.
Rule-of-thumb: uplink is typically about 10-15 dBm lower then downlink
Signal alone is no guarantee of connectivity.
The purpose is to ensure no unexpected interfering sources especially non-WiFi.
If signal is good but throughput is not, then it is an abnormal condition (lack of airtime).
Needs further investigation with spectrum analyzer or air packet capture.
Every device has different receive sensitivity (reception)
Laptops usually have better reception than smartphones
Decide on one and document it. That is the reference. Describe it in your documentation.
Take measurements with a smartphone as worse-case scenario
Downlink signal: -65 dBm@2.4GHz RSSI
Uplink signal: -75 dBm RSSI; >=25 SNR
Downlink throughput: 20-30 Mbps minimum
Uplink throughput: 10-20 Mbps minimum
WiFi is very application specific and client device specific.
Site survey is to collect hard data
Knowing what you are dealing with goes a long way to help you evaluate WLAN designs by different vendors and make the right judgement
High Density Areas
Convention centers, ballrooms, meeting rooms, etc
Planning by capacity not coverage.
Estimate expected maximum number of client devices
Challenging RF conditions especially at 2.4GHz due to
- Exposed node issue
- Co-channel interference
Always use concurrent dual-band APs. Use high performance APs such as ZF7982
High Density Guidelines
1. Use "ofdm-only" settings to reduce cell size
2. Enable background scan and client load balancing*
3. Enable bandsteering
4. Disable intrusion prevention
5. Use 20MHz channel width on 2.4GHz
6. Use 40MHz channel width on 5GHz
7. Limit broadcast traffic
8. Enable ChannelFly on all channels in both bands
9. Try non-traditional channel plan (e.g. 1,4,7,11)
Design for Elevator Coverage
A pair of ZF7731 to create a mesh link spanning the elevator shaft.
Tx power settings depends on length of shaft and nature of shaft.
Connect back-to-back to a 7055 or 73xx low-profile AP.
Tx power may need to be reduced.
large meeting rooms 1 AP to 60-100* clients
smaller meeting rooms of 20-30 pax can share AP
* - depends on bandwidth requirements
- using ZF7982
Case Study - MACUL Education Conference
Mar 20-22, 2013, Cobo Convention Center (Detroit, Michigan)
1. 2 x ZD3500
2. 60 x 7982
3. 1 x Flexmaster
1. 5,811 unique clients
- 802.11a/n: 4,081
- 802.11g/n: 3,376
- 802.11b/g: 130
- 802.11a: 0
3. Mar 20: 655 clients
4. Mar 21: 4,737 clients
5. Mar 22: 4,532 clients
Depends on elevator shaft design.
What about 5GHz?
Lesser congestion and more capacity. More reliable.
Throughput is not as high a concern in hospitality.
But range is shorter than 2.4GHz. Not practical to deploy in corridor. Must be deployed in-room.
5GHz signal is about 15-20 dBm lower than 2.4GHz signal. So wireless SLA must be defined separately for 5GHz.
E.g. -65 rssi @ 2.4GHz; -75 rssi @ 5GHz LOS
5GHz vs 2.4GHz - a comparison
Demand and review this readings during site survey.
So that you can counter-check against heatmap generated.
2.4GHz Heatmap with -65 rssi Threshold
5GHz Heatmap with -65 rssi Threshold
5GHz Heatmap with -75 rssi Threshold