The VeriWave System
How many stateful 802.11 clients can I create using VeriWave hardware?
With the VeriWave Traffic Generator and Analyzer, up to 500 stateful clients can be created on each test port. The 802.11 a/b/g test blade as well as the 802.11 a/b/g/n SISO test blade, has 4 independent test ports on each. Thus, a fully-loaded WT90 VeriWave chassis, housing 9 (nine) test blades is capable of creating up to 18,000 independent 802.11 clients. Each client implements an independent MAC 802.11 state machine, as well as independent implementation of L3 (IP), authentication and encryption and stateful L4 (TCP / UDP) and L7 (FTP / HTTP / VoIP / IPTV etc.) protocol state machines.
top 
How is it possible to create so many wireless clients from a single RF port?
VeriWave utilizes custom-built FPGA and DSP based hardware to concurrently run fully stateful 802.11 MAC state machines for each client. A flow scheduler enables the combination of various traffic types, including Constant Bit Rate voice/video traffic and Best Effort data traffic onto each client. Every client independently updates its state based on channel conditions and backoff timers, and, utilizing a separate client scheduler, determines the timing of transmit or receive activity for each client on the shared channel. This offers optimum channel utilization while still creating behaviors typical to any shared medium, such as hidden-node collisions.
top
What kind of traffic can I send using VeriWave’s traffic generator?
VeriWave applications let you choose from a large variety of traffic flow types. Here is a sample list:
- Voice over IP (various codecs supported)
|
- MPEG-2 Video (various codecs supported)
|
|
|
|
|
- Barcode App (scanner traffic)
|
|
|
|
|
|
|
|
| |
|
| |
|
| |
|
Traffic is hardware generated and can be unidirectional, bidirectional or multicast. Custom flows can be created by the user as well. Each flow can be tagged with QoS fields (layer 2 and/or layer 3 QoS). Each flow can be configured with pass/fail criteria. Reports are generated to help the user understand which flows passed and which did not.
top
I want to test a WLAN client device. What kinds of tests can I perform using a VeriWave test solution?
VeriWave has a number of predefined tests that are ready to go. The test system is very flexible and allows for thousands of custom tests to be created by the user. Popular tests include Throughput, Roaming, Rate vs. Range, triple play, and real-world tests offered by the WiMix application. See the WaveClient description for more information.
top
I just received my new VeriWave gear. Now what?
- How do I install the WaveBlades into the chassis?
Make sure the power is turned off. Make sure your environment is static free. Insert the management card into slot 0 of the chassis. The management card has the word “Management” on the front panel. To insert the card into the chassis, align the card with the top and bottom notches so the card slides in smoothly. Make sure the card is fully inserted by observing that the black handles have locked into place.
Now insert the rest of your cards into any chassis slot as described above. Any card can go into any chassis slot (with the exception of the management card which must go into slot 0). For more information, please refer to the WT90/WT20 page.
After all cards are inserted, the power may be turned on.
- How do I physically connect to the chassis to manage it?
Locate the port labeled “Ethernet” on the back of the chassis. Connect a PC to this port using an ethernet cable. Ideally, this connection should be a direct connection. However a direct connection is not required.
By default, the VeriWave chassis attempts to get a DHCP address when it boots up. If it does not get a DHCP address, it uses a default static IP address of 192.168.1.1. The DHCP timeout can be as long as 5 minutes, so please be patient.
- How do I view the chassis status?
Open a web browser and input the IP address of the chassis. The default address is 192.168.1.1. You should be presented with the status page of the chassis. After all cards have fully booted, the status of each card should be “running”.
- How do I change the IP address of the chassis?
To change the IP address, go to the status page of the chassis (see above). Now select “Login”. You will be prompted for the password (default password is “VeriWave1”). Now you will be re-directed back to the status page and a new tab will appear that says “Config”. Select the Config tab to have access to the network settings of the chassis.
- How do I get updated software?
To download the latest VeriWave software log into the Customers section of our web site. Your username and password should have been mailed or emailed to the person who sent the purchase order to VeriWave. If you would like this information to be re-sent, you can send a request to support@VeriWave.com. Please indicate the serial number of the chassis for which you need information. The serial number is found on the back of the chassis.
- How do I get my license keys?
Your license keys (as well as your username and password) should have been mailed or emailed to the person who sent the purchase order to VeriWave. If you would like this information to be re-sent, you can send a request to support@VeriWave.com. Please indicate the serial number of the chassis for which you need license keys. The serial number is found on the back of the chassis.
top
How do I upgrade my hardware and software applications to the latest VeriWave release?
Be sure to close down all VeriWave applications before beginning the upgrade. Upgrade your software applications first. The first software application to be installed / upgraded should be WaveTest. The other software applications can be installed in any order. The software applications have standard installers and are very easy to install.
After the software has been installed, the hardware can be upgraded. Open the status page of the chassis (see above for instructions). Now select “Login”. You will be prompted for the password (default password is “VeriWave1”). Now you will be re-directed back to the status page and a new tab will appear that says “Firmware”. Select the Firmware tab. Now select “Browse” and browse to C:\Program Files\VeriWave\upgrades. Select the firmware image with the correct firmware version number. Select Upgrade.
After a few seconds you will be redirected back to the Status page where you can observe the progress of the upgrade. The upgrade can take anywhere from 20-60 minutes. Use the “Refresh” button to get updated reports on the upgrade status. Be sure not to reboot or do anything else that would disturb the progress of the upgrade until it is complete.
top
My AP supports 802.11n. How do I connect the RF interface on the VeriWave WaveBlade to my AP?
For 802.11n, the RF connection must be cabled using RF cables with a standard SMA adaptor. Note that there are 3 RF ports on the WaveBlade. Each port is an antenna interface which is actually a female SMA connector. Start out by fixing a 20db attenuator to each port. Connect an RF cable to each attenuator.
Now connect the other end of each cable to the antenna ports of the AP. Preferably, the AP’s diversity antenna should connect to Port C of the VeriWave blade. If the AP has a dominant antenna, it should be connected to Port A of the VeriWave blade.
When making SMA connections, be sure each connection is matched as male/female. Any mismatch (female/female is a common mistake) can result in erratic results.
If the AP does not have a female SMA connector built in, some adaptor must be used to convert to SMA. There are many APs with many different interfaces. If you are unable to locate an SMA adaptor for your AP, please call or email our support team (support@VeriWave.com) with a description (or picture) of the interface.
top
My AP supports 802.11a/b/g. How do I connect the RF interface on the VeriWave WaveBlade to my AP?
For 802.11a/b/g, the connection can be cabled or open air. If open air, simply connect an antenna to the SMA adaptor on the WaveBlade. Note however that performance results will be impacted by open air testing. Also, test results will be difficult to reproduce.
If cabling, start by connecting a 20db attenuator to the WaveBlade. Connect an RF cable to the attenuator. Connect the other end of the RF cable to the antenna port of the AP. If the AP has 2 antenna ports, disable or terminate the port which is not in use. If possible, disable diversity on the AP.
top
How do I verify if the RF connection between the VeriWave system and the AP under test is good?
To verify the connection is good you must verify the AP transmission power is within the limits of the VeriWave receiver. Also, you must verify the transmission power of the VeriWave clients are within the limits of the AP’s receiver. Verifying the AP transmission power is described first.
To verify the AP’s transmission (TX) power is with the limits of the VeriWave receiver, you must first know what the limits of the receiver are. Currently the limits are different for the 802.11a/b/g port than they are for the 802.11a/b/g/n port. Ideal limits are:
- 802.11a/b/g blade: -25 to -35dBm
- 802.11n blade: -5 to -20dBm
Max and min limits are:
- 802.11a/b/g blade: -20 to -60dBm
- 802.11n blade: 0 to -40dBm
To identify the signal strength of the AP, open WaveDynamix and connect to the chassis. Right click the 802.11 port which is connected to the AP. Select “Scan all channels”. After the scan is complete, go into the properties tab and open the “Visible APs” bullet. Select your AP and continue opening the tree until you discover the RSSI of the AP (note the columns can be resized if you are having difficulty reading the data). Make sure the power of the AP is within the ideal limits described above. Perform the scan multiple times and verify that the power level is consistent. If the power level is bouncing more than 10db in range, you probably have an RF issue or diversity issue. If the power level is outside of the ideal range, you should try adjusting the transmission power of the AP, or the attenuation on the cable. Note however that if you add too much attenuation to the cable you could have issues with the AP’s ability to receive VeriWave transmissions.
To verify the VeriWave transmission power levels are within the limits of the AP, it is best to connect a client and monitor the client from the AP’s management interface. Verify the client’s power level is within specified range for the AP. If this range is unknown you must use trial-and-error to determine which power levels offer the best performance. The ideal application for such trial-and-error is WaveDynamix. Simply connect a client and start a flow for that client. Gradually increase and decrease the power level for that client until you get the performance levels you expect. Be sure to validate upstream and downstream flows separately.
top
How do I connect the Ethernet port on the VeriWave WaveBlade to my network?
This can be as simple as connecting directly into the LAN or WAN port of the AP, or it can connect into a switch or controller which connects to the AP. The Ethernet port represents “the internet”, so connect it into whatever port you want your network to receive Ethernet traffic. Ideally, your test network should be separate from any other network. In particular, you want management traffic (traffic that configures the VeriWave chassis) to be on a separate network.
top
When I start a flow I see an ARP timeout. How do I address this?
ARP timeouts are generally due to a configuration error. The error may be in the VeriWave configuration, or it may be in the configuration of the test network. First, verify the following items with the VeriWave configuration:
- MAC addresses of all clients do not conflict
- IP addresses of all clients do not conflict
- Take a packet capture and verify the ARP is being sent from the source client.
- Verify the ARP is being forwarded to the destination client and that the destination client is sending an ARP response.
- Verify the ARP response is being forwarded to the source client.
Below is a screen shot of a capture that shows a correct ARP and ARP response exchange on a wireless interface:
If the ARP or ARP response is not being forwarded, it is likely that there is a configuration issue in the test network. Here are a couple things to verify:
- Physical connectivity is established from end to end
- There are no MAC address or IP address conflicts on the network with any of the test clients.
- VLAN configuration (if applicable) is correct.
- Verify the IP address of every test client is routable throughout the network. Also verify the netmasks and gateway addresses. Some APs require client’s addresses to be on a certain subnet.
If none of the above has fixed the problem, feel free to send the results directory with captures to support@VeriWave.com. Also, sometimes a quick work-around is simply to enable DHCP for the test clients.
top
How do I configure a quick Throughput test using WaveApps?
Assuming the hardware is connected to your network correctly, let’s walk through the steps needed to start a quick throughput test.
- Open WaveApps
- Select Throughput
- Select Setup
- Select Ports
- Input the IP address of the VeriWave chassis and select Connect
- Click the check boxes next to the VeriWave ports that you wish to reserve for the test.
- For each wireless port:
- Select the wireless port
- Set the Band and Channel that that port should use
- Verify the AP’s SSID appears after the scan is complete.
- Select Clients
- A client group is already created by default in row 1. Let’s make this the ethernet group. Double click the Group_1 tag and change the name of this group to “eth-group1”.
- In the “Interface” column, single click the box and select “802.3 Ethernet” from the drop down.
- In the “Port” column, make sure the port is correctly set to the ethernet port you reserved.
- Now let’s create a wireless group. Select the “+” button to add a client group. Another client group will appear in row 2.
- Double click the Group_2 tag and change the name of this group to “11g_group1”.
- In the “Interface” column, single click the box and select “802.11 a/b/g”.
- In the “Port” column, make sure the port is correctly set to the wireless port you reserved.
- In the SSID column, make sure the SSID is set to the SSID you intend to use for the test.
- While row 2 is selected, select the Security tab.
- Set the security method to the security method that matches what is configured on the AP.
- In the Advanced Tab, make sure 802.11g is selected.
- Now select the Test Tab.
- Select the Test Setup tab.
- Change the trial duration to 5 seconds (quick test).
- Select the Throughput Properties tab.
- Select the “Custom” bullet in the Frame Size list box.
- Enter a single frame size to test (i.e. 64)
- Select the Mapping tab.
- In the drop down, select Wireless to Ethernet. Then click Apply.
- Now select the green button to start the test. Console logs will appear as the test runs. At the end of the test, a PDF report will appear with the summary of results. The console logs will print a message at the end that points to the directory where detailed results can be found.
top
How do I configure a quick WaveQoE / WiMix test that emulates an enterprise environment across my WLAN?
Assuming the hardware is connected to your network correctly, let’s walk through the steps needed to start a quick WaveQoE / WiMix test.
- Open WaveQoE / WiMix
- Select Enterprise
- Select the Ports tab.
- Input the IP address of the VeriWave chassis and select Connect
- Click the check boxes next to the VeriWave ports that you wish to reserve for the test.
- For each wireless port:
- Select the wireless port
- Set the Channel that that port should use
- Verify the AP’s SSID appears after the scan is complete.
- Select the Clients tab.
- Three client groups are defined by default. Select row 1 and do the following:
- In the SSID column, single click the drop down and select the SSID you wish the client group to connect to.
- Select the Security Tab and choose the security type that matches the security defined on the AP.
- Select the 802.11 Client Options tab and set the client type to 802.11ag.
- Set the DHCP drop down to enable or disable DHCP as desired (disable to speed things up).
-
Repeat the above step for each client group defined.
- Select the Servers tab.
- Select the first server in the list. Select whether the IP address of the server should be DHCP or fixed. Repeat for each server in the list.
- Select the Test tab.
- Set the trial duration to 5 seconds (quick test).
- Now select the green button to start the test. Console logs will appear as the test runs. At the end of the test, a PDF report will appear with the summary of results. The console logs will print a message at the end that points to the directory where detailed results can be found.
top
How do I interpret the 802.11n block ack to understand which MPDUs are being ACK’d and which ones are not?
The 802.11n block ack is used to acknowledge up to 64 data frames. Frames are acknowledged based on their 802.11 sequence number. The block ack contains two important elements: the “Starting Sequence Number” and the “Block Ack Bitmask”.
The Starting Sequence Number refers to the first data frame being acknowledged by the block ack.
The Block Ack Bitmask is a 64 bit field. Each bit refers to a sequence number which is offset from the Starting Sequence Number. If the bit is 1, the referenced frame is acknowledged. If the bit is 0, the referenced frame is not acknowledged and should be retried. Here is an example of an AMPDU with a block ack response.
Inspect the block ack packet. In this case we see that the starting sequence number is 0. The block ack bitmask is displayed in hex as FF 03 00 00 00 00 00 00.
The first octet of the Block Ack bitmask corresponds to the first 8 frames, but read right to left. The second octet corresponds to the second 8 frames, also read right to left, and so on. The table below gives some examples. For these examples, assume the starting sequence number is zero:
- 01 00 00 00 00 00 00 00 – frame 0 was ACKed
- 02 00 00 00 00 00 00 00 - frame 1 was ACKed, but only that frame
- 03 00 00 00 00 00 00 00 - frames 0 and1 were ACKed
- ff 00 00 00 00 00 00 00 - frames 0-7 were ACKed
- ff 03 00 00 00 00 00 00 - frames 0-9 were ACKed.
- ff f0 00 00 00 00 00 00 – frames 0-7 and 12-15 were ACKed.
- 01 00 ff 00 00 00 00 00 - frame 0, and frames 16-23 were ACKed.
- 01 00 00 00 00 00 00 80 - frame 0 and frame 63 were ACKed.
So, you can see that if a frame is not acknowledged, the sequence numbers can keep moving forward while the sending station keeps retrying that frame. However, when the span between the sequence number of the next frame to be sent and the retry frame gets to be 64, the sending unit has to decide what to do. It can stop aggregating while it keeps retrying the old frame, or it can simply drop the frame.
top
