3.3.a Describe Layer 1 concepts, such as RF power, RSSI, SNR, interference noise, band and channels, and wireless client devices capabilities
RF Power
For an RF signal to be transmitted, propagated through free space, received, and understood with any certainty, it must be sent with enough strength or energy to make the journey. This strength can be measured as the amplitude, or the height from the top peak to the bottom peak of the signal’s waveform, as shown.
The strength of an RF signal is usually measured by it’s power, in watts(W). A wireless LAN transmitter usually has a signal strength between 0.1W (100mW) and 0.001W (1mW).When power is measured in watts or milliwatts, it is considered to be an absolute power measurement.
Power and Receiver Sensitivity
Many people want to know how far wireless signals will go. Knowing this is important for planning a network, as the power of the routers will affect the design of the network, and how much equipment is needed.
Different Wi-Fi routers can have very different power levels. Some are much stronger: they have more speaking or transmitting power than others. Some are very good listeners: they have what is called a better receive sensitivity. These two elements define how well wireless devices will connect, and how far away a receiving Wi-Fi router can be.
Manufacturers do not usually publish information about their router’s transmit power or receive sensitivity. Instead, the manufacturer will give a generic “range” rating to their routers, usually relative to each other. In some cases, usually with more business or professional oriented equipment you can find the information for transmit power and receive sensitivity.
COMPARING POWER LEVELS BTWEEN TRANSMITTES(dB)
A router’s transmit power can be measured with two scales — milliwatts (mW) or dBm:
Sometimes you might need to compare the power level between two different transmitters as we see in this example. T1 is transmitting at 1mW, while T2 at 10mW and T3 at 100mW. Simple subtraction tells us that T2 is 9mW stronger than T1 and simple division tells us that T2 is 10 times stronger than T1. Same goes for T3 and T2, where T3 is 90mW stronger than T2 in subtraction but 10 times stronger than T2 in division.
So which are we going to use???
There are therefore, three(3) important dB laws that we can consider in making a mental comparison of power-level using db;
- Law of Zero – A value of 0 dB means that the two absolute power values are equal.
- Law of 3s – A value of 3db or +3dB means the power value of the interest is double the reference value. Meaning, when a power level doubles, it increases by +3dB and when it is cut in half it decreases by -3dB.
- Law of 10s –A value of 10 dB means the power value of interest is 10 times the reference value and power value of -10dB means the value of interest is 1/10 of the reference.
RULE OF THUMB
Whenever absolute power values multiply the dB value is positive and can be added. When absolute power values divide, the dB value is negative and can be subtracted. See table below for better understanding….
Examples to compare 2 power values using dB
Anywhere you can multiply to get the next value, replace the multiplication with +3 and anywhere you can divide to get the value, replace with -3, lets go!!!!
- Source A,B,C has power values of 4,8,16 respectively;
- Source B is double A OR B is 2xA, so, B=A+3dB,
- Source C is double B OR C is 2xB, so, C=B+3dB,
- ALSO
- To get from A to C
- Source C is Ax2x2, so, C=+3dB+3dB =6dB greater than A
- Source D,E has power values of 5mW, 200mW respectively;
- Source E is 5×2 =10, 10×2 =20, 20×10 =200
- therefore, Dx2x2x10 =5x2x2x10 =200
- E=D+3+3+10 (Remember the tale above)
- Answer =E is 16dB greater than D
Power Comparison Q&A
ANSWER is (A) = means 0dB. Remember, when you compare two absolute power levels, there difference is zero db. Check the power change table
ANSWER is (C) 3dB, remember the table, when you double it is +3dB
Explanation:
100mW = 10 x 10 (from the rule of 10, multiplication is =+10, while division = -10)
Therefore, 100mW = 10 + 10 = 20dBm
and
40mW = 10 x 2 x 2 = 40mW
40mW = 10 + 3 + 3 = 16dBm
Now how can you get 16dBm from 20dBm, just simply add -4dB (note that when you measure a difference we use term dB to indicate it is a relative value)
20dBm + (-4dB) =16dBm
COMPARING POWER AGAINST A REFERENCE dBm
A Network Engineer must be concerned about the RF signal propagating from a transmitter to a receiver.
(1) Calculate Net Loss
When there is a transmit power, it looses strength before reaching the receiver and the loss is called the net loss.
NOTE: To calculate that Net loss, it will be;
- receiver /transmit power (Rx /Tx)
- 0.000031623/100mW
- -65dB
(2) Calculate Received Signal
In Wireless Networks, the reference power level is usually (1mW) and the unit is (dBm). Here, the dBm values of the transmitter is added to the net loss to get the received signal in dBm .
NOTE: To calculate the received signal, it will be; (20dBm -65db )= -45 dBm
(3) Antenna -Calculate EIRP (Effective Isotropic Radiated Power)
The actual power that will be radiated from the antenna is called (EIRP); the resulting signal power level measured in dBm is the combination of the transmitter, cable and an antenna. This is regulated by government agencies and as such, a system cannot radiate higher than a maximum allowable EIRP
To calculate EIRP of a system, add the transmitter power level and the antenna gain and subtract the cable loss;
- Transmitter power level is 20dBm
- Cable loss of 10dB
- Antenna gain 5dBi
- EIRP =20dBm – 10dB + 5dBi
- Result = 15dBm
- Note: Even though the units are different, you can safely combine them for the sake of calculating the EIRP
(4) Calculate Link Budget
The cumulative sum of gains and losses measured in dB over a RF signal path to make sure that the transmitted signal has sufficient power for it to effectively reach and be understood by the receiver. See formula in the diagram….
Rx =20dBm – 2dB + 4dBi – 69dB + 4dBi – 2dB
Rx= -45
Free Space Path Loss
RF signals transmitted through the antenna propagates through free space as a wave and NOT as a ray or straight line. The wave has a 3 dimensional shape that makes it expand as it travels and this expansion also called SPREADING which causes the signal strength to decrease or weaken. Even if there is no obstacle in the path between the transmitter and the receiver, the amplitude of the wave decreases as it travels through free space, this is known as FREE SPACE PATH LOSS.
Spreading therefore is the main cause of free space path loss.
Just be aware that, FSPL is an important part of link budget, along with antenna gain and cable loss.
Effective Range of 2.4GHz and 5GHz Transmitters
This shows the range difference where both transmitters have an equal EIRP. The dashed circles shows where the effective range ends, at the point where the signal strength of each transmitter are equal.
In most indoor locations, wireless clients are usually less than 50 meters away from the access point they are using. Even at 1 meter away, the effect of free space cause a loss of around 46dBm! The free space loss is also greater in the 5GHz band than in the 2.4GHz band. NOTE: To get a feel of the actual range difference between a 2.4GHZ and a 5GHz, a receiver was carried away from the two transmitters until the received signal strength reached -67dBm. On the 2.4GHz, the range was measured to be 140Feet while on the 5GHz, it was measured to be 80Feet.
Understanding power levels at the Receiver (RSSI)
- With Wireless LAN devices, the EIRP levels leaving the transmitter’s antenna ranges from 100mW – 1mW, which corresponds to 20dBm – 0dBm.
- The power levels at the receiver are much less ranging from 1mW – 0mW and the corresponding range of received signal is 0dBm to -100dBm.
- Receivers usually measure signal’s power level according to the “received signal strength indicator (RSSI) scale”. Its an internal 1-byte relative value ranging from 0-255 with 0 being the lowest and 255 being the strongest and this range can vary from one hardware manufacturer to another. There is no useful units and you will likely see RSSI values being measured in dBm after they have been converted and scaled to correlate to actual dBm values.
–So, what received signal strength value is good enough, assuming that the transmitter is sending RF signal with enough power to reach the receiver?
Every receiver has what is called the “sensitivity level” or threshold. Now, as long as a signal is received with a power level that is greater than the sensitivity level, chances are that the data from the signal can be understood correctly. The RSSI value focuses on the expected signal alone, without regards to any other signals that may also be received . All other signals that are received on the same frequency are considered NOISE. Noise is all other signals that are received on the same frequency as the signal you are trying to receive.
The Noise level or average signal strength of noise is called “Noise Floor”. With RF signal, the signal strength must be greater than the noise floor by a decent amount so that the data being received can be received and understood correctly. The difference between the signal and the noise is called SNR (Signal to Noise Ratio) measured in db and a higher SNR the better and preferred.
See this example:
Received signal strength (RSSI) =-54dBm
SNR = Space between RSSI and Noise floor
SNR =(-54dBm – (-90dBm) (Remember the old math: minus x Minus = +)
SNR= -54dBm + 90dBm
SNR =36dB.
BUT THE NOISE LEVEL GRADUALLY INCREASES TO -65dBm FROM -90dBm
SNR =-54dBm + 65dBm
SNR =11 dB (At 11dB, the signal is too close to the noise that the signal might not be useful, hence, the higher the SNR the better and preferred).
EXAMPLE
A receiver picks up an RF signal from a distant transmitter. Which of the following represents the best signal quality received? Example values are given in parenthesis
- Low SNR (10dB), Low RSSI (-75dB)
- High SNR (30dB), Low RSSI (-75dB)
- Low SNR (10dB), High RSSI (-55dB)
- High SNR (30dB), High RSSI (-55dB) Correct Answer. The higher the SNR and lower the RSSI the better for good quality signal.
3.3.b Describe AP modes and Antenna types
(A) ******AP Modes
Cisco APs can operate in one of two modes; Autonomous and Lightweight mode.
(1.) Autonomous Aps
These are standalone Aps without any connection to Wireless controllers, all the intelligence is built into them and they perform more of switching of traffic than routing and the Aps maps each Vlan to a Wlan and BSS. The connection between the Ap and the switches is a trunk link, meaning all allowed Vlans are transmitted to it from the switches via a trunk link. So, in trunk mode, 802.1q encapsulation tags each frame to the Vlan number it came from. The wireless side of an AP permanently trunks 802.11 frames by marking them with the BSSID of the Wlan they belong to.
Benefits
- It is an affordable entry point solution
- No controllers and licensing costs
- Supports the latest Wi-Fi Standards 802.11a/b/g/n for connectivity
- Uses WPA2 Encryption security
- It’s the industry best range -best of RF bread
- It can be upgraded to controller based architecture.
Limitations/Restrictions
- Each AP is managed individually, which is prone to configuration inconsistencies and errors except you use Cisco Prime and if you must install upgrades to the AP software, you do it individually also.
- You need to configure Vlans and SSIDs on all Aps
- It requires enterprise wide Vlans which may not be desirable.
- No central point in the enterprise network for AP management
- Need to configure RF parameters (example, if you are using the 2.4G range, you have to decide the channels for the Aps, channel1 for Ap1, Channel6 for Ap2, Channel11 for Aps and so on)
- Each AP is configured in RADIUS server.
- It has no guest access.
Where to use and when to use Autonomous APs
- Small businesses /small distributed branch offices
- Small warehouses
Summary
For simplicity, 2 Vlans 100, 200 were created in the diagram above.
You could create as many Vlans as your organization requires.
- All links between the distribution switches where routing occurs and the Aps MUST be trunked.
- Aps must be configured with Management IP address so you can remotely manage it for SSID, Vlan and RF configuration.
- Wireless clients connected to the same Ap can communicate through the AP without going through the wired connection and not directly also.
- As a final note for now on Autonomous Aps, pay close attention to data path between client’s as it may be different for other architectures.
(2) Lightweight Aps
Lightweight AP is not a standalone, it MUST be paired with a Wlan controller. By the way, a cisco AP can operate in either autonomous or lightweight mode. So, in this case operating in lightweight mode, it has to join WLC in order to be fully functional. The cooperation between both WLC and LWAP is called SPLIT-MAC ARCHITECTURE.
Unlike autonomous AP, where the link is trunked to the switches, the link for lightweight Aps are access links and as such all Vlan being transmitted by the Wlan to the Aps are all carried over the CAPWAP tunnel between the lightweight Aps and the WLC.
The Split-Mac Architecture has two main functions, one for the WLC and the other for the LWAP.
- WLC- Wireless Controller is where some of the intelligence of the AP is moved to and performs the management functions
- LWAP- Here, we have Lightweight AP whose functions have been divided and only performs real-time 802.11 operations
Management Functions- performed on the WLC
- RF Management
- Client Association and Re-association Roaming Management
- Client Authentication
- Security Management
- QOS
Real-time Functions-performed on the LWAP
- RF Transmit and receive
- Mac-Management
- Encryption of data
Benefits Split Mac-Architecture
- Aps are automatically discovered and configured by the Controller
- Autoconfiguration of firmware
- Centralized management either from WLC GUI or Cisco Prime infrastructure
- Easy to deploy and manage
- Advanced security, can detect and mitigate rogue APs.
- Guest access available
- Consistent configuration across all APs
- Highly customizable
Limitations/Restrictions
- Traffic from end users is forwarded to the WLC
- It’s a single point of failure
Where to use and when to use Split Mac-Architecture
- Campuses
- Hospitals
- Manufacturing plants
- Enterprise where significant customization is needed
Summary
The Wireless LAN Controller (WLC) and Lightweight Access Point (LWAP) use tunneling protocol between them to carry the 802.11 messages and the client data. The WLC and the LWAP can be in different locations and can also be in different IP subnet as well.
CAPWAP
The protocol is called the “Control and Provisioning Wireless Access Point (CAPWAP)
CAPWAP – It is an RFC standard, it encapsulates all the data between the WLC and the LWAP. Focus your attention more on the differences between the architectures for the exam, for further reading RFC5415, 5416,5417, 5418. CAPWAP has 2 tunnels between the Controller and the AP;
- Tunnel for control messages (information containing the WLAN management, configuration and AP management and all messages are encrypted) and
- The other for client data (Packets moving to and from wireless clients associated with a particular AP and the packets are unencrypted by default.)
CAPWAP traffic can either be switched or routed.
Switched –When switched, the AP will connect to the distribution layer switches via a trunk port and other switches too will be connected via trunk port as shown in the autonomous AP connection diagram.
Routed-When routed, a tunnel is formed between the APs and the WLC (as shown above) through the CAPWAP protocol such that the control plane carry the management info and the data plane encapsulates the packets that are sent from/to the wireless client associated with the AP irrespective of the Vlans the clients belong to and is forwarded through the CAPWAP protocol to the WLC which is trunked to the enterprise network and by that, it has access to ALL the Vlans, note that, the AP on which the clients associate is in Vlan 50 using IP 192.168.50.1 and the WLC is using IP 10.10.10.
Types of WLC -LWAP Topologies
(1) Centralized or Unified WLAN topology.
The location of the WLC within the network determine the type of topology. For example, a WLC placed t a central location within the organization like the data center or net the core of the network is called a Centralized or unified WLAN topology. This maximizes the number of AP join to the WLC. The concept of this is that, most of the resources the users need to reach are in a central location like the data center of the organization or the internet. Traffic to and from wireless users travel from APs over the tunnel. It also provides a convenient way to enforce security.
In a large enterprise, you might have thousands of APs in the access layer, with each WLC supporting a maximum number of APs. This means the total number of APs you have associated with a WLC is the total number of CAPWAP tunnel you will have. Each AP MUST have its own tunnel. Each AP has its own Unique management IP address, it connects at the access layer switch with an access link and NOT trunk link and even when multiple VLANS are involved, as we now that VLANs are different subnets, they are all carried over same CAPWAP tunnel to and from the APs. Therefore, the AP will only need ONE single IP address to terminate the tunnel.
Wireless User Mobility: A wireless user moves through the coverage areas of APs and in the process might associate with many APs in the access layer and because all the APs are joined to a single WLC, that WLC can easily maintain the users connectivity to all the areas of the network as her moves around. IF you have more than the maximum number of Aps, then you need to add new WLC with each located centrally. A cisco Unified WLC meant for an enterprise can support up to 6000APs
Traffic in a Unified WLAN topology.
The traffic from one client MUST pass through the AP. The traffic is encapsulated in CAPWAP tunnel and then travel high up into the network to reach the WLC where it is de-encapsulated and examined. Thereafter, the traffic comes down through the tunnel to reach the AP and then back out into the air to the other client as shown with the dotted arrows.
NOTE:
The round trip time (RTT) between the AP and the controller should be less than 100ms so the wireless communication can be maintained in near real time. If there is more latency than the 100ms in the path, the APs may decide that the controller is not responding fast enough and they may disconnect and find another responsive controller.
(2) Embedded WLAN topology.
This is when the WLAN is co-located at the access layer with the access layer switches. This is desirable when the switch platform can also support WLC functions. It is called the embedded Wireless Network topology because the WLC is embedded in the switch hardware.
So, all APs connects to the access switch for network connectivity as well as the slit mac functionality and this makes the CAPWAP tunnel become really short. The tunnel exists only over the length of the cable connecting the AP.
Advantages
–It is cost effective because the same switch platform is what is used for both wired and wireless purposes.
–They typically support up to 200 APs.
–It can be used to solve connectivity problems at branch offices by bringing a fully functional WLC onsite, running on the access layer switches and with this topology, the APs on site can continue to operate without any dependence on WLC at the main site through a WAN connection.
-With a local WLC, wireless devices can reach each other more efficiently instead of travelling to and from through the CAPWAP tunnel where it will be encapsulated and travel through the WAN link to the Main site WLC where it de-encapsulated, processed and return down through the same WAN link.
(3) Mobility Express Topology.
Moving the WLC below the access layer switches and move it into an AP. This is where a fully functional Cisco AP runs software that acts as a WLC. The topology is useful in small scale environment, midsize or multi-site branch locations where the cost of investment in a dedicated WLC will be high. The AP that host the WLC forms a CAPWAP tunnel with the WLC like any other AP would. The topology can support up to 100 APs.
Cisco Access Point (Lightweight AP) Modes
Local Mode
This is the default mode. It offers BSS for wireless client on a specific channel. When it is not servicing any client, it does the following at the background
- Measures noise
- Measures interference
- Discovers rogue devices
- Checks for matches against intruder detection service events
Monitor
An AP in monitor mode doesn’t transmit any frame. It is a sensor that is dedicated to:
- Checks Intrusion Detection System (IDS) events
- Detects rogue APs
- Determines the position of wireless stations
FlexConnect
With this mode, the APs will keep working even when it cannot reach the WLC. If you have an AP in local mode at the branch office and the WLC is at the HQ, it means the AP will have to encapsulate the wireless client data in the CAPWAP tunnel and go through the WAN link to get to the WLC. If that WAN link is down, it means the APs will not be able to operate, but with FlexConnect, it will continue to operate. The AP can locally switch traffic between a VLAN and SSID when the CAPWAP tunnel to the WLC is down.
Sniffer
An AP in sniffer mode dedicates it’s radios to receive 802.11 wireless traffic from other sources like a packet capture device. The AP becomes a remote wireless sniffer; using an application like Wildpackets, Omnipeek or Wireshark, you can connect to it using your PC. This can be useful if you want to troubleshoot a problem and you can’t be on-site.
Rogue Detector
Rogue detector mode makes the AP detect rogue devices full-time. The AP checks for MAC addresses it sees in the air and on the wired network.
Bridge/Mesh
The AP becomes a dedicated point-to-point or point-to-multipoint bridge. Two APs in bridge mode can connect two remote sites. Multiple APs can also form an indoor or outdoor mesh.
Flex plus Bridge
The AP can operate in either FlexConnect or Bridge/Mesh mode. This AP mode combines the two; it allows APs in mesh mode to use FlexConnect capabilities.
SE-Connect
An AP in SE-Connect mode dedicates its radios to spectrum analysis on all Wi-Fi channels. Applications like MetaGeek, Chanalyzer or Cisco Spectrum Expert can be used to connect from you PC. This is useful if you want to remotely discover interference sources that you can’t solve with a sniffer.
Summary
Lightweight Aps are by default in Local mode and in this mode, it provides BSS and allows clients to associate to the wireless LAN. When the mode is changed to any other than Local, the local mode and BSS is disabled.
(B) ********Antenna types
In providing Wireless LAN coverage in a building, indoors, outdoors, park, lobby, classroom, hospital, city streets, whatever the type of space or location, and because of the shapes and sizes of these spaces, it is obvious that one type of antenna cannot meet or fit every application. Antennas therefore come in many shapes and sizes too and each with its own pros and purpose.
Some Antenna Characteristics
- Radiation Pattern -When an alternating current is applied, an RF signal is produced and the electromagnetic waves are radiated EQUALLY IN ALL DIRECTIONS. The energy produced takes the form of ever expanding sphere.
- Antenna Gain -Antennas DO NOT amplify a transmitter’s signal, instead, they amplify the gain. Antenna gain is measured in dBi (decibel-isotropic) and it is a measure of how effectively it can focus the RF energy in a certain direction. The gain of an Antenna in dBi is measured relative to an isotropic antenna. An isotropic antenna DOES NOT EXIST because it is ideal, perfect and impossible to construct. It is shaped like a tiny round point. When an isotropic antenna is compared with itself, the result is a gain of zero (0) dBi. So, think of the antenna gain of zero being a perfect sphere like we see here. If its something you could press into different shapes, then, it means, as the shape is compressed /deformed, it expands into other shapes in other directions. The omnidirectional shape covers a wider area, hence, its antenna gain is lower +4dBi while the directional shape is higher as its built/shaped to cover a more focused area. The only way to find an antenna’s gain is to look at the manufacturer’s specification. The antenna gain is more suited for link budget calculations.
- Beamwith of an antenna- This is what the manufacturer list as the measure of an antenna focus and not the antenna gain.it is listed in degrees as shown in this example. The beam is determined by finding the strongest point which is usually somewhere on the outer circle, next the plot is followed in either direction until the value decreased by 3dB (which means, its been halved ).
4. Polarization
When an alternate current is applied to an antenna, electromagnetic wave is produced. The wave obviously is made of two components; an electrical field and a magnetic field wave. The electrical field will always leave the antenna in a certain orientation. Example; a vertically pointing wire will produce a wave that oscillates up and down in a vertical direction as it travels through free space. Electrical field wave’s orientation is called the antenna polarization. Antennas that produced vertical oscillation are vertically polarized. However, antenna polarization is not of critical importance but the antenna polarization at the transmitter must be matched at the receiver. if there is a mismatch, the received signal can be degraded. CISCO ANTENNAS ARE DESIGNED TO USE VERTICAL POLIRIZATION.
For example, if you mount a transmitter with its antenna pointing upward and while you are away, someone mistakenly knocks it sideways, it automatically changes the radiation pattern and the polarization too. meaning, the polarization does not match for the diagram below and the result can be a degraded signal
Types of Antennas
(1) Omnidirectional ( Example-Dipole)
Diagram 1 Diagram 2
This propagate signal equally in ALL directions and the result of this is a donut like shaped pattern. This type of antenna is well suited for broad coverage of a large room or floor area with the antenna located in the center and because it distributes its RF energy across a wide area, the gain is usually low between +2 to 5dBi. An example of an Omnidirectional antenna is a dipole. Depending on the mounting orientation, some dipole design can be folded up or down and some are fixed.
In order to reduce the size of the Omnidirectional antennas, many cisco APs have integrated the antennas into the case of the device as shown with this Cisco AP. Integrated Omnidirectional antennas usually have a gain of +2dBi in the 2.4GHz band and 5dBi in the 5GHz Band.
Some, wireless LAN adapters used in laptops and mobile devices are small and as a result they often have a antenna gain of 0 or even sometime a minus value and this does not mean that the antennas are not radiating signals, they are, its just that, not as much as the APs.
(2) Directional (Example -Patch)
Directional antennas have more gains because their RF energy is directed towards a particular direction; a typical application is elongated indoor areas, like rooms along long hallways, isles in a Wearhouse . They can be used to cover outdoor areas or spaces along two buildings.
Patch antenna have a flat rectangular shape and can be attached to the ceiling and they produce a broad egg like shape pattern.
They have a typical gain of
6 to 8dBi under the 2.4GHZ band and
7 to 10dBi GHz in the 5GHz band.
Another Example -Dish
Dish use a parabolic dish to focus the signal received unto an antenna mounted at the center. Parabolic shape is important because any wave arriving from the line of sight will be reflected onto the center antenna element that faces the dish. Transmitted waves on the other hand is the opposite, they aimed at the dish and reflected away from the dish along the line of sight. The focused pattern gives the antenna a gain of between 20 – 30dBi, the highest gain of all the wireless LAN antenna.
——————————
3.3.c Describe access point discovery and join process (discovery algorithms, WLC selection process)
AP Discovery and Join Process
From the time and AP powers up until it offers a fully functional BSS (Basic Service Set), it goes through about 8 different states and it goes through in a specific order and these sequence is called STATE MACHINE;
(1.) AP Boot Up: Once powered up, it boots up to load a small image that it will use for communication and gets an IP address either from a DHCP or statically.
(2.) WLC Discovery: The AP goes through series of steps to discover a WLC. See details below.
(3.) CAPWAP Tunnel: The AP attempts to build a CAPWAP tunnel between one or more controllers.
(4.) WLC Join: The AP selects a WLC from a list of candidates and then sends a CAPWAP join request.
(5.) Download Image: The WLC informs the AP of its software release. If the AP’s software is different from that of the WLC, the AP will download, reboot and start from step one, else, no download is done.
(6.) Configuration image: The AP pulls down configuration parameter from the WLC.
(7.) Run State: Once the AP is fully initialized, the WLC places the AP in a run state and they can start providing BSS information and start accepting clients.
(8.) Reset: If an AP is reset by a WLC, the existing client associations and CAPWAP tunnels is removed from the WLC, the AP reboots and starts the process from the begining again.
NOTE ON AP -IMAGE DOWNLOAD
The following are the reasons why an AP will need to download a new image from the WLC;
- If an AP joins a WLC but has a version mismatch.
- If a code upgrade is performed on the WLC itself, requiring all the APs attached the WLC to upgrade too.
- If the WLC fails, causing associated APs to be dropped and join elsewhere.
Note also, if you downloaded a new code release on the WLC but not yet rebooted for it to run the new code, you can also predownloaded the new codes to the controller’s Aps and it will not be loaded until a reboot is performed
WLC Discovery
Ap uses the following methods to discover live candidate controllers that are available;
- Prior knowledge of WLCs
- DHCP and DNS information
- Broadcast on the local subnet.
- An Ap send a unicast CAPWAP discovery request to the local wired subnet and any WLC that exist on the subnet responds with a CAPWAY discovery response.
- if the AP and the Controller are on different subnets, a local router can be configured to relay any broadcast request over UDP 5246 to the specific controller address.
- An AP can be primed with three controllers; a primary, secondary and tertiary controller and these are stored in non-volatile memory, so that the AP can remember them when there is a reboot.
- The DHCP server that supplied the IP address to the AP can also send DHCP option 43 to suggest a list of WLC addresses.
- The Ap attempts to resolve the Cisco-CAPWAP-Controller.localdomain and if he name is resolved to an IP address, the controller attempts to contact a WLC at that address.
- If none of these steps work, the AP will restart and start the discovery process again.
WLC Selection
The discovery process of an AP is a building of a list of Live candidate controllers and then next, it must begin the next phase which is to select one of the controllers and join it. The joining process involves sending a CAPWAP join request and waiting to get a CAPWAP join response and from that point a tunnel is built between the AP and the WLC.
- If an AP has previously joined or has been primed with the primary, secondary or tertiary controllers, it tries to join those in succession.
- If the AP does not know any of the WLC controllers, it tries to discover one and if the controller has been primed as the master controller, then it responds to the APs request.
- The Ap attempts to join the least-loaded WLC in order to load-balance, this is possible because, the controllers report their loads during the discovery phase.
Reasons why an AP is rejected by a WLC
- The platform as its called or license defines the max number of Aps a WLC can support and if it already has the max, the next AP will be rejected.
- APs with priority values configured as critical, high, medium or low also get rejected based on the fact that the controller accommodates as many high-priority Aps as possible. s, the ones with the lowest priority will be rejected to make room for the ones with the highest priority.
3.3.d Describe the main principles and use cases for Layer 2 and Layer 3 roaming
Overview
When client moves around with their wireless devices, they are often ignorant of the wireless network infrastructure making it happen. In an ideal setting, roaming should have a good coverage and be seamless. Cisco has several strategies for roaming;
(1.) Roaming between Autonomous APs
A Wireless client must always associate and authenticate with an AP before it can use the BSS of the AP. Recall that, an autonomous AP has all its intelligence within it, therefore, it keeps a table of all client’s associations with it, while the mac addresses of the client are kept with the switch, while both AP and switch are linked by a wired connection. So in an environment where autonomous APs are deployed, it means, each Ap is configured individually with same SSID.
The wireless client continuously evaluates the quality of the wireless signal and as the client moves away from the AP servicing its request to another location and as the signal degrades, the clients will begin looking for another AP whose signal is better; the client actively scans and send probe request to discover candidate APs and once it finds the right one, it tried to re-associate with it(this process is very quick and simple). Note: Association requests are used to form new association while re-association requests are used to roam from one AP to another.
The Roaming Process:
So, here in diagram 1, the client is associated with AP1 and is moving to AP2. Each autonomous Ap has its own association table and Ap1 has 1 client while Ap2 has none. As the client begin to move into Ap2’s cell coverage area, getting into the border area where the signal degrades, the clients starts to look elsewhere for better and stronger signal and finds one in Ap2, the client decides to roam and associate with Ap2 and both Aps will update their association tables shown in diagram 2 to reflect the move of client-1.
If AP1 has leftover data for client-1, the leftover data is passed to Ap2 over the switched wired network for onward transmission to client-1 who is now associated with AP2 and this accounts for the buffering experienced by the client and the same applies to the client-1 moving through the entire wireless network as shown in the diagram with “O” being the client-1 roaming.
(2.) Intra-controller Roaming
The Roaming Process is similar to that of an autonomous Ap; the only real difference is that the association is handled by the controller and not the Ap like we saw in the autonomous AP.
So, here in diagram 1, lightweight APs are bound to a WLC through the CAPWAP tunnel and the client must still re-associate with a new AP as it roams/move about.
Client-1 is associated with AP1. Here, the controller maintains the detailed database that contain information on how to reach and support the client.
The WLC-1 diagram shows a list of APs, the associated clients and the WLAN being used .
It also contains the MAC and IP addresses, QoS parameters and other information, a lot more than what is displayed above.
Process:
So, when a client-1 starts moving/roaming, not much will change as compared to autonomous Aps except that when it eventually roams into the area of AP2, the controller will update the client’s association from Ap1 to Ap2 and this is known as INTRACONTROLLER ROAMING.
The controller updates the client association table so it knows which CAPWAP tunnel to use to reach a client. When Aps are bound to same controller, the roaming process is simple and efficient. The roaming here takes less than 10ms to complete. Efficient roaming is very important especially if time sensitive applications are being run on the Wireless network like wireless phones, video streaming. Note: when roaming occurs, there could be a brief moment when the client is not fully associated with the AP but as long as the the time is held at a minimum, the user may not even notice. Some other processes that can occur during association;
- DHCP-the client may be programmed to renew DHCP lease on its IP address or request a new IP.
- Client Authentication-The controller may have been configured to use 802.1x method of authentication to authenticate every user/client on the WLAN.
Three (3) Techniques Achieving Efficient Roaming
The client authentication process posses the biggest challenge in achieving this objective of efficient roaming because the interaction between the controller and the RADIUS server take take a considerable amount of time to accomplish. Cisco controllers offer three (3) techniques to minimize the time and efforts spent on key exchange during roams;
- Cisco Centralized Key Management (CCKM): One controller maintains the database of the clients and the keys on behalf of the APs and provides them to other controllers and their Aps as needed during a client’s roam.
- Key Caching: Each client maintains a list of keys used with prior AP associations and presents them as it required when roaming. The destination Ap must be present in the list. The disadvantage of this method it that it is limited to just 8 APs, which means, just 8 Keys.
- 802.11r: This is an amendment to the 802.11 which addresses fast BSS transition. Here, a client can cache a portion of the server key and present to future APs as it roams.
The client MUST have a supplicant or a driver that is compatible with fast roaming for this to work.
(3.) Inter-controller Roaming-(Between Centralized Controllers)
Layer 2 Roam
When a client roams from one Ap to another Ap on a different controller, it is called INTERCONTROLLER ROAMING. The process of the roaming is like the intra-controller roam where the WLC is in charge of keeping the details of the re-association and other details. The difference here is that, both controllers are now involved in the roam and they coordinate the roam.
Before roaming, client-1 is connected to AP1, attached to WLAN Staff bound to VLAN 100 on subnet 192.168.100.0/24 where the client gets its IP address from; when the client-1 roams to another AP on a different controller, it can either continue with its current IP address or work with a DHCP server to renew its lease on the current IP or get another ip address. Controller-2 is also attached to WLAN staff and bound to same VLAN 100 on same subnet 192.168.100.0/24. Therefore, the client-1 roams from Ap1 to AP 2 and stayed on the same VLAN100 and subnet without changing IP address. For this reason, the client is said to have made a layer 2 inter-controller roam. This is also called local-local-roam.
Layer 2 Roaming Process
When client initiate an inter-controller roam, the two controllers compares their VLANIDs and if they are the same as seen in local-local roam, then nothing happens, the roam becomes seamless and the client is said to have undergone a layer 2 inter-controller roam.
Advantages
- Client keeps the same IP address
- It is fast, usually less than 20ms
Layer 3 Roam
When a client roams from one Ap to another Ap on a different controller, it is called INTERCONTROLLER ROAMING. The process of the roaming is like the intra-controller roam where the WLC is in charge of keeping the details of the re-association and other details. The difference here is that, both controllers are now involved in the roam.
Process
When client initiate an inter-controller roam, the two controllers compares their VLANIDs and if they are different, the controller arranges a layer 3 roam (local-to-foreign roam) that will allow the client to keep using its current IP address.
With a layer 3 roam, an extra tunnel is built between the two controllers. The tunnel carries data to and from controllers. The original controller from which the client is roaming from(WLC-1) is called the Anchor-controller (and is determined when a client joins an AP and a controller), while the one it’s roaming to is called the Foreign-controller.
NOTE: Sometimes, you might not want the first controller to be the anchor controller; example, guest client should not be allowed to associate with just any controller and for that reason, you can configure a STATIC ANCHOR CONTROLLER so that, other controllers will direct clients towards it through a layer 3 roaming tunnels. This is useful for guest WLANs
Mobility with Mobility GROUP
- Cisco controllers can be organized into mobility groups to facilitate inter-controller roaming.
- In a mobility group, each controller maintains a mobility list that contain its own mac-address and that of other controllers in the same group.
- Each controller in a group also share a mobility group name.
- If two controller are configured to belong to the same mobility group, client can roam quickly and efficiently both for layer 2 and layer 3 using the 3 techniques mentioned above for achieving efficient roaming.
- if two controller are not listed in each other mobility list, then they are unknown to each other and client will NOT be able to roam between them. Clients will have to associate and authenticate from scratch.
Locating a wireless Client in a Network
Locating a user or a device in a wireless network is an important part of Asset Tracking.
Remember that for a client to use a wireless service (BSS), it must first authenticate and be associated to an AP. Therefore, at the most basic level, you can locate a client /device by the Ap it is associated to. This is a good way to track BUT not granular enough especially APs with very large coverage or for a client device that does not roam aggressively, this means, it (the client device)stays associated with an AP that is now far away and yet there is another AP close by with a better signal.
To Locate a device More Accurately
An AP can more accurately locate a device by using the the Received Signal Strength (RSS) on the client device as a measure of the distance between them. Measuring the distance from a single Ap is more difficult to determine where the client is situated in relation to the AP. The best approach is to measure from at least three AP.
In the case of an AP with an omnidirectional antenna, the client could be located anywhere along the circular path of fixed distance because the RSS would remain fairly consistent at all points along the circle. So, if it’s 25meters away, you will not know which direction of the 25meters you should be looking at since its a circle, it could be any point on the circle. So, this device can be at any point on the circle.
A better solution is obtaining the same measurement from three (3) different Aps in an overlapping manner and determine where they intersect
3.3.e Troubleshoot WLAN configuration and wireless client connectivity issues
In troubleshooting a wireless client, the first course of action is to gather information. You can begin with a broad perspective and then move to specific and direct questions.
For example;
- -when you get a report from many people in the same area, it might be that an AP is misconfigured or malfunctioning.
- -Report from many areas or a single SSID may be a a problem with a controllers configuration.
- -A report from a single wireless user having problems, you might want to spend time on the users device and its interaction with the AP. It will be a bad idea to start troubleshooting the controller
In troubleshooting a single user, you need to think of all the requirement a user needs to meet in order to successfully associate with an AP and use the network.
- Client must be within the RF range of an AP
- Client must authenticate
- Client must request and receive an ip address.
Troubleshooting Clients from WLC
The WLC GUI is an important component in troubleshooting wireless client issues. As wireless clients probes and attempts to associate with an AP, it is indirectly communicating with the WLC. Don’t forget that the AP are connected to the WLC and the WLC coordinates the association of the client with the AP. Most of the details of the clients are actually stored on the WLC. Therefore, with just the mac address of the client, a wealth of information concerning the client can be obtained from the WLC.
There are two (2) GUIs in the WLC
- One for monitoring
- For advanced configurations and monitoring
When you connect to a WLC through the management address from a browser, the default screen shows summary information of the network activities on the right side while the troubleshooting tools are on the left.
To access the advance GUI, click the advance option on the top right side and to search for a client, input the mac address of the client as shown and press enter and the result shown are the divided into tow parts; the left part shows details about the client while the right side shows the applications the client is connected to and the connectivity status.
Checking Client’s Connection Status
The most important information about the client is the connectivity status which is shown as a series of green dots. The clients must go through these series of dots and each state refers to a policy that the client must meet in order to get serviced and if a state move is not successful, then it will be a black dot. A probing client always begins at the start and moves to the layer 2 policy states, then layer 3 policy state as required. These stages in my own term for easy remembrance is called SAADO
- S = Start
- A = Association
- A = Authentication
- D = DHCP
- O = Online
Troubleshooting Connectivity at the AP
In the case where you have multiple users in the same area having issues, then you might want to look at the AP in that area. In troubleshooting the AP, if no client is receiving signals, then you might have a defective radio and might need to go on site to confirm is the transmitter is working correctly.
Otherwise, you can also use the wide array of information collected and monitored on the WLN through the AP view link;