Welcome to TP-LINK Tech Support Forum
+ Reply to Thread
Results 1 to 15 of 15
  1. #1
    Administrator
    Join Date
    Mar 2012
    Posts
    171

    Pharos MAXtream introduction

    FAQ: http://www.tp-link.com/en/article/?faqid=694
    Suitable for: CPE510/CPE210/CPE520/CPE220/BS210/BS510

    Pharos MAXtream introduction



    Background

    802.11 WLAN is contention-based, built on CSMA/CD. As network scale increases, competition and collisions among devices will be so enormous that the actual throughput will drop. To mitigate these effects, Pharos series products use MAXtream.



    What is Pharos MAXtream?


    Pharos MAXtream is a proprietary protocol developed on the basis of TDMA by TP-LINK. The MAXtream technology has the following advantages:


    1. Eliminates hidden node collisions & improves channel efficiency


    2. Lower latency, higher throughput, larger network capacity & more stability



    *Notice: Actual throughput depends on the environmental factors such as distances, obstacles, wireless interference, weather, etc.



    MAXtream is ideal for point to multi-point network like the below scenario.




    How does MAXtream Work?


    By dividing the timing of transmission into different time slots, MAXtream allows the Pharos devices to transmit in rapid succession, one after another, each using its own time slot to transmit and receive they own frames, which greatly reduces the chance of collision.





    When to use MAXtream?


    Pharos MAXtream is a proprietary Wi-Fi protocol that is only compatible with TP-LINK’s Pharos series products. If the devices in the wireless network are all Pharos devices, it is recommended to enable MAXtream for better performance and stability.


    Important Notice: Other wireless devices will not be able to connect to a Pharos device with MAXtream enabled.




    How to configure MAXtream on the Pharos CPE?


    This is a typical scenario to use MAXtream:




    Step 1


    On the Access Point, log in the PharOS web UI to select the MAXtream option under the Wireless menu to enable MAXtream, shown as the figure below.




    Step 2


    Configure the station as Client mode. Then the stations will automatically detects whether the AP has enabled MAXtream and then adjusts itself to join the AP accordingly.


    You can also configure the stations as AP Client Router mode (WISP Client mode), but remember to keep Wireless -> Wireless AP Settings -> Wireless Radio disabled.




    Notice for Other Modes:


    For AP Client Router mode (WISP Client mode):


    1. When Wireless -> Wireless AP Settings -> Wireless Radio is disabled, the CPE adjusts itself to join the AP automatically as Client mode does.


    2. When Wireless -> Wireless AP Settings -> Wireless Radio is enabled, the CPE does not support MAXtream and cannot associate with any MAXtream-enabled AP.



    For Repeater mode and Bridge mode:


    Repeater mode and Bridge modedo NOT support MAXtream, when the access point enables MAXtream, the stations in Repeater mode / Bridge mode cannot join the AP.

  2. #2
    Link to a thread http://forum.tp-link.com/showthread....PE-Pharos-5Ghz
    I opened on AP section on Pharos CPE510, I have not posted last review I will,
    So far everything its going on, the 2 links one 7.2 Kms and the other 778 Mts.
    Things I dont like.
    They are not WDS devices, they repeat WDS as long as main ISP router is WDS capable.
    Other is that when you do Spectrum Analysis kills the connection while you do it, competitor does not.
    and everything is done by software not by hardware on this new line of CPEs 510 and 210, competitor does it in hardware.
    as you can read on the picture, recommend you to turn on MAXtream in large scale wireless deployment, so on a PTP that uses only 2 devices its useless.
    Other important tool, competitor implemented, now you can see on web interface, the way your cat 5 cable is doing, if soemthing is wrong with lan cable now you will see it in web interface.

    subir imagen
    Last edited by fotodamaris; 01-24-2015 at 23:49.

  3. #3
    Quote Originally Posted by fotodamaris View Post
    Link to a thread http://forum.tp-link.com/showthread....PE-Pharos-5Ghz

    They are not WDS devices, they repeat WDS as long as main ISP router is WDS capable.
    Other is that when you do Spectrum Analysis kills the connection while you do it, competitor does not.
    and everything is done by software not by hardware on this new line of CPEs 510 and 210, competitor does it in hardware.
    as you can read on the picture, recommend you to turn on MAXtream in large scale wireless deployment, so on a PTP that uses only 2 devices its useless.
    subir imagen
    Hi fotodamaris,

    You can configure the WDS in client/client router/repeater mode. You don't need to configure it in AP/Router mode because it's automatically detected by the device.
    Name:  WDS.jpg
Views: 0
Size:  36.8 KB

    Other competitors like UBNT, also use software Spectrum Analysis and will kill connection while you do it.

    For PTP link, MAXtream is also useful because it will makes the wireless transmission more stable while there is a lot of noise.

  4. #4
    Quote Originally Posted by Madison View Post
    1-You can configure the WDS in client/client router/repeater mode. You don't need to configure it in AP/Router mode because it's automatically detected by the device.
    2-Other competitors like UBNT, also use software Spectrum Analysis and will kill connection while you do it.
    3-For PTP link, MAXtream is also useful because it will makes the wireless transmission more stable while there is a lot of noise.
    1-I never posted you couldn't set WDS, I posted, Tp-Link devices are not WDS they repeat WDS as long as they are connected to a WDS device
    2-update your data, UBNT AC line is not killing the connection any more
    3-no need to on a PTP, new Beta firmware released March 12-2015 here http://forum.tp-link.com/showthread....Hz-to-6100-MHz supports extra frequency spectrum for 5GHz pharos products from 4920-MHz ~ 6100-MHz. It means users can use channels that is seldom used, like driving on a highway alone.
    Last edited by fotodamaris; 03-13-2015 at 14:20.

  5. #5
    Administrator
    Join Date
    Mar 2012
    Posts
    171
    Hi @Madison and @fotodamaris

    Thanks for your interest on this topic, gentlemen.

    I want to give some further explanation on WDS and MAXtream. (I am not saying mine were the authoritative answers, but it is a reliable reference for TP-LINK products. And any corrections and supplement are welcome )

    WDS(wireless distribution system) is a general calling of connecting several wireless stations (APs) together by wireless. But it doesn't define how to realize it in actual application on software level. All the various modes including client/repeater/bridge/bridge with AP and the WDS function on routers are variants of WDS. AP client router mode (WISP client mode) is usually not treated as a pure wds because it has NAT router function.

    Although it doesn't define how to realize wds, the actual AP products(including TP-LINK and other brands) do need to obey some simple principles of how to forward the frames. As a result, 4-addr-WDS and 3-addr-WDS have came out with their own advantages/disadvantages.

    I'll give a brief of 4-addr-WDS and 3-addr-WDS via the following topology, which is a typical topology of WDS network:


    Part1: Standard WDS (4-addr-WDS)

    client1 )))) (((( AP1 )))) (((( AP2 ))) ((( client2

    When we deliver a mail, the sender need to know three person -- the sender himself, the first postman, the recipient. But the postman need to know four person -- the sender, the postman himself, the next postman, the recipients.
    Similarly, WDS communication is just like delivering a mail. Client1 and client2 only need to send frames to the AP they are connecting to, no need to know what happens afterwards. But the AP1 need to handle four addresses -- the AP1 itself, the next AP, the source sender, the destination recipient. So when AP1 receive the frame from client1, it will modify the frame in 4-addr-format and forward it to AP2. AP2 will modify the frame back to 3-addr-format and then forward to client2 so that client2 can recognize the frame. Vice-versa when client2 sends a frame to client1.

    4-addr-format can provide transparency on MAC layer (Layer 2), which means client1 and client2 can see the true MAC address of each other. We regard 4-addr-wds as the standard WDS because it is symmetric. So if an AP( or router) supports 4-addr format when it is being WDS/bridged/repeated/cliented, we call it WDS-capable or 4-addr-capable. So far, all Pharos products and most TP-LINK AP/routers are 4-addr-capable.

    As we can see in the above topology, both AP1 and AP2 need to be 4-addr-capable to make it a standard WDS network. If AP1 is not 4-addr-capable, we won't get a WDS network no matter how we configure AP2 as 3-addr or 4-addr

    Tp-Link devices are not WDS they repeat WDS as long as they are connected to a WDS device
    This is not accurate. In fact, most TP-LINK devices are WDS-capable, but it still depends on the main router.



    Part2: What about 3-addr-wds?


    As we know now, it requires the main router to be 4-addr-capable to make up a standard WDS network. What if my AP1/router1 is NOT wds-capable in the following topology?

    client1 )))) (((( AP1/router1 )))) (((( AP2 ))) ((( client2

    Assuming that AP1/router1 doesn't support 4-addr-wds, which means it cannot handle 4-addr-format frames so that it cannot be WDS/bridged/repeated/cliented by a standard-WDS device.

    To overcome this difficulty, 3-addr-wds is developed. AP2 connects to AP1 in 3-addr-format. When AP2 receives a frame from client2, it uses its own MAC address (or a virtual MAC) to replace AP2's true address in the frame and then forward it to AP1. In this way the frame is still a 3-addr frame so that AP1 can recognize. AP2 acts like a MAC proxy for client2. From the view of Mac layer, AP1 and client1 don't even know client2 exists, they think they are talking to AP2. We call this topology 3-addr-WDS network.

    The advantage of 3-addr-wds is that it doesn't require the main router(AP1) to support 4-addr-WDS, which means it is compatible with most routers/APs. But the disadvantage is that 3-addr-wds is unsymmetrical, the MAC addresses of the clients behind AP2 is hidden from client1 and AP1. Some Layer 2 network protocols and applications like MAC filering/ARP-binding will be affected.

    Appendix:
    -- Universal repeater is 3-addr-wds mode. WDS repeater is 4-addr-wds.
    -- Client/bridge with AP mode is not unified, some products are 4-addr-only, some are 3-addr-only, some support automatically detect. If there is a "WDS option" in the wireless menu, it means the product supports both mode, and users can select as they desire.
    -- So far Pharos products including CPE210/220/510/520, BS210/510 share the same firmware so they will support wds automatic detection and manual selection.

  6. #6
    Administrator
    Join Date
    Mar 2012
    Posts
    171
    As to Pharos MAXtream, although it is ideal for PtMP, it can still help to stabilize the link in PtP scenario.

    The TP_PharOS_1.1.0_150225_Beta makes a big progress by opening extra frequency spectrum in TEST mode. Using 4.9 and 6 GHz will avoid competing with other 5Ghz Wi-Fi devices around you, which will effectively improve the performance and stability. But it is still in a beta stage. And kindly be reminded that users should follow local regulations when using these spectrum.

  7. #7
    Quote Originally Posted by tplink View Post
    @fotodamaris wrote Tp-Link devices are not WDS they repeat WDS as long as they are connected to a WDS device
    @admin answered This is not accurate. In fact, most TP-LINK devices are WDS-capable, but it still depends on the main router.
    My line might not have the right words, but as you answer but it still depends on the main router so if main router is not WDS Tp-Link will not repeat WDS, or I am missing something

  8. #8
    Administrator
    Join Date
    Mar 2012
    Posts
    171
    In the example client1 )))) (((( AP1/router1 )))) (((( AP2 ))) ((( client2
    Only the second AP is 4-addr-capable is not enough. You are right about that if AP1 is not WDS-capable, AP2 cannot repeat WDS.
    I believe you've understood. But I cannot say the AP2 is not WDS for this point. I prefer that AP2 is WDS-capable, but AP1 makes it impossible.

  9. #9
    Quote Originally Posted by tplink View Post
    I believe you've understood. But I cannot say the AP2 is not WDS for this point. I prefer that AP2 is WDS-capable, but AP1 makes it impossible.
    I totally understand your position.
    WDS has been a discussion with Tp-Link Mexico and the rest of the country for centuries, some say it works others that it makes the link slower, bla, bla, bla.
    UBNT USA forum speaks about WDS as, the WDS grail, imagine ethernet cable on your link
    Mexico & South America UBNT users, post in forums WDS makes their link slower.
    So WDS is a whole mystery, 2 dif. opinions, the main issue I see its that people always enable WDS but they dont know if their ISP router supports WDS, the easiest is to set a Tp-link router after ISP router, but in Mexico that wont happen.
    Regards, from Acapulco Mexico.

  10. #10
    Administrator
    Join Date
    Mar 2012
    Posts
    171
    Thank you for the information about the users' experience on WDS function, which is helpful and impressive.

    I believe in the future manufacturers will have more in common in popular and basic functions when designing products so that devices from different brands will become generally compatible.

    Best Regards

  11. #11
    Quote Originally Posted by tplink View Post
    Hi @Madison and @fotodamaris


    I'll give a brief of 4-addr-WDS and 3-addr-WDS via the following topology, which is a typical topology of WDS network:


    Part1: Standard WDS (4-addr-WDS)



    Part2: What about 3-addr-wds?


    As we know now, it requires the main router to be 4-addr-capable to make up a standard WDS network. What if my AP1/router1 is NOT wds-capable in the following topology?

    client1 )))) (((( AP1/router1 )))) (((( AP2 ))) ((( client2

    Assuming that AP1/router1 doesn't support 4-addr-wds, which means it cannot handle 4-addr-format frames so that it cannot be WDS/bridged/repeated/cliented by a standard-WDS device.

    To overcome this difficulty, 3-addr-wds is developed. AP2 connects to AP1 in 3-addr-format. When AP2 receives a frame from client2, it uses its own MAC address (or a virtual MAC) to replace AP2's true address in the frame and then forward it to AP1. In this way the frame is still a 3-addr frame so that AP1 can recognize. AP2 acts like a MAC proxy for client2. From the view of Mac layer, AP1 and client1 don't even know client2 exists, they think they are talking to AP2. We call this topology 3-addr-WDS network.

    The advantage of 3-addr-wds is that it doesn't require the main router(AP1) to support 4-addr-WDS, which means it is compatible with most routers/APs. But the disadvantage is that 3-addr-wds is unsymmetrical, the MAC addresses of the clients behind AP2 is hidden from client1 and AP1. Some Layer 2 network protocols and applications like MAC filering/ARP-binding will be affected.

    Appendix:
    -- Universal repeater is 3-addr-wds mode. WDS repeater is 4-addr-wds.
    -- Client/bridge with AP mode is not unified, some products are 4-addr-only, some are 3-addr-only, some support automatically detect. If there is a "WDS option" in the wireless menu, it means the product supports both mode, and users can select as they desire.
    -- So far Pharos products including CPE210/220/510/520, BS210/510 share the same firmware so they will support wds automatic detection and manual selection.
    Hi, Thanks heaps for your explanation. That really helps alot for me to understand the mechanism behind each mode.

    As you said, Universal repeater is 3-addr-wds mode. And the AP2 will replace the actual MAC of client2 with a virtual MAC.

    My question is, when this client2 roaming back to an area under AP1/router1's coverage, apparently, the MAC will now turn to client2's real MAC. In this case, should the client2 get the new IP address? Or will it be treated as an IP address conflict (as two mac addresses hold the same IP)?

    Thanks & Regards

  12. #12
    Administrator
    Join Date
    Mar 2012
    Posts
    171
    @Jason

    Hi Jason

    Glad to hear it helps. Here is further information for your queries.

    1. When client2 roams from AP2 to AP1, it keep using the original IP address as long as the DHCP lease is not expired. Client2 never knows its MAC has been replaced and it thinks it is always using its own MAC and the same IP address. If the DHCP lease happens to be expired when it roams, client2 should renew its IP address by sending a request to the DHCP server. But there won't be a situation that two hosts with different MAC requesting for the same IP address. The DHCP server handles the request in the regular way and it won't feel anything strange.
    2. Every hosts in LAN will keep updating the neighbor information by constantly using ARP operation. Every time some one attempts to communicate with Client2, it will ask for its MAC address first. Afterwards client2 will answer with its own MAC address. In this way the neighbors always will maintain the current MAC information of the others.

    I currently have W8980 router and WA850RE extender. What I want to achieve is: the wifi client should keep its IP address and all the network sessions won't be interrupted, no matter which wifi AP it connects, even when the client is roaming from one to the other.
    Yes, "the wifi client should keep its IP address" is doable. But that "all the network sessions won't be interrupted" is not guaranteed. You can treat this roaming behavior as a quick switch. If the switch is quick enough, the client's network probably get no interruption. However, the switch operation depends on many factors, including the channel condition, the client and the AP(router)'s conditions. In addition, the range extender should not be too far from the router otherwise there won't be a smooth transitional area for roaming.

  13. #13
    Quote Originally Posted by tplink View Post
    @Jason

    Hi Jason

    Glad to hear it helps. Here is further information for your queries.

    1. When client2 roams from AP2 to AP1, it keep using the original IP address as long as the DHCP lease is not expired. Client2 never knows its MAC has been replaced and it thinks it is always using its own MAC and the same IP address. If the DHCP lease happens to be expired when it roams, client2 should renew its IP address by sending a request to the DHCP server. But there won't be a situation that two hosts with different MAC requesting for the same IP address. The DHCP server handles the request in the regular way and it won't feel anything strange.
    2. Every hosts in LAN will keep updating the neighbor information by constantly using ARP operation. Every time some one attempts to communicate with Client2, it will ask for its MAC address first. Afterwards client2 will answer with its own MAC address. In this way the neighbors always will maintain the current MAC information of the others.



    Yes, "the wifi client should keep its IP address" is doable. But that "all the network sessions won't be interrupted" is not guaranteed. You can treat this roaming behavior as a quick switch. If the switch is quick enough, the client's network probably get no interruption. However, the switch operation depends on many factors, including the channel condition, the client and the AP(router)'s conditions. In addition, the range extender should not be too far from the router otherwise there won't be a smooth transitional area for roaming.
    Hi Tplink,

    Thanks heaps for your clarification.

    As you said, the client2 won't be aware that the mac address has been changed under universal repeater mode. But how about the ARP cache on the router and the other hosts ends. Eg: the router keeps the arp cache for client2 with the old MAC address. If there is no Gratuitous ARP reply sending from the client2, the router will probably encapsulate the packet with the old MAC address.


    BTW, have you tried to setup WDS between 8980 and 850RE? I got a problem with this setup. I turn on the WIFI on my smartphone, when I am closed to the 850RE. My phone couldn't get DHCP IP address assigend from main router. Any clue for this? Of course, I've disabled the DHCP on 850RE end. and also configured WDS on 850RE end only with same SSID, same security setting.

    Thanks so much.

    Best Regards

  14. #14
    Administrator
    Join Date
    Mar 2012
    Posts
    171
    @jasonchen

    In actual communication, GARP and ARP is used very frequently. Once the smartphone roams from the range extender to the router, it will instantly broadcast ARP requests which asking for its own IP address. Others who receive these requests will immediately update the ARP cache for the smartphone. So in fact the ARP cache won't be a problem in daily use.


    As to the configuration problem:

    1. You need to configure range extender mode1(3-addr) on TL-WA850RE. Model TD-W8980 does not support being WDSed in 4-addr format.

    2. Try to re-configure the TL-WA850RE step buy step:
    Put the TL-WA850RE closed to the router, power it on.
    Connect a PC to the TL-WA850RE through an Ethernet cable. Set a static IP address 192.168.0.100 on the PC.
    Log into the web interface via 192.168.0.254
    Run Quick Setup to configure the TL-WA850RE. Disable DHCP.
    After configuration, wait for 2 minutes. You will see the RE light and signal lights become on, which means the TL-WA850RE has successfully connected to the router.
    Then check whether you PC can obtain IP address through DHCP or not. If the PC can renew a IP address from the router, it means the configuration is finished.
    Then relocate the TL-WA850RE to where you want. Power it on and wait until RE light and signal lights become solid on. Connect your phone to the TL-WA850RE, check whether internet is available.

  15. #15
    Quote Originally Posted by tplink View Post
    @jasonchen

    1. You need to configure range extender mode1(3-addr) on TL-WA850RE. Model TD-W8980 does not support being WDSed in 4-addr format.
    Thanks heaps tplink. That explains why WDS between 8980 and 805RE doesn't work properly.
    Will setup universal repeater mode on 805RE then.

    Thanks again.


 

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts

Copyright © 1996-2016 TP-LINK Technologies Co., Ltd. All rights reserved.