Something doesn't work

Article can be viewed at http://www.moskaluk.com/something_doesn.htm

Well I have been using dev 88 for a few months now and have notice a few problems. But the worst problem is that the broadband block in Wiana fails after a while.

If you look at node 0-63-Ridgemount and O-27-Newell 200 there is a link between them. Actually both have each other IP address setting in Wiana and everything seemed to be working; however, you can see that connection has linked.

Worst off is that O-27-Newell 200 node stop checking in as a result. I issue a broadunblock command to see if it clear it up and sure enough it did. So I believe that broadunblock command needs some more refinement.

Also the link from S-105-0702-La-R and O-27-Newell 200 has also disappeared again. Some days it works and some days it doesn't. I though it was swaying trees or a six lane roadway but no it comes and goes when it wants too.



In addition a new link has established it self between W-123-1620 La R and O-39-Ridgemount. These links currently have are blocked and should not be linked.

Since these new links occurred I keep getting at O-27-Newell 200 node "new gateway" message every few minutes. Once I though I had a solution but it sure seems like it has a mind of its own.

Bandwidth and Coverage using multiplexer

Article can be view at www.moskaluk.com/multiplexer.htm

Very densely populated areas become a challenge for Wireless Mesh. Not that the coverage is affective but the amount of bandwidth becomes paramount. If you need to put in bandwidth in a given area running multi channels, what could help is Super Aggregation. However, Super Aggregation is still theoretical. There is a practical solution today that one could use off the shelf called Wi-Fi Multiplexer.

As Wi-Fi and devices that support Wi-Fi installations matures new and supporting technology immerges. One of those technology has recently been added to that triple Mesh capacity by phasing three channels onto one antenna. This Multiplexer is not a wireless Mesh solution but more a solution of Wi-Fi.

A Wi-Fi Multiplexer is a device that Injects Bandwidth into a area on 3 different channels, 1 , 6. and 11. They used the same antenna and have a noise filtering system that illuminates any additional wireless overlap that may occur. In essence, it would be adding 3 additional channels.
With the Open Source solution finding practical solution with off the shelf components like this can provide solution that proprietary systems could not. In this scenario one could add 3 MeshAP Gateway nodes and inject in a single high gain antenna. Assuming the person is using 802.11b or greater the amount of bandwidth would triple. 

Cost saving of the antenna and cable would be significant but not really something that you would want to but a Return on Investment calculation on. The ability to put more bandwidth from the same source is the objective. For example if you were injecting 11 mbps into a initial channel and then added a second and third channel you would be providing over 33 mbps; however, not in an aggregate solution but as 3 -11 mbps pipes.

With Super Aggregation a number of test have shown interference between channels with the off the shelve Wi-Fi multiplexer the problem of channel overlap and the noise problem is eliminated. This gives the Mesh 3 channels to choose from in a very small area.  

Population density in a Urban environment can overwhelm mesh technology. Bandwidth solution like a Multiplexer way be a solution for wireless Mesh.

Temperature in the Out Door Unit

The article with picture can be viewed at http://www.moskaluk.com/temperature_in_out_door_unit.htm

Moskaluk MeshAP model MODU-3 is an Plastic Nema enclosure with a ATX power supply and a VIA EPIA motherboard. Temperature is being constantly monitored on the CPU. It seems like there is sufficient heat that is generated to keep the MeshAP running.

Actual temperature - 21 degree C at 6 AM with a wind chill of -40 degree C

ODU temperature indicates above 3 C but less than 4 C.

Two heat sources CPU and Power Supply

Check List for New Build

Articles with pictures can be view at www.moskaluk.com/check_list_for_new_build.htm

As I build out new MeshAP there are a number of items I go through prior to customer installation. One of the biggest problem is to ensure that the hardware that you provide is going to work. One of the worst thing you can do is quickly put together a node and hope it actually works with out testing.

If you are building more than one box you will have to make equal time to ensure that every box works. The time it takes to test a MeshAP is as little as a few hours. However, using name brand PC as Dell, Compaq etc. go through the process to ensure that every component is tested. In most case the quality of a use PC is very good. In almost all my PC that I use for MeshAP no components were required to be swapped out due to failure.

The problem with most of the open source solution is the software is flashed from a CD onto your IDE drive. When making 20 machines the process can be very time consuming!

When you are building your own MeshAP from off the shelf components it takes extra time to determine if your device is actually working properly. In the Eighties a term called "Burn test" was used to verify if the PC was work properly prior to delivery. Similarly every MeshAP that is made should run a burn test.  
Here is my version of a Burn test
Building out a MeshAP is quite simple. Most people are able to construct the basic unit as shown below.

However the problem is before final assembly some testing should be done to verify that all the components are working properly. This may sound stupid with all the off the shelf parts but many parts are faulty. It is better to catch the problem prior to assembly rather than after installation.
First the assumption is that you have assemble the basic pieces and the IDE drive has been flash with the latest version of your software.  

Connect a basic PC connection. If you can connect a standard monitor, mouse, keyboard and internet connection. This is the time to do it. For Wrap boards you may only have a power and internet connection available.
The purpose is to setup your MeshAP as a gateway. If you are starting out for the first time or even if you have a network this basic step is required.  
Before you turn on the unit ensure the unit is grounded properly and that the motherboard is not touching a metallic surface.

Initial Test

- Turn on Monitor and mother board.
- Watch the screen boot up.
- When the initial unit boots up successful it normally goes into GUI mode.
- Register the unit into Wiana.
- Once Registered configure setting in Wiana.
The unit will reboot after the setting have downloaded. This normally takes about 30 minute's.

Testing communication between Wiana and your MeshAP in Gateway node

First test, you are going to run is remotemanagement. This command informs Wiana that you up and running and to exchange any updates it has for you. This command normally happens every 30 minutes.

Next test is a leechtest that takes approximately 10 Meg files and downloads this to your MeshAP. Here you can witness the internet gateway speed.
Most MeshAP either use a hard disk, CF Flash card with IDE adapter or a DOM (Disk on Module or chip). What you are trying to test here is if the IDE drive can upgrade and downgrade.

To test for upgrade type command getandverify tobuild25dev88 (as an example)
You may need to watch the monitor because messages may appear indicating that you must first install another component.

Once install you can verify that you are on that build by looking in Wiana and see what it reports.

Testing communication between Mesh nodes, Gateway and Wiana
If you have Mesh network you way want to determine if the performance of your radio card. The assumption is that you MeshAP is configure properly to your network.

Install the antenna

Remove the RJ45 connection
- Be with in distance of another MeshAP and boot up the MeshAP
- Run reporter. This will show you many things. If everything's running well it will show you that you are meshed with another node.
- Ping the gateway to see connectivity. Ping outside the gateway to a known url. like www.yahoo.com
- Run leechtest to see connection speed.
- Run sigspy to see the noise and antenna strength. To mesh you should have at least 8 differences between signal strength and noise.

Test hardware reliability

Most hardware is reliable however there are times when you turn on your MeshAP and it starts failing after a while of operations.
What you are trying to do is turn on your MeshAP and ensure it mesh or is gateway node. To want it to run through many cycles with Wiana.

You are trying to see continuous operation with out rebooting. The longer you can test the better.

Testing MeshAP is not difficult however it is time consuming. That time you spend testing does help and provides more confidence when you actually are deploying your network. You use the testing of the MeshAP as a benchmark prior to deploying it at a customer site.  

Site Testing

Article with pictures can be view at http://www.moskaluk.com/site_testing.htm

Well I mention in a previous article that I installed two MeshAP in January to take advantage of the mild weather. I also mention that the first installation, shown below, went extremely well. I spoke to soon. Since I had problems with the first installation I went back to test the second.

Below is a picture of the Power Box with JB-Mesh Unit (JB-Mesh is MeshAP in a attaché case.)

The antenna position is located on the side tower and is pointing slightly downward. The reason for this is I started to climb the antenna tower, It is very old and rusty. As I got up to the near the top it start to sway. I continue up and then the tower start to bend. I step down and decided this was a good spot.

The neighbour also has a home built antenna in about the same area. Not good to have to antennae so close to each other. The idea was to link this node with one down the street; unfortunately, that node has too many trees between them. The site survey showed that two other links were in communication range and that the noise ratio in sigspy had a difference of 8. This is sufficient to start to mesh and the node did start to mesh. Had I done a proper survey and gone up higher on the pole I believe I would have been able to get the connection that I wanted.

Sometimes you have to be flexible about the situation. Obviously the right way would have been to remove the pole and place a new tower. Also to ensure that radio reception would have taken in account for budding leaves in the spring and summer.

I have to put another node to ensure that this one stays up during the summer months.

Site Survey Problem

Articles with photo can be view at http://www.moskaluk.com/sitesurvey.htmSo I did the unspeakable. I installed a few Mesh nodes with out doing a site survey. One worked extremely well and the other well, lets say I'm writing about it. The problem is I had a small window of opportunity to install an antenna on my buddies chimney. 
It was January in Toronto and the temperature went up to 12 C that day. Since the snow melted off the roofs I figure it was a good time to install the antenna. And I did. No problem, I had a visual sitting of the antenna that I wanted to mesh with so I went and install it. Since it was a beautiful warm day I went and installed a second antenna. It was days after that I went and turn on the MeshAP.
The temperature change to -12 C and it started to snow. Well winging it only got me into a worse situation. Here is why. First the clear line of site wasn't that clear. Many trees and telephone poles are in the way. Normally this isn't a big deal but I also I notice a building near the Fresnel zone. I try to get back on top of the roof to see but since it snowed, and it was too slippery to chance. So I took out my maps and photographs and will try to illustrate the problem.
The Map shows the distance and 3 areas.  

On The Map 1, 2, 3 are specific locations that corresponds with photos below.
Distance is 1.37 Km. They are 200 mW radio cards on 8 dbi antenna.
Position 1 shows clear line of site

Position 2 also shows clear line of site

Position 3 there is a clear line of site unfortunately there is no picture.
Looking with binoculars from the house towards top of the building, 
the only obstacles are a few deciduous trees (trees with leafs that fall.)
To solve this problem will be working on 3 scenarios
Change the configuration of the antennas to direction with 
15 degrees and 20 db 
Add amplifier on both antennas.
Add an additional hop between these locations.
The Panoramic Shot of the Area Facing West-North to the Airport 
(use horizontal scroll bar to view picture.)
The tip of the blue arrow is pointing to behind some trees and beside 
a large evergreen. Just a couple of notes the arrow should be pointing about 1 cm to the left of the #3.

Mobile Mesh First Flight

Arcticle with Pictures can be seen at Mobile Mesh First Flight.

It's not everyday that some one bends the rules, go where no man or MeshAP has gone before. Yes this is a historic moment that the open source MeshAP has taken flight on January 21, 2005 in a little know place in England called Lincolnshire.

The team at South Witham Broadband project launch in a mini ultralite airplane, the first MeshAP named "condor" that went mobile via wings.

As you can see a black MeshAP is strapped to the back seat of the ultralite with a portable battery unit. The setup consisted of an 8db patch antenna mounted inside the microlite.

Prior to take off a symbolic picture was taken of the Pilot Brian and MeshAP

It has been reported that the Pilot Brian wouldn't answer his VOIP phone. I guess he was too busy to talk with ground control.

The Airborne MeshAP linked with three other nodes on its maiden flight.

Roy Eddleston has the putty outputs, showing in maxispy and reporter, and the Wiana screenshots showing the "condor" and another with the link strengths. It linked with 2x15db Omnis and a 7db Yagi,

The purpose of have mobile mesh is extremely important. Having a MeshAP airborne opens up a whole new way of linking Mesh. But it is the extension of the Mesh Cloud that makes this such a unique achievement.

Most MeshAP are fixed and that the actual coverage area does not grow. With a "winged" MeshAP the coverage grows geometrically and extends the mesh beyond its limits. With multi winged MeshAP the mesh can cover areas where existing infrastructure is limited to time and space. More importantly the Winged Mesh is only one hop from the closes MeshAP. When an application is required to further the Mesh and linked to a ground stations a Winged MeshAP is the ideal solution.

Every time I activate a new MeshAP I get this feeling of success and can just imagine how the team at South Witham Broadband project are feeling about this achievement. Innovation like this may seem novel in today's fast past world. Some times we just have to step back and admire the innovators!

I understand that there was a team made up of Tom, Roy, Brian and Alan. I would like to thank Tom for sending the information and pictures. Further information can be seen at http://groups.yahoo.com/group/south_witham_bb/

Price Drop

I have been searching and looking on the net for Disk on Memory DOM or Disk on Chip prices and found out that there has been a tremendous price drop. The DOM actually replace the CF Flash cards and you don't have to buy and ide adapter and flash card. The DOM are single unit with a connector to the power. There are many places to get these DOM however people always ask where I buy my stuff. Well I try to buy mostly from Canadian Vendors; however, in this case the cheapest ones I found were at
http://www.memory.com/item.asp?UID=2004052317425216&ITEM=CF014

Naming Convention for MeshAP

Article can be viewed at http://www.moskaluk.com/naming_convention_for_meshap.htm

In many of my articles I illustrate my mesh topology. Many have enquired to the funny looking names. Some suggested that I use Greek Gods names or Names of the person that are using the MeshAP. Neither one is satisfactory when you are planning a large installation.

Here is my method.

The topology map show very little of the actual name of the node. I try to
name my nodes based on function and asset number, and identification example:

O-39-Ridgemount 20030911-Compaq-1

O - is the direction of the antenna is facing, in this case
O - Omni Directional. E is for East facing, W is for West facing,
S is for South Facing, N is for North facing and D is for Diversity.
Next the 39 is the location address. It could be also the apartment
number.
Ridgemount is the street name or area where the units are located.
20030911 is date when the unit became a live.
Compaq is the Model
1 is the asset tag.

Recently Wiana receive a new feature call Notes. Most of this information can be put into the note section; however, when you are actual "surfing the mesh" and trying to determine where the signal is coming from, it's good to see it in the splash page. On the bottom of the splash page the node name appears. Many times I think I connect to a node but in reality I connect to another node. I planning to have over 1000 nodes so that having a node identification based on place location was cool.

I just wish the new system navigation system would also be added to the splash page. It would probably help.

Swimming Compaq

Article with pictures can be view at http://www.moskaluk.com/swimming_compaq.htm

Did you ever have one of those installation when what every you did it just didn't work out. I did. I was testing a year ago my slim line Compaq computer. This was a used PC and it was working wonderfully for me. I have a few computers that run in the outdoors but they are protected from rain. They are not protected from the temperature or wind. 

Well with this installation I decided that a inexpensive plastic large container should easily stop the rain from getting in. The unit was located on a balcony of an apartment building. The owner used the container as a drink stand for himself. You know dual role MeshAP and Drink stand. The owner notice one day that the plastic was getting hot so the owner open the container to let the heat out. He left it open for a few days, then weeks. I notice in Wiana that my unit stop working. So I went to investigate it. Lets say it was swimming in a mixture of Coke, Rain Water and something that smelt like Rye. 

I remember a story of a guy who dried out his components and the unit work. So I unplug the thing and took it back to the Lab to dry it out. Well sure enough. It worked - it actually work for a few days. When it came back on line then it started to smell. What and odour! Then finally, it just stopped working. 

I tried to salvage the parts ram, hard drive etc. Nothing but nothing worked. I think it was the Coke.

Ping Inactive MeshAP

See article at http://www.moskaluk.com/ping_inactive_meshap.htm

What is Ping?
Packet Internet Groper or Ping is a network utility used to determine whether a particular computer is currently connected to the Internet. It works by sending a packet to the specified IP address and waiting for a reply.

Ping and Radio Cards
When you Ping and IP address using radio cards the first signal line which is 0 always gives a higher value if the radio card is inactive. The radio card in stand by mode. If the card detects network activity it will wakeup and go to full power.

Seq. line refers to sequence and the time is in milliseconds. This example show the extra time the radio card require to wake up. In this example the first line is a bad time but I have seen it higher that 1000 ms!
64 bytes from 1.x.x.x: icmp_seq=0 ttl=62 time=55.213 ms
64 bytes from 1.x.x.x: icmp_seq=1 ttl=62 time=11.612 ms
64 bytes from 1.x.x.x: icmp_seq=2 ttl=62 time=12.901 ms
64 bytes from 1.x.x.x: icmp_seq=3 ttl=62 time=11.928 ms
64 bytes from 1.x.x.x: icmp_seq=4 ttl=62 time=12.086 ms
64 bytes from 1.x.x.x: icmp_seq=5 ttl=62 time=9.519 ms


Since my network has very few people on it I have to ensure that the radio cards don't go to sleep. I keep them working 24 hours a day by issuing a ping command from the last node to the first node . The other nodes that have more traffic on them do not require this constant wake up call.

As a result of constant pings what actually happens to the reporting? Well if I don't ping inactive MeshAP the sometimes fail to log into Wiana.

So what do I actually ping? From the furthest node out I issue a ping command to the gateway MeshAP.

Ping as a monitor tool
Although Wiana is good monitoring to show topology and any changes to your node, configuration etc. Waiting for 30 minutes to see if your nodes are up is acceptable for most situation; however there are times when you need to monitor the nodes on a continuing basis. There are a number of commercial and shareware version but if you ssh into your node and issue the command pingtest you can see all your nodes that are connected and their ping times. If you issue watch pingtest then you can monitor all your nodes continuously.

Ever evolving Mesh

Article with pictures can be view at http://www.moskaluk.com/ever_evolving_mesh.htm

Two nodes W-123-1620 La R and 0-63-Ridgemount lost there ability to link. Further in my investigation to determine why these to boxes stop meshing I use the command sigspy. The result of using this command showed increase in noise between nodes. Then the noise stopped and the connection came back. The setup of both these nodes are a clean line of sight. The area showed over 30 different wireless channels being used.

The area is densely populated which makes me wonder if all these 802.11 devices can really intermix? I know some pundits had once said that even leaky microwave ovens can make noise. I wonder if it is something more like white noise. A machine like a generator ?

Blue Arrow shows the missing link.

You may wonder why I don't think it is a hardware or software problem. Because in all cases if I take the equipment back to lab environment everything will work.

The link between these two points has been poor. I changed out equipment and antennae many times. To solve the problem I changed the antennae to be directional; although, this improve the quality of link both units; this restricted other client from getting on to the network. The omni direction not only provides a link to other nodes and to the gateway but they also provide connection to the client site. Changing the radio pattern then affects the ability for the client to communicate with these nodes!

I will look into getting a splitter in order to put in directional antenna and omni antenna on the same node. It may help.

Understand your topology.

Origial article can be viewed http://www.moskaluk.com/attempting_to_connect_gateway.htm

January 13, 2005
Since upgrading to Locustworld build 25 dev 88 from dev 83, I have been getting repeated messages of the gateway issue command to all repeater nodes. Here is the kicker, I only have one gateway.

I though it was a Wiana setting problem, you set up a prefer gateway or secondary gateway, etc. It doesn't matter what the Wiana setting are, it still gives the same problem. What actually happens is the network stops performing for that few seconds then it starts finding the new gateway again. The performance would be considered to be poor.

To illustrate the problem I "Ping" from the 4th hop to the gateway node. I continue to get a root message that interrupts pings and stop network traffic as seen below:

64 bytes from 1.XXX.XXX.XXX: icmp_seq=26809 ttl=62 time=12.549 ms

Broadcast Message from root@meshbox
(somewhere) at 21:45 ...

Attempting to connect gateway 1.XXX.XXX.XXX


64 bytes from 1.XXX.XXX.XXX: icmp_seq=26810 ttl=62 time=18.269 ms
64 bytes from 1.XXX.XXX.XXX: icmp_seq=26811 ttl=62 time=12.614 ms

Broadcast Message from root@meshbox
(somewhere) at 21:45 ...

New gateway found at 172.XXX.XXX..1


64 bytes from 1.XXX.XXX.XXX: icmp_seq=26812 ttl=62 time=26.199 ms

With help of many people I found that the problems were neither software or hardware related, but more to the point of how my mesh was deployed. To illustrate this, lets examine the MeshAP that are having the problem and look at there relationship to other nodes. I will use my existing an older topology map to illustrate the problem of node setup.

O-39-Ridgemount is the Gateway Node.
E-123-713-La-Ro and O-63-Ridgemount are meshed in a triangle with the Gateway node (O-39-Ridgemount.) A meshed node is a MeshAP that has multiple paths to a gateway node.
S-105-0702-La-R is 2 hops from the gateway. It is not Mesh properly and it is known as a Dog Leg. Dog Legs are defined as single node, that is not meshed, either from a gateway node or meshed node.
Likewise W-123-1620 La R and O-27-Newell 200 are "Linear route" and are 3 and 4 hops away from the Gateway node. Linear route can be define as two or more nodes that are linked off either a mesh node or gateway node and only have a single pathway to a mesh node or gateway.

Dog Legs and Linear are not considered to be Meshed nodes but are considered to routed nodes. My assumption is Locustworld software can tolerate Dog Legs and Linear routes; however, up to a point. Determining the point is not a magic number like "3" hopes but more of how far the nodes are from the actually gateway and other meshed nodes.

As an example S-105-0702-La-R is a dog leg that is 2 hops from the gateway node but only 1 hop from a meshed node. Currently this dog leg node show no sign of the "Attempting to connect to new gateway" message. But it once did.

S-105-0702-La-R node try to mesh itself with the gateway node; however, the location of this node is physically on the other side of a large building to the gateway node. The signal was being reflect off another building. The reflection signal is as strong as it's closes neighbour. As the MeshAP tried to connect to the gateway it would stop and fail and revert back to the closest node. This back and forth connection was creating problems.

The right solution to this problem was to setup a minimum signal for the node, in other words the node would only mesh with a stronger signal. The problem occurred is both the reflected signal and the closet node had the same signal strength. The difference was that the noise or the gateway node had poor signal quality. Please note that all MeshAP do not have the same strength radio cards. The gateway node has 200 mW radio card while the dog leg and the first hop node have regular 32mW radio card. This led to a more complex solution.

The complex solution was to:

- issue a blocknode of the gateway node at the dog leg node and

- also issue a blocknode of the dog leg node at the gateway node.

This create a new problem. In Wiana there is a setting to unblock nodes. It can be set from 10 minutes onward. What happens is a the interval of time a unblocknode command is issued and everything that was blocked would release and process would again start up.

To resolve the unblocknode problem a new blocknode Wiana setting was introduced to permanently block these nodes. This solution solve the unblocknode problem; however, the actual syntax of the setting is complex. similar the manual blocknode command, in Wiana, both nodes that required to be block must have each other's IP address.

This created the dog leg topology. The node is stable. The signal from the gateway node to the first mesh node then to the gateway node is clear and is noise free; however, as a result it is not self healing, in the event of a node failure, it is not meshed.

In my Linear route I continue to have problems. Similar to the dog leg node the Linear nodes developed as the dog leg due to the reflected and noisy signal problem. What created an additional problem was that I added another node after the dog leg. This create an additional problem known as Mesh Ring. Mesh Ring is define as 4 or more linear nodes that start and end in either a gateway nodes or mesh nodes. They are actually meshed but create a circle route.

As you can see from the above illustration O-27-Newell node and the dog leg S-105-0702-La-R linked together and created a Mesh Ring. The problem of this new link is the effect of change of weather and season.

When the leave fell and the weather is terrible this link appears. It will disappear after a while then reappear again. The problem escalated further when E-123-713-La-Ro started to mesh with O-27-Newell node. Again because of the different strength radio cards blocknode command in Wiana blocknode fixed the problem.

The performance actually improve with a Mesh Ring then after a while it started to become poor. I issued a "sr" command to the gateway from the O-27-Newell node and the result was that ping would get to the gateway in one route then return in another. In theory this should actually be good but in practice, using radio cards, it turn out to be poor.

I would like to say that my nodes are being utilize 100% every minute but the sad part is most of nodes go idle. As a result I sometimes have to send pings from the furthest node to keep nodes up.

As I writing I stop pinging from my 0-27-Newell node. As soon as I get a some noise between nodes I start getting a "Searching for a gateway to use..."message in my console screen. This problem is one that annoys me to no end because I have only one gateway. I see this mostly in linear route where the radio card is not direct line of site with the mesh nodes.

To resolve this problem I tried to setup a directional antenna between the mesh node and the linear route. It didn't work. I setup a mobile mesh using JB-Mesh and notice that the linear route turn into a mesh nodes and the "Attempting to connect to gateway.." only happen a few times after that. Then node stabilized.

What had happen was that the number of hopes was reduce to 2 rather than 4 and performance was excellent. I illustrated the effect of adding two more node (blank ovals) that reduce the problem of linear route.

When adding nodes it important to place them in a triangle as per mesh topology. Dog leg, Linear and Mesh Ring should be avoided.

Linux Radio Cards

I keep getting asked if a particular brand is prism chipset radio cards. I use the following link:

http://www.linux-wlan.org/docs/wlan_adapters.html.gz

This link show many radio cards and supporting Linux drivers. Please ensure that you look for Prism II or Prism 2.5 chipsets.

Meshing between Channel

Article with picture can be view at http://www.moskaluk.com/meshing_between_channel.htm

My Lab results show I have a small mesh called MMN and have 13 nodes 8 are shown. Prior to deploying my mesh nodes I test them out and various scenario to ensure that everything is working. They use to call it "Burn Test" in the 80's.
I setup a channel 1 and channel 8. Nodes O-lab 20040601- and D-Experimental are on dev 94.

Wiana reporter that Channel 1 is Meshing with Channel 8. I've checked reporter on O-Lab and sure enough I can see channel 8 on D-Experimental . What I think is happening is either the topology graph is screwing up or something has been implemented that has made the units gone astray.

I did not investigate it any further. If you can duplicate this scenario please let me know.
I will reset the lab back to dev 88.

Power Box Recharging

This is a continuation on the Power Box. Since all the alarms went off on the power box I unplug the MeshAP and let it recharge. The thing took over 36 hours to recharge on a regular 110 outlet. The small adapter isn't big or powerful. If I plug the power box into the truck and ran the truck for 4 hours the unit would be totally recharged. I guess I have to look at the specification of the Trucks alternator to find out why it would take less time than the house outlet.

I hate to buy some solar panel and find that they were too small to recharge the battery.

Just keeping you updated!

Mobile JB-Mesh

Article with pictures can be viewed at http://www.moskaluk.com/mobile_jbmesh.htm

When you look at today network appliances i.e. cellular phones, palm devices, MP3 players et.al. you can think how small, well designed and portable they are. Most of them can fit into a shirt pocket.
Well here is my version, it is a MeshAP in a suitcase with a power box. It sure not small, it isn't elegant and it has a few pounds to it.

My wife bought me a hard cover attaché. Unfortunately it has a brand name of JB over it. My initials are DAM so you can see the dilemma I'm in. I don't want to hurt her feeling so I had to find a use for this case. I got the idea from mini-itx.com website. You know those guys put mini-itx into everything.

I call the device JB-Mesh. When I go to do a test site I bring along the JB with me. The JB-Mesh contains 8 dbi omni directional antenna, with 100 mW radio card running Locustworld dev 88. No key board, no mouse, no portable monitor and it runs as a repeater.

So far I have taken this up the street to see if it will connect with my other MeshAP. Sure it works. I guess the thing to do is buy a portable flat panel monitor and bring a keyboard and mouse.

Power Box and MeshAP

The alarms on the power box started to go off at the 9 hour and 52 minutes mark. Yes I'm disappointed because my researched indicated that MeshAP would be running continous for a couple of days. I will need to reduce the power supply to Morex 60W and try it again. Unfortunately all my Morex are in the field being used!

I believe the 200 mW Radio card and the 250 Watt power supply drain most of the battery. I also backdated the software to dev 88 and ran a lot of traffic through the MeshAP. The software for most of the day was running without GUI then after rebooting a few times the GUI came on.

If I had a few solar panel they would required the battery to rapidly charge while running the MeshAP. My assumption is also that the battery many need to be replaced; however, it is only eight months old and has been utilize a few times.

The power box is recharging while powering the MeshAP. The lenght of time it takes to charge may will increase.

While writing this blog the alarm on the battery is indicating a problem. The battery charger cannot provide enough electricity to charge the battery while running the MeshAP.

I will repeat the test again tomorrow.

Portable Power Supply

Article with Pictures can be seen at http://www.moskaluk.com/portable_power_supply.htm

The number one problem that keeps requiring in wireless mesh is not the software or hardware but the continuous flow of electricity. Once again this problem keeps coming up and once again I keep going back to auxiliary power supplies. There are a number of commercial applications that take solar power and provide continuous amount of electricity; however, many of these commercial applications are being mass produce and the cost are dropping.

My wife purchase a portable power box that can be recharged from either a house outlet, automobile or solar. The reason my wife bought it was that the dogs refuse to pull the sled and we had to charge up the snow machines (Yeah that was suppose to be Canadian humour, eh?) They have enough amps to start a vehicle with a dead battery and they have regular North American power outlets. This particular unit can run an appliance while it is being charged. It automatically has a resister built in that would prevent a solar panel from draining the battery when not charging. Basically ready to go unit!

I'm about to test and see how long a MeshAP with regular ATX power supply will run with out a recharge. The specifications indicate it should go for a few days. If I used the Morex 60W power supply it may go for a week.

I will be talking to a local guy about setting up off the shelf solar panel to recharge the power box. If this works out then I have to get a bigger Nema enclosure to house both hardware and power box. My vision is that I will have an antenna assembly with two solar cells flanking either side of the enclosure with the antenna sticking out of the box. An all-in-one combination. It should look a bit like a satellite.

If any of you have a spare solar panels (2 amps) I could use some.

More Savings on Out Door Enclosure.

First I like quality products and when I mention about the Out Door Enclosure- ODE it was based on some research; however, I believe there are less expensive ODE. First I do not endorse this company or have I ordered any items from this company but the price sure looks good for ODE.

http://www.electro-comm.com/Downloads/ecommwireless.pdf

is a copy of their catalogue. I hope this helps people out.

Outdoor Enclosure Deal

I've been overwhelmed with emails regarding the Outdoor Enclosure. Many of you expressed concern regarding a single contract that everyone one could share. I also received many emails on the size and functionality of the Nema box. Unfortunately I tried to get the best deal but many of you would prefer to contact the sales agent yourself. The contact information:

Ken Mills
Ferraz Shawmut
416 253 8534
Ken.Mills@ferrazshawmut.com

Contact him by email I understand you can also pay him buy credit card.

Please note that the full line of enclosures can be viewed at:

http://enclosures.fintix.com/aria.html

Now saying all this I am setting up an account with Ferrax Shawmut to provide me with volume discount. For those of you interested please contact me at:

don@moskaluk.com

Starting Out

Well starting out 2003 with Open Source Mesh Network was an experience. I built some home grown mesh boxes. There are Compaq computers with Locustworld software, prism II 802.11b radio cards with some Tomato Sauce can antennas.

Article can be view on my website at http://www.moskaluk.com/starting_out.htm

Shown is my Gateway server a Mid Tower Compaq Presario and repeater node, a Compaq slim line computer. They continue to operate today however the radio cards are now 200 mW and they use commercial 8 dbi antenna.

First antenna a Primo Tomato and Basil Sauce with "Duct" horn extension.

I remember the cost of a website that indicated it will only take less than 5 US dollars worth of parts to make a great antenna. Funny it cost me over $45 dollars. I guess you have buy the parts in volume?

The final product spray painted White and place a logo of a Toucan pun for two cans. Get it?

In the early days I thought directional antenna was the way to go but later I found that the omni direction is ideal for Wireless Mesh in the 802.11B standard.

To build a omni directional antenna required a little more effort and the cost started to go up. Luckily I found a local manufacturer www.superpass.com . It seems that he provided most of the retailers on the net. What I found is I pay less for an antenna than my Toucan.

Looking back I just have to laugh what I did to save a buck. Today I take "off-the-shelve" parts and assemble the units together. It doesn't take a lot of rocket science to figure this out but it sure helps when you have the right parts!

Merry Christmas

To all the celebrate Christmas on January 7, Merry Christmas!

Best wishes to you and your family; Health, Happiness and Prosperity!

Outdoor Enclosure and Power supply

I been looking for an expensive outdoor enclosure for the last year now. There are a lot available but what I found them to be is extremely expensive, especially when buying from a distributor. I went to my favorite antenna manufacturer in Waterloo Ontario and he sold me the Scepter enclosures. The are fiberglass or plastic type material and are working very well, but the problem is to install the motherboard and power supply I need to put feet on the motherboard and glue it to the back of the enclosure, as shown below.

Article with picture can be view here.

My new ODU Enclosure are manufactured down the street (I really embarrassed about this because I was searching around the world) and the sales guy came to my house and drop off the units with in 12 hours. What service! But in my defense the name of the company and the listing in the yellow pages did not correspond to these types of boxes. They were billed under switch and fuse assemblers.

These new units are less expensive and have a non metallic backing plate. I will be building two units. Both units will have 200 mW radio cards.

I use regular ATX power supply for two reason.

-They are inexpensive and are able to be obtain quickly.
-They are grounded and the mother board and antenna are grounded.

The Morex power supply is extremely expensive, it is low powered and it is not grounded. It really shouldn't be used for this application. I have a third power supply which is located in my Indoor Units which is a micro power supply that came with slim line case. They are OK as well.

Advertisment on the web

Yeah, a few people are asking why I sold out and put advertisement on my web pages. Well I not getting paid to do this work and I need some source of revenue to continue providing the knowledge and research.

But on the other hand I found that some of these advertisement are really neat. Like the link on Performance and Power Supply, there are 5 advertisements on antennas. One on antenna mounts called www.blueskymast.com . I just thought it was really neat to see there video on how it works. Now it may seem that I am plugging their product, no not really, but I never needed or wanted anything like this. So when I need a portable poll I guess I will abandon my idea of having a pole with 3 tent wires and tent stakes. My idea would be cheaper but there looks really good. I guess if I painted the pole black with yellow rope it would look a little better?

Oh I forgot revenue, Ok everyone check out at least 5 different advertisements.

No you don't have too. Really no. Don't do it. Ok if you must. :)

Update Performance and Deployment with Pictures.

I'm new at this blog and my pictures did not come out in the last article. Here is a link to my website regarding performance and deployment with pictures.

Performance and Deployment

Since I wrote the paper on Wireless Topology, I have received many request to build out mesh networks; but my objective is to build my own network first. As a result I started to build a small wireless mesh in my neighbourhood. Linking friends and family together. The problem that I have is my hardware configuration does not match the physical topology of the landscape. Even though I following my plans out I found that the wrong antenna and the placement of that antenna can create some major performance problem.

I have a node that is sitting 16 floors up using 8 dbi down tilt built into the antenna. The problem is that only some of the signal is getting to the actual antenna that is besides the building. What to do? I recently acquired a 200 mW radio card that has antenna diversity. I place a directional antenna pointing upwards to the node that is on the 16th floor; I've place a second regular 8 dbi antenna above the directional antenna i.e. two different antenna on the same radio card.

I also set Wiana Diversity to have a hi pin TX and low pin RX where the omni was on the RX. Great. The link strengthens but the connection to the gateway sucked. When tested the performance of the node on the 16th floor was similar to that of the node below it. I was able to improve the nodes performance by placing the antenna and radio card in a new position.

I replace the 200mW back with the MA311 and left the node on the 16th floor to have poor performance once again. I took a second solution by adding another node to further away from the building and with in tolerance of the down tilt antenna. Guess what. It works much better. Two solution, two different results.

So Where Am I?

I have been working with Locustworld Open Source Software and developing hardware for the past year. I subscribed to Locustworld email forum where I learned many things about building and deploying mesh; however, there was many items that I had to research. As I researched the items I put together many papers at www.moskaluk.com/papers.htm . With in 10 days of publishing, my website exploded with hits and visits. I gave out permission for others to use my work on their site. And the reason for this is not to make the same mistakes as I did. We are suppose to learn from our mistakes but what was happening is that we could not build on our successes. It's now over six months since I have written my last article; however, it looks like I have at least 2 more articles in me.

One will be Quality of Service or Six sigma operations and next paper will be on Best Practices when deploying Wireless Mesh in a Municipality.

I invite you all to contribute to these papers. If you are interested please send me an email don@moskaluk.com

Hopefully we can get message out so people who stop making the same mistakes over and over and over again.

Welcome to Wireless Mesh Blog

Google reported that "Blogs" where growing rapidly and that millions of people used blogs for information purposes. They also reported that over 60% of the people did not know what "Blogs" where. I was one of the 60%. CNN reported that they were getting Eye witness account using blogs and I thought this would be a good venue for Wireless Mesh. Many people will start using blogs in the future.

My intension is to write about Mesh on daily or weekly basis.

So kick back and enjoy!

Don Moskaluk