SimpliciTI protocol

Syed,

Please see http://e2e.ti.com/support/low_power_rf/f/156/t/17043.aspx on the maximum number of end devices.

I don't think repeaters have settings;  they just repeat what they hear.  Or I am misunderstanding the question?

Hi

Thanks for the reply.

The info link  also does not give exact maximum number of end devices when simpliciTI protocol is used.

For the repeater case, we just want to have maximum number of repeater at each depth level of the network.

For more clear,

please give us the following info on simpliciTI,

MAX_DEPTH ?

MAX_ROUTERS at each depth level ?

MAX_CHILDREN at each depth level ?

Thanking you.

Syed

Syed,

SimpliciTI supports a star or point-to-point topology, so there are no routers or children (please refer to http://focus.ti.com/lit/ml/swru130b/swru130b.pdf for more information on the supported SimpliciTI network topologies).  To get "depth" in a network means that end-points are connected to an access point through repeaters (which re-transmit any message they receive except those who comes from other repeaters or those where the frames hop_count is 0).  The number of repeaters is limited by the maximum hop count.  In the smpl_nwk_config.dat file, these lines control the maximum number of hops:

/* max hop count */
-DMAX_HOPS=3

To keep network traffic to a reasonable volume, you'll want to keep this number as small as possible, but it basically controls what I think you mean by depth.

The maximum number of nodes to any access point is restricted by the amount of RAM at that access point;  the post I referred to earlier gives some details about that.

I hope this clarifies SimpliciTI a little.

The maximum number of end devices supported is somewhat ambiguous in SimpliciTI.  This is because it is dependent upon the network topology and also the number of Link ID"s you need per radio platform.

SimpliciTI communicates through links established during network commissioning processes.  Each link will have two effective addresses associated with it, one for each radio on each end of the link.  These effective addresses are a combination of the physical 32 bit address of the radio and an additional 32 possible port numbers.  The key here is that it is possible for the same port number to be used if the physical addresses are distinct.  This provide a total effective address of 37 bits.  Note: the effective 37 bit address is not necessarily the same as port numbers are allocated dynamically and are dependent upon previously allocated ports.

SimpliciTI does not exhaustively optimize port allocations during link processes.  Instead a trade off between fully optimized allocation and dynamic efficiency is used.  Thus, the radio issuing the link request will always increment the port address for the remote radio.  This is because the radio requesting the link cannot know the address of the radio it is linking to and thus increments the port number before sending it in the link request to guarantee a unique effective address.

The radio listening for the link will know the address of the radio generating the link request so it can determine if the port number needs to be updated.  Because of this, port allocation by the listening radio can be optimized to a greater extent.  Effectively, if the listening radio cannot find a Link ID with the 32 bit physical address of the radio requesting the link along with the current port reference then there is no need to increment the port reference before allocating the port as this will create a viable Link ID with the current port reference.

Basically what this all boils down to is that each radio which initiates a link process (calls SMPL_Link) will use up a port reference regardless of whether it is necessary and each radio which responds to a link request (by calling  SMPL_LinkListen) can hand out the same port reference until a radio with an established link makes another link request.

Finally, there are a maximum of 255 Link ID handles available to any radio so this represents the upper limit of links that can be established.

Technically there is no limit to the number of range extenders (RE), repeaters in your terms, in a network but range extenders in SimpliciTI do not replay any packets received from another range extender.  This is purely to limit network flooding as the SimpliciTI protocol is fairly simple so managing this network flooding is done this way.  What this means to you is there is a maximum number of hops a message can take before reaching its destination.  If there are no access points in a network, only 2 hops maximum are capable since the communication would be ED -> RE -> ED.  Note that two range extenders would not work as the second one would ignore any packets from the first.

If there is an access point in a system, it will work like a range extender and because it is an access point, a range extender will repeat its packets and the access point will repeat any range extender packets.  This allows for a maximum communications path of ED -> RE -> AP -> RE -> ED or 4 hops maximum.

Jim Noxon

Hi

Thanks for the detailed info.

We have one following application scenario where we will have many star network with AP and End devices for each network.

1. Each Star Network will have both normal END devices and Special END devices.

2. These Special END devices should be able to communicate to its AP ( as I learned it is possible using SimpliciTI) and also to the other Special END devices in other Star Network.

Hope this explains the scenario clearly.

Please suggest me a way to code for this scenario in the application.

FYI... I am already able to do a toggle between 2 AP's from an END device point of view, means that the END device can choose an specific AP to communicate by setting the relevant JOIN and LINK token using "SMPL_Ioctl" API.  Ref: http://e2e.ti.com/support/low_power_rf/f/155/p/113558/403723.aspx#403723

Thanking You

Syed

All SimpliciTI networks are fundamentally peer to peer underneath.  This means that once you have a link token, you can make a connection with any radio which will accept that link token.  Therefore, if you can swap between two access points by changing the Join and Link tokens, then after you swap them merely create a link between the two special end devices to form a bridge between the two networks.

Jim Noxon