Project details or expliantion

[Matthew Blackstone]

I'm really interested in this project and maybe using or adapting it for a text and image only system. I'm working my way through the git repo but do you have any YouTube videos of you describing the discrete parts of the system and how they interact? I'm also interested in the on air details modulation encoding etc of the 900mhz mesh radio.

[Paul Gardner-Stephen]

The 900MHz radios we currently use are the RFD900+ from rfdesign.com.au, and use GFSK. We are looking at moving to CSS (chirped spread spectrum) for the link budget improvements that it can offer.

We don't have any singular diagram that covers the whole system, however the main software components architecturally are:

1. Mesh Datagram Protocol / Mesh Stream Protocol, our UDP/TCP equivalents, that use public cryptographic keys as network addresses.

2. Serval Rhizomee, our store-and-forward delay-tolerant networking protocol, which is designed for scalable mesh solutions. It is the basis of MeshMS (SMS-like text messaging) and MeshMB (Twitter-like micro-blogging) as well as general file distribution services on the mesh.

Perhaps if you can tell us more about what you are trying to achieve, we can point you in the direction of the most useful information.

Paul.

On 13 September 2017 at 15:12, Matthew Blackstone <matthew.b...@gmail.com> wrote:

I'm really interested in this project and maybe using or adapting it for a text and image only system. I'm working my way through the git repo but do you have any YouTube videos of you describing the discrete parts of the system and how they interact? I'm also interested in the on air details modulation encoding etc of the 900mhz mesh radio.

--
You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-developers+unsub...@googlegroups.com.
To post to this group, send email to serval-project-developers@googlegroups.com.
Visit this group at https://groups.google.com/group/serval-project-developers.
For more options, visit https://groups.google.com/d/optout.

[Matthew Blackstone]

Is there a concern about the much lower bandwidth of a chirp protocol vs GFSK.  I have seen mention of adapting to HF, a NVIS approach would be neat however bandwidth limitations pop up gain.

I've also wondered about how Rhizome communicates between peers, how does it tell when an "artifact/message" needs to be propagated between peers once a link is established, when it should stop being propagated, how the link is initially established, etc. 

I appreciate using the public key as a node address, two birds with one stone.

I would be very appreciative if you just recorded the next time you blackboard this out and posted it to youtube.  

I'm looking for a system to transfer/distribute SMS/MMS type messages (several hundred kilobyte max size, enough for medium resolution photos and pdf's) in a cryptographically secure way while also allowing verification of source(offline key hash verification) and also maintain an ability for broadcast messages across several hops separated in time and space. 

Two example uses: 

1. I am at a large event where wireless data capacity is at saturation. I want to send a message to a friend, they don't need to get it immediately. I send message from app on my phone, app relays to mesh node i have associated with(say its in my backpack).  Node stores message. Node continuously seeks out and establishes links with other nodes. Once link is established message buffer is synchronized between nodes then link is terminated. process repeats as i walk around venue. This process repeats and the message propagates among all nodes that are within range of each other over time until my message arrives at a node my friend is linked to or message expires due to hop count or time since transmission.

2. I am in an area with no wireless data due to lack of infrastructure(remote) or where power is unavailable or equipment is damaged. I need to send a pdf/form/image to a higher level for evaluation/action. I send data from app where it is transferred to the node on my vehicle. Once uploaded the node continuously seeks out and establishes links with other nodes as i carry on moving around and collecting data and completing reports. Once link is established message buffer is synchronized between nodes then link is terminated. as above this carries on as people move about the affected area until my messages get to their destination or time out.

Ideally these functions would be abstracted from the physical transmission, frequency modulation encoding. I have a personal affinity for hoperf and SI Labs. But country to country frequency availability and power limits being what they are i would not want the system tied to one specific baseband solution. I would like these functions to be able to live on a low cost device target sub 100usd running on a c.h.i.p. or raspi zero type platform interfacing with client app/devices over 2.4g wifi.