Git Product home page Git Product logo

lbard's Introduction

LBARD: Low-Bandwidth Asynchronous Rhizome Demonstrator

Serval Project, March 2016

Overview

LBARD is a prototype content transport protocol designed for a network of two or more Serval nodes connected by a low-bandwidth data link such as a UHF packet radio or SMS. LBARD efficiently synchronises Rhizome content, such as MeshMS text messages, between the nodes.

Rhizome's standard synchronisation protocol was not designed for scarce bandwidth, so it does not perform well in such low-bandwidth networks. LBARD improves on the standard protocol by prioritising bundles more carefully and by using a novel “tree synchronisation” technique based on the XOR of content hashes to rapidly prune commonly-held content from the negotiation stage.

Architecture

LBARD's synchronisation protocol is intended to eventually replace the standard Rhizome synchronisation protocol, but for the time being it is implemented as a separate daemon process upon which Serval DNA does not depend.

In software architectural terms, LBARD acts as a content bridge between two or more Serval DNA daemon processes (one running on each node). An lbard daemon runs on each node, and uses the Rhizome REST API to query and update the local Rhizome content held by its local servald daemon. For example, the LBARD daemon uses the GET /restful/rhizome/bundlelist.json REST request to fetch a list of all local content, then communicates this list over the low-bandwidth link with other LBARD daemons.

Each LBARD daemon prioritises the remote bundles that it does not already possess, and requests the remote daemon to transmit those bundles, in order of priority.

Download

LBARD is available as a Git repository from GitHub: https://github.com/servalproject/lbard. It can be downloaded using the Git clone command:

$ git clone https://github.com/servalproject/lbard.git
Cloning into 'lbard'...
remote: Counting objects: 1752, done.
remote: Compressing objects: 100% (32/32), done.
remote: Total 1752 (delta 14), reused 0 (delta 0), pack-reused 1720
Receiving objects: 100% (1752/1752), 417.68 KiB | 238.00 KiB/s, done.
Resolving deltas: 100% (1172/1172), done.
Checking connectivity... done.
$ _

The LBARD Git repository includes the Serval DNA repository as a submodule. After cloning, download this submodule:

$ cd lbard
$ git submodule init
Submodule 'jni/serval-dna' (https://github.com/servalproject/serval-dna.git) registered for path 'jni/serval-dna'
$ git submodule update
Cloning into 'jni/serval-dna'...
remote: Counting objects: 20115, done.
remote: Total 20115 (delta 0), reused 0 (delta 0), pack-reused 20115
Receiving objects: 100% (20115/20115), 21.77 MiB | 3.22 MiB/s, done.
Resolving deltas: 100% (12822/12822), done.
Checking connectivity... done.
Submodule path 'jni/serval-dna': checked out '2da3f12cfa1e4114072a9f1a94a8d477a4955f65'
$ _

Build

To build the LBARD daemon and other executables:

$ make
echo "#define VERSION_STRING \""`./md5 main.c rhizome.c txmessages.c rxmessages.c bundle_cache.c json.c peers.c serial.c radio.c golay.c httpclient.c progress.c rank.c bundles.c partials.c manifests.c monitor.c timesync.c httpd.c energy_experiment.c status_dump.c fec-3.0.1/ccsds_tables.c fec-3.0.1/encode_rs_8.c fec-3.0.1/init_rs_char.c fec-3.0.1/decode_rs_8.c bundle_tree.c sha1.c sync.c`"\"" >version.h
clang -g -std=gnu99 -Wall -fno-omit-frame-pointer -D_GNU_SOURCE=1 -o lbard main.c rhizome.c txmessages.c rxmessages.c bundle_cache.c json.c peers.c serial.c radio.c golay.c httpclient.c progress.c rank.c bundles.c partials.c manifests.c monitor.c timesync.c httpd.c energy_experiment.c status_dump.c fec-3.0.1/ccsds_tables.c fec-3.0.1/encode_rs_8.c fec-3.0.1/init_rs_char.c fec-3.0.1/decode_rs_8.c bundle_tree.c sha1.c sync.c
clang -g -std=gnu99 -Wall -fno-omit-frame-pointer -D_GNU_SOURCE=1 -DTEST -o manifesttest manifests.c
clang -g -std=gnu99 -Wall -fno-omit-frame-pointer -D_GNU_SOURCE=1 -o fakecsmaradio fakecsmaradio.c
$ _

To build the Serval DNA daemon, see the Serval DNA INSTALL instructions.

Configuration

The Serval DNA daemon must have Rhizome operations enabled:

$ ./serval-dna/servald config set rhizome.http.enable 1
$ _

The REST API of the Serval DNA daemon must be set up an authentication username and password for the LBARD daemon; for example, to set up the username “lbard” with the password “secretsquirrel”:

$ ./serval-dna/servald config set api.restful.users.lbard.password secretsquirrelp
$ _

These same credentials must then be provided to the LBARD daemon as its second command-line argument, so that it can access Serval DNA's REST API:

$ ./lbard 127.0.0.1:$PORT lbard:secretsquirrel $SID $RADIO pull

Testing

LBARD uses a test framework derived from the Serval DNA test framework. All tests can be run by running the following command:

$ tests/lbard

Logs from each test will be created in the testlog/ directory

Support for different radio types

At present RFD900 (and RFD868) radios from RFDesign.com.au are the primarily supported radios. They MUST run the custom firmware developed for them in order to work with LBARD. That firmware allows the radios to operate in a true ad-hoc model, without any clock or other coordination.

There is also highly experimental preliminary support for Codan and Barrett HF radios using ALE 2G text messages as the transport.

Adding support for new radio types

*** This information is very much a work in progress.

There are several steps required:

  1. Create a simulation driver for your radio in src/drivers as fake_.c. This will be used by the fakecsmaradio program that is the heart of the test framework.
  2. Add an auto-detection test for your new radio to tests/lbard that instructs fakecsmaradio to instantiate your new radio type, and then probe whether LBARD can detect the radio type. This test will of course at this point fail, because you have not yet implemented the LBARD driver for your radio.
  3. Create a radio driver for your radio in src/drivers as lbard_.c. This is what LBARD will use to manage radio of this new type. This consists of several main components: (1) Auto-detection of radio; (2) sending of packets, and (3) parsing of input from the radio. At present, all radios are expected to be interfaced with via a standard serial UART interface.

Copyright 2016-2018 Serval Project Inc.
CC-BY-4.0 This document is available under the Creative Commons Attribution 4.0 International licence.

lbard's People

Contributors

gardners avatar zwettemaan avatar quixotique avatar lakeman avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.