I an seeing several discussions and presentation touting HAMWAN over MESH, but no counter arguments. Does anyone have a writeup to refute any of the HAMWAN claims? or at least give the AREDN POV?
Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer
I did see the slides of a presentation from last year on HamWan (presentation was not given by HamWan core team) that was filled with out of date and inaccurate info about ARED but funny enough actually pushed people towards AREDN.
HamWan is a powerful method of high speed data networks, but it requires a technical knowledge and skill to setup because it is (at last check) not prepackaged and required (at last check) to pre-coordinate address space with your local 44/8 IP coordinator every time you need a new device(this could be an issue in a disaster when you need to add several new devices as links). If you and all your local users had the skills to run a Linux Server HamWan can be great because of its minimal restriction placed on the node operator and the freedom they give to change anything you want. AREDN is designed to be easier to setup and to remain very standardized (but this means we have traded out some enhance configuration options in favor of ease and standards)
There isn't going to be one that is better than the other for every case. If I had users in my area that were skilled enough to run HamWan I might of considered it, but locally I have trouble getting people to even enter a callsign correctly.
But I will leave it to others if someone wants to make detailed comments.
Bob W8ERD
My personal opinion...
I'm sadden there isn't more of a spirit of relishing within the ham community all the technologies we have and opportunity to learn and use. From my perspective, we have many tools in the tool-bag. They all have positives and negatives, let's see how they can work together to server our purpose and where they best fit. I love ham radio over ham radio :) . For example, I can see use of SoCal back bone links incorporating HamWan ideas -- using vendor firmware and TDMA more efficient on the backbone long distance link -- typically operated by those with strong IT skillset -- integrating with AREDN at area distribution and cell coverage areas.
Here's a summary dump of the notes I had taken ~9 months ago from the HamWAN vs MESH PPT I saw in Dec 2016. I'd be happy for anyone to clarify any of my misunderstandings of HamWan:
1) Slide 2 &3: hmmm. 802.11 Wifi Behavior related to CSMA:
AREDN incorporates 802.11n csma with RTS/CTS. The transmitters use this protocol to gain timed channel access and coordinate across hidden transmitters and foreign networks contending on the same channel (coordinate with different SSID or AP/Clients). today, it is common use of RTS/CTS with regular 802.11 wifi behavior.
2) Slide 2: performance problems in any noisy env?
Both AREDN and HAMWAN rely on sufficient SNR or quality links to work effectively. 'The' main noise source comes from part 15 wifi channels that all groups avoid like the plague.
3) Slide 5: Mesh does not scale:
HamWan is technically a "mesh' using OSPF at IP layer 3. AREDN uses OLSR mesh protocols also at layer 3 with an algorithm better characterizing RF links to route traffic.
HamWan and AREDN have equal opportunity to share same channel or segment radios on different channels. Firmware could be swapped back and forth for any radio topology with equivalent channel settings. CSMA vs. TDMA is at the link layer 2 level with no Mesh protocols involved. Both 'scale' proportionally with TDMA increasingly winning on 'throughput' as distances increase. Mikrotik WISP operators recommend no more than 20 clients in the Mikrotik forums to retain performance as well.
4) Slide 6 & 7 & 8: HamWan is Cellular, Mesh does not scale, and sectorized.
Both technologies can be deployed in the same cellular design and sectorized design. The difference is after you select which channels the nodes will share (what links you want to be isolated), ARDEN packaging requires less IT skillset and knowledge to deploy. Both technologies deal with the same noise and retransmit issues, and again TDMA wining out on throughput as distance increases. The idea of all nodes being on the same channel would not scale regardless of technologies involved.
5) slide 9 &10, 11, 12 : Mesh is insecure, Security focus, public access
HamWan and AREDN have equal ability to incorporate user authentication and message authentication using IPsec 'AH' packet authentication (with out message encryption). Packet authentication can be used (or not) with any combination of OSPF, OLSR, CSMA, TDMA protocols. AREDN will continue to add security capabilities with growth--and regularly patches the firmware with reported vulnerabilities. We're all network security cognizant in today's WannaCry world, right?
6) slides 12 ham network.
AREDN will continue to use Part 97 frequencies to stay out of wifi noise. FCC action has driving vendors like Mikrotik to prevent end users from operating outside part 15 rules and channels. HamWan use of vendor firmware means part 97 channels are no longer available with all new equipment now hitting the market. AREDN is based on Opensource and we customize the firmware to work in part 97 channels. While opensource is under threat, it is receiving FCC support (see TP-Link FCC settlement) and AREDN will continue to modify the firmware of latest devices to use part 97 frequencies accordingly.
7) slide 13 mesh does not do internet
All businesses use 10.0.0.0/8 address space per RFC definitions for their company networks. This slide is like saying it's difficult for a business to interact with the internet, when it's everywhere. Until everyone migrates to IPv6, the AREDN address space does not need coordinated across network islands. They are connected with internet tunnels. Millions of nodes would need to exist to have a meaningful probability of IP address collision (and then can be manually corrected).
Security and part 97 purposes are the primary issues. AREDN's approach is opening up access as needed for emcomm purposes and support of an incident. Any device on an AREDN mesh could be setup to be accessed from the internet (it's linux, we can do anything we want in software). I do this all the time. However let's keep in mind the purpose of ham radio--we're not here to be an extension of the internet. We're here to substitute when the internet goes down and maybe our reach to the internet in the next county or state can give a local incident improved options. We all want to be cognizant that we're not too dependent on the internet.
Over the course of this thread, there'll be discussion as to the benefits/ weaknesses of both. Most of my thinking with regards to both revolves around emergency communications exclusively. Those of you experimenting with these modes, thank you. You're making the whole system far more robust for all of us.
As one thinking about EmComm, the answer I need is not either/ or; it's both/ and. I need layers of reliable communications. In hill-and-tree-infested areas, HAMWAN has dead spots. MESH has greater flexibility to deal with that issue and is self-healing. Getting out of the area: add an RMS Gateway on HF.
Again, thank you for all of the experimenting with the systems. You're making my job easier. I plan to use all I can. In my area, we have both... and the trees/ hills/ valleys. We need it all.