World's First Fully Programmable and Open-Source Massive-MIMO Platform(renew-wireless.org)
renew-wireless.org
World's First Fully Programmable and Open-Source Massive-MIMO Platform
https://renew-wireless.org/
36 comments
This is a really cool, valuable, important project. It's a little bit of a bummer that they went down the USRP Route (https://powderwireless.net/equipment), as USRPs are really not accessible to many people (price point). Something like a Lime Microsystems radio, or the BladeRF 2.0 Micro (https://www.nuand.com/bladerf-2-0-micro/), would have been really nice.
There is a reason everyone uses Ettus Radios (USRP). You will quickly pay back the extra cost though saved engineering time over by not buying cheap hobbyist hardware.
The Skylark Wireless radios are based on the Lime Microsystems chip. The lead developer at Skylark is the guy who originally wrote Gnu Radio Companion and went on to write SoapySDR and Pothosflow, as well as the first (useful) driver for the LMS chips. Super smart guy at the core of a super smart team. They’re big on open source and I know firsthand that the University of Utah is using their radios. I believe that page is just old, there’s a picture on that page that shows a box that I recognize as very early Skylark equipment.
(disclaimer: I used to work at Skylark.)
(disclaimer: I used to work at Skylark.)
I guess what frustrates me is stuff like googling "ad9361 sync" gets you:
https://wiki.analog.com/resources/eval/user-guides/ad-fmcomm...
Whereas googling this about the LMS7002 seems to show you can buy a very expensive board to do this: https://www.crowdsupply.com/fairwaves/xync#synchronization-p...
Also, on the limesdr too: https://discourse.myriadrf.org/t/synchronize-two-limesdr/171...
it looks like you can chain reference clocks to get a fixed offset? https://wiki.myriadrf.org/LimeSDR-USB_hardware_description and then get some external device to do reset on the digital side for sync.
----
Independently, I'd love to know more about Skylark's SDR product and play with it.
Whereas googling this about the LMS7002 seems to show you can buy a very expensive board to do this: https://www.crowdsupply.com/fairwaves/xync#synchronization-p...
Also, on the limesdr too: https://discourse.myriadrf.org/t/synchronize-two-limesdr/171...
it looks like you can chain reference clocks to get a fixed offset? https://wiki.myriadrf.org/LimeSDR-USB_hardware_description and then get some external device to do reset on the digital side for sync.
----
Independently, I'd love to know more about Skylark's SDR product and play with it.
For those confused what it is the YouTube video in the page breaks down what it is, what they hope to do with it, and how they are testing with it at a more general level than the text. It also breaks down some of the terms like massive MIMO.
The page is really bad. I am working in telecommunications and needed some time to find what it exactly is...
Do they make a chip without the diplexer built in, so you can use different antennas for RX and TX?
What's MIMO? I thought it was MMO (Massively Multiplayer Online).
It's used to increase data rates in radio transmissions. Rough intro: if you have a cabled connection and you want to double your data rate, you can use a second cable. Likewise, if you have a radio transmission between two antennas, you can increase the data rate by adding one more antenna on each side (hence Multiple Input Multiple Output). The signals of both transmitters will arrive overlapped on both receivers, but it is usually possible to detangle them with a system of equations.
Trying to read this on iOS, but the site is completely obscured by an expanded menu that I can't seem to close in any way. Could anyone give a TL;DR of what this is?
Reconfigurable Eco-system for Next-generation End-to-end Wireless
Argos was first developed at Rice in 2011 as a 64-antenna base station, becoming the first Massive MIMO base station ever built. Since then, Argos has evolved with two newer versions. The most recent version of Argos (ArgosV3) has been already deployed around Rice campus as part of the ArgosNet project. Argosv3, now known as Faros, is developed by Skylark Wireless and now commercially available.
The RENEW platform aim is to serve three goals: 1) Integrate and deploy on the POWDER platform, at the University of Utah campus and areas around the campus. 2) Empower custom testbeds, using Iris and Faros, to enable novel applications. 3) Work closely with other open-source efforts, like OpenAirInterface, to empower next-generation of wireless equipment.
The RENEW team will be part of the 1st international workshop on open software defined wireless networks (colocated with MobiSys 2020)>>See more
Our team is launching the RENEW Visitors Program inviting the community to come to Rice to work with us on the massive MIMO platform>>See more
Ashu Sabharwal will give a talk at the IEEE 5G World Forum's Workshop on 5G Applications and Services in Dresden Germany>>See more
https://renew-wireless.org/
The RENEW platform aim is to serve three goals: 1) Integrate and deploy on the POWDER platform, at the University of Utah campus and areas around the campus. 2) Empower custom testbeds, using Iris and Faros, to enable novel applications. 3) Work closely with other open-source efforts, like OpenAirInterface, to empower next-generation of wireless equipment.
The RENEW team will be part of the 1st international workshop on open software defined wireless networks (colocated with MobiSys 2020)>>See more
Our team is launching the RENEW Visitors Program inviting the community to come to Rice to work with us on the massive MIMO platform>>See more
Ashu Sabharwal will give a talk at the IEEE 5G World Forum's Workshop on 5G Applications and Services in Dresden Germany>>See more
https://renew-wireless.org/
There is a fairly high density of acronyms and domain specific jargon on this page which makes it quite hard to understand what this is about and what is cool about it.
Could someone ELI5?
Could someone ELI5?
It is a first attempt to make a full solution open source radio that uses multiple antennas.
I believe all current high performance radios in this space are super duper proprietary, with near no data on how they work in open access.
I believe all current high performance radios in this space are super duper proprietary, with near no data on how they work in open access.
It's not so much that the high performance radios are proprietary as it is that building one of these systems is expensive and requires lots of domain specific knowledge. You could somewhat easily build one of these systems for ~$250k using CoT's hardware, in fact Lund University did a few years ago for some of their MASSIVE MIMO research.
If you don't want to use CoT's hardware for some reason then you can roll your own transceivers. You have a choice of Analog Devices chips or Lime Microsystems. Xilinx has a new platform that would also work but it hasn't been widely released yet. Lime's chip is cheap but it's performance is crap and its drivers are even worse. ADI's chips are easy to use and have fantastic performance but you pay for it. You also need to shell out for the massive FPGA's to handle 64*6.4GB/s of streaming data and then you need a system capable of processing it which will still probably be an FPGA or a DSP.
If you don't want to use CoT's hardware for some reason then you can roll your own transceivers. You have a choice of Analog Devices chips or Lime Microsystems. Xilinx has a new platform that would also work but it hasn't been widely released yet. Lime's chip is cheap but it's performance is crap and its drivers are even worse. ADI's chips are easy to use and have fantastic performance but you pay for it. You also need to shell out for the massive FPGA's to handle 64*6.4GB/s of streaming data and then you need a system capable of processing it which will still probably be an FPGA or a DSP.
Even this way, try to find a schematic in open access with software, and etc
Thank you !
It's free grid internet or what?
I'm sorry, but what's this? Am Open Hardware RF initiative? Is it just antennas? I'm sorry but the info is all over the place and I'm a bit confused. I'm not native english speaker but I work on the ISP space. It would be easier to compare it to standard ISP stack.
Their project is more focused on (but not limited to) the physical layer, particularly for cellular wireless networks.
Usually in cellular (e.g. LTE, 5G) both the mobile phone and the cell towers implement all of their PHY using a mixture of custom SoCs, FPGAs and DSPs, and all the code behind of this is as closed as it gets (even the drivers).
What they are doing is basically implementing as much as possible of their cellular network with code that is open. Literally getting the signal as close as possible from the antennas, sampling, and processing everything with code that can be studied and changed.
This openess makes it easier for researchers to develop new ideas for wireless networks (just reprogram your CPU, instead of having to beg to a manufacturer to provide SDKs for their super secretive equipment).
Usually in cellular (e.g. LTE, 5G) both the mobile phone and the cell towers implement all of their PHY using a mixture of custom SoCs, FPGAs and DSPs, and all the code behind of this is as closed as it gets (even the drivers).
What they are doing is basically implementing as much as possible of their cellular network with code that is open. Literally getting the signal as close as possible from the antennas, sampling, and processing everything with code that can be studied and changed.
This openess makes it easier for researchers to develop new ideas for wireless networks (just reprogram your CPU, instead of having to beg to a manufacturer to provide SDKs for their super secretive equipment).
>This openess makes it easier for researchers to develop new ideas for wireless networks
Could it be used as part of the stack to create city wide non-cellular wireless network for Internet[1], Are there any such current projects anywhere?
[1]https://needgap.com/problems/51-non-cellular-network-mobile-...
Could it be used as part of the stack to create city wide non-cellular wireless network for Internet[1], Are there any such current projects anywhere?
[1]https://needgap.com/problems/51-non-cellular-network-mobile-...
Not really. The same economic and spectrum constraints apply. By the time you've extended the Wi-Fi PHY and built out a network to cover most of a city you've effectively reinvented LTE.
>The same economic and spectrum constraints apply
Single Indian Telecom (Airtel) paid ~ 2 Billion USD(current price) in 2016 for 173.8 Mhz spectrum across 1800/2100/2300 Mhz bands. Where as small YC startup(Link in my parent comment) has created Bangalore city wide WiFi Internet(2.4 Ghz/ 5 GHz has been made unregulated and free).
So, definitely telecom spectrum cannot be called same as WiFi as one requires Billion dollar oligarchy to enter the space and the latter can be entered with 4 member startup team.
Single Indian Telecom (Airtel) paid ~ 2 Billion USD(current price) in 2016 for 173.8 Mhz spectrum across 1800/2100/2300 Mhz bands. Where as small YC startup(Link in my parent comment) has created Bangalore city wide WiFi Internet(2.4 Ghz/ 5 GHz has been made unregulated and free).
So, definitely telecom spectrum cannot be called same as WiFi as one requires Billion dollar oligarchy to enter the space and the latter can be entered with 4 member startup team.
Super helpful breakdown, thank you! And to OP for asking
Don't know much of that. The ISP I work for just buys stuff from vendors and puts interface layers on top of it, which what I work with. I've seen some ground work and people just works with modules.
I don't deny the value of such project as this modules cost some big money (and licenses), but unless they really do a very good job it will introduce too much complexity IMO. You can also call someone at $vendor and cry in their shoulder.
It's already tiresome to work with different vendor quirks.
I don't deny the value of such project as this modules cost some big money (and licenses), but unless they really do a very good job it will introduce too much complexity IMO. You can also call someone at $vendor and cry in their shoulder.
It's already tiresome to work with different vendor quirks.
Does this mean a real possibility for user-defined, community 5G networks? If so, that could usher in a new age of mesh networks to democratize the long-range wireless network space. If the hardware is cheap and a protocol were used that's anonymous and encrypted by default, this could be a game changer.
From an idealistic newbie's perspective, I imagine the possibility of users buying routers with this tech, plugging them into their fiber or copper connection, and creating or expanding an existing 5G meshnet + a home net. Maybe it could drastically drop costs for 5G network operators as they wouldn't need to put up extra receivers anywhere. They could simply offload that burden to the consumer. Most routers are lent out anyway, so not only would they make money, consumers wouldn't need to worry about which operator they chose because they'd all be in the same meshnet anyway.
But eh... I'm sure there's some big thing I'm missing besides human greed.
From an idealistic newbie's perspective, I imagine the possibility of users buying routers with this tech, plugging them into their fiber or copper connection, and creating or expanding an existing 5G meshnet + a home net. Maybe it could drastically drop costs for 5G network operators as they wouldn't need to put up extra receivers anywhere. They could simply offload that burden to the consumer. Most routers are lent out anyway, so not only would they make money, consumers wouldn't need to worry about which operator they chose because they'd all be in the same meshnet anyway.
But eh... I'm sure there's some big thing I'm missing besides human greed.
Well, a LimeSDR based E-UTRA (LTE) Band 38 2x2 eNodeB could be had for a couple hundred USD, depending on how cheap you can get the Host computer and clock source.
I'd consider just injecting the clock via the fiber (running a dedicated single-mode fiber from the next PoP (up to ~10 km away) to the eNodeB, potentially exposing Ethernet service from the base station, just like the typical monolithic DOCSIS wifi router), via one of two ways:
(1) run a separate wavelength via CWDM and just AM-modulate a sinusoidal reference oscillator output onto it, or
(2) anchor the clock for the data link from the PoP to the eNodeB to your reference oscillator, and just tap your reference clock (to feed into the local VCXO-based master-PLL) off of the data link receiver's clock recovery circuity. Some mild phase noise is acceptable thanks to the VCXO.
I'd consider just injecting the clock via the fiber (running a dedicated single-mode fiber from the next PoP (up to ~10 km away) to the eNodeB, potentially exposing Ethernet service from the base station, just like the typical monolithic DOCSIS wifi router), via one of two ways:
(1) run a separate wavelength via CWDM and just AM-modulate a sinusoidal reference oscillator output onto it, or
(2) anchor the clock for the data link from the PoP to the eNodeB to your reference oscillator, and just tap your reference clock (to feed into the local VCXO-based master-PLL) off of the data link receiver's clock recovery circuity. Some mild phase noise is acceptable thanks to the VCXO.
Well, for one you're not going to get a very wide coverage range or throughput with the ~1 watt of power you're limited to in the ISM bands. Your router will also have to accept interference from 802.11 routers, Bluetooth, etc devices. There is a lot of on going research into dynamic spectrum allocation and cooperative scheduling which this platform is actually supporting but that still requires federation which your use case seems to be against.
That sounds an awful lot like what Fon Wireless is doing.