7 ms·
I'd love to setup a small scale network for personal use, but the elephant in the room is licensing.... Is it actually possible to formally license a DIY LTE ne
by pomatic 5y ago
I'd love to setup a small scale network for personal use, but the elephant in the room is licensing.... Is it actually possible to formally license a DIY LTE network? (Or AMPS, 2G, 3G). In the UK, for example it is possible to get experimental licenses in principle, but I doubt it would be possible for an individual to legally operate a permanent or long term network on/near commercially allocated frequencies?
- alexfromapex 5y agoI think there needs to be a push to make it possible. It's kind of tiring having big corporations being the only ones who can do things.
- karteum 5y agoThere are already a lot of things that are feasible without any license. In Europe/CEPT, you might have a look at all bands under the provisions of ERC Recommendation 70-03 (https://docdb.cept.org/download/2464 https://docdb.cept.org/download/2464). However, those provisions are made in order to ensure a good/fair access to anyone, and therefore to prevent a single user or single technology from overusing those bands which are meant to be shared. For that purpose, there are associated restrictions (in terms of power/EIRP, duty-cycle) and/or mandatory sharing approaches (Listen-before-talk, detect-and-avoid, etc.). In the case of Wi-Fi, CSMA/CA is a form of listen-before-talk. Unfortunately, mobile technologies defined at 3GPP (GSM, HSPA, LTE, NR) are not designed to be used in such a way (i.e. they don't have any sharing mechanism such as LBT and they require _by design_ dedicated/licensed bands), which by the way implies some kind of specific coordination at the country borders where two operators are using the same channels... (you might look at ECC recommendation 15-01 for an example of PCI sharing). LAA is a way to have an LTE carrier within the (shared) 5 GHz band, but it has to rely on an anchor carrier for signaling, which requires licensed spectrum. Multefire is a fully-unlicensed solution, but I doubt many UEs (smartphones) support it, and anyway because it must implement the same power limitations and LBT as wi-fi in order to comply with regulations I doubt it would be much better than wi-fi... (maybe it would in some specific case where deterministic QoS is important) One more thing : keep in mind that a typical 3G/4G/5G macrocell site (e.g. around 65 dBm EIRP per carrier) is something very expensive : your mileage may vary but it can easily be around 100000 € / site when some construction is required.
- joecool1029 5y ago>LAA is a way to have an LTE carrier within the (shared) 5 GHz band, but it has to rely on an anchor carrier for signaling, which requires licensed spectrum. I want to add, as I said in my last comment that 5G NR allows for 5Ghz to be used as primary carrier, it was controversially included in the standard. A study from earlier this month showed LAA apparently doesn't play too well with wifi nearby: https://www.cs.uchicago.edu/news/article/laa-wifi/ https://www.cs.uchicago.edu/news/article/laa-wifi/
- g_p 5y agoIn general, cellular technologies have been designed against assumptions of a clean (or at least exclusively used) RF channel. 4G and 5G are deployed with frequency reuse between base stations, which implement the same standard and can coordinate their emissions and scheduling of clients (both in time and frequency) to minimise interference. WiFi is a whole different kettle of fish - it's designed to be used by multiple independent access point operators simultaneously, with the ability to change frequency if needed based on the interference observed. It's designed to try to deliver good performance by listening before transmitting etc, to avoid transmitting over another device, to avoid a tragedy of the commons scenario where selfish devices end up rendering WiFi unusable for everyone (including themselves), through refusing to yield time to devices transmitting on other networks. NR-U and LAA etc don't generally play according to the same rules, as they're standards arising from the world of exclusive spectrum access, and coordination of base stations by one operator - in the world of cellular, the base stations allocate uplink channels for their clients. That doesn't work in WiFi with multiple networks in the same approximate location, hence they need to try to prevent interference and cross-talk.
- g_p 5y agoIn the UK there are 2 ways to get access to long-term spectrum. It's also usable from a bands and handets perspective! 1. Shared access licence. There's 3.3 MHz (duplex paired) of former DECT guardband available for local use at the top of the 1800 MHz band. This should work fine to run 2G or 4G. There's 10 MHz of TDD at 2390 to 2400 MHz, indoor use only I believe, and most useful for 4G. There's also 3.8 to 4.2 GHz which allows outdoor use, and is dedicated shared spectrum for 5G. Transmit power limited aren't enough to run a big commercial network, but are enough to deploy a private network on a campus or private site. The former two should be widely supported on handsets. The latter (5G option) is aligned with band N77. 2. If you're in a rural area, get a local access licence to "take" existing operator spectrum and use it legally, with 3 years max tenure. It can be renewed if the operator has no intention to use it. Operators don't like this, as they see spectrum as their owned property, but this isn't the case, and they only have a right to use it... The process of getting a local access licence is deliberately complicated by operators as they don't want you doing it, but it can be done. You need to understand the technology and commercials of the industry though to realistically be successful. Some operators want you to talk to them first, others want you to talk to Ofcom first. If you know what you're doing, and make a strategically selected request for spectrum, it can be done. You can operate a commercial service in this spectrum, but you have to be clear to users the time-limited nature of spectrum access. Going forward though, it's clear from Ofcom's own priorities that a more dynamic nature of spectrum allocation and utilisation authorisation is a priority and likely to be coming down the line - it's been a work item the last couple of years. Once that comes, we could see easier ways to access spectrum. Realistically though, operators will do what they can to make spectrum access complex, to preserve a final moat in a market of 10+ years of continued reduction in average revenue per user.
- ackbar03 5y agoWhat would be the cost of this though? I thought these things tend to get bid up to the range of billions of dollars
- g_p 5y agoThese licenses are in set-aside bands specifically for sharing. You're talking about 80 GBP per year for up to 10 MHz, and 80 GBP per year for each additional 10 MHz. https://www.ofcom.org.uk/manage-your-licence/radiocommunication-licences/shared-access https://www.ofcom.org.uk/manage-your-licence/radiocommunicat... This isn't nationally auctioned spectrum (which gets bid up to billions, then operators have to pay those fees into government coffers).
- vxNsr 5y agoYes this was my question… The FCC is pretty aggressive about keeping people off frequencies they don’t have a license for, how does that work in these types of environments? Especially something like a college campus, where you have 10,000+ students who would all need their phones to be reprovisioned… not to mention how would it work if you walk off campus? LTE/5G isn’t really great at working with multiple networks… it just wasn’t set up that way…
- varispeed 5y agoDoes it mean that the protocol has big corporation monopoly built in by design? I am not familiar with the details, but by the sound of it why is this even legal?
- vxNsr 5y agoNot really, anyone can own spectrum, in the last few years its gotten really expensive as it's become more important to building a robust network but historically it wasn't that hard to license an small band in your area.
- PeterisP 5y agoRadio spectrum is a shared resource that for many use cases needs exclusive allocation to work well - if one person/organization is using a frequency band in an area, they need to be the only ones who can use that frequency band there. Think TV or radio stations; if someone else is transmitting in the same channel, then it just disrupts the broadcast. So most such spectrum is allocated (often through auction) and then whoever has that allocation is legally granted a monopoly on the use of that spectrum, enforced by government who will prevent anyone else from using that spectrum with fines and even physical interference (finding, disconnecting and confiscating the violating transmitters) if needed. Transmitting over radio waves is a highly regulated privilege. There are a few ranges that are open to public (i.e. the bands used for Wifi and BT) with devices that need to be certified to ensure that they transmit only in the narrow permitted ranges, but in most frequency ranges (including the LTE/5G ranges) it is illegal to operate a transmitter without an explicit license.
- joecool1029 5y ago5G NR allows for 5Ghz to be used as primary carrier in band 46. (as opposed to LTE that only allows it as a secondary carrier.) This wouldn't require a license to use. I believe CBRS band 48 (3.5Ghz) is lightly licensed but I don't remember if you can set it up as a primary carrier on LTE.
- ksec 5y agoDo you know how does the patent licensing works on 5G NR-U? I have been trying to find an answer but literally everyone in the industry are mum about it. I really really want to see real world usage of NR-U and as a possible replacement of WiFi.
- g_p 5y agoI understand that NR-U is a full implementation of a 5G base station (and client handset/CPE if you can find one that actually implements NR-U!), and therefore you will require patent licenses for the full implemented stack. In an ideal world, your radio manufacturer and software vendors would have appropriate patent licenses in place. For this and other reasons (handset support being commercially problematic since carriers want to preserve their dominance), I don't see NR-U really replacing Wi-Fi. The friction to joining a device to a network is also far higher, and there's much more complexity in running the network and architecting a suitable core network and user plane functions and keeping it all running. WiFi really is a lot simpler in many ways.
- ksec 5y agoWell NR-U is pretty much a Qualcomm technology ( MultiFire ) brought to 3GPP for standardisation. So I expect it to be supported in all 3GPP Rel 16 modem and Qualcomm X60+ modem as well. I dont think Handset support would be a concern unless I am missing something. I was hoping for some exemption or special arrangement on the patent issues for consumer private 5G Network. WiFI just sucks, comparatively speaking. Including WiFI 6E. And judging from the ways they are doing things I have very little hope for 802.11ay and 802.11be ( WiFi 7 ).
- Sephr 5y agoYou could simply host your network on unlicensed spectrum.
- dsr_ 5y agoAFAICT -- and I would be happy to see counter-evidence -- all the unlicensed spectrum suitable for LTE or 5G is in the 5-6 GHz frequencies. The physical characteristics of transmission don't change, so one might as well just deploy the cheaply available 5GHz wifi systems -- you don't get an advantage by using LTE or 5G protocols. i suppose if you only had relatively low data rate applications you could use 900MHz, but there's a dearth of CPE. 900MHz unlicensed has slightly more bandwidth than a single 2.4GHz wifi channel.
- dilyevsky 5y agoYou can probably drop it down some more and use tvws spectrum. Probably will be alone there
- g_p 5y agoTVWS channel allocations often don't align neatly with duplex paired 3GPP bands, meaning you might need multiple adjacent TVWS channels. In addition, in many countries the TVWS regulations only cover specifically authorised devices which adhere to TVWS standards (which 3GPP doesn't), and create a whole host of challenges for the client devices (handsets), which don't understand TVWS rules, and need their transmit power to be controlled by the base station, which isn't necessarily aware of the client location, or authorised within TVWS rules to query the TVWS database on behalf of the client.
- cozzyd 5y agoMight as well use 8.0211ah there. It's not like you can get consumer equipment that supports the 900 MHz ISM band for LTE.
- pgorczak 5y agoRules are different in each country of course but most unlicensed frequencies can only be used with mechanisms that leave space for other users (networks) like listen before talk or restricted duty cycles. Cellular technology isn’t built with shared access to spectrum in mind although there is/was some effort to add it (LTE-U, 5G-NR-U).
- cozzyd 5y agoI recently deployed an LTE network at a research station in Greenland. Licensing ended up much easier than I expected (the Greenlandic government almost immediately granted the license for LTE band 8), but this is in a place in the middle of the ice sheet hundreds of miles from the nearest village.
- gregsadetsky 5y agoI’d love to hear more about this, if you can share. How complex / large is it of an installation? How comparable (or not at all) to setting up a large scale wifi network? How complex is the maintenance? Team size for installing / operating vs number of people served? Everyone’s phone has a custom sim card? Thanks :) Truly curious about this.
- g_p 5y agoNot OP you replied to, but have done similar before. Answering your questions in order (at least the ones not specific to that scenario) to share some thoughts, which are certainly not answers by any stretch: - setting up a mobile network is quite different to a WiFi network in some ways, but perhaps a little more in line with a complex large-scale WiFi network. Mobile networks are architected around a "core network" which manages mobility and user traffic, and a "radio access network" which connects the base stations to your core network. The "core" is a fair bit more involved than a large-scale WiFi network setup, at least in my experience. You will need domain knowledge of 3GPP networks, and that's sometimes hard to get hold of, at least outside of traditional mobile operators and vendors. That can be a barrier. - In terms of maintenance, a well-built RAN should "run itself" for the most-part. You'll want some monitoring on equipment. If you are using modern software-based base stations, they will run on Linux. Treat this like any large at-scale fleet deployment of Linux servers. Keep them patched and plan your maintenance windows for reboots etc. Your core network itself will realistically be a bunch of Linux servers to provide network functions. You'll have the usual periodic maintenance issues keeping up a bunch of complex services that you don't always fully understand, and the temptation to never update anything, as "it's working right now". - The number of people you need for installing depends on skill sets. You'll need some experienced riggers to install macro sites on masts by climbing the masts. But you also need people who can pour concrete to get masts into the ground. And people who can install masts. You'll need someone who knows some DC electricals and can get everything powered up. You'll need some radio knowledge to plan out the network and check the antennas are pointed in the right directions and with the correct tilts. If you outsource this to a subcontractor, they can probably get the same people who do commercial networks to do it, but that will cost you dearly, and you'll never quite know what happened or how to fix it if something goes wrong! - To operate the network I guess it's not a huge amount different to any other complex IT system. If you build out your core network well, you can serve a few thousand users without too much trouble, with minimal people. Once you start to scale beyond a few thousand users, you have to scale up the core and architect it a bit better. There's some routine monitoring and maintenance, and the usual things you need to do in any production environment, but in general it's not too horrendous. If you go down the "big vendor box" route like carriers do, they'll take big complex managed services to keep everything in check, but have far fewer really good network people on-staff. The biggest issue you'll have in terms of staffing up to manage it is getting the breadth of knowledge and skills needed to have access to the right range of skills as needed - legacy telecoms can get complex fast. If you are having an issue with IMS for 4G calling (VoLTE), you might only need 1 person, but that's in addition to the 1 3GPP core expert you have, the 1 IP networking expert you have, etc. - You need to put a custom SIM into every device, and set up a PLMN identity for the network, which is just a 5 or 6 digit number that identifies the network to handsets. The SIM tells the phone what network it should try to join, and contains the crypto keys used to do authentication with the network. You can often get a PLMN allocated by your national telecoms regulator, or use one in the 999/xx range, which are set aside for private, uncoordinated use.
- numpad0 5y agoAt least on paper the Japanese sXGP standard is close to what you want, it’s reportedly just TD-LTE Band 39 and equipments are licensed the same way as Wi-Fi APs and dongles are. Maybe if there’s enough demand for workplace private phone network, such laws could be passed to run 5G or 4G as Wi-Fi alternatives.
- ffk 5y agoCheck out CBRS which is "licensed-by-rule." You should be able to use CBRS to deploy private 5G.
- TaylorAlexander 5y agoIf anyone knows the answers to this for USA I'd be very curious (at this time I don't see one). People are mentioning possibly using a 5ghz band? Very weird that the ubuntu article talks about how easy it is to do this and makes no mention of spectrum license concerns.