If you wonder about real world use cases for open source LTE stacks like srsLTE: They’re great for deploying small ad-hoc cells e.g. in vehicular and rescue scenarios like in our work from earlier this year: https://arxiv.org/abs/1802.09262
It's estimated that private LTE-networks will become $27 billion market by 2024. That's roughly 330 thousand LTE base stations for private networks.
5G is specially designed this kid of industry in mind, so more advanced applications come in addition to the LTE markets. For example, many factories can replace their physical networks with 5G private networks for automation and robotics.
I haven't read the "5G" 3GPP spec yet. I was super interested in using LTE to replace WiFi in offices, because centralized multiplexing seems like such an attractive approach.
I have tested a device developed by the company I work for on a private LTE network. I was provided SIM cards, which were programmed for their particular network.
In Android devices everything worked as expected.
However, none of the iOS devices associated with this network, although the devices did support the LTE band for that network.
Speculation was that there is a PLMN (cellular provider identifier) whitelist in Apple devices, which could not be worked around.
Yes you need to burn a SIM card, but that's it basically. srsLTE and other "homebrew" eNodeB software is standards-compliant and your smartphone will most likely connect.
In practice, you need to worry about legal restrictions -- you're not allowed to transmit on common LTE bands.
Not sure how this will actually look in the future. Is the office mobile network physically separated, or do you use the public network with special SIM cards that connect to a virtually separated mobile network?
One idea is that commercial installations rely on unlicensed spectrum (like 6Ghz) to transmit LTE. Check out MuLTEfire for this. As far as I know, there are no LTE base stations transmitting these frequencies available for purchase.
With 5G some countries have also taken into concern a more liberal spectrum policy, which could result in more mobile network operators. One idea follows is that the spectrum would be "oversold" by selling the same frequency to organizations tied to particular cities, for example.
Finally, there is the notion of virtual SIM cards. And as far as I know, alternatively you could also just hack the LTE server to accept each incoming connection (with any SIM) and use a challenge after the Internet connection is established, and then just kick out any device which doesn't send the magic packet.
Yeah, the idea was to use ISM band. The problem was finding MTs that supported these bands. The LTE base station was much easier.
I was hoping 5G had taken this use case into account, and thought to optimize for those bands, or alternatively allow for use of 60 GHz, or something where i can get a local license for usage of that band.
You cannot accept unknown SIM cards due to mutual authentication -- the provider must know the shared secret that is burned into the SIM.
You can achieve network separation APNs, so there is no need for such a hack.
The question I have in mind is: Does your office advertise it's own PLMN with WiFi-like equipment, or do you outsource your office wireless network to an established operator which handles authentication and provides a virtual, separate network?
Regarding your question, I recall reading a paper which compared both cases: the office owning the whole infrastructure, and one in which it is rented/outsourced to an established MNO. While the paper suggested that both use-cases are plausible, in practice your preference is likely to vary depending on local regulation and use-case (e.g. latency requirements etc.).
Chiming in from GP: I think mobile setups like base stations mounted on drones or high altitude platforms are really interesting. We’re currently looking at providing cellular connectivity to rescuers and rescuees at sea using a fixed wing drone.
Not much actually. You need to make auth work which means knowing some secrets stored on the SIM card. You can buy your own custom cards or probably use eSIM (however I haven’t tried option 2).
That is the first question that jumped into mind for me. How is this useful to me now. I assume that the LTE chip in our devices will handshake with these networks seamlessly.
I used to volunteer for a first aid organisation and often we would be providing support for large events like county shows or sports events.
The radios we used were simple VHF radios on licensed frequencies. There was no encryption which meant anyone could listen in, so obviously we weren't allowed to discuss personal details over the radio. Depending on the actual event we may be covering a large area or uneven terrain, so the radios may not always work.
For backhaul we'd either use our phone if we needed to call the emergency services or for larger events we'd have them onsite so they would forward the message on - for example to warn a hospital of an incoming patient (often it was in remote areas where the nearest hospital had limited A&E facilities) or to call an air ambulance.
Huawei's got a portable LTE Network that happens to be just for things like this. They call the technology eLTE (enterprise LTE) and the portable solution rapid eLTE. For backhaul you can use Fiber, Radio, Ethernet or Satellite, pretty much anything. the solution supports sending multimedia over the handys, voice, video, gps and data. They also offer a trunking solution over eLTE.
Perhaps I’m missing something but what is the benefit of LTE over 802.11ac? I know that traditionally LTE is only on licensed bands and 802.11ac mostly on unlicensed. However, it seems 802.11ac gear is significantly cheaper. Both technologies can be run across whatever spectrum you’d like.
What characteristics does LTE have that makes it better than 802.11ac for some of the applications mentioned?
Mobile/Cell technologies work better when not all stations are "visible" from each other - as the channel access control in 802.11 is fundamentally peer-to-peer (even with an AP)
Some of the major outdoor WiFi companies (Ubiquiti, Miktrotik) etc started out with 802.11 then moved to cellular-style (TDMA etc.) methods for this reason.
I thought distance was more tied to effective radiated power, installation, and frequency selections. Off the shelf 2.4 GHz 802.11ac gear can easily do 30+ mi.
Side question: there's no way to get low-level access to the WiFi adapter in a MacBook Pro, right? It would be nice if I could do SDR stuff with the radio antenna that came with my computer... though I'm guessing it's not the right type of antenna anyway.
Yes, the wifi antennas are very specifically sized for the 802.11 frequency ranges. They also often have band pass filters built in to prevent transmissions outside of those bands, as required by FCC regulations and such. That said, if you are really interested in that kind of thing, and you are willing to switch to Linux, the wireless card drivers are very hackable.
Source: I forked the driver for my wireless card many years ago and still maintain it, tho these days it's mostly the upstream kernel driver plus a few hacks to get the most out of the hardware: https://github.com/FreedomBen/rtl8188ce-linux-driver
Thank you! I'm sorry I'm not totally sure what you are asking. Are you asking how much does it drain the laptop battery? To be honest I've never really tested that, but I'm curious now :-) If you are asking how much the driver affects the transmit power of the wireless card, it takes it from 20 dB up to 33 dB, and each 3 dB is a doubling of power, so it cranks it up by a significant amount.
Usually the radio is fixed-band, but the receive and transmit digital side will usually be some combination of fixed-function pipeline and actual DSP software. The problem is that the information needed to reprogram them is highly proprietary.
Yep, you need to be DJI or similar to get access to them. Very discouraging as a small fish. (I've looked into the broadcom-based project above, unfortunately it's not suitable for continuous reception/transmission, and you would still need processing power anyway).
nah, and rtl-sdr units are under $20 so you should look into picking one up, they're a blast. You can actually create a transmitter out of a sound card though, it's kind of a fun experiment https://dspillustrations.com/pages/posts/misc/using-your-sou...
Some people have also tried to create a wifi interface from SDR transceivers, but they're extremely inefficient.
I have started development for low power IOT devices and am very keen to use LTE bands from 450 MHz to 2.7 GHz
LTE Cat-M1/NB1/NB2 protocol layers L1-L3
to leverage the low power modems that are being developed on chip. I want to know if there is a way to make a local transmitter, there is no LTE-M cellular spectrum where I work from
