Does Purism Plan on Adding 5G as an Option for the Librem 5?

Most of the world is focusing on 5G for sub-6GHz frequencies, so their cell phones can use the same antennas as 4G. The problem with the mmWave frequencies is that they can be blocked by obstructions, so you need direct line of sight with the cellular tower. Phones designed for mmWave 5G need antennas along all 4 edges of the phone to deal with people’s hands blocking the signal when holding the phone, and many of them have 6 or 7 different antennas inside them.

Amother thing is that the frequencies over 95 GHz are basically useless for cellular communication because the signal is too short and too easily blocked. The US is only using 24 - 47 GHz for its mmWave 5G, and most carriers outside the US have decided that 5G over 6GHz is not worth it, so they aren’t bothering with mmWave 5G.

5 Likes

Fully agree. For daily use 5G is more a marketing thing than really needed on a smart phone. (maybe for competitive gaming …?)

2 Likes

You have nothing to worry about. 5G is designed to share spectrum space with 4G, so it can be implemented without shutting down existing 4G networks. There is still massive investment in 4G and the amount of 4G coverage is actually expanding, because the cellular carriers know that it won’t be disappearing any time soon. Ericsson, which is the 2nd largest Telecom equipment maker in the world, says that 85% of the world’s population had 4G coverage at the end of 2021, and expects that percentage to grow to 95% in 2027.

According to the Global mobile Suppliers Association (GSA), by the end of 2021, there were 906 carriers in 243 countries, which were investing in LTE (i.e. 4G), compared to 487 carriers in 145 countries, which were investing in 5G networks. Of those investing in 5G, only 99 carriers in 50 countries were investing in 5G standalone networks, meaning that the vast majority are planning to operate dual 4G and 5G networks.

I expect 4G to still be around for at least the next 15 years, and probably longer.

6 Likes

Given the Librem 5’s already massive power consumption with huge battery to compensate, would a more power-hungry modem even make that big of a difference?

2 Likes

Absolutely.

2 Likes

To add to the confusion, Starlink is licensed now to deploy 20K new Starlink satellites in to orbit. Spacex is currently at about 3K starlink satellites currently deployed and are adding new satellites at the rate of around 30 new satellites per launch and with several launches taking place per week every week now.

Starlink has a deal with T-Mobile as well. Within the next few years sometime, your T-Mobile phone service will work everywhere on earth. When you make a call, the phone will look for the T-Mobile nwtwork first. If it can’t find a T-mobile tower, it will look for roaming via other carriers. If no roaming is available, it will reach out directly to a Starlink satellite in low earth orbit. So whether you’re at the south pole, in the middle of the Amazon jungle, or in the middle of the ocean, your cell phone will still work. I don’t know how they get around the requirement for a satellite dish and how they can blanket the cell tower frequencies from space without causing havoc.

2 Likes

More power drain is to me not something I’m willing to accept from my L5.
Also, 4G has for me more then enough data troughput. I do not see the need for 5G or higher.

2 Likes

The last time I looked at this issue, there were no 5G M.2 modems in the 30 x 42 x 2.5 mm form factor, because the 5G modems were too power hungry and they needed a larger heat spreader than could fit on that size of M.2 card. However, the industry has figured out how to make 5G modems that consume less power and need less heat dissipation, so there are now 30x42mm 5G M.2 modems that support Linux and USB on the market. (I don’t think the L5 supports PCIe over M.2.) For example: Cinterion MV31 Ultra High Speed IoT Modem Card (5G)

At this point, all carriers at still supporting 4G and probably will continue to for the next decade at least, so there isn’t a compelling reason to upgrade to a 5G modem in my opinion, because it will consume more energy than a 4G modem, but maybe some people need the lower latency and high bandwidth of 5G. mmWave 5G won’t be possible on the L5 without a significant redesign of the phone, because it will need additional antennas on every edge of the phone.

Given that it is now technically possible, it would be interesting to know whether Purism plans to offer a sub-6GHz 5G modem card for the L5. Frankly, I think it is low priority, but maybe others feel differently.

@StevenR , SpaceX says that their Direct To Cell (D2C) service will work on normal 4G LTE modems, so it should work on the L5. It is designed for low bandwidth communications when outside the range of the carrier’s cell towers.

3 Likes

Let’s get a RELIABLE 4G modem before we look at anything else. If anything they should look for alternative 4G modems that are maybe not as troublesome?

1 Like

And hopefully also with free firmware :wink:

3 Likes

Sounds like a poll needs to be created to have a better idea of the available data.

I’m in canada and had lots of problems, ironically after my old modem got replaced with the same firmware.

1 Like

Whereas I have no problems in Canada using my modem, so it sounds like a hit and miss.

1 Like

It is achievable provided that the chip or chipset still enforces the legal requirements. For example, you should be able to have a PHY chip that enforces the legal requirements but implement the MAC layer and higher layers in free software.

That may not achieve full freedom but it would be an improvement over a completely blackbox modem. For example, some of the reliability or connectivity issues in some countries with some carriers could probably be more readily fixed, without a need to go into the PHY layer.

I believe that’s correct - because it is difficult to guarantee security if you have an M.2 card and PCIe is available on the M.2 socket and used by the M.2 card. It might need some kind of IOMMU.

I didn’t confirm that with the schematics though.


As it stands today, 4G - as a technology - offers more than enough bandwidth for my needs. So I am in no rush to upgrade to a 5G modem were one to become available. (5G is available on my nearest tower and in other nearby areas - I can see that my iPhone does 5G if I switch off WiFi.)

Note that the BM818 modem claims a marketing speed of 150 Mbit/s and even that is well short of what 4G is capable of. So, were a better compatible modem to become available, there is scope to get higher bandwidth while staying with 4G.

(However looking at a different 4G device that I have, I don’t get anywhere anywhere near 150 Mbit/s anyway - so I think I am being limited by my telco, implicitly or explicitly.)

1 Like

I am more interested in what is practical instead.

1 Like

That reminds me, whatever happened to the Gemalto PLS8?

2 Likes

I think that free firmware on 4G and 5G phones is not only possible, but that it should be and is inevitable to happen eventually. Many ham radios on the market allow ham radio operators to easily transmit on police, fire, ambulance, and several government frequencies. Sometimes your new radio will transmit anywhere, right out of the box, new and un-modified. There are probably hundreds of thousands, if not millions of ham radio operators in just the US alone. On top of that, you don’t need to show any license when buying these ham radios. You can walk in to many local retail outlets, buy these radios anonymously, and leave without anyone knowing what you bought. So the controls on the potential abuse of the technology is not found in keeping people out. That doesn’t work and is bad for society in the long run if free software is not allowed by law.

Using ham radios as just one example, you can get yourself in a lot of big trouble very quickly if you cross certain lines. After I modified my two-meter handheld radio several years ago, I grew over time, to be afraid to use that radio at all. With no safetys, it’s too easy to accidentally transmit on police and fire frequencies. I used to also monitor the main fire dispatch channel. I was never stupid enough to set the CTCSS tones to open the fire department repeater input frequency to any transmissions that I might make and to set a frequency offset to allow me to transmit in to that repeater. But a few times I accidentally switched channels on the radio and somehow ended up giving my callsign by mistake over the fire dispatch output frequency (no big deal as without a repeater, you won’t get out very far). Even so, after that I always kept looking to verify the frequency every time before I keyed-up that radio after that. I was a bit relieved to toss that radio in to the landfill a few years later because of several unintended issues that came up after I modified that radio to allow out-of-band transmissions. I couldn’t reverse the modification at the time because the original modification involved using tweezers to scrape a 0402 sized (near microscopic) capacitor off of the circuit board inside of the radio. Without access to a $5K surface-mount solder rework station, I couldn’t replace that capacitor. Most people don’t realize how ignorant they are about their own assumptions. Using that same radio at a hamfest a few years later, a presenter there said to me “key up your radio and unkey it, but don’t say anything”. So I did. Then he gave me my own callsign. He was presenting on radio fingerprinting technology. No two oscillators are exactly the same. The oscillator in my radio had already been catalogged and attached to my callsign in his database at one point as I had previously used the local ham radio repeaters, giving my callsign each time. If I had intended any mischief say, by hasassing the police on their own radio channels (a stupid thing that I have never done), the police could have provided my radio fingerprint pattern to a repeater owner to retrieve my callsign. I didn’t even know about that technology until a few years after getting my license. The point is that most people who are up to some kind of no good will be caught because there are invisible traps. Most people who are smart enough to hack in to bleeding edge technology either know they will be caught, so they don’t do it, or they get caught. When you do get caught, those penalties can be severe.

So as with ham radio equipment, the honor system actually does work. I guess it’s not exactly the honor system in one way. It can also be like there is a gun to your head if you mess-up. But in exchange, there is always a path made available that you can follow if you want to experiment with the technology and build your own stuff, complete with metaphoric street signs, guard rails, and forgiveness of reasonable mistakes. That is what ham radio does. One unofficial rule is to not modify your commercially-made radio hardware. If you do, you’re on your own. But if you build your own radio from scratch, some forgiveness might be in order if you accidentally violate a rule in the learning process. If you’re capable of writing 4G or 5G cell phone opensource firmware and make an honest mistake, you’re more likely to be offered a high paying job than anything else. So the technology should be made public as much as is possible.

The cell phone carriers probably don’t want people experimenting on their networks. So if Verizon wants to share intellectual property with Samsung so that Samsung can build cell phones to work on the Verizon network and they don’t let anyone else in on those secrets, it will be really difficult for any opensource firmware to be written for that hardware. That is the biggest problem. But if you respect licensing requirements, you should be able to write your own firmware. Finding the radio chip provider who will share their datasheets with you without making you keep that information a secret might be the hardest thing to do. By nature, that requirement to keep those secrets would prohibit you from publishing your opensource code. So there is a lot of hoops to jump through. But opensource cell phone firmware should be possible.

4 Likes

Their customers don’t want people experimenting on the carriers’ networks either.

Reportedly, the use of unauthorised mobile network repeaters is a bit of a problem in my country. (It can be a death spiral, with someone having poor mobile signal, buying a cheap dodgy unauthorised repeater from some overseas supplier, using the repeater, making everyone else’s signal worse, … REPEAT. :wink:)

However obviously the rules vary from country to country, and within one country, the rules vary from one frequency band to another. I am talking in the previous paragraph about spectrum that has been licensed to the carrier by the government for the carrier’s (exclusive) use in the mobile network and so naturally both government and carrier are unamused and unforgiving about unauthorised use of that spectrum c.f. licensed for non-exclusive use c.f. available for unlicensed use with a class licence provided that you stick to the power limits and other restrictions (e.g. WiFi).

1 Like

Beside of the problems to get to the point where we have an open source firmware … as consumer I don’t care if this is legal or not. Who should control it as long as it does its job?

On the other hand there is a even cheaper Amazon-like company with worse products and all risk to its customers (I think it’s called Temu) that sells a lot of untrustworthy stuff like things that send on military frequencies as I read. It’s so easy to import forbidden devices. In contrary open source firmware is harder to abuse (you need some skills to code and compile or at least to flash your device) and it has important reasons to exist. There is no real reason to not allow open source firmware modems. It even can have regulations like “only licensed people are allowed to apply MRs” etc. The most important part is, that we can verify security. I think it’s possible on common smartphones to get hacked via modem exploits.

1 Like

What carrier are you with?

1 Like