The unmonitorable phone, is it possible?

Hey lain, i've wanted a phone for a while, but most of my computing revolves around the PC so i'm pretty illiterate when dealing with phones and android. my threat model is >interceptors >carrier >google My plan is to simply get a phone compatible with whatever custom rom i find suitable (get rid of google) then install some OTR software (get rid of carrier spying and intercepting risk), is this enough or is the more into it in securing your phone? what do lainons run on their phones? Another fun question, is it possible to to counter the GSM based location tracking?

>>5753 Your best bet is to to snip the antenna and do everything through the IP stack. Then you can worry about securing the software. If you really want a true cellphone, you're either going to need to reverse the DRM infested radio interface layer or stick a custom radio into your phone.

the other lainon is correct >Another fun question, is it possible to to counter the GSM based location tracking? hypothetically, but are you willing to risk years in prison to do so using some methods? (certain methods also make the phone unusable) the phone cannot to my knowledge be tracked if the battery is out of it, since it cannot transmit/receive carriers were essentially designed from the ground up to gather data. this even more so as 5g becomes commonplace (the carriers call it "better content delivery" aka targeted advertising in more than just your cell phone life)

>>5755 >but are you willing to risk years in prison to do so using some methods? well you can't just leave me hangin', tell us about them

>>5755 >but are you willing to risk years in prison to do so using some methods? Legality of methods depends on where you live.

short answer: no, you should consider the hardware compromised. The DSP chip probably has DMA access to your phone's entire address space and it probably has a backdoor, and it has the ability to send info out however the manufacturer wants it to.

>>5759 How the hell can backdoored hardware work if the firmware is flashed?

>is it possible to counter basic laws of physics No. Avoiding triangulation is not using radio.

How about building a phone? I have visions of SDR hardware, a directional antenna to target one tower, and a budget of a few hundred dollars.

so you could technically just stick phone in a faraday cage and only use when needed. that's really the best you're going to be able to easily do

>>5763 >so you could technically just stick phone in a faraday cage and only use when needed. see that's the problem, i already thought about that, but what if someone needs to call you?

>>5753 Nearly all closed-source apps and some open-source apps will have their own spying capabilities. So be very hesitant to install any extra apps. Beware of spying through the browser too. >>5760 The baseband firmware is separate from the operating system and installing a custom OS does not replace it. Also, a backdoor can be implemented entirely in hardware without using any firmware. Anything that can be done in firmware or software can also be done directly in hardware at the cost of not being able to update it later and of making the development more expensive. There are persistent rumors that the NSA sends chips to device manufactures with orders to include them in the device and keep it secret. For example, see: http://www.loper-os.org/?p=2433 >>5762 It's illegal to operate cellular equipment without regulatory approval and also against the carriers' terms. But, there is an open-source GSM stack implementation available. It could be modified for use on the public networks at your own risk. See: https://osmocom.org/projects/baseband/wiki/LegalAspects#Using-modified-phones-on-carrier-networks >>5765 Pager service still exist and allows receiving without revealing your precise location. You could arrange for your contacts to send you a page telling you to turn on your phone when needed. Or, set up a VOIP server, give out that number, and configure it to page you when it receives a call.

whatever happened to being able to ping data off iridium constellations and have it be re-transmitted? is there still a way to do it, or is it too labor intensive? >>5756 doing so involves a number of either signal obfuscation methods not limited to either jamming or somehow spoofing your location using a variety of methods. the problem with that is it either requires a number of people or a means to appear to be in multiple locations at once, or spamming junk signals all around. anything that deals with spoofing locations (to transmitters) in many places (or jamming) will end you in prison getting pounded in the ass, because it is no less than a minor act of electronic warfare against a civilian population (or so you would be labeled) which is more or less how many of the laws are written >but the signals threw off legitimate commerce/transportation so that makes the broadcaster a terrorist

>>5766 most devices are assumed to be hacked from the factory, since there is virtually no security between the manufacture and the distribution of said devices, people have been known to walk into laptop packing plants and install viruses onto them

>>5763 there are products that do this. They're expensive but they definitely work. Thing is that still wouldn't protect you from most attacks, just gives you a little extra piece of mind

>>5760 the DSP isn't controlled by the phone firmware, it's a seperate computer.

>>5761 what about directional antenna connecting to just one station?

i was running copperhead os with apps like aimsicd, yalp store and newpipe. But since copperhead os is no more i dont know what to do. im not a fan of other roms like lineage os. i guess best option is to use iphone and restrict my mobile use. i heard the original copperhead os developer planning to make something new though. so i will keep an eye on that too.

>>5779 Apple is part of PRISM, you're still just as fuarrrked.

>>5779 > copperhead os is no more Yeah it really is a shame > i heard the original copperhead os developer planning to make something new Any source on that? It seems like he went dark, is there any way to follow the guy?

>>5781 nvm he apparently still tweets: https://twitter.com/DanielMicay/

>>5781 here it is. rattlesnake os. https://www.reddit.com/r/CopperheadOS/comments/8ugj2i/rattlesnakeos_update/

>>5753 They use triangulation. So say your phone is within range of three (or more) towers, they can basically map out an area of where you could be based on these three points and your cell speed (i.e. which can be roughly converted to a distance). You can't really avoid GSM tracking without ripping out the chip.

>>5753 If your threat model includes carrier, you need a radio that works point to point LPWAN is your best bet Low power, long range Something like LoRa It's low bandwidth, but unregulated spectrum with a range in kilometers You're not going to stream video on it but you could order a pizza from a nearby restaurant with no middleman

I had and still have somewhere a GSMK CP500, their solution I believe was to quarantine the baseband and have a built in killswitch to shut it and a firewall to constantly monitor it.

it seems like the main issue here is that most people want to be available to others under a single phone number (and through an insecure telephony system) everywhere they go. there are cellphone towers and satellite relays everywhere but In order to interface with them you need a SIM card (or digital SIM) with even more stateful identifiers on it. It would seem to me that addressing this issue somehow would be the only way to remove the threat of carriers. either through compromises such as 'i'll only interface with the telephony system in areas with public wifi' or through adding complexity (i.e. through proxies which will slow everything down or multiple phones)

>>5892 Is it necessary for cellular service to be bidirectional all the time? Is that how the network knows which tower to route your calls to?

>>5948 Yes and yes. The nearest commercial alternative to this arrangement would be pagers and those are receive only.

>>5949 I was actually thinking about pagers when I made that comment. I knew they're one direction but I figured they operated off of conventional cell towers (which is apparently not the case). Correct me if I'm wrong, but the options for pagers out there these days have extremely limited coverage, right?

>>5951 >options for pagers out there these days have extremely limited coverage, right? Iridium satellite pagers exist. Every year there's a c3 talk about a project to intercept them with sdr so they must still be in service. That might be worth looking into.

The only unmonitorable phone is pic related.

>>5952 >>6031 IIRC before the lainchan split, there was a discussion about using HAM radio to send data, essentially negating the need for cell service

>>6033 There are a few projects that use 2.4GHz transceivers for mesh networking/bypassing last mile ISPs. I think sending data streams over a ham radio is illegal.

>>6035 Data transmission is 100% legal. It's just encoded information. What's illegal is encryption and more broadly codes "obscuring the meaning of any communication."

>>6036 so basically >internet over HAM = fine >encrypted internet over HAM = super bad criminal activity

>>6045 It is an old law where people thoight that spies would use HAM and CB for encrypted messages Also it is against the law to record any of transmissions but if you don't give them evidence to convict you, that law is literally braindead stupid one to ignore

>>6046 Isn't there an issue with the HAM radio community being full of miserable cunts? I've heard that they'll go out of their way to track people down and tattle on them to the authorities. I can only imagine a sniff of encrypted data would be enough to attract them.

>>6047 >I can only imagine a sniff of encrypted data would be enough to attract them. implying they would know the difference between encrypted and encrypted data if it was encrypted prior to conversion to an audio signal, if all you did was broadcast junk data all day they could not imply it was encrypted (maybe you were generating random signals to test a receiver)

>>6048 >if all you did was broadcast junk data all day they could not imply it was encrypted Depends how fair the authorities want to play it. Even Western governments lose their soykaf over encryption. Look up key Disclosure Law. You can't prove this random looking data on your harddrive is not encrypted so you go to jail. It's not impossible that they would treat data transmissions the same if they wanted to. >(maybe you were generating random signals to test a receiver) Maybe for like 10 minutes. If every single day you transmit nothing but "junk" then you're going to need a better lie to convince the jury. The judicial system wasn't born yesterday.

I'm a windows pleb with little knowledge of how any of this soykaf works. 1. I want to get a new phone. If I want to be more secure, would it be a good idea to swap a smartphone for devices that fill the same purpose? >Getting a less smart, smart phone to make calls, a DS for games, etc. To what extent could I do this before I'm carrying around a backpack full of devices?

>>6144 *soykaf

>>6144 not necessarily. Literally any phone you buy will have some kind of tracking capability.

>>6144 >1. I want to get a new phone. >If I want to be more secure, would it be a good idea to swap a smartphone for devices that fill the same purpose? I suppose, if you take it far enough. You could get a flip phone and remove the battery when you're not using it, as an example. You could always do what I do and just not have a phone at all, though. I don't need one and if you must have a phone you could just get a landline put in, which I don't believe will be able to turn on and spy on you as a cell phone easily can be made to. >>Getting a less smart, smart phone to make calls, a DS for games, etc. >To what extent could I do this before I'm carrying around a backpack full of devices? It depends on what you're doing. A Sansa Clip with Rockbox is very small and convenient. Would you really want more than a music player and perhaps a handheld game?

Get the Librem 5 when it's out: https://puri.sm/shop/librem-5/

>Another fun question, is it possible to to counter the GSM based location tracking? No. Your carrier needs to know which tower you are connected to so that they can route your line correctly. There is nothing you can do to stop them from using this to track you. It seems theoretically possible to counter fake LE honeypot towers, but I'm not sure. IMO phone privacy/security is all about taking this fundamental compromise and figuring out how to mitigate it. For instance if nobody, including your carrier, can find out who you are or what you are using your phone for, there's a limit to how much damage they can do.

>>6150 Not untraceable. If you've been reading this thread you'd know that there's basically no way to counteract the tracking of a device connected with a SIM: >>5755 >>6155 >>5892 (Don't mean to sound like a dick here, just saying).

>>6158 >Not untraceable Only while the cellular modem is enabled. You can leave it disabled with the hardware switch for its entire life if you wish. Also, the modem is located on the USB bus instead of the RAM bus, which drastically reduces the attack surface of having a cellular modem on a device. If compromise via modem is your concern, but you must use a cellular device, the Librem 5 seems to be the best option.

Can apps can have hidden permissions and activity even when turned off? Hard to believe that apps won't do some background spying after I turn off their suspicious permissions, especially on android. Many android apps have around 30 permissions that are visible and can be turned off, but on ios there are few and are they really off and can there be hidden ones? Google apps can even turn them back on.

>>6198 There are some permissions that are always enabled on android. For example: Internet access, info about network/wifi/bluetooth, or the nfc reader. Your clipboard can be read at any time, even when the apps are in the background. Permissions control has gotten better over the past few years, but I would still consider permissions broken on android. My pi-hole reveals a lot of background activity from apps, even if I am asleep and the phone hasn't been touched for several hours. Some more info here: https://developer.android.com/guide/topics/permissions/overview

>>6199 Yes, some background info leakage can't be turned off easily. But what I'd like to find out is which os is safer for having social media or messenger apps that have privacy issues or to safely use app like keepass. It seems like ios apps are more securely sandboxed than android if I turn off all the permissions for each app but I'm not sure. On android there seems to be more uncontrollable background activity and more dangerous permissions for some apps than on ios.

>>6166 At that point why would you even buy a phone? Sure you can mitigate the somewhat remote possibility of compromise through the modem, but you can't change the fact that whenever you are connected to a cellular network you are being tracked. This would still be a fundamental compromise even if you got a USB modem for your PC and somehow got it on your carrier's network. If you aren't going to use GSM service, say all you need is a phone number, then there are plenty of soft-phone options you could set up on any old laptop or tablet. The point is that you have to compromise the key feature of a cell phone, the one thing that makes it different from any old computer, if you want to be untraceable.

>>6201 I mostly agree, but I do think the Librem 5 has an extra edge due to form factor, UX, the other hardware switches, and the care they're putting into the hardware production. Personally, I would ideally set the device to use VoIP service and not use the cellular modem except for emergencies, which would negate needing to have a secondary device in a faraday bag when traveling. Beyond the cellular modem, making any device dedicated to communication untraceable takes a lot of work. Wifi MAC address needs to be randomized regularly, multiple phone numbers used that need to be isolated by use-case and will, in worst case, be compromised by the aforementioned... etc, etc. If your threat model is really that high, you should know better than to invest in any cell phone for anything but a brief moment, if at all. I'm sure the rest of us don't need to go that far, and are just trying to satisfy our paranoid hobby.

>>5753 Yes, but it wouldn't really be a phone afterward.

>>5753 First you need to buy the phone out right with cash/anonymously. make sure the Sim card doesn't have your name/info linked to it either. Make sure you have a phone you can root. >interceptors Do you mean the stingrays the cops have or just randoms intercepting cell data. https://opensource.srlabs.de/projects/snoopsnitch sorry, that the best I got. I do think this is an issue that should be taken more seriously though. FARADAY CAGE? >carrier First make sure to buy you phone with cash and don't use a Primary network carrier. Most require more personal info and are used to doing credit checks. some will let you do prepaid deals with minimal info however I don't trust them and they give soykafty deals. use a third party carrier that piggie backs off there networks. (ST,Ting, ect) typically these are easier to give less/false info too because they just want customers and data collection is less important in there business model. Mind you there still selling any info you give them. After that if you manage to get a phone thats not linked to your name/info. There still is the cell towers tracking you. Thats kind of a bitch. Basically Use some sort of Faraday cage (I use an ammo can). >google Make sure the phone you buy is root-able, that means not just what type of phone (nexus 5) but the literal model number. if your buying from a carrier (use ebay/amazon) make sure you lookup with model phone they sell, some lock down the boot loader. I've made this mistake thinking I was buying a phone easy too root. Also make sure your phone has a decent custom rom to load. https://wiki.lineageos.org/devices/ when you get the phone ROOT IT then install a custom rom. get rid of GAPPS. I Listed LIneageos as a reference, However it is not google free by default. You can make it google free though and it supports a fuarrrkton of phones After that use the /etc/hosts file to block google and google related urls. There is an app in fdroid for that, although I forget the name. or use ADB https://www.theandroidportal.com/edit-modify-hosts-file-block-websites-android-root/ ALL of what you asked is possible, NONE of it is easy.

>>5753 >The unmonitorable phone, is it possible? yes. center it well in a copper box or well-made copper-lined bag. >my threat model is >>interceptors orbot (tor), afwall+ (firewall), overridedns, adaway, protonvpn (free tier), >>carrier take out the sim card and use it via wifi over tor/vpn. linphone for calls. tor browser for surfing. conversations for im. >>google don't install gapps. don't have a google account. >what do lainons run on their phones? lineageos with addonsu, no gapps, no xposed framework, lots of system apps frozen with performancecontrol. files encrypted with openkeychain or even just 7zip. logs and other vulnerabilities deleted with a script. there are still unknowns, radio partition, hardware backdoors, blobs, who knows what. but you can do a lot to close some of the more obvious gaping holes. and nothing is going to get through your handy copper container, since it's a miniature faraday cage.

>>6264 >how to make an unmonitorable phone >get rid of the thing that makes it a phone and use it like a soykafty computer the only advantage here is the form factor. and since you can only use it on wifi, that part isn't even all that important. if you're not going to use cellular service, why not just get a tablet or something? hell, you can probably do all this with PSP homebrew. no need to root and worry about trying to lock everything down on untrusted hardware (well, not to the same degree as with a mainstream smart phone).

>>6267 true, but you can put the sim card back in and use it normally as a phone at any time. afwall+ comes in handy here for deciding which apps are whitelisted for cellular, which for vpn, which for wifi, which are blocked completely from all communications, etc. it's kind of amazing how few things need whitelisting for the phone to be fully functional. just ntp and download manager, along with anything else you want.

Any of you know or use https://cryptocloud.mobi ? Would you recommend it? I am concerned with privacy and security, as everyone should be, and need a smartphone. I have looked at projects like RattlesnakeOS and LineageOS, both help to go around Google a bit but not so much. Android has many security flaws and 0days. iPhones seem to be a tad better in that regard, not sure I want to jump on the proprietary train though and they've been fined in Italy for slowing down older models with updates... The market for a private phone seems fuarrrked up in all kind of ways, so I'm open to your suggestions. If I need to pay a premium or use a proprietary solution I'm open to it if there's proof that the security and privacy parts are great.

>>6299 I can't vouch much for the product itself, but it smells like bullsoykaf. Any product that claims to be unbreakable/unhackable, impossible to wiretap and jadda jadda is something I'd be wary of. They make some very bold claims for something that appears to simply be customized android with a sim card. I guess the sim card might give some extra privacy though.

>>6303 also, what's allowing them to give away a free unlimited data plan?

>>6299 That's for drug dealers who are too stupid to put Lineage and a few apps on a phone. It's laughable that you would even ask. >>6307 There's nothing free about this. You're going to pay an arm and a leg for a phone up front and probably a subscription as well.

>>6308 >Laughable that you would even ask >Having this attitude towards other people >ever >>>/trash/ with your elitism

>>6309 No. That site is a mountain of a snake oil. There is a place for ridicule and this is it.

>>6310 Alright, well I see what you're saying. Point taken.

>>6310 The PDFs are not detailed enough, but just enough to think it would be possible to do. Really locking the phone doesn't seem unrealistic, if you modify the hardware and kernel. Add onto that full disk encryption and it should be pretty secure. Choosing a ROM and apps that I can do, compiling the kernel gets tougher and really I don't want to mess with hardware. Same for a SIM with encrypted communications, sounds pretty easy to do. Really everything sounds like snake oil when it pertains to security, the defending side will always be vulnerable. I'm just getting desperate that we can't even find a device to communicate privately and securely...

>>6313 And as a side note, among the snake-oil products they did a better job. Unhackable etc are just marketing bullocks but it doesn't that hacking their product is not harder.

>>6313 IIRC there was a mobile phone encryption system that was essentially "Unhackable" and then someone realized that you could solder a wire from two other parts of the phone and essentially negate all security measures. I dont know if there is a way to prevent a determined person from doing something like that, but if you lose a phone and dont want someone finding all of your dick pics, I guess it is ok

>>6314 >>6313 Right, so, let's begin, shall we? Firstly - the constant use of the word "encrypted". BUY AN ENCRYPTED ANDROID. EVERYTHING IS ENCRYPTED. YOU'RE SAFE. I agree with using crypto all over the device - but they literally cannot promise all app dbs are encrypted, for example - some apps just will store their soykaf in plaintext. It happens. These guys are marketing and selling a solution you could make yourself (ie, by using the built-in FDE and only using apps with crypto). I've also never heard of Knox, but visiting the website tells me it's some kind of kernel hardening/secure enclave/secure boot package. That's cool, but probably not the main threat when it comes to phones. Honestly, I also trust Samsung about as far as I can throw the entire employee-base of the company. Corporations are retarded and will backdoor their own products - see RSA, Juniper, Cisco, Huawei, etc. That's probably unjust of me, but still. Big promises about a security product solving all your woes make my skin crawl. Trust me, if any of them really worked, they'd be so widely deployed you wouldn't stop hearing about them. The ones that do are so fuarrrking detrimental to usability that using them is generally considered the pursuit of autists (see Qubes). Also, lastly, these cryptocloud chucklefucks keep talking about PGP - anyone who knows soykaf about cryptography these days would prefer to avoid PGP, and definitely wouldn't recommend it to a layman. Nothing within it is broken, but it and all the implementations contain a large number of weird design choices and clunk that identify it strongly as being a relic from the 90s. It's a bit like showing up to a gunfight with a Model 1855 - you feel fuarrrking cool as soykaf, but it's not exactly the easiest thing to use, and consequently, it's actually not very clever to use it. I'm sure the CRYPTO NERD CONGREGATION will flock in to defend PGP, but the fuarrrking creator doesn't use it.

>>6324 Fair enough about PGP. Knox basically mean putting a custom ROM on it is tough. Really. I find your sperging out on encryption marketing gimmick is going too far. We know that to protect data you need to encrypt. Encrypt data, encrypt the disk, use encrypted protocols... Repeating it for a marketing effect sounds reasonable if you're able to pull off what you say. Anyway, seems like we're fuarrrked.

>>6329 My problem is that to encrypt all the data they'll inherently only be installing a subset of apps you could install yourself. Nobody's fuarrrked. The Signal protocol, or whatever it goes by this month, is an excellent crypto protocol. I disagree with certain implementation choices (y u phone number to identity, moxie), but it's probably the best example of a functional, easy-to-use, secure cryptosystem.

>>6324 idk man, criticizing PGP for its flaws is one thing, and you aren't wrong there, but so called "replacements" tend to be fragmented and limited to some particular use case. in any particular niche, there will probably be better options than PGP (e.g. OMEMO, OTR, and signal's algorithm for e2e encrypted chat). however if your use case demands flexibility, then i'm not aware of any alternative cryptosystem that functions as a de facto baseline/fallback the way PGP does. to extend your metaphor, PGP is the gun you bring if you don't know what you're going to be hunting, how long you're going to be out, or what kind of ammunition will be available. if you had more control over the situation, you could pick a better gun, but PGP at least ensures you'll have something halfway functional in the majority of situations. ...and unlike guns, you can later bootstrap to something more appropriate.

>>6332 Signal does nothing in the way of physical security though. Guess phone disk encryption is the best choice we have.

>>6336 The problem with FDE is that you routinely decrypt it for normal use. Mounting encrypted containers on fly mitigates the problem of mixed use devices, but I don't think anyone has that working nicely on Android.

All phones post-9/11 are bugged

I found a setup I think you guys might like: Pixel 2/3 + GrapheneOS VoIP Number (soykaf calls) Riot.im (comms) VPN Karma Go with Refuel plan (internet/calls outside of WiFi) A few Mint Mobile trial SIM cards(when non-VoIP number is needed) And maybe change the IMEI whenever a SIM card is used and discarded. Thoughts?

>>7694 >Thoughts? Same as any other "secure" phone. https://www.devever.net/~hl/nosecuresmartphone

>>7695 Ugh, dude... First, It's insane to assume that this applies for untargeted surveillance. Yes I'm sure the NSA/Mossad have a dozen 0-days exploiting the baseband chip, but if those were used for mass dragnet, that resulting ubiquitous and unexplained traffic would get noticed immediately. Your argument is like saying you shouldn't drink water, because you could theoretically drown. Second, most modern phones DO have IOMMU. Third, why the hell are you on the grid? That's all the food you're gonna get, troll.(Always argue in good faith and avoid using personal attacks.)

>>7697 >First, It's insane to assume that this applies for untargeted surveillance. It sounds like you didn't read the OP. >>5753 >my threat model is >carrier Carrier can push arbitrary code to baseband whenever they want. Baseband processor has unrestricted access to RAM. So nothing you do on the application processor can protect you in this thread model. >Second, most modern phones DO have IOMMU. Link one. >Third, why the hell are you on the grid? I'm behind 7 proxies.

>>7698 https://www.reddit.com/r/CopperheadOS/comments/6wtul0/on_sensationalism_about_basebands_all_future/

>>7699 >You can disregard anything portraying the baseband as being unique in that it runs a secondary operating system or has direct access to memory. It's something that applies to components like SSDs, WiFi, batteries, touchscreens, fingerprint readers and many others. Those components all effectively run a secondary OS and many have direct access to some of the memory including the GPU, image processor, WiFi, Bluetooth and cellular baseband. What makes the baseband processor unique is the long range radio attached to it. It's basically always in contact with your carrier, ready to receive commands and exfiltrate data. >For components that aren't part of the SoC, it's the responsibility of other vendors to set up IOMMU isolation rather than Qualcomm. The quality of the isolation will vary. It's much more complicated that a black and white issue where there's either an IOMMU or no IOMMU. There can be an IOMMU with very tight restrictions providing proper isolation or a setup where the IOMMU is effectively not doing anything and permits access to all of the memory. Determining that requires real research. Even the CopperheadOS guys couldn't figure out if their IOMMU is configured correctly. I believe this was already covered by the first link from >>7695