Something like the Hong Kong protests for example. If a protestor wanted to use his personal computer on a network that is monitored, how can he change its properties enough so that it is not traced back?

For example, I would assume the MAC can be used to try to identify the user of the device. If the same MAC address was used on the person's home network, and the network that was being monitored, the user can easily be identified. So using a throw-away external wireless adapter to change the MAC might be a good idea.

The main concern is not about hiding the activity done on the network, but rather hiding the fact that the same device was previously used on another network, to avoid the person using it from being identified. Basically, how can the device be made to look like a completely new or random device with no way to trace it back whatsoever?

  • MACs don't route, so only the first router will see your MAC where it will be stripped out. Tracking and identification involves a myriad of functions that could be, and probably is, a book: IPs, Cookies, FingerPrinting, Browsers, Plugins, Operating Systems, Networks, Protocols, GEOtagging, Bluetooth, Wifi Beacons, Cellular, Others. A specific device might be adapted, but there is no generic answer. May 4, 2021 at 16:49
  • 1
    @user10216038 if the home router is provided by the ISP, then MAC tracing can be done (and has been done in the past)
    – schroeder
    May 4, 2021 at 20:08
  • @schroeder - MACs are not part of the layer 3 IP protocol, at all. They are not present. If you have a reference to how MAC tracing is being performed, I'd love to see it. May 4, 2021 at 22:24
  • @user10216038 I never said they were. But, as I mention, if the ISP owns the router, then the device has access to layer 2 data and can trace that data.
    – schroeder
    May 5, 2021 at 7:00

3 Answers 3


Short Answer: It's not possible to hide from your ISP.

ISPs will not be giving free internet access to everyone on the streets, so they have to know something about you to know if you are their customer or not. You must have a login and password, or your router have one, or they run a cable from their router inside your home. So if you access something restricted, and your ISP gets a letter asking to identify who are using that IP, that is right on their records.

Now if the ISP have information on your devices, that depends on your ISP. They usually don't, because the device does not matter for them, and they don't need device information to have information on you.

If you want to be fairly anonymous, buy some popular Android phone, factory reset it, don't activate Google services on it, and use it without a SIM card, and only connect to open wifi. Factory reset it from time to time.


Like the previous answer mentioned, MAC addressees are strictly limited to the layer 2, which means that they will never get routed on Internet out of your personal network. However, if you feel unsafe about the security of your own local network and want to modify the hard-coded properties of your network interface card, it is possible to use MAC spoofing techniques on your operating system. There is also some software tools to do the same thing with a graphical interface.


You can easily change the MAC address reported by most network stacks, and however that address is not routed, per se, past your network gateway.

What could give you off might be:

  • the software on the device. Browsers and the like all have a specific signature, and the Electronic Frontier Foundation even has an "estimating" gadget to tell you how unique your "fingerprint" is.
  • hidden software on your device (i.e. spyware).
  • your navigation pattern. Most people are not aware of this, but they usually tend to adhere to a broad navigational routine. This might be enough to restrict a search using Big Data and a powerful access to an area's networking infrastructure (i.e. you need to be the government or a large ISP).
  • the network itself might not be trustworthy. It might be easy for a large equipment manufacturer to embed backdoors and surveillance code, and even implement an "internet over the internet" using, for example, damaged IP packets that would be dropped by normal routers, but have significance for devices of a specific manufacturer. This of course would require a large equipment manufacturer to be essentially the property and toy boy of a large Nation-state. Then, seeing the significant strategic value of such an operation, the manufacturer might be directly or indirectly subsidized by the State (e.g. with large tax cuts on its suppliers, so that they can afford lower prices), driving the prices down and letting the compromised equipment flood the market, until most cheap devices are running it, or running on it, or communicating through it.


Basically, how can the device be made to look like a completely new or random device with no way to trace it back whatsoever?

For large values of "whatsoever", you need a new device. One that has never been used for any other purpose, and whose usage is itself randomized using some reliable technique (dices?).

Otherwise, you can run a virtual machine inside a reliable hardware, and reassign new information - MAC, hard disk serial number, DMI etc. - every time. Maybe, by comparing a new, virgin VM and a new, virgin VM with changed information, you can come up with the offsets where that information resides, and at that point you can script a complete machine regeneration using a blank template and random values poked at those offsets; creating a "new" machine then becomes just the matter of a few clicks.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .