China recently announced that it is creating a “quantum communication satellite” to provide a means of transmitting data without the risk of eavesdropping by hackers. While this is an interesting concept, I see it as more of a novelty than a practical solution to the eavesdropping problem. In other words, I don't think it will ultimately be successful.
The problem with this sort of “ivory tower” security model is that it assumes that everything works as planned, and that every link in the chain of data custody can be trusted. Unfortunately, that is seldom the case.
For example, there is a high likelihood that this satellite communication system will be connected to a traditional, terrestrial network at some point. If that happens, the game is over – even if the data cannot be intercepted from the quantum satellite, it will likely be vulnerable while it sits on the terrestrial network.
Sure, traditional networks can be protected, such as designing an “air gapped” network structure to prevent cross-contamination of networks, but I predict that the limited utility of such an isolated network will ultimately cause China to make decisions which will result in a less secure chain of custody for the data. Often decisions made to satisfy a one-off usage scenario can introduce weaknesses into the system that can never be removed – after all, once you provide the means to move one tidbit of data out of a secure system, you provide the means to move other data out of that system.
However, history has shown us that the biggest weakness in this communication strategy is the same one that plagues every network on earth: Humans will secure and use this network.
When humans are involved and attackers find the technological safeguards too difficult to overcome, they will resort to other means – they will attack and exploit the human weaknesses in the system. After all, attackers routinely get what they want by taking advantage of human factors such as errors, carelessness, and greed.
If you want to increase the security of any network (without building a quantum satellite), there are steps you can take:
- When you design a process, a network, or a security strategy, remember that users often go to great lengths to make copies of data and move them around – frequently, with good intentions. This makes the data even more difficult to secure and contain, and increases the likelihood that an authorised user will expose the data to an unauthorised user. Monitor both data at rest and data in motion to ensure that you know how data is being used and how it is moving through (and possibly out of) your network.
- Use data classification, coupled with network and data segregation to limit the loss when your network is compromised. Using these methods, you can greatly increase the difficulty for hackers to acquire your data, and limit what they can see when they are on your network. This is analogous to the adage that you shouldn't put all your eggs in one basket – if you break up your data into discrete enclaves, you can prevent an attacker from easily grabbing a dump of all of your data at once.
- Use encryption to limit the value of your data if someone misappropriates it. A lot has been said of encryption, but it is extremely effective in making stolen data useless, if is used effectively. Study encryption best practices to ensure you're doing it right.
- Understand and tightly control all access to your network and data – including all users, applications, and processes. That includes authorised third parties, as well.
These are just a few of the basics, but they can make a big difference in your security.
Of course, another big issue with China's plan is that they've announced this new quantum communication channel to the world. That means it is now on the radar of hackers and nation-states, and you can rest assured that these potential attackers will be looking for ways to compromise the security of this system.
That brings up another lesson: If you want to keep something a secret, issuing a press release about it may not the best strategy.
Contributed by Dwayne Melancon, chief technology officer, Tripwire Inc.