Jamming Attacks on Bluetooth and ZigBee technology designed in NS-3

[Nusrat Fatema]

Hello All,

I need help regarding the subject stated above. Is there anyone who can help me to implement this on NS-3? I am a new user in NS-3.

For Bluetooth and ZigBee there is no such modules available. Is it possible to help regarding this issue?

Also I need to make an energy consumption model for the Bluetooth jammer.

The website like:
http://www.nsnam.org/wiki/index.php/NS-3_wireless_jamming_model
has already been searched.

Please kindly help me out. Thanks in advanced.

[Tommaso Pecorella]

Hi,

ZigBee and Bluetooth are extremely complex protocols. Simulating them is no easy task.

Take ZigBee. There's the MAC/PHY (802.15.4), then the routing protocol, the addressing, the upper layers, etc. Bluetooth, even more complex, as it depends on the version.

Jamming attack. if it's "simple" jamming (e.g., white noise spamming over a wide frequency range), there's the spectrum model (check the Microwave Owen model). If you want a more specific jamming (e.g., one trying to disturb specific packets by following the Bluetooth channel jumping), you'll have to model exactly the attacker's capabilities, which isn't easy either.

As another example, try implementing a jammer with the UberTooth module. You're going to have a bad time (it seems easy, but it's not).

Summarizing... my suggestion is: you wanna really evaluate how easy (or hard) is jamming a protocol, don't simulate it. Do it.

Cheers,

T.

PS: I work with ns-3 a LOT, but I also work with real devices. And sometimes you have to use the real stuff.

[Nusrat Fatema]

Hello Tommaso, 

Thank you for your reply. I know it is one of the hardest project which I have got as a Masters thesis. And I have only 2 months in hand. I am completely new with NS-3. That is why I really need help. 

My task is to implement Constant, Deceptive, Random and Reactive jamming on the Bluetooth and as well as on ZigBee technologies. 

According to the demand of my thesis, I won't be able to do that on any hardware/ testbed. I need to simulate it on NS-3 and find out the result that how well it can jam the technologies. I have seen people are doing on Ubertooth. But that is not my work. Anyhow I need to simulate it. 

I am an Electronics student I would love to work on real things instead of using programming language in which I have less knowledge. 

I need to find out the hopping sequence of the user and jam accordingly. 

Can you also tell me how do I download all the models from NS-3 website and keep them in the NS-3 folder and run those new downloaded modules? 

Please suggest me something. Looking forward for help. 

Is it okay if I get your email id to ask there properly? 

--

Nusrat.

[Tommaso Pecorella]

Hi,

I'd rather use the group, as it could be beneficial for everybody.

About your task, I'm deeply sorry, but I strongly believe that it's next to impossible to do in 2 months.
The reason is: there is no bluetooth stack in ns-3.

What you could do (but, I repeat, it's extremely far from evaluating for real such a thing) is to develop a new module with just the bare bone thing you need: the PHY layer and the channel hopping.

Then you can use the Spectrum model to simulate a jammer.

However, even if you do this, the problem is: what you will find out ?

The result of a jam is always 100% success, if you can inject enough noise in the channel. The attack success, however, depends on how "good" is the attacker. As an example, doing a jamming with Ubertooth is extremely hard, because the hardware isn't able to "lock" the channel hopping for enough time.

Basically the clock skew of the sender and receiver are "beneficial" against the attack. One of the few cases where imprecision is good. However, simulating this imprecision is hard, as the time in ns-3 is "perfect". So you'll overestimate the attack success probability.

Anyway, you'll find out by yourself I guess.

Practical suggestions. Read the ns-3 tutorial and manual, especially the Channel and Spectrum models. Study their code and how they're used.

Then study a simple MAC from any of the current modules. You'll need to implement a new one.

Then it will be time to design and implement a simple Bluetooth PHY layer. Keep it simple.

By then, you'll have enough knowledge to figure out how to do the attack, hopefully.

Good luck, you'll need it. Also because 2 months are definitely too few.

Cheers,

T.

[Nusrat Fatema]

Hello Tommaso,

Sorry for the late reply. Thank you for your reply. As there is no Bluetooth stack available in NS-3, is there any other simulator which consists ZigBee and Bluetooth stack except NS-2? By using that stack I can fulfill the requirement of the project. 

--

Nusrat.