Aircrack-ng can recover the WEP key once enough encrypted packets have been captured with airodump-ng. This part of the aircrack-ng suite determines the WEP key using two fundamental methods. The first method is via the PTW approach (Pyshkin, Tews, Weinmann). The default cracking method is PTW. This is done in two phases. In the first phase, aircrack-ng only uses ARP packets. If the key is not found, then it uses all the packets in the capture. Please remember that not all packets can be used for the PTW method. This Tutorial: Packets Supported for the PTW Attack page provides details. An important limitation is that the PTW attack currently can only crack 40 and 104 bit WEP keys. The main advantage of the PTW approach is that very few data packets are required to crack the WEP key.
Recover Keys 9.0 Crack
The other, older method is the FMS/KoreK method. The FMS/KoreK method incorporates various statistical attacks to discover the WEP key and uses these in combination with brute forcing. It requires more packets than PTW, but on the other hand is able to recover the passphrase when PTW sometimes fail.
SSE2, AVX, AVX2, and AVX512 support is included to dramatically speed up WPA/WPA2 key processing. With the exception of AVX512, all other instructions are built-in Aircrack-ng, and it will automatically select the fastest available for the CPU. For non-x86 CPUs, SIMD improvements are present as well.
The first method is the PTW method (Pychkine, Tews, Weinmann). The PTW method is fully described in the paper found on this web site. In 2005, Andreas Klein presented another analysis of the RC4 stream cipher. Klein showed that there are more correlations between the RC4 keystream and the key than the ones found by Fluhrer, Mantin, and Shamir and these may be additionally used to break WEP. The PTW method extends Klein's attack and optimizes it for usage against WEP. It essentially uses enhanced FMS techniques described in the following section. One particularly important constraint is that it only works with arp request/reply packets and cannot be employed against other traffic.
By using a series of statistical tests called the FMS and Korek attacks, votes are accumulated for likely keys for each key byte of the secret WEP key. Different attacks have a different number of votes associated with them since the probability of each attack yielding the right answer varies mathematically. The more votes a particular potential key value accumulates, the more likely it is to be correct. For each key byte, the screen shows the likely secret key and the number of votes it has accumulated so far. Needless to say, the secret key with the largest number of votes is most likely correct but is not guaranteed. Aircrack-ng will subsequently test the key to confirm it.
Looking at an example will hopefully make this clearer. In the screenshot above, you can see, that at key byte 0 the byte 0xAE has collected some votes, 50 in this case. So, mathematically, it is more likely that the key starts with AE than with 11 (which is second on the same line) which is almost half as possible. That explains why the more data that is available, the greater the chances that aircrack-ng will determine the secret WEP key.
However the statistical approach can only take you so far. The idea is to get into the ball park with statistics then use brute force to finish the job. Aircrack-ng uses brute force on likely keys to actually determine the secret WEP key.
This is where the fudge factor comes in. Basically the fudge factor tells aircrack-ng how broadly to brute force. It is like throwing a ball into a field then telling somebody to ball is somewhere between 0 and 10 meters (0 and 30 feet) away. Versus saying the ball is somewhere between 0 and 100 meters (0 and 300 feet) away. The 100 meter scenario will take a lot longer to search then the 10 meter one but you are more likely to find the ball with the broader search. It is a trade off between the length of time and likelihood of finding the secret WEP key.
For example, if you tell aircrack-ng to use a fudge factor 2, it takes the votes of the most possible byte, and checks all other possibilities which are at least half as possible as this one on a brute force basis. The larger the fudge factor, the more possibilities aircrack-ng will try on a brute force basis. Keep in mind, that as the fudge factor gets larger, the number of secret keys to try goes up tremendously and consequently the elapsed time also increases. Therefore with more available data, the need to brute force, which is very CPU and time intensive, can be minimized.
For cracking WEP keys, a dictionary method is also included. For WEP, you may use either the statistical method described above or the dictionary method, not both at the same time. With the dictionary method, you first create a file with either ascii or hexadecimal keys. A single file can only contain one type, not a mix of both. This is then used as input to aircrack-ng and the program tests each key to determine if it is correct.
The techniques and the approach above do not work for WPA/WPA2 pre-shared keys. The only way to crack these pre-shared keys is via a dictionary attack. This capability is also included in aircrack-ng.
With pre-shared keys, the client and access point establish keying material to be used for their communication at the outset, when the client first associates with the access point. There is a four-way handshake between the client and access point. airodump-ng can capture this four-way handshake. Using input from a provided word list (dictionary), aircrack-ng duplicates the four-way handshake to determine if a particular entry in the word list matches the results the four-way handshake. If it does, then the pre-shared key has been successfully identified.
It should be noted that this process is very computationally intensive and so in practice, very long or unusual pre-shared keys are unlikely to be determined. A good quality word list will give you the best results. Another approach is to use a tool like john the ripper to generate password guesses which are in turn fed into aircrack-ng.
You can specify multiple input files (either in .cap or .ivs format) or use file name wildcarding. See Other Tips for examples. Also, you can run both airodump-ng and aircrack-ng at the same time: aircrack-ng will auto-update when new IVs are available.
Next, we look at cracking WEP with a dictionary. In order to do this, we need dictionary files with ascii or hexadecimal keys to try. Remember, a single file can only have ascii or hexadecimal keys in it, not both.
When running aircrack-ng, it will load the fastest optimization based on what your CPU supports. For package maintainers, it is very useful as they don't have to target the one supporting all the CPU which would be the slowest.
It will create and/or update a session file saving the current status of the cracking (every 10 minutes) as well as all the options used, wordlists and capture files used. Multiple wordlists can be used and it works with WEP and WPA.
The overriding technique is capture as much data as possible. That is the single most important task. The number of initialization vectors (IVs) that you need to determine the WEP key varies dramatically by key length and access point. Typically you need 250,000 or more unique IVs for 64 bit keys and 1.5 million or more for 128 bit keys. Clearly a lot more for longer key bit lengths. Then there is luck. There will be times that the WEP key can be determined with as few as 50,000 IVs although this is rare. Conversely, there will be times when you will need mulitple millions of IVs to crack the WEP key. The number of IVs is extremely hard to predict since some access points are very good at eliminating IVs that lead the WEP key.
While aircrack-ng is running, you mostly just see the beginning of the key. Although the secret WEP key is unknown at this point, there may be clues to speed things up. If the key bytes have a fairly large number of votes, then they are likely 99.5% correct. So lets look at what you can do with these clues.
If the bytes (likely secret keys) are for example: 75:47:99:22:50 then it is quite obvious, that the whole key may consist only of numbers, like the first 5 bytes. So it MAY improve your cracking speed to use the -t option only when trying such keys. See Wikipedia Binary Coded Decimal for a description of what characters -t looks for.
And if the first few bytes are something like 74:6F:70:73:65, and upon entering them into your hexeditor or the links provided in the previous sentence, you see that they may form the beginning of some word, then it seems likely an ASCII key is used, thus you activate -c option to check only printable ASCII keys.
As a side note, Windows WZC only supports fixed length hex or ascii keys, so the shortest inputable key is 5 characters long. See the table above on this page regarding how many characters are needed for specific key lengths.
As you have seen, if there are multiple networks in your files you need to select which one you want to crack. Instead of manually doing a selection, you can specify which network you want by essid or bssid on the command line. This is done with the -e or -b parameters.
There will be times when key bytes will have negative values for votes. As part of the statistical analysis, there are safeguards built in which subtract votes for false positives. The idea is to cause the results to be more accurate. When you get a lot of negative votes, something is wrong. Typically this means you are trying to crack a dynamic key such as WPA/WPA2 or the WEP key changed while you were capturing the data. Remember, WPA/WPA2 can only be cracked via a dictionary technique. If the WEP key has changed, you will need to start gathering new data and start over again.
Capturing WPA/WPA2 handshakes can be very tricky. A capture file may end up containing a subset of packets from various handshake attempts and/or handshakes from more then one client. Currently aircrack-ng can sometimes fail to parse out the handshake properly. What this means is that aircrack-ng will fail to find a handshake in the capture file even though one exists. 2ff7e9595c
Comments