Authentication and Encryption

Now that you understand some of the methods used to authenticate users, it’s time to ex- plore some encryption methods. The beginning of this chapter discussed WEP. The prob- lem with WEP is that it can be broken easily. Therefore, other methods have been established in an effort to provide more strength in encryption. In the following sections, you will learn about Wi-Fi Protected Access (WPA) and Wi-Fi Protected Access 2 (WPA2).


WPA Overview

WPA was introduced in 2003 by the Wi-Fi Alliance as a replacement for WEP. WPA uses Temporal Key Integrity Protocol (TKIP) to automatically change the keys. TKIP still uses RC4; it just improves how it’s done. This is a major improvement over static WEP. WPA can optionally support Advanced Encryption Standard (AES), but it’s not mandatory. WPA is based on 802.11i draft version 3. WEP uses RC4 encryption, which is very weak. The better alternative was to use AES encryption, but that would have required an equip- ment upgrade. To avoid an equipment upgrade, WPA was developed to use TKIP and a larger IV than WEP. This would make it more difficult to guess the keys while not requir- ing new hardware. Instead, you could simply perform a firmware upgrade in most cases.

WPA offers two authentication modes:

• Enterprise mode: Enterprise mode WPA requires an authentication server. RADIUS is used for authentication and key distribution, and TKIP is used with the option of AES available as well.

• Personal mode: Personal mode WPA uses preshared keys, making it the weaker op- tion, but the one that is most likely to be seen in a home environment.

Figure 17-15 shows the process of WPA authentication.


At the beginning of negotiations, the client and AP must agree on security capabilities. After the two agree on the same level of security, the 802.1x process starts. This is the standard 802.1x process, as outlined previously. After successful 802.1x authentication, the authentication server derives a master key and sends it to the AP. The same key is de- rived from the client. Now the client and the AP have the same Pairwise Master Key (PMK) , which will last for the duration of the session.

Next, a four-way handshake occurs (see Figure 17-16), in which the client and authentica- tor communicate and a new key called a Pairwise Transient Key (PTK) is derived. This key confirms the PMK between the two, establishes a temporal key to be used for message encryption, authenticates the negotiated parameters, and creates keying material for the next phase, called the two-way group key handshake.


When the two-way group key handshake occurs, the client and authenticator negotiate the Group Transient Key (GTK), which is used to decrypt broadcast and multicast trans- missions.

In Figure 17-16, you can see that the AP first generates a random number and sends it to the client. The client then uses a common passphrase along with this random number to derive a key that is used to encrypt data to the AP. The client then sends its own random number to the AP, along with a Message Integrity Code (MIC), which is used to ensure that the data is not tampered with. The AP generates a key used to encrypt unicast traffic to the client. To validate, the AP sends the random number again, encrypted using the de- rived key. A final message is sent, indicating that the temporal key (TK) is in place on both sides.

The two-way handshake that exchanges the group key involves the generation of a Group Master Key (GMK) , usually by way of a random number. After the AP generates the GMK, it generates a group random number. This is used to generate a Group Temporal Key (GTK) . The GTK provides a group key and a MIC. This key changes when it times out or when a client leaves the network.