Methods and modular cryptographic device with status determination

<b>Background of the Invention</b>

Security is an extremely important consideration in network communications. With the ever-increasing utilization of the Internet, most networks now have Internet gateways which open them up to external attacks by would-be hackers. Further, the popularity of wireless networks has also increased dramatically as technology has enabled faster and more reliable wireless communications. Yet, wireless communications are inherently less secure than wired communications, since wireless communication signals are typically much easier to intercept than signals on cables which are often difficult to access.

As a result, cryptography is often used to encrypt private or secret communications to reduce the likelihood that they will be deciphered and used by malicious individuals or organizations. By way of example, wireless local area networks (WLANs) and WLAN devices are widely used and provide a convenient and cost-effective approach for implementing network communications where it may be difficult or otherwise impractical to run cables. One of the more prominent standards which has been developed for regulating communications within WLANs is promulgated by the Institute of Electrical and Electronic Engineers' (IEEE) 802 LAN/MAN Standards Committee, which is the 802.11 standard. In addition to providing wireless communications protocols, the 802.11 standard also defines a wireless equivalent privacy (WEP) cryptographic algorithm which is used to protect wireless signals from eavesdropping.

WEP relies on a secret key that is shared between wireless stations and an access point. The secret key is used to encrypt data packets prior to transmission, and an integrity check is used to ensure that packages are not modified during the transmission. Nonetheless, it has recently been discovered that the WEP algorithm is not as immune to external attacks as once believed. For example, in an article entitled "Intercepting mobile communications: The Insecurity of 802.11" by Borisov et al., MOBICOM, Rome, Italy, July 2001, the authors set forth a number of vulnerabilities in WEP. In particular, it was noted that a significant breach of security occurs when two messages are encrypted using a same initialization vector (IV) and secret key, as this can reveal information about both messages.

Moreover, WEP message ciphertext is generated using an exclusive OR operation. By exclusive ORing ciphertext from two messages generated using the same IV, the key streams cancel out and it is then possible to recover the plain text. As such, this key stream re-use is susceptible to a decryption dictionary attack in which a number of messages are stored and compared to find multiple messages generated with a same IV.

As a result, more robust network security is often required for many network applications. One example of a network security device to be connected between a protected client and a network is disclosed in U.S. Patent No. 6,240,513 to Friedman et al. The network security device negotiates a session key with any other protected client. Then, all communications between the two clients are encrypted. The device is self-configuring and locks itself to the IP address of its client. Thus, the client cannot change its IP address once set and therefore cannot emulate the IP address of another client. When a packet is transmitted from the protected host, the security device translates the MAC address of the client to its own MAC address before transmitting the packet into the network. Packets addressed to the host contain the MAC address of the security device. The security device translates its MAC address to the client's MAC address before transmitting the packet to the client.

Even more robust cryptographic devices may be required to secure sensitive or classified communications. More particularly, in the U.S. the communications of government entities that include sensitive (but unclassified) information must comply with the Federal Information Processing Standards Publication (FIPS) publication 140-2 entitled "Security Requirements For Cryptographic Modules." Classified communications, which are typically referred to as Type 1 communications, must comply with even stricter standards.

One example of an encryptor which is certified for Type 1 communications is the TACLANE Encryptor KG-175 from General Dynamics Corp. The "classic" version of the TACLANE encryptor has Internet Protocol (IP) and Asynchronous Transfer Mode (ATM) interfaces, and an E100 version has a fast Ethernet interface. The classic version may also be upgraded to fast Internet by replacing the IP/ATM network interface cards therein with two new E100 interface cards.

Despite the security benefits provided by such devices, many of these encryptors are fairly bulky and may consume significant amounts of power. One particularly advantageous cryptographic device which provides both space and power saving features is the Sierra module from Harris Corp., Assignee of the present application. The Sierra module is an embeddable encryption device that combines the advantages of high-grade security (e.g., Type 1) with the cost efficiency of a reprogrammable, commercially produced, FIPS 140-2 level 3 or 4 encryption module. The Sierra module can take on multiple encryption personalities depending on the particular application, providing encryption/decryption functionality, digital voice processing (vocoding) and cryptographic key management support functions. The Sierra module also provides the user with the capability to remove the Type 1 functionality, allowing the device to be downgraded to an unclassified device. Also, because of its relatively small size, low power and high data rates, this device is well-suited for battery sensitive applications.

By way of example, the Sierra module has been implemented in a Secure WLAN (SWLAN) personal computer (PC) card called SecNet 11, which is also produced by Harris Corp. The SecNet 11 card allows rapid communication of multimedia information (data, voice, and video) in a secure environment. The SecNet 11 card may be used as a wireless network interface card for WLAN "stations," for wireless bridges, and for access point (APs), for example. The SecNet 11 device is more fully described in U.S. published application nos. 2002/0094087 and 2002/0095594, both of which are hereby incorporated herein in their entireties by reference.

Accordingly, the SecNet 11 card provides numerous advantages in terms of size, power requirements, and flexibility in WLAN environments. However, it may be desirable to provide such benefits in other network environments as well.

<b>Summary of the Invention</b>

In view of the foregoing background, it is therefore an object of the present invention to provide a cryptographic device that provides high level security and is relatively easily adaptable to numerous network environments and related methods.

This and other objects, features, and advantages in accordance with the present invention are provided by a cryptographic device which may include a cryptographic module and a communications module removably coupled thereto. More particularly, the cryptographic module may include a first housing, a user network interface (e.g., a user Local Area Network (LAN) interface) carried by the first housing, a cryptographic processor carried by the first housing and coupled to the user network interface, and a first connector carried by the first housing and coupled to the cryptographic processor. Further, the communications module may include a second housing, a second connector carried by the second housing and being removably mateable with the first connector of the cryptographic module, and a network communications interface (e.g., a network LAN interface) carried by the second housing and coupled to the second connector. The communications module may also include at least one logic device for cooperating with the cryptographic processor to determine a status of the communications module.

In addition, the communications module may be a predetermined one from among a plurality of interchangeable communications modules each for communicating over a different communications media. Thus, the same cryptographic module may advantageously be used for numerous network applications simply be interchanging the appropriate communications module for the desired application. As such, a user needs only one cryptographic module for a given LAN device, but can readily adapt the one cryptographic module for different networks by interchanging communications modules.

This may be particularly important since the cryptographic module includes the cryptographic processor and associated cryptographic algorithms and keys. That is, the evaluation process to have such a device certified for use with sensitive or classified communications can be quite lengthy and extensive, and thus expensive. However, since the various communications modules merely provide interfaces for different types of networks and do not transmit "red" (i.e., unencrypted) confidential/classified data, they may not require the same certification scrutiny. Accordingly, interchanging the communications modules may be less expensive than having to obtain an entirely new cryptographic device with a different network interface for each desired application.

In particular, the status of the communications module may be at least one of a type of communications module (i.e., an identifier of the type of communications module coupled to the cryptographic module) and an operating status thereof. Moreover, the at least one logic device may also permit the cryptographic processor to configure the network communications interface of the communications module. In addition, the communications module may further include at least one status indicator carried by the second housing and coupled to the at least one logic device, and the at least one logic device may be a complex programmable logic device (CPLD), for example.

Additionally, the network communications interface may be a wireless LAN (WLAN) communication circuit, a wireline LAN communication circuit, or a fiber optic LAN communication circuit, for example. Also, the user network interface may be an Ethernet interface, for example. The cryptographic processor may include a host network processor coupled to the user network interface, and a cryptography circuit coupled to the host network processor. The cryptographic processor may further include an unencrypted data buffer circuit coupled between the user network interface and the cryptography circuit, and an encrypted data buffer circuit coupled between the cryptography circuit and the network communications interface.

To provide still further security features, the cryptographic module may also advantageously include a tamper circuit for disabling the cryptographic processor based upon tampering with the first housing. By way of example, the tamper circuit may include one or more conductors substantially surrounding the cryptographic processor, and the cryptographic processor may be disabled based upon a break in the conductor(s).

A communications method aspect of the invention may include providing a cryptographic module, such as the one described briefly above, and removably coupling the user interface thereof to a network device. The method may further include providing a communications module, such as the one described briefly above, with the second connector thereof removably mated with the first connector of the cryptographic module. The method may further include using the network interface to communicate with a network, and causing the at least one logic device to cooperate with the cryptographic processor to determine a status of the communications module.

A communications system in accordance with the invention may include a plurality of network devices coupled together to define a network (e.g., a LAN), and a cryptographic device, such as the one described briefly above, coupled to at least one of the network devices.

<b>Brief Description of the Drawings</b>

<ul id="ul0001" list-style="none" compact="compact"><li>FIG. 1 is perspective view of a cryptographic device in accordance with the present invention.</li><li>FIG. 2 is an exploded view of the cryptographic device of FIG. 1 illustrating the various modules thereof.</li><li>FIG. 3 is top plan view of the cryptographic device of FIG. 1.</li><li>FIGS. 4 through 9 are schematic block diagrams illustrating the various components of the cryptographic device of FIG. 1 in greater detail.</li><li>FIG. 10 is a timing diagram illustrating status and configuration operations for the communications module of the cryptographic device of FIG. 1.</li><li>FIG. 11 is a block diagram of a cryptographic packet generated in accordance with the present invention.</li><li>FIGS. 12 and 13 are perspective views illustrating the connector configurations of the communications module and cryptographic module, respectively, of the cryptographic device of FIG. 1.</li><li>FIG. 14 is another exploded perspective view showing the bottom of the cryptographic device of FIG. 1 and further illustrating coupling of the various modules thereof.</li><li>FIGS. 15 through 20 are flow diagrams illustrating various communications method aspects in accordance with the present invention.</li></ul>

<b>Detailed Description of the Preferred Embodiments</b>

The present invention will now be described more fully hereinafter with reference to the accompanying drawings, in which preferred embodiments of the invention are shown. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Like numbers refer to like elements throughout, and prime notation is used to indicate similar elements or steps in different embodiments.

Referring initially to FIGS. 1 through 9, a communication system <b>29</b> in accordance with the present invention illustratively includes a cryptographic device <b>30,</b> a plurality of network devices <b>40,</b> and a network such as a wireless Local Area Network (WLAN) <b>48.</b> The cryptographic device <b>30</b> illustratively includes a cryptographic module <b>31</b> coupled to one of the devices <b>40</b> and a communications module <b>32.</b> As shown in FIG. 2, the communications module <b>32</b> is removably coupled to the cryptographic module <b>31,</b> as will be discussed further below. Generally speaking, in accordance with the present invention, a plurality of interchangeable communications modules <b>32</b> may be connected to the cryptographic module <b>31</b> for communicating over different communications media. While in the illustrated embodiment the communications module <b>32</b> is a WLAN module which includes dual tri-band antennas <b>33,</b> it will be appreciated based upon the following discussion that the cryptographic device <b>30</b> can be used with numerous types of wired and wireless networks.

By including the appropriate chip sets/ interface circuitry in different communications modules <b>32,</b> each of these modules may interface with a different network medium (e.g., WLAN, wireline medium, fiber optic medium, etc.), yet all interface with the same cryptographic module <b>31.</b> That is, the same cryptographic module 31 may be used for numerous network applications simply by coupling the appropriate communications module <b>32</b> thereto for the desired application. Examples of various types of communications modules <b>32</b> that may be used include WLAN modules, plain old telephone service (POTS) modules, tactical radio modules, E1/T1 modules, in-line network encryptor (INE) modules, a VersaModule Eurocard (VME) bus module, etc.

The modular design and ease of interchangeability not only provides a convenient way to quickly configure the cryptographic module <b>31</b> for different applications, but it may also be particularly useful for high level security applications such a Type 1, FIPS 140-2 level 4, etc. This is because the evaluation process to have a cryptographic device certified for use with sensitive or classified communications at these levels can be quite lengthy and extensive, and consequently costly. Thus, to manufacture respective cryptographic devices for different network applications can be cost prohibitive since each one will have to individually undergo the rigorous and costly certification.

Yet, since in accordance with the present invention the cryptographic module <b>31</b> preferably includes all of the sensitive cryptographic circuitry and associated cryptographic algorithms/keys, the various communications modules <b>3</b>2 merely provide interfaces for different types of networks. That is, they do not process or transmit "red" (i.e., unencrypted) confidential/classified data, and thus they will likely not require the same certification scrutiny as the cryptographic module <b>31.</b> Accordingly, the communications modules 32 may provide significant cost savings over having to purchase an entirely new cryptographic device with a different network interface for each desired application.

In particular, the cryptographic module <b>31</b> illustratively includes a first housing <b>34,</b> a user network interface <b>35</b> carried by the first housing, a cryptographic processor <b>36</b> carried by the first housing and coupled to the user network interface, and a first inter-module connector <b>37</b> carried by the first housing and coupled to the cryptographic processor. The user network interface <b>35</b> may be an Ethernet physical layer (PHY) interface compatible with the IEEE 802.3 standard, for example, as will be appreciated by those skilled in the art. Various connectors <b>38</b> are also carried by the first housing <b>34</b> for coupling the cryptographic module <b>31</b> to different network devices <b>40</b> (e.g., personal computers (PCs), servers, portable communications devices, etc.).

By way of example, the connectors <b>38</b> may be wireline connectors, such as an RJ45 connector <b>85</b> (FIG. 8), or fiber optic connectors, such as an LC fiber optic connector <b>86.</b> Caps <b>39</b> may also be included for protecting the connectors <b>34.</b> A power switch <b>41</b> and LED status indicators <b>42</b> (i.e., power, link state, fill, and alarm) are also carried by the first housing <b>34.</b>

It should be noted that the term "user" is used with relation to the user network interface <b>35</b> simply to indicate that this interface is for the user network device side and not the communications network side of the cryptographic device <b>30.</b> That is, "user" does not mean that the interface <b>35</b> is only for individual user devices such as PCs. Instead, the user network interface may be connected to a variety of different LAN devices (e.g., servers, bridges, access points, etc.), as noted above.

The communications module <b>32</b> illustratively includes a second housing <b>45,</b> a second inter-module connector <b>46</b> carried by the second housing and removably mateable with the first connector <b>37</b> of the cryptographic module <b>31,</b> and a network communications interface <b>47</b> carried by the second housing <b>45</b> and coupled to the second connector. In the present example, the network communications interface <b>47</b> includes a WLAN communication circuit (e.g., an 802.11 chip set) for cooperating with the antennas 33 to wirelessly communicate with a network (e.g., LAN) <b>48,</b> as will be discussed further below. Yet, as noted above, the network communications interface <b>47</b> may be a wireline LAN communication circuit, a fiber optic LAN communication circuit, etc., for example.

The various circuit components of the cryptographic module <b>31</b> may be implemented in a cryptographic circuit card (CCA) <b>50,</b> for example, as will be appreciated by those skilled in the art. The circuitry of the communications module <b>32</b> may similarly be implemented in a CCA <b>51.</b> The cryptographic module <b>31</b> may also include a power CCA <b>52</b> carried by the first housing <b>34</b> and including power supply/filtering circuitry <b>53</b> for powering the cryptographic processor <b>36,</b> the user network interface <b>35,</b> and the communications module <b>32.</b>

The cryptographic processor <b>36</b> may include a host network processor <b>54</b> connected to the user network interface <b>35,</b> and cryptography circuitry 55 connected to the host network processor. More particularly, the cryptography circuitry <b>55</b> illustratively includes an unencrypted (i.e., "red") data buffer 56 connected to the host network processor <b>54,</b> a cryptography circuit <b>57</b> connected to the unencrypted data buffer, and an encrypted (i.e., "black") data buffer <b>58</b> connected between the cryptography circuit and the first connector <b>37.</b>

By way of example, the unencrypted and encrypted data buffers may be first-in, first-out (FIFO) buffers implemented using field-programmable gate arrays (FPGAs), and the cryptography circuit <b>57</b> may be implemented in an application specific integrated circuit (ASIC). One cryptography ASIC that is particularly well suited for use with the present invention is the above-noted Sierra (and Sierra II) device from Harris Corp. Of course, it will be appreciated by those skilled in the art that other suitable circuitry may be used as well.

The host network processor <b>54</b> illustratively includes a plurality of modules which may be implemented using hardware and/or software, as will be appreciated by those skilled in the art. Generally speaking, the host network processor <b>54</b> includes a first 802.3 medium access controller (MAC) controller 60 for interfacing the user network interface <b>35,</b> a second 802.3 MAC controller <b>61</b> for interfacing the cryptographic processor <b>36</b> and network communications interface <b>47,</b> as will be described further below, and a processor <b>62</b> coupled between the MAC controllers. The host network processor <b>54</b> and user network interface <b>35</b> may communicate via dedicated lines for Media Independent Interface (MII) communications, as will be discussed further below, and a management data input/output bus (FIGS. 6 and 8), for example.

More specifically, the processor <b>62</b> may include a hypertext transfer protocol (HTTP) server module <b>73,</b> a simple network management protocol agent <b>63,</b> a firewall/routing module <b>64,</b> an over the air re-keying/over the network rekeying (OTAR/OTNR) module <b>65,</b> and an over the air zeroization/over the network zeroization (OTAZ/OTNZ) module <b>66.</b> Moreover, the processor <b>54</b> also illustratively includes a mode controller <b>67</b> for providing proper configuration based upon the particular mode or media with which the cryptographic module <b>31</b> is to operate (e.g., WLAN access point (AP) mode, ad-hoc mode, infrastructure mode, etc.). The mode controller <b>67</b> may also perform other configuration/monitoring functions, such as for service set identifiers (SSIDs), channel, transmission level, data rate, 802.11 band selection (i.e., a, b, g) depending upon the particular application the cryptographic module <b>31</b> is to be used for, as will be appreciated by those skilled in the art. Additional modules such as an Internet protocol (IP) security protocol (IPSec)/high-assurance IP encryption (HAIPE) module <b>68,</b> a key management module <b>69,</b> and/or a device discovery module <b>70</b> may also be included depending upon the given implementation, as will also be appreciated by those skilled in the art. The cryptographic module also preferably includes respective memory devices <b>71, 72</b> for the host network processor <b>54</b> and cryptography circuit <b>57.</b>

The power circuitry <b>53</b> illustratively includes external power interface (I/F) circuitry <b>75,</b> which may be connected to a DC source (e.g., battery), a wall wart AC adapter, an Ethernet power source, etc. Of course, it will be appreciated that other power sources may be used in different implementations. The power circuitry 53 further illustratively includes cryptographic/communications module power isolation/filtering circuitry <b>76</b> coupled to the external power I/F circuitry <b>75.</b> A cryptographic module power circuit <b>77</b> and a communications module power circuit <b>78</b> are coupled to the power isolation/filtering circuitry <b>76</b> for respectively supplying the cryptographic and communications modules <b>31, 32.</b> Further, a data filter/electrostatic discharge (ESD) protection circuit <b>79</b> is included for filtering signals communicated between the cryptographic module <b>31</b> and communications module <b>32,</b> as will be appreciated by those skilled in the art.

To receive high level certification (e.g., level 4 FIPS 140-2, Type 1) for classified and/or secret communications, cryptographic devices typically have to include some degree of physical tamper protection to prevent malicious individuals or organizations from physically compromising the device and discovering the secret key or algorithm being used. In accordance with the present invention, the cryptographic module <b>31</b> also illustratively includes a tamper circuit <b>80</b> for disabling the cryptography circuit <b>57</b> based upon tampering with the first housing <b>34.</b> By way of example, the tamper circuit <b>80</b> preferably includes one or more conductors substantially surrounding the cryptography circuit <b>57</b> so that the cryptographic processor is disabled based upon a break in any one of the conductors.

More particularly, the conductors may be relatively thin printed circuit traces printed on the inside of the first housing <b>34</b> and attached to the cryptographic processor <b>36.</b> Since the conductors substantially surround the cryptographic processor <b>36</b> (or some portion thereof), if someone attempts to drill through the first housing <b>34</b> to access the cryptographic processor then one or more of the printed traces will be broken. The same holds true if someone opens the first housing, as the traces will be pulled away from the cryptographic processor <b>36</b> also causing breaks therein.

In either event, the open circuit condition resulting from the broken conductor(s) causes power to a cryptographic power interface circuit <b>81</b> to be disrupted to be discontinued. That is, power from a dedicated encryption algorithm/secret key battery <b>82</b> is prohibited from flowing to the cryptographic power interface circuit <b>81</b> via the cryptographic module power circuitry <b>77.</b> As a result, the algorithm and secret key, which are preferably stored in a volatile memory, are permanently and instantly erased so that they cannot be discovered by malicious individuals or organizations. The tamper circuit <b>80</b> may thus provide tamper protection from all angles, if desired.

As noted above, the cryptography circuit <b>57</b> implements a desired encryption algorithm to provide a predetermined security level (e.g., Type 1, FIPS 140-2 levels 1 through 4, etc.). By way of example, Advanced Encryption Standard (AES), Baton, or Medley encryption algorithms may be used to provide such high level security. Of course, other high level security algorithms known to those skilled in the art may be used as well. Additionally, other cryptographic algorithms which are considered to be less secure than those noted above may also be used in accordance with the present invention when the cryptographic device <b>30</b> is to be used in less sensitive environments (e.g., general commercial or corporate applications).

The cryptography circuitry <b>55</b> also illustratively includes a plurality of modules which may be implemented using hardware and/or software. Referring particularly to FIG. 8, the unencrypted data buffer (i.e., red FPGA) <b>56</b> illustratively includes a host interface/FIFO control module <b>90</b> for communicating with the host network processor <b>54</b> via the MII protocol, and traffic and command (CMD) FIFOs <b>91, 92</b> receiving outputs of the host interface/FIFO control module. It should be noted that various data paths in FIG. 8 are labeled as "red" and/or "black" to indicate whether they convey unencrypted or encrypted data, respectively, or both, to aid in understanding of the present invention.

The output of the traffic FIFO <b>91</b> is connected to a buffer <b>93,</b> which is connected to a first high speed parallel interface <b>94</b> of the cryptographic circuit <b>57.</b> The output of the command FIFO <b>92</b> is connected to a first external bus interface unit (EBIU) 106 of the cryptographic circuit <b>57.</b> This EBIU <b>106</b> is also connected to control registers <b>95</b> and a multiplexer <b>96.</b> Another input of the multiplexer <b>96</b> is connected to the output of a second high speed parallel interface <b>97</b> of the cryptographic circuit <b>57.</b> The output of the multiplexer <b>96</b> is passed to a cyclic redundancy check module <b>98,</b> the output of which is passed through an output FIFO <b>100</b> back to the host interface/FIFO control module <b>90.</b>

The first high speed parallel interface <b>94</b> of the cryptography circuit <b>57</b> has a respective word counter <b>101</b> associated therewith. A cryptographic processing module 102 of the cryptography circuit <b>57</b> interfaces the first and second high speed parallel interfaces <b>94, 97</b> and one or more cryptographic engine modules <b>103</b> via a bus controller <b>104.</b> The cryptographic processing module <b>102</b> also communicates with a fill circuit <b>105</b> for the loading of cryptographic keys. The EBIU 106 also interfaces the cryptographic processing module <b>102</b> with the memory <b>72.</b> A second EBIU <b>107</b> interfaces the cryptographic processing module <b>102</b> with control registers <b>110</b> and a multiplexer <b>111</b> of the encrypted data buffer (i.e., black FPGA) <b>58.</b> The signal path between the second EBIU <b>107</b> and the multiplexer <b>111</b> provides a command signal path.

Various components of the host network processor <b>54,</b> red FPGA <b>56,</b> cryptographic circuit <b>57,</b> and black FPGA <b>58</b> also communicate via one or more general purpose input/output (GPIO) busses as shown, as will be appreciated by those skilled in the art. Additional circuitry <b>112</b> may also be coupled to the cryptography circuit <b>57</b> in certain embodiments for over/undervoltage detection, temperature detection, and/or panic zeroizing as required for a particular implementation, as will also be appreciated by those skilled in the art.

An output of the second high speed parallel interface <b>97</b> is passed via a buffer <b>113</b> to an input interface <b>114</b> which includes protection gating to prohibit red data from entering the black FPGA <b>58.</b> The output of the input interface <b>114</b> is connected to a second input of the multiplexer <b>111</b> defining a traffic (i.e., data) path thereto. The output of the multiplexer <b>111</b> is provided to a cyclic redundancy check module <b>115,</b> the output of which is provided to an output FIFO <b>117.</b> An output of the MAC interface/FIFO control module <b>118</b> is provided to the input of the traffic FIFO <b>116.</b> The output of the traffic FIFO <b>116</b> is passed via a buffer <b>120</b> back to the input of the first high speed parallel interface <b>94</b> of the cryptographic circuit <b>57,</b> and the output of the output FIFO <b>117</b> is connected to the MAC interface/FIFO control module <b>118,</b> which communicates with the communications module <b>32,</b> as will be discussed further below.

The various circuitry of the communication module <b>32</b> will now be described in further detail with particular reference to FIGS. 5 through 7. As noted above, the various circuitry of the communications module <b>32</b> is implemented in the communications CCA <b>51.</b> In particular, the communications (or radio in the present WLAN example) CCA <b>51</b> illustratively includes a power interface <b>126</b> for cooperating with the communications power circuit <b>78</b> to supply the various communications circuitry components. Additional filter/ESD circuitry <b>127</b> may also be included in the signal path from the cryptographic module <b>31,</b> if desired.

More particularly, the signal path between the cryptographic module <b>31</b> and communications module <b>32</b> includes a plurality of lines for MII communications, as well as a three-wire serial interface (3WSI), as seen in FIG. 6. Generally speaking, the MII lines are for transferring encrypted data between the cryptographic module <b>31</b> and the communications module <b>32,</b> and the three wire serial interface is for status/configuration operations of the communications module, as will be discussed further below.

More particularly, the MII lines pass through the filter/ESD circuitry <b>127</b> to the network communications interface <b>47.</b> In the present WLAN example, the network communications interface <b>47</b> includes an 802.11 a/b/g AP/MAC chip set <b>128</b> connected to the MII lines, and an associated 802.11 a/b/g radio <b>129</b> connected to the 802.11 a/b/g AP/MAC chip set for wirelessly communicating with a WLAN. One or more memories <b>130</b> may be provided for the 802.11 a/b/g AP/MAC chip set <b>128.</b> The 802.11 a/b/g AP/MAC chip set <b>128</b> illustratively includes a processing module <b>141,</b> an Ethernet MAC module <b>142</b> for communicating with the cryptographic module <b>31,</b> and a WLAN MAC module <b>143</b> for performing the appropriate 802.11 WLAN interface and processing operations, as will be appreciated by those skilled in the art.

The communications CCA <b>51</b> also illustratively includes a logic device <b>131,</b> such as a complex programmable logic device (CPLD), which is connected to the above-noted three wire serial interface. Generally speaking, the CPLD <b>131</b> cooperates with the cryptographic processor <b>36</b> to detect, status, and configure different types of communications modules <b>32.</b> More particularly, the host network processor <b>54</b> polls the CPLD <b>131</b> to determine what type of communications module <b>32</b> is connected to the cryptographic module <b>31</b> (i.e., WLAN, wireline, fiber optic, etc.), as well as its operational status, as will be appreciated by those skilled in the art. The CPLD <b>131</b> also permits the host network processor <b>54</b> to configure the network communications interface <b>47</b> for operation in a given application, as will also be appreciated by those skilled in the art.

Referring additionally to FIGS. 9 and 10, the three lines of the three wire serial interface respectively carry clock signals, data signals, and enable signals between the cryptographic and communications modules <b>31, 32.</b> The clock signal is provided to a sixteen bit (although other sizes may also be used) serial to parallel data converter <b>135,</b> an output register <b>136,</b> a sixteen bit parallel to serial data converter <b>137,</b> and control logic <b>138.</b> More particularly, control data coming from the cryptographic processor <b>36</b> via the data line is written to the serial to parallel data converter <b>135</b> to be output by the output register <b>136.</b>

More particularly, the communications module <b>32</b> may further include one or more status indicators <b>140</b> (e.g., light emitting diodes (LEDs)) carried by the second housing <b>45</b> for indicating operational mode, band, or other appropriate status information. The LEDs <b>140</b> receive multiple bits (e.g., eight) from the output register <b>136.</b> Another set of bits (e.g., seven bits) from the register <b>136</b> are for enabling/disabling the communication module transmission circuitry (e.g., radio power amplifiers (PA)), and the remaining bits of the sixteen bit output is for providing a reset signal for the communications module <b>32.</b>

The input buffer <b>139</b> receives multiple bits (e.g., eight) of status (e.g., radio status for a WLAN implementation) information and multiple bits (e.g., eight) of hardware information from the 802.11 chip set <b>128</b> (or other network communications interfaces in other embodiments) to pass along to the cryptographic processor <b>36</b> via the parallel to serial data converter <b>137</b> and the data line of the three wire serial bus. Read and write data buffers <b>150, 151</b> may also be connected to the data line, if desired. Furthermore, the control circuitry <b>138</b> also receives the enable signal and enables the output register <b>136</b> and input buffer <b>139.</b>

A read or write operation occurs when the enable signal goes high, as seen in FIG. 10. The format of the command packets sent from the cryptographic processor 36 to the CPLD <b>131</b> are as follows. The first four address bits (A15-A12) of a packet instruct the CPLD <b>131</b> whether it is to receive data from the cryptographic processor <b>36,</b> or whether it is to supply requested data thereto. The remaining address bits (A11-A0) provide the address for the appropriate component or operation being requested, while the data bits (D15-D0) are reserved for data. As such, thirty-two bit serial words are exchanged between the cryptographic processor <b>36</b> and CPLD <b>131.</b>

An exemplary read/write addressing scheme is to use 0110 for the bits A15-A12 for a write operation, and 1011 for a read operation as shown, although other addressing schemes may also be used. Both the cryptographic module <b>31</b> and communications module <b>32</b> preferably clock data out on falling edges of the clock signal and clock data in on the leading edges, although other timing arrangements may be used in different embodiments.

A particularly advantageous approach for transferring the command packets from the cryptographic processor <b>36</b> to the communications module <b>32</b> will now be described. The host network processor <b>54</b> generates cryptographic processor command packets for the cryptographic processor <b>36.</b> These packets each include an Ethernet address portion for addressing the cryptography circuit <b>57</b> and an IP packet that encapsulates a cryptographic command. In accordance with the present invention, the host network processor <b>54</b> encapsulates a command packet to be operated upon by the communications module <b>32</b> within the cryptographic command, as shown in FIG. 11. By using the second EBIU <b>107,</b> for example, the communications module command packets may be passed to the communications module <b>32</b> without processing (i.e., encrypting). This provides a convenient way to transcend the red/black data boundary (FIG. 6) without potentially compromising security.

More particularly, the format of a cryptographic processor command packet is as follows. The Ethernet address portion of the packet is addressed to the cryptography circuit <b>57.</b> More particularly, the address portion may include Ethernet header addresses, an IP header, and cryptographic command information, as will be appreciated by those skilled in the art. The communications module command packet destined for the communications module is encapsulated in the data portion of the IP packet. Accordingly, when the cryptography circuit <b>57</b> receives such a cryptographic processor command packet, it will recognize the packet as a cryptographic command. As such, the cryptography circuit <b>57</b> will strip its own address information from the packet and transfer the remaining portion (i.e., the encapsulated communications module command packet) to the communications module <b>32.</b> Preferably, the host network processor <b>54</b> formats the data portions of the IP packets (and, thus, the command packets for the communications module <b>32)</b> based upon the simple network management protocol (SNMP), although other protocols may also be used.

The above-described approach may be used for sending communications module command data via the MII lines or the BWSI, and this approach may be used in reverse to communicate information back to the host network processor <b>54,</b> as will be appreciated by those skilled in the art. Since typical prior art cryptographic devices include all of the cryptography and communications circuitry within the same housing, the formatting of status/configuration commands for the communications circuitry is typically not an issue. However, as will be appreciated by those of skill in the art, the above-described approach provides a convenient and secure way to perform such command/control operations despite the separation between the cryptographic and communications modules <b>31, 32.</b> Of course, it will be appreciated that other approaches for formatting and/or encapsulating such command packets may also be used, as will be appreciated by those skilled in the art.

The above-described interchangeability of the communications modules <b>32</b> and the ability to pass the command packets through the red/black boundary is facilitated by using a same, predetermined interface protocol, i.e., an MII protocol, along the entire signal path between the user network interface <b>35</b> and the network communications interface <b>47.</b> That is, the cryptographic processor <b>36</b> not only communicates with the user network interface <b>35</b> using an MII-based protocol, it also communicates with the network communications interface <b>47</b> using the same MII-based protocol. The MII protocol may be based upon the original MII standard set forth in the IEEE 802.3 standard, or it may be a variant thereof such as reduced MII (RMMI) or gigabit MII (GMII), for example, although other protocols may be used as well.

Maintaining the consistent use of the MII protocol through the chain of circuitry from the user network interface <b>35</b> to the network communications interface <b>47</b> allows the cryptographic module <b>31</b> and the communications module <b>32</b> both to operate using a unique external MAC addresses, while at the same time using fixed internal MAC addresses. More particularly, the Ethernet MAC modules <b>60</b> and <b>143</b> operate using a unique external MAC addresses for each individual cryptographic module <b>31</b> and communications module respectively, while the Ethernet MAC modules <b>61</b> and <b>142</b> use fixed MAC addresses which are the same for every cryptographic device <b>30.</b>

Thus, the cryptographic circuitry <b>55</b> essentially becomes transparent to the communications module <b>32,</b> as it appears to the communications module that it is connected directly to the Ethernet MAC module <b>61.</b> Moreover, the "hard-coded" MAC addresses used by the Ethernet MAC's in both modules <b>61</b> and <b>142</b> provide for the transfer of command packets as described above, as well as a controlled transmission of encrypted data packets, as will be appreciated by those skilled in the art.

Another particularly advantageous feature of the invention is that different communications modules <b>32</b> may not only be used to allow a single cryptographic module <b>31</b> to be used with multiple media types (e.g., wireless, wireline, fiber optic, etc.), but the communications modules may also be used to provide multi-mode operation for a given media, such as in the case of a WLAN. More particularly, a WLAN communications module <b>32</b> may advantageously use an 802.11 a/b/g chip set <b>128</b> that is switchable between wireless LAN modes (i.e., access point (AP) mode, infrastructure mode, and ad-hoc mode) by the cryptographic module <b>31</b> using the above-described command packets, for example.

Thus, a same WLAN communications module <b>32</b> in accordance with the present invention may advantageously be used with any advice in a WLAN to provide desired functionality, such as individual station operation, bridging to a wired network, peer-to-peer communications, etc., as will be appreciated by those skilled in the art. Moreover, mode changes can be accomplished "on the fly" as desired using command packets. It will therefore be appreciated that with such a WLAN communications module <b>32,</b> the cryptographic device <b>30</b> provides complete 802.11 functionality in a single unit while also providing a wireless bridge that can be used to access a secure network. The cryptographic module <b>30</b> may advantageously be configured to allow selection and configuration of 802.11 modules of operation via a standard Web browser, for example.

Alternately, switching between WLAN operational modules may also be accomplished by using different types of 802.11 chip sets <b>128</b> for respective WLAN operational modes in different WLAN communications modules. That is, a different WLAN communications module <b>32</b> would be used depending upon whether an AP, infrastructure, or ad-hoc mode was desired for a given LAN device <b>40.</b>

Turning to FIGS. <b>12-14,</b> the coupling structure of the cryptographic and communications modules <b>31, 32</b> will now be further described. More particularly, the first housing <b>34</b> of the cryptographic module <b>31</b> may include a first body <b>180</b> and a first extension <b>181</b> extending outwardly therefrom, and the second housing <b>45</b> may include a second body <b>182</b> and a second extension <b>183</b> extending outwardly therefrom. As such, the first and second extensions <b>181, 183</b> may be aligned in overlapping relation when the first and second connectors <b>37, 46</b> are removably mated together.

The first connecter <b>37</b> is illustratively carried by the first body <b>180</b> adjacent the first extension <b>181,</b> and the second connector <b>46</b> is carried by the second extension <b>186.</b> Although other arrangements may be used in accordance with the present invention, this arrangement is particularly advantageous in that it allows the cryptographic CCA <b>50,</b> which has more circuitry than the power supply CCA <b>52,</b> to be positioned to take advantage of the extra length (and, therefore, surface area) of the first extension <b>181.</b> Similarly, the communications CCA <b>51</b> is positioned to take advantage of the additional length of the second extension <b>183.</b>

Each of the first and second extensions <b>181, 183</b> may also have surface features on opposing surfaces thereof to slidably engage and guide the cryptographic and communications modules <b>31, 32</b> together in mating relation. By way of example, the surface features may include rails <b>185</b> and corresponding channels <b>186</b> which define one or more slidable interlocking (e.g., dovetail) joints therebetween (two are shown in the exemplary implementation). One or more fasteners, such as captive screws <b>187</b> which mate with corresponding threaded holes <b>188,</b> are also preferably included for removably fastening the cryptographic and communications modules <b>31, 32</b> together.

As shown in the illustrated example, the first and second connectors <b>37, 46</b> are multi-pin electrical connectors, although various electrical connector styles known to those skilled in the art may be used. Also, one or more seals <b>190</b> may be positioned between the cryptographic module <b>31</b> and the communications module <b>32.</b> It will therefore be appreciated that the above-described electrical/mechanical structure provides a robust yet simple interconnection that is capable of providing desired EMI shielding and environmental sealing. Various materials (e.g., metal, plastic, etc.) may be used for the first and second housings <b>37, 45,</b> as will also be appreciated by those skilled in the art.

Based upon the foregoing description, numerous advantages of the present invention will be apparent to those skilled in the art. For example, the cryptographic device <b>30</b> is interoperable with standard commercial 802.11 and 802.3 networking equipment. More particularly, it may be used with any computing platform with an Ethernet interface (e.g., LINUX/UNIX, VxWorks, Windows, Macintosh, etc.). As such, independent developers may advantageously be able to develop applications without the need to write special drivers to communicate with the user network interface 35. Likewise, independent developers may advantageously be able to develop communications modules <b>32</b> for various and/or specialized communications applications since they will interface with the cryptographic module <b>31</b> via a well-defined, controlled electrical/mechanical interface. Furthermore, the coupling structure not only provides for easy interchangeability of different communications modules <b>32</b> with a single cryptographic module <b>31,</b> the rugged housing and connector design allows for operation over a wide range of climates and conditions.

Turning additionally to FIG. 15, a first communications method aspect of the invention will now be described. Beginning at Block <b>250,</b> the user network interface <b>35</b> of the cryptographic module <b>31</b> is coupled to a LAN device <b>40,</b> at Block <b>251.</b> Further, the communications module <b>32,</b> once attached to the cryptographic module <b>31,</b> may then be used to communicate with various networks (i.e., LAN) <b>48,</b> thus concluding the illustrated method, at Block <b>254.</b>

Referring to FIG. 16, another communications method aspect of the invention begins (Block 260) with coupling the cryptographic module <b>31</b> to the network device <b>40,</b> at Block <b>261,</b> with the communications module <b>32</b> being coupled to the cryptographic module as described above. The method further includes using the cryptographic processor <b>36</b> to communicate with the user network interface <b>35</b> and the network communications interface <b>47</b> using a same predetermined protocol (e.g., MII), at Block <b>263,</b> as discussed above, and also communicating with the network (i.e., LAN) <b>48,</b> at Block <b>264,</b> thus concluding the illustrated method (Block <b>265).</b>

Two additional method aspects for WLAN operation are now described with reference to FIGS. 17 and 18. Beginning at Block <b>270,</b> the cryptographic module <b>31</b> is coupled to the network device <b>40,</b> at Block <b>271,</b> with the communications module <b>32</b> being removably coupled to the cryptographic module <b>31,</b> as described above. If during the course of operation it is determined that a different WLAN mode of operation is required, at Block <b>273,</b> if a multi-mode network wireless network interface <b>274</b> is included in the WLAN communications module <b>32,</b> as discussed above, the interface may be switched to the desired wireless LAN mode, at Block <b>274.</b> Thereafter, or if a new WLAN mode is not required, wireless communications with the network (i.e., LAN) <b>48</b> may be conducted, at Block <b>275,</b> thus concluding the illustrated method (Block <b>276).</b> If different 802.11 modes are implemented in respective WLAN communications modules <b>32,</b> as discussed above, the step illustrated at Block <b>274</b> may be replaced with the step of removably coupling a new communications module providing the desired WLAN operational mode to the cryptographic module <b>31,</b> at Block <b>280'.</b>

Still another communications method aspect of the invention is now described with reference to FIG. 19. The method beings (Block <b>290)</b> with coupling the cryptographic module <b>31</b> to the network device <b>40,</b> at Block <b>291,</b> with the communications module <b>32</b> being removably coupled to the cryptographic module, and using the communications module to communicate with the network (i.e., LAN) <b>48,</b> at Block <b>293,</b> as described above. The method also includes using the logic CPLD <b>131</b> in cooperation with the cryptographic processor <b>36</b> to determine a status of the communications module <b>32,</b> at Block <b>294,</b> thus concluding the illustrated method, at Block <b>295.</b> Of course, it will be appreciated that status may be obtained (and/or configuration performed) prior to commencing communications with the network (i.e., LAN) <b>48,</b> and that repeated status updates may continue to be obtained through the communications process.

Another communications method aspect of the invention will now be described with reference to FIG. 20. The method begins (Block 300) with coupling the cryptographic module <b>31</b> to the network device <b>40,</b> as described above, at Block <b>301,</b> with a communications module <b>32</b> being removably coupled to the cryptographic module. The method may further include causing the host network processor <b>54</b> to generate cryptographic packets for the cryptographic circuit <b>57</b> each including an address portion and a data portion, and to encapsulate command packets for the network communications interface <b>47</b> in the data portions of the cryptographic packets, at Block <b>302,</b> as previously described above. Thus, if the cryptographic circuit <b>57</b> determines that a command packet is encapsulated in the cryptographic packet, the cryptographic circuit passes the command packet to the communications module <b>32</b> without performing cryptographic processing thereon, at Block <b>304,</b> as also discussed above. Otherwise, cryptographic processing is performed on the data in the cryptographic packet, at Block <b>305,</b> thus concluding the illustrated method (Block <b>306).</b>