U.S. patent application number 12/394016 was filed with the patent office on 2012-01-19 for enhanced multi factor authentication.
Invention is credited to Steve Dispensa.
Application Number | 20120017268 12/394016 |
Document ID | / |
Family ID | 41381543 |
Filed Date | 2012-01-19 |
United States Patent
Application |
20120017268 |
Kind Code |
A9 |
Dispensa; Steve |
January 19, 2012 |
ENHANCED MULTI FACTOR AUTHENTICATION
Abstract
In one embodiment, a network element comprises one or more
processors, and a memory module communicatively coupled to the
processor. The memory module comprises logic instructions which,
when executed by the processor, configure the processor to receive,
via a first communication channel, a primary authentication request
transmitted from a user from a first device, process the primary
authentication request to determine whether the user is authorized
to access one or more resources, in response to a determination
that the user is authorized to access one or more resources,
initiate, a secondary authentication request, and transmit the
secondary authentication request from the network element to the
user via a second communication channel, different from the first
communication channel.
Inventors: |
Dispensa; Steve; (Leawood,
KS) |
Prior
Publication: |
|
Document Identifier |
Publication Date |
|
US 20090300745 A1 |
December 3, 2009 |
|
|
Family ID: |
41381543 |
Appl. No.: |
12/394016 |
Filed: |
February 26, 2009 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11862173 |
Sep 26, 2007 |
|
|
|
12394016 |
|
|
|
|
60866068 |
Nov 16, 2006 |
|
|
|
60939091 |
May 21, 2007 |
|
|
|
61031768 |
Feb 27, 2008 |
|
|
|
Current U.S.
Class: |
726/7 ; 726/3;
726/5 |
Current CPC
Class: |
H04L 63/0869 20130101;
H04L 63/18 20130101; H04L 63/166 20130101; H04L 63/168
20130101 |
Class at
Publication: |
726/7 ; 726/3;
726/5 |
International
Class: |
H04L 9/32 20060101
H04L009/32 |
Claims
1. A method to authenticate a resource access request, comprising:
receiving, in an authentication server in a communication network,
a primary authentication request transmitted from a user via a
first communication channel; processing the primary authentication
request to determine whether the user is authorized to access one
or more resources; in response to a determination that the user is
authorized to access one or more resources, initiating, in a
network element, a secondary authentication request; and
transmitting the secondary authentication request from the network
element to the user via a second communication channel, different
from the first communication channel.
2. The method of claim 1, wherein the primary authentication
request is originated from a first computing device coupled to the
communication network and includes a username and a password.
3. The method of claim 2, wherein processing the primary
authentication request comprises searching a data file maintained
by the authentication server for a corresponding username and
password.
4. The method of claim 2, wherein the primary authentication
request further comprises an identifier associated with a specific
network resource, and wherein processing the primary authentication
request comprises searching a data file maintained by the
authentication server for a corresponding username and password
associated with the specific network resource.
5. The method of claim 4, wherein transmitting the secondary
authentication request from the network element to the user via a
second communication channel comprises originating a call from a
network element to a telephone number associated with the
username.
6. The method of claim 5, wherein the call requests an
authentication response from the user.
7. The method of claim 6, wherein originating a call from the
network element to a telephone number associated with the username
comprises playing a prerecorded message approved by the user.
8. The method of claim 1, further comprising: receiving, in the
network element, a response to the secondary authentication
request; processing the response to the secondary authentication
request to determine whether the user is authentic; in response to
a determination that the user is authentic, granting access to one
or more network resources.
9. The method of claim 8, wherein the response to the secondary
authentication request comprises at least one of a key sequence
entered into a communication device or a voice message entered into
the communication device.
10. The method of claim 8, further comprising caching a successful
response to the secondary authentication request for a
predetermined period of time.
11. A network element, comprising: one or more processors; a memory
module communicatively coupled to the processor and comprising
logic instructions stored in a computer readable medium which, when
executed by the processor, configure the processor to: receive, via
a first communication channel, a primary authentication request
transmitted from a user from a first device; process the primary
authentication request to determine whether the user is authorized
to access one or more resources; in response to a determination
that the user is authorized to access one or more resources,
initiate, a secondary authentication request; and transmit the
secondary authentication request from the network element to the
user via a second communication channel, different from the first
communication channel.
12. The network element of claim 11, wherein the primary
authentication request is originated from a first computing device
coupled to the communication network element and includes a
username and a password.
13. The network element of claim 12, further comprising logic
instructions stored in a computer readable medium which, when
executed by the processor, configure the processor to search a data
file maintained by the authentication server for a corresponding
username and password.
14. The network element of claim 12, wherein the primary
authentication request further comprises an identifier associated
with a specific network resource, and further comprising logic
instructions stored in a computer readable medium which, when
executed by the processor, configure the processor to search a data
file maintained by the authentication server for a corresponding
username and password associated with the specific network
resource.
15. The network element of claim 14, further comprising logic
instructions stored in a computer readable medium which, when
executed by the processor, configure the processor to originate a
call from the network element to a telephone number associated with
the username.
16. The network element of claim 15, wherein the call requests an
authentication response from the user.
17. The network element of claim 16, further comprising logic
instructions stored in a computer readable medium which, when
executed by the processor, configure the processor to play a
prerecorded message approved by the user.
18. The network element of claim 17, further comprising logic
instructions stored in a computer readable medium which, when
executed by the processor, configure the processor to: receive, in
the network element, a response to the secondary authentication
request; process the response to the secondary authentication
request to determine whether the user is authentic; in response to
a determination that the user is authentic, grant access to one or
more network resources.
19. The network element of claim 18, wherein the response to the
secondary authentication request comprises at least one of a key
sequence entered into a communication device or a voice message
entered into the communication device.
20. The network element of claim 18, further comprising caching a
successful response to the secondary authentication request for a
predetermined period of time.
Description
RELATED APPLICATIONS
[0001] This application is a continuation-in-part of U.S. patent
application Ser. No. 11/862,173 by Steve Dispensa, entitled MULTI
FACTOR AUTHENTICATION. This application claims the benefit of U.S.
Provisional Application No. 61/031,768, filed Feb. 27, 2008,
entitled ENHANCED MULTI FACTOR AUTHENTICATION.
BACKGROUND
[0002] Internet access has become ubiquitous. In addition to
traditional dial-up and Local Area Network-based network access,
wireless access technologies including IEEE 802.11b and 802.11g
(WiFi), WiMax, Bluetooth.TM., and others are being widely deployed.
Many public locations, such as airports, bookstores, coffee shops,
hotels, and restaurants have free or fee-based access to wireless
Internet service. Some locations, such as hotel rooms, also offer
internet access via Ethernet ports. In addition, businesses offer
visiting professionals access to Internet service while they are on
the premises.
[0003] Such Internet access services typically are not secured at
the datalink layer. It is often possible for network
administrators, other users, or even criminals to capture and view
network transmissions made on these networks. The "last mile", or
the few hops on the network that are closest to the end user, are
often only lightly secured, if at all, and are particularly
vulnerable to traffic snooping. Enhanced communication security
would find utility.
[0004] In addition, authentication remains a persistent technical
problem in the information technology industry. With the
proliferation of untrusted applications and untrusted networks, and
the increasing use of the Internet for business functions, the
authentication issues have become prominent. Authentication refers
to a process by which a user makes his or her identity known to a
system or application which the user is attempting to access, and
occasionally, also the process by which the user verifies the
identity of the system being accessed. A common authentication
technique involves the use of a shared username and password
combination. This style of authentication is vulnerable to a number
of weaknesses. For example, passwords must be made long enough to
be secure while being short enough to be memorable. Additionally,
the loss of the password is sufficient to allow an attacker to gain
access to the system by impersonating the user. Therefore,
additional authentication techniques would find utility.
BRIEF DESCRIPTION OF THE DRAWINGS
[0005] The detailed description is provided with reference to the
accompanying figures. In the figures, the left-most digit(s) of a
reference number identifies the figure in which the reference
number first appears. The use of the same reference numbers in
different figures indicates similar or identical items.
[0006] FIG. 1 is a schematic illustration of a networked computing
environment in accordance with an embodiment.
[0007] FIG. 2 is a schematic illustration of a server in accordance
with an embodiment.
[0008] FIGS. 3-8 are flow diagrams of embodiments of methods for
secure network computing.
[0009] FIG. 9 is a schematic illustration of a networked computing
environment in accordance with an embodiment
[0010] FIG. 10 is a flow diagram of embodiments of a method for
multifactor authentication.
[0011] FIG. 11 is a schematic illustration of and embodiment of a
data file which may be used in a multifactor authentication.
DETAILED DESCRIPTION
[0012] Described herein are exemplary systems and methods for
secure network computing and multifactor authentication. The
methods described herein may be embodied as logic instructions on a
computer-readable medium. When executed on a processor, the logic
instructions cause a general purpose computing device to be
programmed as a special-purpose machine that implements the
described methods. The processor, when configured by the logic
instructions to execute the methods recited herein, constitutes
structure for performing the described methods.
[0013] In the following description, numerous specific details are
set forth in order to provide a thorough understanding of various
embodiments. However, various embodiments of the invention may be
practiced without the specific details. In other instances,
well-known methods, procedures, components, and circuits have not
been described in detail so as not to obscure the particular
embodiments of the invention.
[0014] FIG. 1 is a schematic illustration of a networked computing
environment in accordance with an embodiment. In the exemplary
architecture depicted in FIG. 1, one or more client computing
devices 110a, 110b, 110c establish a communication connection with
a point of presence (POP) server 130, which in turn communicates
with one or more target servers 140, 142, 144 via a network 120.
Target servers 140, 142, 144, in turn, provide access to one or
more computing resources such, as, e.g., internet services,
electronic mail services, data transfer services, and the like.
[0015] Client computing devices 110a, 110b, 110c may be any
computer-based communication device, including a personal computer
110a, a personal digital assistant (PDA) 110b, or a terminal device
110c. Client computing devices 110a, 110b, 110c establish a
communication with POP server 130 via a communication network,
which may be the same network 120 or a separate communication
network. The particular form of communication network is not
important. Communication network may comprise one or more direct
communication links (e.g., a dial-up connection) between respective
remote access devices 110a, 110b, 110c. Alternatively, the
communication network may comprise a private data network such as,
e.g., an X.25 network, a local area network (LAN), a wide area
network (WAN), or a public network such as, e.g., the Internet.
[0016] In one embodiment, POP server 130 may be implemented by a
general purpose computing device such as, e.g., a server, that
executes logic instructions which cause the processor to execute
various methods for performing secure network computing. FIG. 2 is
a schematic illustration of an exemplary computer system 200
adapted to perform secure network computing. The computer system
200 includes a computer 208 and one or more accompanying
input/output devices 206 including a display 202 having a screen
204, a keyboard 210, other I/O device(s) 212, and a mouse 214. The
other device(s) 212 can include a touch screen, a voice-activated
input device, a track ball, and any other device that allows the
system 200 to receive input from a developer and/or a user. The
computer 208 includes system hardware 220 and random access memory
and/or read-only memory 230. A file store 280 is communicatively
connected to computer 208. System hardware 220 includes a processor
222 and one or more input/output (I/O) ports 224. File store 280
may be internal such as, e.g., one or more hard drives, or external
such as, e.g., one or more external hard drives, network attached
storage, or a separate storage network.
[0017] Memory 230 includes an operating system 240 for managing
operations of computer 208. In one embodiment, operating system 240
includes a hardware interface module 254 that provides an interface
to system hardware 220. In addition, operating system 240 includes
one or more file systems 250 that managed files used in the
operation of computer 208 and a process control subsystem 252 that
manages processes executing on computer 208. Operating system 240
further includes a system call interface module 242 that provides
an interface between the operating system 240 and one or more
application modules 262 and/or libraries 264.
[0018] In operation, one or more application modules 260 executing
on computer 208 make calls to the system call interface module 242
to execute one or more commands on the computer's processor. The
system call interface module 242 invokes the services of the file
systems 250 to manage the files required by the command(s) and the
process control subsystem 252 to manage the process required by the
command(s). The file system 250 and the process control subsystem
252, in turn, invoke the services of the hardware interface module
254 to interface with the system hardware 220.
[0019] The particular embodiment of operating system 240 is not
critical to the subject matter described herein. Operating system
240 may be embodied as a UNIX operating system or any derivative
thereof (e.g., Linux, Solaris, etc.) or as a Windows.RTM. brand
operating system.
[0020] In one embodiment, memory 230 includes one or more network
interface modules 262, 268, one or more secure tunnel modules 264,
and one or more communication management modules 266. Network
interface modules may be implemented as web browsers such as, e.g.,
Internet Explorer, Netscape, Mozilla, or the like. Secure tunnel
module 264 comprises logic instructions which, when executed by a
processor, configure the processor to generate a secure
communication tunnel between the POP server 130 and a client
computing device such as, e.g., one or more of client computing
devices 110a, 110b, 110c. Communication management module 266
comprises logic instructions which, when executed by a process,
configure the processor to manage communications between the POP
server 130 and one or more client computing devices 110a, 110b,
110c and between the POP server 130 and the one or more servers
140, 142, 144.
[0021] In embodiments, POP server 130 receives a service request
from a client computing device such as, e.g., one or more of client
computing devices 110a, 110b, 110c, identifying one or more
resources available on a server such as 140, 142, 144. For example,
the service request may be embodied as a Uniform Resource Locator
(URL) transmitted to POP server 130 from a browser executing on a
client computing device. In response to the service request, POP
server 130 establishes a first communication link between the POP
server 130 and the one or more resources available via a computing
network identified in the service request. In one embodiment, POP
server 130 may launch an independent request for the resource
request for the resource identified in the service request from the
client computing device. POP server 130 may further establish a
first, secure communication link between the POP server 130 and the
client computing device 110a, 110b, 110c, and connect a network
interface module on the client computing device to the secure
communication link.
[0022] POP server 130 may further manage communication activity
between the client computing device and the one or more resources
available via a computing network at the POP server. In one
embodiment, managing communication activity may include passing
information received from a server 140, 142, 144 in response to a
resource request from the POP server 130 to a client computing
device 110a, 110b, 110 c via a secure communication link.
[0023] Operations implemented by the various modules 262, 264, 266,
268 and by client computing devices 110a, 110b, 110 c are explained
with reference to FIGS. 3-8. FIGS. 3-8 are flow diagrams of
embodiments of methods for secure network computing. FIG. 3 is a
flow diagram illustrating high-level operations executed by a
computing device such as one of client computing devices 110a,
110b, 110c in a method for secure network computing. In one
embodiment, at operation 310 a secure communication link is
initialized between POP server 130 and the client computing device
110a, 110b, 110c. At operation 315 the client computing device
sends outbound data via the secure communication link and at
operation 320 the client computing device receives inbound data via
the secure communication link. If, at operation 325 there are more
outbound data requests, then control passes back to operation 315.
Similarly, if at operation 330 there is more inbound data control
passes back to operation 320 and the inbound data is received. If
there are no further data requests or inbound data remaining, then
control passes to operation 335 and the secure communication link
may be terminated. Operations illustrated in FIG. 3 are explained
in greater detail in FIGS. 4-8.
[0024] FIG. 4 is a flow diagram illustrating operations in a method
for initializing a secure communication link between POP server 130
and a client computing device 110a, 110b, 110c. In one embodiment,
POP server 130 implements an application tunneling technology,
referred to herein as an AppTunnel, to construct a secure
communication tunnel between POP server 130 and a client computing
device 110a, 110b, 110c.
[0025] Referring to FIG. 4, at operation 410 POP server 130
receives an address of one or more resources such as, e.g., a
website or other resource available on network 120. In one
embodiment, the address received may represent a URL received in a
service request from a web browser executing on client computing
device 110a, 110b, 110c. In one embodiment, POP server 130
transmits an ActiveX control comprising launch data to client
computing device 110a, 110b, 110c. In alternate embodiments, the
ActiveX control client may be replaced with a plug-in module
compatible with other protocols such as, for example, the JAVA
architecture or a CORBA architecture.
[0026] At operation 420 the client computing device 110a, 110b,
110c receives the ActiveX launch data from server 130. If, at
operation 425 the client computing device is not compatible with
ActiveX technology, then control passes to operation 465 and a
rewriter host module may be activated. This procedure is explained
in greater detail below. By contrast, if the client computing
device 110a, 110b, 110c is capable of implementing an ActiveX
control, then control passes to operation 430 and the client
computing device 110a, 110b, 110 c, activates an ActiveX control
host page. The Active X control host page instructs the browser how
to load the ActiveX control, where to retrieve it from (if it is
not already locally cached), and with what parameters the control
should be started. A similar host page may be used to embedding
plug-ins into other browsers such as Mozilla, Netscape, etc. This
information may be contained in an HTML <OBJECT> tag.
[0027] At operation 435 the client computing device 110a, 110b,
110c initiates the ActiveX control received from the server 130.
The ActiveX control causes the client computing device 110a, 110b,
110c to launch a new incidence of a browser or other network
interface software (operation 440). If additional ActiveX controls
are necessary to enable the client computing device 110a, 110b,
110c, then one or more additional ActiveX controls maybe
transmitted between the server 130 and the client computing device
110a, 110b, 110c.
[0028] At operation 450 the client attaches a secure channel module
to the browser. In one embodiment, the secure channel module may be
embodied as an AppTunnel client module. The AppTunnel secure
channel module is described in detail in U.S. patent application
Ser. No. 10/737,200, incorporated by reference here above. Portions
of that description are excerpted in the following paragraphs.
[0029] Application tunneling is a method for transporting data from
a user's computer to a third party computer (a "proxy"). In one
implementation, Application data may be intercepted as soon as it
is sent (i.e., above layer 4 of the OSI model), before it is
encapsulated by Internet protocols such as TCP, UDP, or IP. Data
may then be transported across the network to the proxy. In other
implementations, this application-level data is acquired
differently (perhaps, for example, through the use of a virtual
adapter). In all cases, it is application-level data (OSI layers
5-7, depending on the application and the protocol) that is
tunneled via AppTunnels.
[0030] AppTunnels is application tunneling technology that tunnels
data from an application on a first computer to a second computer,
perhaps over a secure tunnel, for further processing and proxying
at that other computer. AppTunnels technology may be implemented in
a Static form or in a Dynamic form. Static AppTunnels technology
requires manual (or pre-configured) establishment of the
application tunnels and listeners. By contrast, dynamic AppTunnels
implements on-the-fly tunnel creation using hook mechanisms.
[0031] Encryption and/or other security technologies may be applied
to the tunnel to add security to the data being transported,
although this is not strictly necessary.
[0032] AppTunnels differs from existing tunneling technologies such
as Generic Routing Encapsulation (GRE) in that it is designed to
operate at the end user's computer, in such a way that the end
user's applications do not have to be informed about the existence
of the tunneling technology. By contrast, tunneling technologies
such as GRE, on the other hand, are designed to be implemented in
network elements such as routers.
[0033] AppTunnels differs from host-based encryption technologies
such as SSL in several ways. First, SSL technology must be directly
supported by the application in order to be applied. AppTunnels,
however, does not require application awareness in order to be
applied. Furthermore, SSL is closely tied to a particular security
and encryption architecture. AppTunnels may be used with or without
security technologies, and imposes no requirements on the
underlying security technologies. AppTunnels has been used in
conjunction with the WTP security protocol and with SSL.
[0034] To intercept the network traffic of an application a local
listener is created and a tunnel is established between the first
computer and the second computer. In one embodiment, a local
listener may be implemented as listening TCP, UDP, or other
socket(s) bound to a local host address (e.g., 127.0.0.1) on a
computer. The port number, where applicable, may be determined by
the tunneled application, and may be arbitrary in some protocols.
Local listeners may be created either in response to instructions
from dynamic AppTunnels hooks or by static configuration received
in advance. It is also possible for a user to manually initiate the
creation of a listener (and its associated tunnel). A data tunnel
may be created between the AppTunnels client software and a
compatible tunnel module on a server. In one embodiment, the tunnel
may be implemented in accord with the WTP protocol described in
U.S. patent application Ser. No. 10/737,200.
[0035] Once an AppTunnel has been initialized, the end user
application can be directed to connect to the AppTunnel. The
process for this varies per application. In the case of Dynamic
AppTunnels, no other action is necessary; data simply starts
flowing from the application to the AppTunnels software, which in
turn tunnels the data across the connection to the WTP
concentrator.
[0036] In the case of Static AppTunnels, one additional step is
required. The application is tricked into connecting to the local
listener instead of to the target server, as would otherwise
naturally be the case. To trick the client, the DNS system is
configured to return the localhost address (127.0.0.1, usually) to
requests for the destination server's IP address. This is usually
done by changing the locally-present "hosts" file that the
computer's DNS system consults before returning an IP address. This
hosts file is modified, with an entry being inserted for the name
of the target server with the IP address of localhost. In one
embodiment, the AppTunnels client may be implemented as an
executable component that is incorporated into and run by the
user's web browser such as, e.g., an ActiveX control. For other
browsers or other platforms, a Netscape Plugin may be used.
[0037] In one embodiment, AppTunnels implements a method of
intercepting traffic in order to tunnel it is as follows: First,
the computer's DNS resolution system is modified to re-route
traffic for target network servers to the local computer. Next, the
AppTunnels Client establishes itself as a server on the port that
the application would expect to connect to. Once established,
applications transparently connect to the Client on the local
computer rather than to their natural target-network servers. No
configuration or modification of the tunneled application is
necessary.
[0038] The AppTunnels method can be used to tunnel any user-mode
data over a tunnel to the server 130. This includes TCP and UDP on
all platforms. Other protocols may be available for tunneling,
depending on the platform. The AppTunnels architecture supports
complex network protocols such as FTP, RPC, H.323, and other
proprietary multi-connection protocols. It provides this support by
inspection of protocol data at the server. A protocol may be termed
`complex` if it requires more than a simple client-to-server TCP
connection.
[0039] In AppTunnels mode, the client computing device 110a, 110b,
110c receives a module called the AppTunnels Client, which attaches
to the web browser. The AppTunnels client enables the web browser
to access target network web pages by proxying requests through the
AppTunnels client. The AppTunnels client forwards the requests to
the server 130, which retrieves the requested document and returns
it to the AppTunnels client, which in turn returns it to the web
browser.
[0040] In brief, the client computing device 110a, 110b, 110c first
reads the proxy settings configuration from the user's web browser.
The client computing device 110a, 110b, 110c stores the proxy
settings and configures the browser to use a proxy
auto-configuration file. This file instructs the browser to request
its new proxy settings from the AppTunnels client. The request is
made and replied to, and the new settings cause all further
requests for documents to be proxied through the AppTunnels
client.
[0041] The AppTunnels client then establishes a connection to
server 130 and authenticates itself. After authentication, the
AppTunnels client establishes itself as a server application and
listens for incoming requests from the browser. As requests are
received, they are forwarded to server 130. Responses are read in
from the server and are sent back to the waiting browser.
[0042] The browser is monitored by Dynamic AppTunnels for any
network communication attempts. In one embodiment, these attempts
may be monitored using function call hooks of the Microsoft Winsock
library, but other hooks are possible, as are other monitoring
architectures (such as, for example, Winsock Layered Service
Providers). When new network traffic is detected, it is intercepted
by the Dynamic AppTunnels code for further processing.
[0043] In one embodiment, child processes created by the browser
may be injected with the Dynamic AppTunnels monitoring code. Thus,
network traffic generated by child processes send will be
encapsulated in the AppTunnel. Child process monitoring may be
implemented using an API hook for all of the CreateProcess( )
family of functions. When a call to CreateProcess( ) is made, the
Dynamic AppTunnels code receives it first, and ensures that the
monitoring code is injected in the resulting new process.
[0044] In one embodiment, dynamic AppTunnels technology is
implemented using API hooks. When dynamic AppTunnels receives a
request to start a tunneled process, it creates a new process using
the CreateProcess( ) API call. A new process may be created as
suspended, so that the process does not run after it is initially
created. At this point, one or more imported functions are
substituted, or hooked, such that they point to wrapper functions
that are part of the Dynamic AppTunnels software. The hooked
functions are of two classes: network functions and process
management functions.
[0045] In particular, the CreateProcess( ) function (and its
relatives) may be hooked so that any child processes that are
created can have the Dynamic AppTunnels monitoring code injected as
well. This code is responsible for signaling the browser plugin
that it should inject the rest of the hooks into the newly created
process. Any child processes of that child are treated in the same
way.
[0046] The networking functions that are hooked are related to
connection requests and to name resolution requests. In general,
all functions that are called during initial connection setup are
intercepted. Once hooked, these functions receive any connection
requests, and use this information for two purposes. The first is
to coordinate with the tunneling protocol on the AppTunnel server
to create an additional AppTunnel between the client and the
AppTunnel server, and to create a local listener that is attached
to that tunnel. The second is to re-direct the requesting
application to the local listening socket, so that connections are
made it instead of to the original target server. This process
allows network traffic generated by the client to be captured by
the browser plug-in and tunneled.
[0047] Alternatively, the plug-in may choose to examine the
connection request information and make a decision at runtime as to
whether the traffic should be tunneled or allowed to go straight
out to the network as it otherwise would have. These decisions can
be based on names (such as DNS or WINS names), network addresses,
port numbers, or other identifying information. While this
description has largely been written in the context of TCP sockets,
it should be pointed out that other kinds of network traffic may be
supported, including UDP and raw IP packets.
[0048] Referring back to FIG. 4, at operation 455 the browser
instantiated at the client computing device 110a, 110b, 110c loads
the requested resource, which may be displayed on a suitable user
interface such as, e.g., a computer screen or the like. At
operation 460, subsequent data transfer operations between the
client computing device 110a, 110b, 110c and the server 130 are
conveyed through the secure tunnel. When the user of the client
computing device is finished with the browsing session, the browser
may be closed. Closing the browser also closes any dynamic
AppTunnels constructed between the client computing device 110a,
110b, 110c and the server 130.
[0049] FIG. 5 is a flow diagram illustrating operations in a method
for implementing an AppTunnel on a client computing device 110a,
110b, 110c. In one embodiment, the ActiveX control transmitted from
the server 130 to the client computing device operations of FIG. 5
may cause the client computing device to perform the operations
illustrated in FIG. 5.
[0050] Referring to FIG. 5, at operation 510 one or more static
listeners and AppTunnels are initialized on the client computing
device. At operation 515 the AppTunnels plugin (i.e., the ActiveX
control) receives a request to start an application. For example,
the plugin may receive a request to start an instance of a
browser.
[0051] At operation 520 it is determined whether the device
supports dynamic AppTunnels. As described above, AppTunnels can
operate in either a static mode or in a dynamic mode. The
difference is in the way the data is acquired for tunneling. Static
AppTunnels uses a static local listening TCP/IP socket for each
pre-configured service. If a user wants to use a web browser over a
static AppTunnel, for example, there must be a configured
application tunnel listening on TCP port 80 on the local host.
Furthermore, the destination address for the AppTunnel must be
specified. To cause the user's application to connect to the
AppTunnels socket instead of trying to use Internet routes in the
usual way, the DNS system of the client computing device is
configured (usually using the computer's hosts file) to change the
IP address of the server in question to point to the local host
(usually 127.0.0.1), thereby fooling the application into making a
local connection to the listening socket.
[0052] Thus, referring to FIG. 5, if at operation 520 dynamic
AppTunnels is not enabled, then control passes to operation 525 and
the DNS system is configured, and at operation 530 a local
listening socket is created. At operation 535 a new process is
created.
[0053] By contrast, if at operation 520 the device accommodates
dynamic AppTunnels, then control passes to operation 540 and a new
process is created on the client computing device. In one
embodiment creating a new process may involve a user clicking an
HTML link on the web page in a browser executing on the client
computing device. The ActiveX control then launches the new process
and injects the Dynamic AppTunnels application monitoring code into
the newly started process (operation 545). At operation 550 a new
AppTunnel is created between the client computing device and the
server 130.
[0054] Following either operation 535 or 550, control passes to
operation 560 and the application receives data in the listening
socket generated by the AppTunnel. At operation 565 the data
received in the AppTunnel is removed and passed to the process
(i.e., the web browser) for further processing and presentation to
a user via a suitable interface such as, e.g., a display.
Operations 560-565 may be repeated until, at operation 570, there
is no more data to tunnel, whereupon operations terminate.
[0055] FIG. 6 is a flow diagram illustrating operations in a method
for processing outbound traffic from a network interface module
such as, for example, a web browser executing on a client computing
device 110a, 110b, 110c. The operations of FIG. 6 depict traffic
processing for a browser that has an AppTunnel module attached to
the browser. Referring to FIG. 6, at operation 610 the web browser
receives an address of a resource available on network 120. In one
embodiment, the address may represent a Uniform Resource Locator
(URL) of a resource available on network 120. At operation 615 the
browser initiates a service request for the resource. At operation
620 the service request is intercepted by the AppTunnel module. At
operation 625 the AppTunnel client secures the request data and at
operation 630 the AppTunnel client forwards the request to the POP
server 130. At operation 640 the secured request is received at the
POP server 130. At operation 645 the secure tunnel module 264
(i.e., the AppTunnel server) extracts the request data from the
secure tunnel.
[0056] At operation 650 the POP server forwards the service request
to the address identified in the service request. In one
embodiment, the service request is received in a first network
interface module 262 instantiated on POP server 130, and POP server
130 instantiates a second network interface module 268 and launches
a service request from the second network interface module.
[0057] FIG. 7 is a flow diagram illustrating operations in a method
for processing inbound traffic such as, for example, one or more
resources returned from a service request. Referring briefly to
FIG. 7, at operation 710 the data returned by the resource request
is received at the POP server 130. In one embodiment, the data is
received in the second network interface module 268 instantiated on
the POP server 130. At operation 715 the response data is secured.
In one embodiment, response data is operated on by the secure
tunnel module 264. At operation 720 the response data is placed in
the secure tunnel to the client computing device 110a, 110b, 110 c
via the first network interface module 262.
[0058] At operation 725 the client receives the response data
transmitted to the client by the server. At operation 730 the data
is removed from the secure tunnel established by AppTunnels. In one
embodiment, the AppTunnels client attached to the browser in the
client computing device 110a, 110b, 110c removes the received data
from the tunnel and, at operation 735, forwards the data to the web
browser. The web browser may present the data on a user interface
such as, e.g., a display, for viewing by the user.
[0059] Referring back to FIG. 4, if at operation 425 the client
computing device is not capable of executing a plugin such as,
e.g., an ActiveX control, then control passes to operation 465 and
a URL rewriter technique is activated. URL rewriting is a method of
providing a reverse proxy server for use in secure remote access
systems and other applications without the need for any locally
installed code. Most web browsers can exchange traffic with web
servers using a protocol known as secure HTTP, or HTTPS. However,
most websites do not support HTTPS. To add security to every
website the user requests, special HTTPS requests are made to a URL
rewriter server instead of the target web server, using HTTPS to
the rewriter. The rewriter then forwards the request to the target
web server by proxy, using the expected destination protocol of the
web server (typically HTTP). On return, the web data is returned
over HTTPS to the client's browser. It should be noted that any
other form of browser-based security, or no security whatsoever,
could be used in the communications link to the rewriter.
[0060] It should be noted that references to other web data will
now cause the user's browser to make direct requests to the target
server, rather than ask the rewriter server. Thus, references to
other web content (hyperlinks, images, java applets, and so on) may
be rewritten in the webpage so that they refer to the rewriter
instead of to the target server to ensure that all web traffic is
routed through the rewriter, and not via direct connections.
Accordingly, any web references in the returned document are
rewritten with references to the URL rewriter server.
[0061] FIG. 8 is a flow diagram illustrating operations in a method
for processing inbound traffic such as, for example, one or more
resources returned from a service request. Referring briefly to
FIG. 8 at operation 810 a rewriter host page on the client
computing device 110a, 110b, 110c launches a second web browser
using instructions embedded in the HTML and JavaScript code that is
a part of the page. At operation 815 the rewritten URL is written
into the second browser. At operation 820 the second browser
requests the specified resource using the rewritten URLs.
[0062] At operation 825 the server 130 receives the service request
from the client computing device 110a, 110b, 110c. At operation 830
the server 130 un-rewrites the URL, and at operation 835 the server
retrieves the requested resource from the server 140, 142, 144
hosting the resource on the network 120. At operation 840 the
server rewrites the embedded URLs in the retrieved resource, and at
operation 845 the server returns the rewritten resource to the
client.
[0063] At operation 850 the client computing device 110a, 110b,
110c receives and processes the requested resource. In one
embodiment, processing the requested resource may include
presenting the resource on a suitable display. If at operation 855
there are more requests to be processed, then control passes back
to operation 820, and the browser requests the specified
resource(s). By contrast, if at operation 855 there are no further
resource requests, then the process terminates.
[0064] Thus, the operations described in FIGS. 3-8 enable a client
device such as one or more of client computing devices 110a, 110b,
110c to establish a secure "last mile" communication link, thereby
enabling secure communication with resources hosted by one or more
servers 140, 142, 144 in a network 120. The systems and methods
described herein are agnostic regarding the type of encryption
applied (if any) to communication links between POP server 130 and
servers 140, 142, 144.
[0065] The systems described above are browser-based systems. In
alternate embodiments, a client computing device 110a, 110b, 110c
may download and install a permanent piece of client encryption
module onto the end user's computer. The permanent client provides
encryption services between the client computing device 110a, 110b,
110c and the servers 140, 142, 144. With a permanent client
encryption module, the end user does not have to navigate to the
service provider's website in order to turn on secure surfing.
Further, a client-based approach can support all Internet-based
applications.
[0066] In another embodiment, an ActiveX control installs and
initializes a virtual VPN Miniport, as described in U.S. patent
application Ser. No. 10/737,200. The ActiveX control captures and
processes any relevant traffic on the computer. In this way, end
user traffic is directed into the secured tunnel described
above.
[0067] A virtual VPN Miniport can support non-TCP protocols, while
maintaining the convenience of a web-based environment. It can also
secure applications that have already been started. For example, if
the end user were running an instant messaging client, that
client's communications would be secured from the time of
connection forward, without a need to re-launch the client.
Installation of a VPN driver can be silent and automatic, and only
needs to occur one time. From that point on, the ActiveX control
activates the VPN driver and manages network routes to cause
traffic to be directed into the VPN driver for encryption.
[0068] Access to the POP server 130 may be provided by a number of
methodologies. In one embodiment access to POP server 130 may be
provided on a pay-for-service business model. In this embodiment,
POP server 130 may include a transaction processing module to
process payment transactions, e.g., by a credit card or other
payment mechanism. Charges may be levied on the basis of bandwidth
consumed, or by a time parameter (i.e., minutes, hours, days,
years, etc.).
[0069] In alternate embodiments access to server 130 may be
implemented on the basis of advertising revenue. For example,
pay-per-click advertising can be implemented, using
randomly-selected advertisements, pay-per-click advertising can be
implemented, using information gained by inspecting the user's
traffic during secure surfing to select targeted advertisements, or
traditional Internet banner advertisements can be inserted, either
randomly or based on inspection of the secured data. Advertisements
can be inserted in the initial service provider web page, in the
refreshed web page that hosts the ActiveX control, or even inline
with the displayed web pages.
[0070] Multiple levels of service may be defined. For example,
low-quality service might be provided free of charge, while
high-quality service might be provided for a fee. Service levels
may be differentiated by one or more facts such as, for example
throughput (i.e., a performance aspect might be controlled),
bandwidth (i.e., the total number of bytes transferred might be
capped), data transfer (i.e., a transfer cap might be imposed per
day, per month, or per year), or some combination thereof. Other
options include allowing only a limited amount of time to use the
system or bandwidth in a tier or a certain amount of throughput for
a certain amount of time, with decreased throughput after the
elapse of that time, or restricting access to certain resources
based on the service level.
[0071] Last-mile encryption service may be provided to a user with
no pre-existing account. The web-based interface can be used simply
by entering the address of a website that is to be securely
accessed. In the installed client case, the user logs in
anonymously using the supplied anonymous login method. Anonymous
users may be granted a different tier of service, as described
above.
[0072] In an anonymous user case, cookie-based, form-based, or IP
address-based information may be used to correlate the anonymous
user's browsing activities, for purposes including providing
advanced service features (such as browsing history, enhanced
status reporting, etc), selecting relevant advertising, or for
other purposes.
[0073] Users that desire temporary top-tier service may be given
the option of paying electronically for a one-time use of the
service without creating an account.
[0074] Users that wish to use the service repeatedly may wish to
create a user account. The user account could then be used to track
browsing history, make more intelligent decisions about advertising
content to be presented, or offer other value-added services. Users
of the service would then be prompted to log in to the website (or
to the downloaded client software) using the established
authentication credentials. Optionally, a cookie-based login
persistence mechanism can be supported, allowing the user to go for
a period of time without the need to log in.
Multi-Factor Authentication
[0075] As described above, authentication remains a technical
challenge in the information technology sector. Described herein
are multifactor authentication techniques which may be used alone,
or in combination with other security techniques described herein
to provide secure access to resources in a computer network.
Embodiments of multifactor authentication techniques will be
described in the context of a computer network similar to the
network described in with reference to FIG. 1. It will be
understood, however, that authentication techniques as described
herein may be implemented in a wide variety of computer
networks.
[0076] FIG. 9 is a schematic illustration of a networked computing
environment in accordance with an embodiment. In the exemplary
architecture depicted in FIG. 9, one or more client computing
devices 910a, 910b, 910c, 910d, 910e establish a communication
connection with an authentication server 930, which in turn
communicates with one or more target servers 940, 942, 944 via a
network 920. Target servers 940, 942, 944, in turn, provide access
to one or more computing resources such, as, e.g., internet
services, electronic mail services, data transfer services, and the
like.
[0077] Client computing devices 910a, 910b, 910c, 910d, 910e may be
any computer-based communication device, including a personal
computer 910a, a personal digital assistant (PDA) 910b, a terminal
device 910c, a mobile telephone 910d, or a land-line telephone
910e. Client computing devices 910a, 910b, 910c, 910d, 910e
establish a communication with authentication server 930 via a
communication network, which may be the same network 920 or a
separate communication network. The particular form of
communication network is not important. Communication network may
comprise one or more direct communication links (e.g., a dial-up
connection) between respective remote access devices 910a, 910b,
910c, 910d, 910e. Alternatively, the communication network may
comprise a private data network such as, e.g., an X.25 network, a
local area network (LAN), a wide area network (WAN), or a public
network such as, e.g., the Internet.
[0078] Authentication server 930 may be embodied as a computing
device, substantially as described in with reference to FIG. 2,
above. Referring briefly to FIG. 2, the computing device 200 and
comprise one or more authentication modules 269 which may execute
when as an application module and the memory 230 of the computing
system 200. In some embodiments, the authentication module 269 and
implemented logic instructions which, when executed by a processor
such as the processor 222, cause the authentication module 269 to
implement multifactor authentication procedures to manage access to
one or more resources of the computer network 920, such as for
example, resources provided by servers 940, 942, or 944.
[0079] In some embodiments, the authentication server 930
implements a first authentication process in response to an
authentication request from a client computing device such as one
of client computing devices 910a, 910b, 910c, 910d, 910e. If the
first authentication process is successful, then the authentication
server 930 originates a second authentication request to a client
device such as one of client computing devices 910a, 910b, 910c,
910d, 910e. In some embodiments, the authentication request from
the client is transmitted through a first communication channel and
the second authentication request originated by the authentication
server 930 is transmitted using a second communication channel,
different from the first communication channel. The authentication
server 930 may process the response to the second authentication
request and allow or deny access to a resource based on the
response. In some embodiments the first communication channel and
the second communication channel may be across separate
communication networks. For example, the first communication
channel may be across the computer network, while the second
communication channel may be across a telephone network.
[0080] One embodiment of multifactor authentication will be
explained with reference to FIG. 10, which is a flow diagram of
embodiments of a method for multifactor authentication. Referring
to FIG. 10, at operation 1010a first client initiates a primary
authentication request for access to a resource provided by
computer network 920. In some embodiments, the primary
authentication request may include a username and password
combination associated with a user and/or a device from which the
primary authentication request is originated. The primary
authentication request may be transmitted to the authentication
server 930 to a first communication channel.
[0081] Operation 1015 the authentication server 930 receives the
primary authentication request, and at operation 1020 the
authentication server 930 processes the primary authentication
request. In some embodiments, the authentication server 930
performs a centralized authentication function which manages
authentication to one or more resources and network 920. For
example, authentication server 930 may maintain a data file
comprising username and password combinations which may be
associated with one or more resources of the computer network 920.
FIG. 11 is a schematic illustration of and embodiment of a data
file which may be used in a multifactor authentication. Referring
briefly to FIG. 11, the illustrated data file 1100 includes a
column for usernames, a column for passwords, and a column for
approved resources. Usernames and passwords may be logically
associated with the approved resource indicated in the table 1100.
A single username may be associated with multiple passwords for
different approved resources. In one embodiment, processing the
primary authentication request may comprise searching the data file
1100 maintained by the authentication server for a username and
password combination that corresponds to the username and password
combination receipt in the primary authentication request.
[0082] If, an operation 1025, the primary authentication request is
unsuccessful, i.e., if there is no corresponding username and
password combination in the data table 1100, then the
authentication server 930 denies the requestor access to network
resource(s). in some embodiments, the authentication server 930 may
transmit an error message to the requestor indicating that the
username and password are invalid. Control that returns to
operation 1015 and the authentication server 930 continues to
monitor for another primary authentication request.
[0083] By contrast, if that operation 1025 primary authentication
request is successful, i.e., if there is a corresponding username
and password combination in the data table 1100, then control
passes to operation 1035 and the authentication server 930
initiates a secondary authentication request. The secondary
authentication request is transmitted from the authentication
server 930 to the user via a second communication channel,
different from the first communication channel. In some
embodiments, the secondary authentication request is transmitted
from the authentication server 930 to a second client in the user's
possession. For example, the user may initiate the primary
authentication request from a computing device such as a desktop
computer or laptop computer and the authentication server 930 may
transmit the secondary authentication request by initiating a
telephone call to a telephone registered to the user. Referring
briefly again to FIG. 11, a user may register, via a suitable user
interface, a contact number to which the secondary authentication
request may be transmitted from the authentication server 930. In
response to a successful primary authentication request, the
authentication server 930 may initiate a call to contact number
indicated in the data table 1100. It should be noted that a user
may provide different contact numbers for different resources.
[0084] At operation 1040 the secondary authentication request is
received at the second client. The secondary authentication request
may comprise a voice message which makes a request for information
to authenticate the user. In some embodiments, the system may allow
a user to prerecord a customized secondary authentication request
in the user's own voice. For example, the user may record a message
requesting a specific sequence of keystrokes or requesting the user
to speak to a specific word or group of words. Having a user
recorded message in the second authentication request helps to
authenticate the system to the user, thereby eliminating or at
least reducing the likelihood of a "man in the middle" attack on
the system. In alternate embodiments, rather than using a
prerecorded message in the user's voice, a user may select one or
more tokens to be presented with a secondary authentication
requests. For example, a token may include a predetermined word or
character or numeric sequence selected by the user.
[0085] At operation 1045 the user response to the secondary
authentication request initiated by the authentication server 930.
For example, in some embodiments the user may respond by pressing a
predetermined sequence of keystrokes on a telephone keypad, or by
pressing the pound key or the star key. In alternate embodiments
the user may respond by speaking a predetermined word or series of
words. In alternate embodiments the user need not provide an
affirmative response; simply answering the telephone call may
suffice as a response.
[0086] In some embodiments, the secondary authentication request
initiated by the authentication server 930 may be implemented as a
text message rather than a telephone call. Accordingly, the
response to the secondary authentication request may also be
implemented as a text message in which the user transmits a
predetermined character or series of characters back to the
authentication server 930.
[0087] In alternate embodiments, the secondary authentication
request may require a user to initiate a call back in order to
authenticate the user. For example, the secondary authentication
request may transmit a text message or a voice call requesting the
user to call back to the system to authenticate the user. In some
embodiments, a return phone number may be included with the
secondary authentication request, while in other embodiments a user
may be required to call a predetermined phone number. As described
above, the user may be required to provide one or more codes in the
secondary authentication response.
[0088] At operation 1050 the authentication server 930 receives the
response to the secondary authentication request, and at operation
1055 the authentication server 930 processes the response. In some
embodiments, authentication server 930 maintains authentication
codes which represent the anticipated response to the secondary
authentication request in the data table 1100. The response to the
secondary authentication request received from the user may be
compared with the authentication code stored in the data table 1100
in order to determine whether the user is authentic.
[0089] If, at operation 1060, the response to the secondary
authentication request fails to successfully authenticate the user
then access to network resources is denied at operation 1065 and
control passes back to operation 1015 in the authentication server
monitors for additional incoming primary authentication requests.
In some embodiments, the authentication server 930 may implement an
error routine in response to a failed a secondary authentication
request. The authentication routine may transmit an error message
to the user via the first communication channel, the second
communication channel, or both. The error message may instruct the
user that authentication has failed and they provide the user with
an opportunity to restart the authentication process.
[0090] By contrast, is that operation 1060 the response to the
secondary authentication request successfully authenticates the
user then control passes to operation 1070 and the user is granted
access to the network resource or resources associated with the
username and password in the data table 1100. Control then passes
back to operation 1015 and the authentication server 930 continues
to monitor for additional primary authentication request.
[0091] Thus, the operations depicted in FIG. 10 enable the network
infrastructure depicted in FIG. 11 to implement a multifactor
authentication process. In some embodiments described herein, the
multifactor authentication process utilizes two separate network
devices, i.e., a computing device and a telephone. In some
embodiments, the multifactor authentication process may utilize a
single network device, i.e., a computing device, which executes two
or more logical network devices. For example, a user may initiate a
primary authentication request from a first application executing
on the computing device, and the second authentication request may
be directed to a second application executing on the computing
device. For example, the second application may be an Internet
Protocol (IP) telephony application.
[0092] Various features may be added to the functionality of the
basic authentication process described herein. In some embodiments,
the authentication server 930 may store in a memory module such as
cache memory the results of a primary authentication request
initiated by a user, alone or in combination with the results of a
secondary authentication response provided by the user. The results
may be stored in for a predetermined period of time or for a
dynamic period of time. Thus, when a user has successfully
authenticated himself or herself to the system additional
authentication may not be required during the time period. The
authentication server 930 may require that subsequent primary
authentication requests be initiated from the same network address
in order to bypass the secondary authentication request. Thus, in
some embodiments the authentication server 930 may detect the
network address from which the primary authentication request is
initiated and may store the network address in a memory module.
[0093] Further, there may be circumstances in which secondary
authentication requests may not be necessary. For example, if a
user is located on a trusted network in the secondary
authentication request may be bypassed. Thus, in some embodiments
of the authentication server 930 may detect the network address
from which the primary authentication request is initiated and may
compare the network address with a list of approved network
addresses stored in a memory module.
[0094] Still further, there may be circumstances in which the
authentication server 930 declines to initiate a secondary
authentication requests. In some embodiments, in the event of a
predetermined number of failures for a primary authentication
request the authentication server 930 may flag a user as a suspect
for fraudulent access and may decline to initiate a secondary
authentication request unless further conditions are met. In some
embodiments, in the event multiple primary authentication requests
are received from different network addresses within a
predetermined period of time the authentication server may flag a
user as a suspect for fraudulent access and may decline to initiate
a secondary authentication request unless further conditions are
met. For example, a user may be required to reset passwords or to
speak personally with an administrator.
[0095] In some embodiments, the authentication server 930 may
provide a user interface that enables users to register one or more
telephone numbers or contact addresses for the network device
intended for use for the secondary authentication request. The user
interface may further permit users to select one or more
authentication codes or personal identification numbers (PINs) for
both the primary authentication request and the secondary
authentication response.
[0096] Various alternate embodiments may be implemented. For
example, n some situations it may not be possible to submit a
user's primary authentication credentials to the authentication
server without incurring unwanted side-effects. For example,
logging into a web application using primary authentication
credentials may cause the application to take actions such as
creating a user session, logging a message, or the like that may be
undesirable.
[0097] In such situations, a the authentication server may
implement a pre-authentication process. After receiving the primary
authentication credentials, the authentication system can attempt
to pre-authenticate them using a different API interface, rather
than pre-authenticating to the target server itself. For example,
in a web application such as Microsoft's Outlook Web Access, the
system may pre-authenticate the user by calling the Windows
LogonUser( ) API, which checks the user's username and password
against the Windows password database. Alternatively, in a Citrix
environment, the system could pre-authenticate the user using the
Citrix authentication APIs.
[0098] If the pre-authentication step is successful, the secondary
authentication may be implemented as described before. Only if that
is also successful are the user's credentials submitted to the
application in question for final log-in. This becomes a
three-phase login, but it has the benefit of allowing compatibility
with applications that would not otherwise support the two-phase
approach.
[0099] In addition, the strength of the authentication process can
be increased using voice-print technology during the confirmation
call. During the secondary authentication call, the system asks the
user to repeat a series of words. The user repeats the words, and
the system makes a determination of whether the user is who he
claims to be by evaluating the user's voice against a voice
database, using voice matching algorithms.
[0100] Again, this makes the system three-factor: the primary
authentication is something the user knows, the secondary is
something the user has (phone), and the tertiary authentication is
something the user is (his voice).
[0101] This system can also be used for multi-person
authentication. For example, in situations which require the
approval of more than one to allow an action to complete, multiple
secondary authentication calls can be placed. For example, in the
case of a bank transfer requiring two people to agree to the
transaction, the system may place multiple confirmation calls, one
to every person authorized to approve the transaction. It could
then play back details of the proposed transaction to each user
(which can happen simultaneously), and if a minimum number of those
users confirm the transaction, the system returns success. This
system can scale to an arbitrarily large number of required
confirmations.
Enhanced Multi-Factor Event Confirmation System
[0102] Security-related events occur continuously on a day-to-day
basis. Security events include authentications to secure computer
systems, financial transactions through a bank or brokerage, or
particular line-of-business events such as the submission of source
code to a source control system or the accessing of a patient's
medical records. These events provide an opportunity for fraud,
both on an individual and on a bulk scale, and consequently,
securing these events is of the greatest importance.
[0103] Existing event confirmation systems typically rely on a
single authentication factor being presented by the person
triggering the event. For example, login to a company's remote
access system is typically secured using a username and a password
assigned to the user logging in. Or, in the case of a credit card
transaction, the card number (which is, essentially, a secret known
only to the cardholder) is presented, along with the associated
expiration date and sometimes a signature.
[0104] Each of these existing single-factor systems carries with it
the implicit weakness associated with the single factor in
question: secrets can be lost or stolen, cards can be cloned, and
so on. Introducing a second factor into the verification process
can significantly increase the security of event confirmations. The
PhoneFactor system provides such a second factor.
[0105] PhoneFactor operates by placing a telephone call over the
public telephone network to a user's pre-registered phone number
(or one of several pre-registered phone numbers). The phone call
includes any relevant details of the event in question and prompts
the user to confirm the event. The user confirms the event by
entering a series of digits and/or symbols using the phone's
keypad.
[0106] Generally:
[0107] 1) User initiates an event
[0108] 2) PhoneFactor verifies any required business rules (correct
username/password, for example)
[0109] 3) If successful--PhoneFactor places a call to the user's
pre-registered phone number
[0110] a. If successful--PhoneFactor plays back event data to the
user and prompts the user to confirm the event
[0111] b. If the data match the user's expectations, the user
confirms the event by entering a pre-arranged set of keystrokes
[0112] ) If a failure occurs at any point during the above
procedure, the event is deemed not to be confirmed.
[0113] The out-of-band nature of PhoneFactor dramatically
complicates the attacker's job, since he now must successfully
subvert two entirely different networks based on two entirely
different technologies.
[0114] For example, one event type is login to a computer network
through a remote access system. After the user's username and
password are entered and confirmed, PhoneFactor generates a phone
call to a pre-registered phone number including relevant details of
the event, such as the geographic location from which the login is
being attempted. If the information provided by the PhoneFactor
phone call matches the user's expectations, the user enters a
pre-registered sequence of keypresses into the phone, confirming
the event.
[0115] Another example is in the realm of online banking. Say, for
example, a bank customer initiates a wire transfer for $1,000 to an
account ending in 1111. The PhoneFactor system makes a phone call
to the user's registered phone number and reads off this
information. If the information matches the user's expectations,
the user confirms the event in the usual way.
[0116] Another example of event confirmation is in the area of
healthcare. A healthcare worker may log into a medical records
system, triggering a login event confirmation similar to the one
described above. Later in the session, the user may add a
prescription to the patient's record, prompting an additional event
confirmation. In this way, multiple event confirmations can be
mixed and matched, allowing implementers of the technology to make
tailored decisions about which events benefit from confirmation in
which circumstances.
[0117] PhoneFactor can add event confirmations to a variety of
existing systems in a variety of circumstances. One kind of
integration involves the use of third-party fraud scoring systems.
These systems evaluate a group of factors, such as time of day,
network addressing information, and so on, to determine if the
transaction taking place is likely to be legitimate or fraudulent,
and if it is suspected to be the latter, an event confirmation
using PhoneFactor can be initiated.
[0118] There are several cases in which the user would refuse to
confirm the event, or even signal a fraud alert. One such case is
if the user receives a phone call that was not expected--for
example, the user is driving down the road, nowhere near a
computer, and the phone rings. The user would refuse to confirm the
event because he did not trigger it.
[0119] Fraud alerts can be sent in real-time at the request of the
user to alert the appropriate company representatives or
authorities that a fraudulent event has been triggered. This kind
of "hot lead" may improve investigators' chances at locating the
fraudster.
[0120] The PhoneFactor event confirmation system may be implemented
as follows. Administrators of the system to be protected define in
advance a variety of events for which confirmation is desired.
These events are configured using a computer-based confirmation
interface that allows the user to create an appropriate event
template, consisting of: [0121] One or more pre-recorded outgoing
messages to the user, selected from a chosen message set (perhaps
per-language, per-region, etc.) [0122] Zero or more digit strings,
which are rendered in spoken language during the event confirmation
call using pre-recorded numerical recordings, selected from a
language number set [0123] Zero or more text-to-speech fields,
rendered in the user's selected language [0124] The order in which
these elements are to appear in the outgoing PhoneFactor message is
supplied
[0125] Pre-recorded static messages may also be included in the
outgoing message, for purposes such as marketing or user
information.
[0126] Event templates may be customized on a per-language,
per-region, or other basis. These message sets contain one message
for each requirement in the event template. For example,
administrators may define an English message set and a Spanish
message set. During submission of the PhoneFactor request, the
submitter requests a specific message set, and messages from that
message set are chosen to be played in the outgoing PhoneFactor
message.
[0127] The combination of event templates, message sets, and the
specified response keystroke sequence allows for a variety of
implementation alternatives. For example, the user could be
prompted using the Wire Transfer template, with the English message
set, and the confirmation code could be the last four digits of the
user's bank account number.
[0128] PhoneFactor event confirmation can be applied to a wide
variety of problem spaces and contexts, and can play an important
role in fraud prevention when dealing with sensitive events and
information.
[0129] Reference in the specification to "one embodiment" or "an
embodiment" means that a particular feature, structure, or
characteristic described in connection with that embodiment may be
included in at least an implementation. The appearances of the
phrase "in one embodiment" in various places in the specification
may or may not be all referring to the same embodiment.
[0130] Also, in the description and claims, the terms "coupled" and
"connected," along with their derivatives, may be used. In some
embodiments, "connected" may be used to indicate that two or more
elements are in direct physical or electrical contact with each
other. "Coupled" may mean that two or more elements are in direct
physical or electrical contact. However, "coupled" may also mean
that two or more elements may not be in direct contact with each
other, but may still cooperate or interact with each other.
[0131] Thus, although embodiments of the invention have been
described in language specific to structural features and/or
methodological acts, it is to be understood that claimed subject
matter may not be limited to the specific features or acts
described. Rather, the specific features and acts are disclosed as
sample forms of implementing the claimed subject matter.
* * * * *