8.3 Identity Management
One of the fundamental aspects needed for security for IoT (and for that matter many Internet-based services) is the ability to identify, with certainty, users or devices that make up the system. For most Internet services, especially the free ones, there is no confirmed identity of a user. For other systems, the identity is based on either a bank account or credit card verification.
However, with a proliferation of services comes a proliferation of identity systems. For example, one Internet service provider, Comcast’s Xfinity, enables individuals to use their Facebook account as identity for signing-in to the Xfinity services and management Web site. The vulnerability here is obvious, and, in the very least, it doubles the opportunity for a hacker to obtain login credentials of an account. Of course, Xfinity is also an IoT provider, and associated with your account is the ability to add home security and home automation. With a compromise of a Facebook account, someone could come home after vacation to find his or her house cyber vandalized (with the heating system turned-off potentially causing the pipes to freeze) and/or to receive a follow-up phone call from the vandal (who now knows you are home because he can see you on your own security camera).
In addition to identifying users, IoT providers must have a system that enables them to identify their devices and for the device to identify the provider. In general, this is done using Public Key Infrastructure and is used to set up trusted connections between the device and provider for patches, software upgrades, and information exchange. Security here is maintained to the extent that the IoT manufacturer and service provider have good code and maintain good key discipline. Of course, how did they establish identity in the first place?
Recommendation: We need to rethink the approach to identity. For example, is there a role for government? Specifically, can U.S. Postal Service provide an identity vetting service?23
8.3 Identity ManagementOne of the fundamental aspects needed for security for IoT (and for that matter many Internet-based services) is the ability to identify, with certainty, users or devices that make up the system. For most Internet services, especially the free ones, there is no confirmed identity of a user. For other systems, the identity is based on either a bank account or credit card verification. However, with a proliferation of services comes a proliferation of identity systems. For example, one Internet service provider, Comcast’s Xfinity, enables individuals to use their Facebook account as identity for signing-in to the Xfinity services and management Web site. The vulnerability here is obvious, and, in the very least, it doubles the opportunity for a hacker to obtain login credentials of an account. Of course, Xfinity is also an IoT provider, and associated with your account is the ability to add home security and home automation. With a compromise of a Facebook account, someone could come home after vacation to find his or her house cyber vandalized (with the heating system turned-off potentially causing the pipes to freeze) and/or to receive a follow-up phone call from the vandal (who now knows you are home because he can see you on your own security camera).In addition to identifying users, IoT providers must have a system that enables them to identify their devices and for the device to identify the provider. In general, this is done using Public Key Infrastructure and is used to set up trusted connections between the device and provider for patches, software upgrades, and information exchange. Security here is maintained to the extent that the IoT manufacturer and service provider have good code and maintain good key discipline. Of course, how did they establish identity in the first place?Recommendation: We need to rethink the approach to identity. For example, is there a role for government? Specifically, can U.S. Postal Service provide an identity vetting service?23
การแปล กรุณารอสักครู่..