Granted Patent B2

US 12,579,225 · App. 18/261,920 · Granted Mar 17, 2026

Inventors: Tetsuya Okuda (Tokyo, JP); Koji Chida (Tokyo, JP); Yuichiro Dan (Tokyo, JP)

Assignee: NTT, Inc.

G06F21/10H04L63/101

Abstract

Claims (27)

1 . A remote authorization control system comprising:

a resource access device;

a resource management device; and

an authentication device,

wherein the resource access device downloads an access control list from the authentication device onto a secure region in the resource access device,

wherein the resource access device downloads resources from the resource management device onto the secure region,

wherein in response to receiving a resource use request, the authentication device executes a remote attestation to the secure region, acquires an ID of the secure region, and verifies the resource use request using the ID of the secure region,

wherein in response to a successful verification of the resource use request, the resource access device determines whether a user is allowed to use the resources based on the access control list, and allows the user to use the resources when the resources are usable, and

wherein the resources include a weight parameter for a deep neural network executed in the resource access device, and the weight parameter is set to the deep neural network when the resources are usable.

2 . The remote authorization control system according to claim 1 ,

wherein the authentication device registers a user ID of the user in association with the ID of the secure region in initial registration, and

wherein in response to receiving the resource use request, presence or absence of illegality is determined by confirming whether the user ID of the user and the ID of the secure region are registered as registration information.

3 . A resource access device included in a remote authorization control system, the remote authorization control system including the resource access device, a resource management device, and an authentication device,

wherein the resource access device includes:

a memory; and

a processor coupled to the memory and configured to:

download an access control list from the authentication device onto a secure region in the resource access device;

download resources from the resource management device onto the secure region; and

in response to receiving a resource use request, and executing a successful verification of the resource use request, by the authentication device, using an ID of the secure region acquired through a remote attestation to the secure region, determine whether the user is allowed to use the resources, based on the access control list, and allow the user to use the resources when the resources are usable, and

wherein the resources include a weight parameter for a deep neural network executed in the resource access device, and the weight parameter is set to the deep neural network when the resources are usable.

4 . A remote authorization control method performed in a remote authorization control system, the remote authorization control system including a resource access device, a resource management device, and an authentication device,

wherein the method includes:

downloading, by the resource access device, an access control list from the authentication device onto a secure region in the resource access device;

downloading, by the resource access device, resources from the resource management device onto the secure region; and

in response to receiving a resource use request, and executing a successful verification of the resource use request, by the authentication device, using an ID of the secure region acquired through a remote attestation to the secure region, determining, by the resource access device, whether a user is allowed to use the resources based on the access control list, and allowing the user to use the resources, by the resources access device, when the resources are usable, and

wherein the resources include a weight parameter for a deep neural network executed in the resource access device, and the weight parameter is set to the deep neural network when the resources are usable.

5 . A non-transitory computer-readable recording medium storing a program causing a computer to function as the resource access device according to claim 3 .

Assignments (2)

Continuity (1)

References Cited (11)