International Business Machines Corporation

National Taiwan University

Amcad Biomed Corporation

Ming-Hsun Wu

Cheng-Ta Lee

Wei-Shiau Suen

Chih-Wen Chao

Wei-Hsiang Hsiung

Sheng-Tung Hsu

Rick M F Wu

Gregory Lyle Galloway

Chun-Shuo Lin

Ta-wei Lin

Ming-Pin Hsueh

Ying-Hung Yu

Ting-Jui Hu

Ping-Hung Lee

Ming Da Ho

Ping Min Lin

Li-Ju Chen

Shiming Lin

Argon Chen

An-Bang Wang

Kuo-Chen Huang

Paul Coccoli

I-Chun Lin

Yu-Hsin Wang

Yi-Hua Wang

Yin Lee

Chiung-Nein Chen

King-Jen Chang

Winson C W Chao

Winson Cw Chao

Chia-Chih Chen

King -Jen Chang

Chiung -Nein Chen

Yu -Hsin Wang

A network-based appliance includes a mechanism to set-up and selectively use an 'out-of-band' encryption channel. The mechanism comprises a packet parser, and a packet dispatcher, and it is integrated with an existing network layer stack that typically is not visible to host applications. In lieu of simply encrypting all data it receives, the mechanism instead analyzes one or more attributes, e.g., protocol type, application type, current encryption strength, content payload, etc., associated with a packet transmission to determine whether further encryption is required. The evaluation may include a deep packet inspection (DPI) when the information at the network layer (e.g., IP address, port number, etc.) is not sufficient to determine if the payload in the packet needs to be further encrypted. Based on the result of the analysis, packets are dispatched to the encryption channel as and when necessary. When additional encryption is not necessary, however, packet(s) are instead dispatched through an ordinary non-encrypted channel.

Context-based adaptive encryption

An intelligent network management device including an analytic unit, conducting an analysis according to received packets in order to determine whether a given event is occurred; and a processing unit, generating and sending a control instruction to a SDN controller to change configurations of a SDN switch when the analytic unit determined the given event has been occurred.

Intelligent network management device and method of managing network

Disclosed is a method of customizing an appliance. The method includes steps of pre-storing a public key in the appliance; connecting the appliance to an external storage device; and booting up the appliance to automatically proceed with the following customization process: obtaining a customization file from the external storage device; authenticating the customization file with the public key; and executing customization with the customization file if the authentication succeeds.

Managing the customizing of appliances

A Man in the Middle (MitM) computer receives a first session identifier from a client for a first communication session between the client and a server, and monitors Transport Layer Security (TLS) communication sessions between the client and the server, where the first session identifier is one of an unknown session identifier and an invalid session identifier. In response to receiving the first session identifier from the client, the MitM computer performs one of: requesting a second session identifier from the server for a second communication session if the first session identifier is an unknown session identifier; and transmitting, to the client, an instruction to flush a session cache in the client, where flushing the session cache in the client forces the client and the server to establish a full TLS handshake in order to obtain a session key if the first session identifier is an invalid session identifier.

Seamless abort and reinstatement of TLS sessions

Embodiments provide a system and method for network tracking. Through various methods of packet encapsulation or IP option filling, one or more packets of information can be tagged with a unique security tag to prevent unauthorized access. A user agent can be validated by an authentication server through acceptance of one or more user credentials. The authentication server can generate a security token that can be transmitted to the user agent. The user agent can generate a keystream from the security token, and portions of that keystream can be attached to the packets as the security tag. The tagged packets can be forwarded to an authenticator, who can recreate the keystream from a copy of the security token provided by the authentication server. If the tags generated from the authenticator match the tags on the tagged packet, the authenticator can strip the tag from the tagged packet and forward the packet on to its next network address.

System and method for providing persistent user identification

Methods and systems for high-availability data processing include detecting, at a first data processing system, a change in link state between the first data processing system and a second data processing system. A link state between the first data processing system and a third data processing system is changed responsive to the detection in accordance with a first high availability policy stored at the first data processing system. An identifier of the first data processing system is changed in accordance with the first high availability policy to conform to a second high availability policy stored at the first data processing system. The detection, change of the link state, and change of the identifier are repeated in accordance with the second high availability policy.

High availability in packet processing for high-speed networks

A network-based appliance includes a mechanism to provide TLS inspection with session resumption, but without requiring that a session cache be maintained. To this end, the inspector is configured to cause the TLS client to participate in maintaining the session context, in effect on behalf of the TLS inspector. In operation, when the inspector first receives a session ID from the TLS server, the inspector generates and issues to the client a session ticket that includes the original session ID and other session context information. In this manner, the inspector converts the Session ID-based connection to a Session Ticket-based connection. The session ticket is encrypted by the inspector to secure the session information. When the TLS client presents the session ticket to resume the TLS connection, the inspector decrypts the ticket and retrieves the session ID from it directly. The inspector then uses the original session ID to resume the TLS session.

Transparently converting a TLS session connection to facilitate session resumption

A network-based appliance includes a mechanism to provide TLS inspection with session resumption, but without requiring that a session cache be maintained. To this end, the inspector is configured to cause the TLS client to participate in maintaining the session context, in effect on behalf of the TLS inspector. In operation, when the inspector first receives the session ticket from the TLS server, and in lieu of caching it, the inspector generates and issues to the client a composited ticket that includes the original ticket and session context information that contains the session key. The composited ticket is encrypted by the inspector to secure the session information. When the TLS client presents the composited session ticket to resume the TLS connection, the inspector decrypts the ticket and retrieves the session context from it directly. The inspector then uses the original session ticket to resume the TLS session.

Cacheless session ticket support in TLS inspection

A computer program product for transmitting data flow in a network between two resources using a processing circuit to perform a method which includes obtaining a data record from a first resource, storing the data record and an associated data record identifier in a first memory, transmitting the data record from a first network to a second network, storing the data record and an associated data record identifier in a second memory, determining by an inline service provider whether the data record is suitable for transmission from a first resource to a second resource; based on determining that the data record is suitable for transmission by the inline service provider transmitting only the data record identifier stored in the second memory to the first switch and retrieving the data record stored in the first memory associated with the data record identifier for transmission to the second resource.

Packet flow in an overlay network

In response to receiving an unknown first session identifier from a client for a first communication session between the client and a server, a Man in the Middle (MitM) computer requests a second session identifier from the server for a second communication session between the server and the MitM computer. The MitM computer generates a third session identifier for a third communication session between the MitM computer and the client. The MitM computer generates a fourth communication session between the server and the client using a combination of the second communication session and the third communication session. In response to receiving an invalid session identifier from the client for a fifth communication session between the client and the server, the MitM computer transmits an instruction, to the client, to flush a session cache in the client to force a full TLS handshake between the client and the server.

A first client encryption initiation is intercepted from a client. The first client encryption initiation is intended for a server. Based on the first client encryption initiation, a second client encryption initiation is initiated with the server. Receiving a server response from the server responsive to the initiated second client encryption initiation. A first secure connection is negotiated with the client. The first secure connection is based on the intercepted first client encryption initiation and based on the server response. A session key to perform secure communication with the client is obtained from the first secure connection. A second secure connection is established with the server. The second secure connection is based on the server response and the session key.

Synchronizing secure session keys

An echogenicity quantification method and a calibration method for ultrasonic device using echogenicity index are disclosed. The method includes: receiving an ultrasound image which comprises a plurality of grayscale pixels; choosing a region of interest (ROI); calculating the values of the grayscale pixels in ROI to obtain average value and standard deviation; excluding pixels in ROI with the grayscale value smaller than the sum of the minimum value in the grayscale image and the product of a first scaling factor and the standard deviation, and larger than the sum of the average value and the product of a second scaling factor and the standard deviation; averaging the values of the remaining grayscale pixels in ROI to obtain an average value of interest; choosing a reference region; averaging the values of the remaining grayscale pixels in the reference region to obtain an average value of reference; and calculating the difference between the average value of interest and the average value of reference to obtain an echogenicity index.

Echogenicity quantification method and calibration method for ultrasonic device using echogenicity index

Growing community of inventors

Taipei, Taiwan

Ming-Hsun Wu