The computer implemented system for detecting an anomaly in a set of data gathered progressively in time with inputs and outputs includes qualifying modules for determining if the data qualify for going through the analysis process; overview model modules for determining if the global data are abnormal by means of one or more overview model; detail model modules for determining if one or more individual data is abnormal by means of one or more detail model, in particular if the global data are abnormal; and/or AI modules for analyzing the data based on deep learning/neural networks analysis with autoencoders; and/or machine learning or reinforcement learning and/or Multilayer Perceptron (MLP) procedures to detect patterns of data. The invention aims in particular at finding singular anomalies, in particular in company accounts.

Systems, methods, and devices are disclosed for establishing sessions over a network. A query can be sent, from a mobile device to an authoritative DNS server, requesting a session over a network. In response to the query, a list of available communication protocol options can be received from the authoritative DNS server. A specific communication protocol can be selected from the list of available communication protocol options, where the specific communication protocol is selected based on one or more performance metrics. A connection with a gateway device can then be initiated using the selected specific communication protocol.

This disclosure describes techniques that enable a service provider to synchronize public-private Internet protocol (IP) pairs between a central logging entity (CLE) server that attributes network traffic and network address translation NAT device that logs network traffic. The CLE server includes a communication protocol client communicatively coupled to a communication protocol host. The CLE server, in response to receiving a deregistration message from a registration protocol server, sends a clear request message comprising a private address via the communication protocol client. The deregistration message includes the private address. The NAT device includes the communication protocol host. The NAT device, in response to receiving the clear request message comprising the private address via the communication protocol host, clears a binding that maps the private address to a public address to prevent network address translation between the private address and the public address.

An Internet Protocol (IP) address allocation method and a device to resolve a technical problem that an IP address conflict occurs when a control plane device and a user plane device allocate an IP address for one user equipment (UE) at the same time. When receiving a session establishment request message, a control plane network device may determine whether a first user plane network device is capable of allocating an IP address for a terminal device. If the first user plane network device is capable of allocating the IP address for the terminal device, the control plane network device may send an address allocation indication to the first user plane network device.

A network controller configured to provide network access to client devices, receives a network access request from a client device. The network access request includes a media access control (MAC) address of the client device and information about a first private key. The network controller sends to a server an authentication request, which includes the MAC address of the client device. The network controller receives an authentication response from the server, which includes a second private key. The network controller determines whether the first private key is the same as the second private key. In response to determining that the first private key is different from the second private key, network access is denied to the client device, and in response to determining that the first private key is the same as the second private key, network access is granted to the client device.

A network controller configured to provide network access to client devices, receives a network access request from a client device. The network access request includes a media access control (MAC) address of the client device and information about a first private key. The network controller sends to a server an authentication request, which includes the MAC address of the client device. The network controller receives an authentication response from the server, which includes a second private key. The network controller determines whether the first private key is the same as the second private key. In response to determining that the first private key is different from the second private key, network access is denied to the client device, and in response to determining that the first private key is the same as the second private key, network access is granted to the client device.

A method is disclosed for providing IP access across packet data network gateways (PGWs), comprising: receiving, from a UE, at a coordinating node, an attach request; sending a request to create a first new session to a first PGW; sending a request to create a second new session to a second PGW; receiving, from the first PGW and at the coordinating node, a first request for policies for the UE; receiving, from the second PGW and at the coordinating node, a second request for policies for the UE; opening a first data tunnel from the coordinating node to the first PGW; opening a second data tunnel from the coordinating node to the second PGW without closing the first data tunnel; and opening a data tunnel between the UE and the coordinating node for providing IP access to both the first PGW and the second PGW.

A communication session for a UE can be restored in the event of serving call session control function (S-CSCF) node unavailability and/or application server (AS) unavailability by storing, prior to IMS unavailability, attribute-value pairs (AVPs) at a home subscriber server (HSS). These AVPs can be used independently by individual IMS nodes to restore a communication session for a UE due to an unavailable IMS node. When a first S-CSCF node becomes unavailable, a proxy CSCF (P-CSCF) node can send a SIP request originating from the UE to a second S-CSCF node. The second S-CSCF node can then send a request to the HSS for an identifier of an AS associated with the UE. Upon receipt of the AS identifier (e.g., an active AS name AVP) from the HSS, the second S-CSCF node can send the SIP request to the AS in order to restore the communication session for the UE.

A mobile network operator (MNO) may control WiFi QoS. 3GPP has specified control mechanisms for various levels of quality of service (QoS) over the cellular access and core network. Embodiments described herein provide differentiation of WiFi QoS based on MNO requirements. In particular, extensible authentication protocol (EAP) and diameter messages may be extended to include a wireless local area network QoS parameter. This may be used by user equipment to set the uplink 802.11e User Priority (UP) for offloaded or evolved packet core-routed WiFi traffic.

An IP address allocation method and a device are provided, so as to resolve a technical problem that an IP address conflict occurs when a control plane device and a user plane device allocate an IP address for one UE at the same time. When receiving a session establishment request message, a control plane network device may determine whether a first user plane network device is capable of allocating an IP address for a terminal device. If the first user plane network device is capable of allocating the IP address for the terminal device, the control plane network device may send an address allocation indication to the first user plane network device.