MULTICAST LEAVE METHOD
    3.
    发明公开

    公开(公告)号:EP4440077A1

    公开(公告)日:2024-10-02

    申请号:EP23165957.4

    申请日:2023-03-31

    摘要: A method of managing content delivery to a client device by a proxy, where the content is made up of a sequence of content segments and each segment is encoded at a plurality of bit rates or quality levels. The proxy starts off by receiving content requests from the client device over unicast, forwarding them to a content server, and in response, receiving the content over unicast from the content server before forwarding it onto the client device. At some stage, the proxy joins a multicast channel and receives segments over multicast, and a timer is started. That timer is restarted every time one of the requests from the client device matches that of a segment received over multicast. The proxy leaves the multicast channel once the timer expires, and returns to satisfying segment requests from the client device using unicast. The timer can be adjusted according to the determined bandwidth between the unicast content source and the proxy.

    NETWORK ANOMALY DETECTION
    9.
    发明公开

    公开(公告)号:EP4422127A1

    公开(公告)日:2024-08-28

    申请号:EP23157824.6

    申请日:2023-02-21

    IPC分类号: H04L9/40

    CPC分类号: H04L63/1425 H04L63/20

    摘要: A computer-implemented method of training a network anomaly detection system is disclosed. The method involves generating (301) synthetic benign network data and synthetic anomalous network data and combining (302) the synthetic benign network data and synthetic anomalous network data to generate combined synthetic network data having a predetermined density of anomalous network data. The combined synthetic network data is provided (303) to a trained anomaly detection model, and an accuracy score is determined (304) that is representative of how accurately the trained anomaly detection model recognises anomalous activity in the combined synthetic network data. If the accuracy score is less than a threshold value, the anomaly detection model is trained (305) with additional network data and a new accuracy score is determined. Otherwise, the predetermined density of anomalous network data is reduced (306) and a new accuracy score is determined until a predetermined stopping criterion is met.

    CALLING LINE IDENTITY SPOOF DETECTION IN THE PRESENCE OF REPEAT CALLS

    公开(公告)号:EP4418643A1

    公开(公告)日:2024-08-21

    申请号:EP23156781.9

    申请日:2023-02-15

    IPC分类号: H04M3/22 H04M3/436 H04L9/40

    摘要: A computer implemented method of processing calls in a telephone network, the method comprising: identifying a plurality of calls that have been made by a particular telephone number to destination telephone numbers in the telephone network; dividing the plurality of calls into one to N subsets of calls corresponding to a first to Nth occurrence of the calls to the destination telephone numbers, wherein the first subset of calls includes a sequence of the calls to the destination telephone numbers occurring for a first time; determining whether the sequence of calls in at least one of the subsets of calls corresponds to a correct order for calling those destination telephone numbers, wherein the correct order is determined based on a predetermined rule for ordering destination telephone numbers in the telephone network and wherein a party that is authorised to use the particular telephone numbers follows the correct order; and determining that one or more of the calls were made by spoofing the particular telephone number if the sequence in which the calls in the at least one subset were placed does not correspond to a correct order.