# US 2025/0097155 A1 · Embedding an artificially intelligent neuron capable of packet inspection and system optimization in ipv6 enabled wlan networks > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2025/0097155 A1 - Document ID: US20250097155A1 - Status: Published application - Kind: A1 - Application number: 18/898,233 - Filed: 2024-09-26 - Published: 2025-03-20 - Earliest claimed: 2021-12-31 - Assignees: Fortinet, Inc. - Applicants: Fortinet, Inc. - Inventors: Srinivasa Subbarao Neeli - Web page: https://patentgrep.com/patent/US20250097155A1 ## CPC - H04L 47/127: by using congestion prediction - G06N 20/00: Machine learning - H04L 45/74: Address processing for routing - H04W 84/12: WLAN [Wireless Local Area Networks] - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72) ## Abstract Responsive to matching a site prefix to IPV6 network traffic from clients, the traffic as intended, and responsive to not matching the site prefix, classifying the corresponding traffic as unintended. An initial rate of packet occurrence and predict load caused by intended traffic and predicting load caused by unintended traffic is calculated, based on an initial rate of packet occurrence. The predicted traffic loads are fed back by configuring behavior of network modules according to the predictions of intended traffic load and unintended traffic load. Packet processing traffic at the network modules is based on traffic classification from the outcome of the AI-neuron. ## Claims ### Claim 1 An access point coupled to a plurality of Wi-Fi clients on a private network and coupled to a data communication network, to intelligently filter unwanted IPV6 traffic from Wi-Fi clients, the access point comprising: a processor; a network communication interface, communicatively coupled to the processor and to the data communication network; and a memory, communicatively coupled to the processor and storing source code for: a client learning module to learn IPV6 addresses for the plurality of Wi-Fi clients on a private network; a traffic classification module to attempt to match site prefix of IPV6 addresses to an assigned organizational specific address, responsive to matching the site prefix, classifying corresponding traffic as intended, and responsive to not matching the site prefix, classifying corresponding traffic as unintended; and a load prediction module to calculate initial rate of packet occurrence and predict load caused by intended traffic and predicting load caused by unintended traffic, based on an initial rate of packet occurrence, wherein the load prediction module feeds back the predicted traffic loads by configuring behavior of network modules according to the predictions of intended traffic load and unintended traffic load, wherein the processor processes corresponding traffic at the network modules based on traffic classification from the outcome of the AI-neuron. ### Claim 2 (depends on 1) The access point of claim 1, wherein responsive to the site prefix, determining an outcome for the corresponding traffic. ### Claim 3 (depends on 1) The access point of claim 1, wherein intended traffic is forwarded and unintended traffic is dropped. ### Claim 4 (depends on 1) The access point of claim 1, further comprising deauthenticating a Wi-Fi client sourcing unintended traffic. ### Claim 5 (depends on 1) The access point of claim 1, wherein site prefixes of IPV6 packets are matched at bits 33 to 48. ### Claim 6 (depends on 1) The access point of claim 1, wherein the predicted load is determined using multi linear regression for supervised learning. ### Claim 7 (depends on 1) The access point of claim 1, wherein the predicted load concerns at least one of the processor, a network load, the memory, and network processes. ### Claim 8 (depends on 1) The access point of claim 1, wherein the traffic classification module attempt to match a regional site prefix of IPV6 addresses to an assigned region of the access point. ### Claim 9 A method in access point coupled to a plurality of Wi-Fi clients on a private network and coupled to a data communication network, to intelligently filter unwanted IPV6 traffic from Wi-Fi clients, the method comprising the steps of: learning IPV6 addresses for the plurality of Wi-Fi clients on a private network; attempting to match site prefix of IPV6 addresses to an assigned organizational specific address, responsive to matching the site prefix, classifying corresponding traffic as intended, and responsive to not matching the site prefix, classifying corresponding traffic as unintended; calculating initial rate of packet occurrence and predict load caused by intended traffic and predicting load caused by unintended traffic, based on an initial rate of packet occurrence, feeding back the predicted traffic loads by configuring behavior of network modules according to the predictions of intended traffic load and unintended traffic load; and processing corresponding traffic at the network modules based on traffic classification from the outcome of the AI-neuron. ### Claim 10 (depends on 1) The method of claim 1, wherein responsive to the site prefix, determining an outcome for the corresponding traffic. ### Claim 11 (depends on 1) The method of claim 1, wherein intended traffic is forwarded and unintended traffic is dropped. ### Claim 12 (depends on 1) The method of claim 1, further comprising deauthenticating a Wi-Fi client sourcing unintended traffic. ### Claim 13 (depends on 1) The method of claim 1, wherein site prefixes of IPV6 packets are matched at bits 33 to 48. ### Claim 14 (depends on 1) The method of claim 1, wherein the predicted load is determined using multi linear regression for supervised learning. ### Claim 15 (depends on 1) The method of claim 1, wherein the predicted load concerns at least one of the processor, a network load, the memory, and network processes. ### Claim 16 (depends on 1) The method of claim 1, wherein the traffic classification module attempt to match a regional site prefix of IPV6 addresses to an assigned region of the access point. ### Claim 17 A non-transitory computer-readable media in access point coupled to a plurality of Wi-Fi clients on a private network and coupled to a data communication network, implemented at least partially in hardware for, when executed by a processor, to intelligently filter unwanted IPV6 traffic from Wi-Fi clients, the method comprising the steps of: learning IPV6 addresses for the plurality of Wi-Fi clients on a private network; attempting to match site prefix of IPV6 addresses to an assigned organizational specific address, responsive to matching the site prefix, classifying corresponding traffic as intended, and responsive to not matching the site prefix, classifying corresponding traffic as unintended; and calculating initial rate of packet occurrence and predict load caused by intended traffic and predicting load caused by unintended traffic, based on an initial rate of packet occurrence, feeding back the predicted traffic loads by configuring behavior of network modules according to the predictions of intended traffic load and unintended traffic load; processing corresponding traffic at the network modules based on traffic classification from the outcome of the AI-neuron. ### Claim 18 (depends on 1) The method of claim 1, wherein responsive to the site prefix, determining an outcome for the corresponding traffic. ### Claim 19 (depends on 1) The method of claim 1, wherein intended traffic is forwarded and unintended traffic is dropped. ### Claim 20 (depends on 1) The method of claim 1, further comprising deauthenticating a Wi-Fi client sourcing unintended traffic.