Network architecture providing device identification and redirection using whitelisting traffic classification
A network edge device includes switching circuitry configured to switch traffic from one or more endpoint devices to corresponding application services over a network; and processing circuitry configured to monitor the traffic from the one or more endpoint devices, compare the monitored traffic to classify the one or more endpoint devices into a corresponding trust level of a plurality of trust levels, and route the traffic from each of the one or more endpoint devices based on its corresponding trust level. The network edge element is configured to provide network connectivity to the one or more endpoint devices.
1. A network edge device configured to connect one or more endpoint devices to a network, the network edge device comprising:
switching circuitry configured to switch traffic including packets from the one or more endpoint devices to corresponding application services over the network; and
processing circuitry configured to monitor the traffic from the one or more endpoint devices, analyze the monitored traffic with a model to classify the one or more endpoint devices into a corresponding trust level of a plurality of trust levels, and route the traffic from each of the one or more endpoint devices based on its corresponding trust level including an untrusted level where the traffic is steered, by altering a path of the traffic, outside of the network edge device, away from its intended destination, to restricted zones for evaluation thereof,
wherein a data set of training traffic is obtained from network data that has suspicious/attacking traffic removed therefrom with the suspicious/attacking traffic determined based on known suspicious behavioral patterns, and the data set with the suspicious/attacking traffic removed therefrom is examined and labeled to retrain the model for detection of unknown suspicious behavioral patterns.
2. The network edge device of claim 1 , wherein the traffic is monitored through collection of one or more of network measurements and non-network measurements of each of the one or more endpoint devices.
3. The network edge device of claim 2 , wherein
the network measurements include any of timings and sizes of data packets, timings and headers of data packets, and timings and content of control packets, and
the non-network measurements include CPU, memory, and file system utilizations; host identifiers; operating system logs; classification is done with one of labeled supervised, unlabeled supervised, and unsupervised machine learning.
4. The network edge device of claim 1 , wherein the network edge device is configured to provide network connectivity to the one or more endpoint devices.
5. The network edge device of claim 1 , wherein, for initial connectivity of an endpoint device, the endpoint device is classified in a suspicious trust level and moved based on continuous monitoring.
6. The network edge device of claim 1 , wherein the one or more endpoint devices are classified by comparing behavior relative to a group of similar types of devices.
7. The network edge device of claim 1 , wherein the processing circuitry is configured to continually monitor the traffic from the one or more endpoint devices, update the corresponding trust level based thereon, and reroute the traffic based on an updated trust level.
8. A non-transitory computer-readable medium comprising instructions that, when executed, cause a processor to perform the steps of:
obtaining a model that was trained, with a data set of training traffic, for detection of unknown suspicious behavioral patterns, wherein the data set of training traffic is obtained from network data that has suspicious/attacking traffic removed therefrom with the suspicious/attacking traffic determined based on known suspicious behavioral patterns, and wherein the data set of training traffic with the suspicious/attacking traffic removed therefrom is examined and labeled for retraining the model;
monitoring traffic, including packets, received by a network edge device from one or more endpoint devices destined for corresponding application services over a network, wherein the network edge device is configured to connect the one or more endpoint devices to the network;
classifying, by analyzing the monitored traffic with the model, the one or more endpoint devices into a corresponding trust level of a plurality of trust levels; and
causing routing of the traffic from each of the one or more endpoint devices based on its corresponding trust level including an untrusted level where the traffic is steered, by altering a path of the traffic, outside of the network edge device, away from its intended destination, to restricted zones for evaluation thereof.
9. The non-transitory computer-readable medium of claim 8 , wherein the monitoring traffic includes
collecting one or more of network measurements and non-network measurements of each of the one or more endpoint devices.
10. The non-transitory computer-readable medium of claim 9 , wherein
the network measurements include any of timings and sizes of data packets, timings and headers of data packets, and timings and content of control packets, and
the non-network measurements include CPU, memory, and file system utilizations; host identifiers; operating system logs; classification is done with one of labeled supervised, unlabeled supervised, unsupervised machine learning.
11. The non-transitory computer-readable medium of claim 8 , wherein the monitoring, the classifying, and the causing are performed by a network edge element that is configured to provide network connectivity to the one or more endpoint devices.
12. The non-transitory computer-readable medium of claim 8 , wherein, for initial connectivity of an endpoint device, the endpoint device is classified in a suspicious trust level and moved based on continuous monitoring.
13. The non-transitory computer-readable medium of claim 8 , wherein the one or more endpoint devices are classified by comparing behavior relative to a group of similar types of devices.
14. The non-transitory computer-readable medium of claim 8 , further comprising
continually monitoring the traffic from the one or more endpoint devices, updating the corresponding trust level based thereon, and rerouting the traffic based on an updated trust level.
15. A method comprising:
obtaining a model that was trained, with a data set of training traffic, for detection of unknown suspicious behavioral patterns, wherein the data set of training traffic is obtained from network data that has suspicious/attacking traffic removed therefrom with the suspicious/attacking traffic determined based on known suspicious behavioral patterns;
monitoring traffic including packets, by a network edge device configured to connect one or more endpoint devices to a network, from one or more endpoint devices, destined for corresponding application services over the network;
classifying, by analyzing the monitored traffic with the model, the one or more endpoint devices into a corresponding trust level of a plurality of trust levels; and
causing routing of the traffic from each of the one or more endpoint devices based on its corresponding trust level including an untrusted level where the traffic is steered, by altering a path of the traffic, outside of the network edge device, away from its intended destination, to restricted zones for evaluation thereof.
16. The method of claim 15 , wherein the monitoring traffic includes
collecting one or more of network measurements and non-network measurements of each of the one or more endpoint devices.