Inferring radio type from clustering algorithms
Described embodiments provide systems and methods for inferring a network type and network conditions. The system includes a packet capturing engine configured to capture a plurality of network packets from a plurality of TCP network connections. The system includes a packet analyzer configured to analyze the plurality of network packets to generate a plurality of metrics. The system includes a network classifier configured to infer network types of the plurality of TCP connections based on the plurality of metrics and at least one classification model. The system also includes a conditions ranking engine configured to estimate a level of network congestion for each TCP connection based on the plurality of metrics and the network types.
1. A method for identifying a type of network of a transport layer connection, the method comprising:
(a) establishing first metrics of network traffic traversing one or more devices via a plurality of transport layer connections providing communications with a plurality of types of networks, the first metrics comprising inter-arrival intervals of packets of the network traffic and at least one of a percentage of packets of the network traffic within a predetermined inter-arrival interval, a probability of a packet having the inter-arrival interval within the predetermined inter-arrival interval, or average minimum round trip times of the packets;
(b) classifying, by a network classifier, the plurality of types of networks into a classification model using at least the inter-arrival intervals of the packets of the network traffic and at least one of the percentage of the packets of the network traffic within the predetermined inter-arrival interval, the probability of a packet having the inter-arrival interval within the predetermined inter-arrival interval, or the average minimum round trip times of the packets; and
(c) identifying, by the network classifier responsive to a device receiving one or more packets for the transport layer connection, the type of network of the plurality of types of networks for the transport layer connection based at least on second metrics of the one or more packets and the classification model.
2. The method of claim 1 , wherein (a) further comprises capturing packets from the network traffic of the plurality of transport layer connections traversing the one or more devices.
3. The method of claim 1 , wherein the plurality of types of networks comprises one of a mobile network or a fixed network.
4. The method of claim 3 , wherein the mobile network comprises one of a 2G, 3G, 4G or 5G network.
5. The method of claim 1 , wherein the first metrics comprise one or more of the following: average throughput, average instantaneous throughput, average inter-arrival intervals, average inter-sending interval, maximum round trip time, minimum round trip time, average round trip time, average load delay and average noise delay.
6. The method of claim 1 , wherein the transport layer connection is established with a mobile device via the type of network comprising a mobile network.
7. The method of claim 1 , wherein (c) further comprises inferring, by the network classifier, the type of network for the transport layer connection based at least on comparing the metrics of the one or more packets to the classification model established by the network classifier for the plurality of types of networks.
8. The method of claim 1 , further comprising distinguishing, by the network classifier, between different types of networks based on the inter-arrival intervals of the metrics.
9. A system for identifying a type of network of a transport layer connection, the system comprising:
a network classifier executable on one or more processors, coupled to memory, and configured to receive first metrics of network traffic traversing the one or more devices via a plurality of transport layer connections providing communications with a plurality of types of networks, the first metrics comprising inter-arrival intervals of packets of the network traffic and at least one of a percentage of packets of the network traffic within a predetermined inter-arrival interval, a probability of a packet having the inter-arrival interval within the predetermined inter-arrival interval, or average minimum round trip times of the packets;
wherein the network classifier configured to classify the plurality of types of networks into a classification model using at least the inter-arrival intervals of the packets of the network traffic and at least one of the percentage of the packets of the network traffic within the predetermined inter-arrival interval, the probability of a packet having the inter-arrival interval within the predetermined inter-arrival interval, or the average minimum round trip times of the packets; and
wherein the network classifier is configured to identify, responsive to a device receiving one or more packets for the transport layer connection, the type of network of the plurality of types of networks for the transport layer connection based at least on second metrics of the one or more packets and the classification model.
10. The system of claim 9 , wherein packets are captured from the network traffic of the plurality of transport layer connections traversing the one or more devices.
11. The system of claim 9 , wherein the plurality of types of networks comprises one of a mobile network or a fixed network.
12. The system of claim 11 , wherein the mobile network comprises one of a 2G, 3G, 4G or 5G network.
13. The system of claim 9 , wherein the first metrics comprise one or more of the following: average throughput, average instantaneous throughput, average inter-arrival intervals, average inter-sending interval, maximum round trip time, minimum round trip time, average round trip time, average load delay and average noise delay.
14. The system of claim 9 , wherein the transport layer connection is established with a mobile device via the type of network comprising a mobile network.
15. The system of claim 9 , wherein the network classifier is further configured to infer the type of network for the transport layer connection based at least on comparing the metrics of the one or more packets to the classification model established by the network classifier for the plurality of types of networks.
16. The system of claim 9 , wherein the network classifier is further configured to distinguish between different types of networks based on the inter-arrival intervals of the metrics.