Security for network environment using trust scoring based on power consumption of devices within network
Systems, methods, and articles of manufacture comprising processor-readable storage media are provided for implementing security mechanisms for network environments. For example, a method includes collecting power consumption data of a plurality of devices operating within a network and determining trust scores for the plurality of devices based, at least in part, on the collected power consumption data. The trust score for a device provides a measure of trustworthiness of the device exhibiting normal operating behavior within the network. Each device is assigned to one of a plurality of trust tiers based on the determined trust scores, wherein each trust tier specifies an authentication level for devices assigned to the trust tier. One or more authentication procedures are applied to authenticate a given device operating within the network based on the authentication level specified by the trust tier to which the given device is assigned.
1. A method, comprising:
collecting power consumption data of a plurality of devices operating within a network;
determining trust scores for the plurality of devices based, at least in part, on the collected power consumption data, wherein the trust score for a device provides a measure of trustworthiness of the device exhibiting normal operating behavior within the network;
assigning each of the plurality of devices into one of a plurality of trust tiers based on the determined trust scores, wherein each trust tier specifies an authentication level for devices assigned to the trust tier; and
applying one or more authentication procedures to authenticate a given device of the plurality of devices operating within the network based on the authentication level specified by the trust tier to which the given device is assigned.
2. The method of claim 1 , wherein each trust tier is allocated to a predefined range of trust scores, and wherein each device of the plurality of devices is assigned to a trust tier by determining which trust tier of the plurality of trust tiers has a predefined range of trust scores which includes the determined trust score of the device.
3. The method of claim 1 , further comprising:
assigning a default trust score to a newly provisioned device operating within the network; and
assigning the newly provisioned device to one of the plurality of trust tiers based on the default trust score.
4. The method of claim 1 , wherein determining trust scores for the plurality of devices further comprises:
determining device interaction patterns between the devices operating within the network; and
determining a trust score for each of the devices based, at least in part, on the determined device interaction patterns.
5. The method of claim 4 , wherein the trust score is determined using a trust scoring machine learning model which is trained to determine trust scores using a training dataset comprising features corresponding to device interaction patterns.
6. The method of claim 5 , wherein the features corresponding to device interaction patterns are based on device pairs and at least one feature associated with sets of neighbor nodes associated with the device pairs.
7. The method of claim 6 , wherein the at least one feature associated with sets of neighbor nodes comprises a number of neighbor nodes of each device of a given device pair.
8. The method of claim 6 , wherein the at least one feature associated with sets of neighbor nodes comprises a sum and difference between the number of neighbor nodes of each device of a given device pair.
9. The method of claim 6 , wherein the at least one feature associated with sets of neighbor nodes comprises a number of common neighbor nodes between the devices of a given device pair.
10. The method of claim 6 , wherein the at least one feature associated with sets of neighbor nodes comprises a value indicative of an intersection of the neighbor nodes of the devices of a given device pair divided by a union of the neighbor nodes of the devices of the given device pair.
11. The method of claim 1 , wherein the devices comprise sensor devices operating in a wireless sensor network of an Internet of Things network environment.
12. An article of manufacture comprising a non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code is executable by one or more processors to implement a method comprising:
collecting power consumption data of a plurality of devices operating within a network;
determining trust scores for the plurality of devices based, at least in part, on the collected power consumption data, wherein the trust score for a device provides a measure of trustworthiness of the device exhibiting normal operating behavior within the network;
assigning each of the plurality of devices into one of a plurality of trust tiers based on the determined trust scores, wherein each trust tier specifies an authentication level for devices assigned to the trust tier; and
applying one or more authentication procedures to authenticate a given device of the plurality of devices operating within the network based on the authentication level specified by the trust tier to which the given device is assigned.
13. The article of manufacture of claim 12 , wherein each trust tier is allocated to a predefined range of trust scores, and wherein each device of the plurality of devices is assigned to a trust tier by determining which trust tier of the plurality of trust tiers has a predefined range of trust scores which includes the determined trust score of the device.
14. The article of manufacture of claim 12 , further comprising program code which is executable by the one or more processors to implement a method comprising:
assigning a default trust score to a newly provisioned device operating within the network; and
assigning the newly provisioned device to one of the plurality of trust tiers based on the default trust score.
15. The article of manufacture of claim 12 , wherein determining trust scores for the plurality of devices further comprises:
determining device interaction patterns between the devices operating within the network; and
determining a trust score for each of the devices based, at least in part, on the determined device interaction patterns.
16. The article of manufacture of claim 15 , wherein the trust score is determined using a trust scoring machine learning model which is trained to determine trust scores using a training dataset comprising features corresponding to device interaction patterns, wherein the features corresponding to device interaction patterns are based on device pairs and features associated with sets of neighbor nodes associated with the device pairs.
17. The article of manufacture of claim 16 , wherein the features associated with sets of neighbor nodes comprise:
a first feature comprising a number of neighbor nodes of each device of a given device pair;
a second feature comprising a sum and difference between the number of neighbor nodes of each device of the given device pair;
a third feature comprising a number of common neighbor nodes between the devices of the given device pair; and
a fourth feature comprising a value indicative of an intersection of the neighbor nodes of the devices of the given device pair divided by a union of the neighbor nodes of the devices of the given device pair.
18. A server node, comprising:
at least one processor; and
system memory configured to store program code, wherein the program code is executable by the at least one processor to perform a process which comprises:
collecting power consumption data of a plurality of devices operating within a network;
determining trust scores for the plurality of devices based, at least in part on, the collected power consumption data, wherein the trust score for a device provides a measure of trustworthiness of the device exhibiting normal operating behavior within the network;
assigning each of the plurality of devices into one of a plurality of trust tiers based on the determined trust scores, wherein each trust tier specifies an authentication level for devices assigned to the trust tier; and
applying one or more authentication procedures to authenticate a given device of the plurality of devices operating within the network based on the authentication level specified by the trust tier to which the given device is assigned.
19. The server node of claim 18 , wherein each trust tier is allocated to a predefined range of trust scores, and wherein each device of the plurality of devices is assigned to a trust tier by determining which trust tier of the plurality of trust tiers has a predefined range of trust scores which includes the determined trust score of the device.
20. The server node of claim 18 , wherein determining trust scores for the devices comprises:
determining device interaction patterns between the devices operating within the network; and
determining a trust score for each of the devices based, at least in part, on the determined device interaction patterns;
wherein the trust score is determined using a trust scoring machine learning model which is trained to determine trust scores using a training dataset comprising features corresponding to device interaction patterns, wherein the features comprise:
a first feature comprising a number of neighbor nodes of each device of a given device pair;
a second feature comprising a sum and difference between the number of neighbor nodes of each device of the given device pair;
a third feature comprising a number of common neighbor nodes between the devices of the given device pair; and
a fourth feature comprising a value indicative of an intersection of the neighbor nodes of the devices of the given device pair divided by a union of the neighbor nodes of the devices of the given device pair.