Artificial intelligence-based device recognition
By using AI-based data flow collection and machine learning models in IoT facilities, IoT device information can be automatically identified and acquired, solving the problems of insufficient device documentation and protocol dependence in existing technologies, and achieving non-intrusive and flexible information acquisition from devices.
Patent Information
- Application Number
- CN202080101337.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-05-28
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2040-05-28
AI Technical Summary
In IoT infrastructure, there is a lack of overall setup documentation for IoT devices connected to the network, and existing network management tools require end devices to support specific protocols, resulting in limited device visibility.
By employing an artificial intelligence-based approach, through data flow collection points, data processing modules, and machine learning models, device identification information is derived using data flow fingerprints, enabling automatic identification and information acquisition of IoT devices and avoiding dependence on specific device protocols.
It enables non-intrusive identification of IoT devices, eliminating the need for adjustments on the device side and not relying on specific device protocol support, thus improving the flexibility and accuracy of device information acquisition.
Smart Images

Figure CN115668878B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates generally to the field of network management in an Internet of Things (IoT) environment, and more particularly to techniques for obtaining information related to IoT devices connected to the network. Background Art
[0002] An IoT facility may include a plethora of IoT devices connected to a network. IoT devices may include, for example, one or more of sensors, actuators, controllers, network components, etc. In many IoT facilities, there is not enough documentation of the overall setup (e.g., installed devices, properties of installed devices, etc.) available or the available documentation is outdated. Device information may be collected via the use of network management tools. Known network management tools typically require end devices to support specific protocols, such as SNMP, ICMP, LLDP, etc., to allow visibility of these devices. Summary of the invention
[0003] Briefly, aspects of the present disclosure relate to artificial intelligence-based techniques for obtaining information related to IoT devices connected to a network.
[0004] In a first aspect, a system for obtaining information about an IoT device connected to a network is disclosed. The system includes a quantity flow collection point configured to collect data units transmitted to or from a specified IoT device among a plurality of IoT devices connected to the network. The system also includes a data processing module configured to process quantitative information about data units transmitted to or from the specified IoT device collected over a defined time interval to create a temporal data flow fingerprint of the specified IoT device. The system also includes an artificial intelligence module configured to derive device identification information of the specified IoT device from the temporal data flow fingerprint of the specified IoT device using a machine learning model.
[0005] In a second aspect, a computer-implemented method for obtaining information about an IoT device connected to a network is disclosed. The method includes receiving a collection of data units transmitted to or from a specified IoT device among a plurality of IoT devices connected to the network. The method also includes processing quantified information about the data units transmitted to or from the specified IoT device collected over a defined time interval to create a temporal data traffic fingerprint of the specified IoT device. The method also includes deriving device identification information of the specified IoT device from the temporal data traffic fingerprint of the specified IoT device using a machine learning model.
[0006] Other aspects of the present disclosure implement the features of the above methods in computing systems and computer program products.
[0007] Additional technical features and benefits can be achieved through the technology of the present disclosure. The embodiments and aspects of the present disclosure are described in detail herein and are considered to be part of the subject matter claimed. For a better understanding, reference is made to the detailed description and the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] The foregoing and other aspects of the disclosure are best understood from the following detailed description when read in conjunction with the accompanying drawings. To easily identify discussion of any element or action, the most significant digit or digits in the reference number refer to the figure number in which the element or action is first introduced.
[0009] Figure 1 A system for obtaining device information according to an aspect of the present disclosure is illustrated.
[0010] Figure 2 A known type of Ethernet frame is illustrated.
[0011] Figure 3 An example of a graphical image representing a data traffic fingerprint of a device is illustrated.
[0012] Figure 4 is a flow chart illustrating a computer-implemented method for obtaining device information according to an aspect of the present disclosure.
[0013] Figure 5 An exemplary computing environment is illustrated in which aspects of the present disclosure may be implemented. DETAILED DESCRIPTION
[0014] Various techniques related to systems and methods will now be described with reference to the accompanying drawings, in which the same reference numerals represent the same elements throughout. The accompanying drawings discussed below and the various embodiments used to describe the principles of the present disclosure in this patent document are intended to be illustrative only and should not be interpreted in any way as limiting the scope of the present disclosure. Those skilled in the art will appreciate that the principles of the present disclosure can be implemented in any suitably arranged device. It should be understood that the functions described as being performed by certain system elements can be performed by multiple elements. Similarly, for example, an element can be configured to perform the functions described as being performed by multiple elements. Many innovative teachings of the present disclosure will be described with reference to exemplary non-limiting embodiments.
[0015] Various aspects of the present disclosure are directed to artificial intelligence (AI) based techniques for obtaining information about IoT devices connected to a network. The basic idea here is to use information about the data traffic sent to or from a specified IoT device to derive device identification information, such as device type, and in some cases, device type variants. The device identification information is derived from the data traffic fingerprint of the specified IoT device using a machine learning model. The model can be trained on prior data to learn data traffic fingerprints that characterize the behavior of a large number of device types / device type variants.
[0016] The inventors have recognized that the basic behavior of IoT devices involving data communication clearly characterizes the device type or in some cases even the device type variant. For example, the basic behavior regarding data communication can be analyzed based on the protocol used, or the data priority, or the communication intensity, or the characteristics of the data exchanged (such as packet length, selected content, etc.), or any combination thereof. The concept of the present invention makes it possible to obtain information about IoT devices with connectivity to a network without relying on the capabilities of these devices (such as supporting specific protocols) or requiring adaptation on the device side.
[0017] Now referring to the accompanying drawings, Figure 1 A system 100 is shown according to an illustrative example embodiment. The system 100 is configured to obtain device information about one or more IoT devices connected to a network 110. The network 110 may include, for example, but not limited to, a local area network (LAN), a wireless local area network (WLAN), a wide area network (WAN), etc. The IoT devices 102a, 102b, 102c, 102d may transmit data on the network 110 via one or more network components such as switches, routers, etc. Figure 1 In the illustrative example of , four IoT devices 102a, 102b, 102c, 102d are identified, which are linked to the router switch 104 at corresponding ports 108a, 108b, 108c, 108d via corresponding links 106a, 106b, 106c, 106d. Links 106a-d may include, for example, Ethernet cables. In various embodiments, one or more IoT devices 102a-d may be wirelessly connected to the router switch 104, for example, via a wireless access point (not shown). In some embodiments, the router switch 104 may also be configured to function as a wireless access point. In one embodiment, the router switch 104 may have a modular architecture configured to support connections through several other types of interfaces (such as serial interfaces, T1 interfaces, etc.).
[0018] The router switch 104 (also known as a multilayer switch) is a single device configured to perform switching and routing functions. The router switch 104 is used as a switch to connect devices on the same subnet or virtual LAN, and has built-in IP routing intelligence to double as a router. The router switch 104 has processing capabilities and includes, for example, a CPU 112 and a memory 114. The router switch 104 can therefore support routing protocols, inspect input data packets, and even make routing decisions based on source and destination addresses. In other embodiments, the switch and router can be used as separate devices. In other embodiments, routing and switching functions can be implemented by other network components, including but not limited to network security devices such as firewalls. Those skilled in the art will understand that although a single router switch 104 is shown, in reality, a typical device will include multiple routing and / or switching devices connected to a plethora of IoT devices.
[0019] Network 110 also includes a node that functions as a gateway 116. Gateway 116 may include a router, firewall, server, or any other device that enables data traffic to flow between the devices of network 110 and one or more external networks 118, including but not limited to the Internet.
[0020] As used herein, the term "Internet of Things device" or "IoT device" refers to a device that is capable of sending and / or receiving data over a network without requiring human-to-human or human-to-device interaction. Non-limiting examples of IoT devices include various types of sensors, actuators, controllers (e.g., programmable logic controllers), intelligent electronic devices, servers, and network components such as routers, switches, firewalls, etc. Therefore, it should be understood that although for illustrative purposes, Figure 1 1 and 2. IoT devices 102a-d and network components such as router switch 104 and gateway 116 are identified separately in FIG. 1, but each of these devices is understood to be an IoT device in the context of the present specification and claims.
[0021] According to an embodiment of the present disclosure, the network 110 is connected to a network manager 120, which may be configured to obtain device information of connected IoT devices, among other functions. The network manager 120 may include a computing system including one or more computing devices, such as Figure 5 In a non-limiting example embodiment, the network manager 120 may be implemented in a cloud computing environment. In certain other embodiments, aspects of obtaining device information of connected IoT devices may be implemented in a network node having the required computing capabilities, including, for example, a router switch 104.
[0022] The system for obtaining device information according to the present disclosure includes multiple elements / modules as described below. These elements / modules can be distributed across multiple computing devices, or can be embedded in a single device such as a router switch 104.
[0023] Broadly speaking, the system 100 includes one or more quantity flow collection points 122, a data processing module 124, and an AI module 126. The quantity flow collection point 122 collects data units transmitted to or from a specified IoT device among a plurality of IoT devices connected to the network 110. The term "or" herein refers to an inclusive meaning and is understood to mean "and / or". That is, "data units transmitted to or from a specified IoT device" should be understood to mean data units received by the specified IoT device and / or data units sent by the specified IoT device. For the purpose of illustration, the specified IoT device is identified herein as device 102a, and it should be understood that the described techniques can be applied to other devices connected to the network 110 accordingly. The data processing module 124 processes quantitative information about data units transmitted to or from the specified IoT device 102a collected over a defined time interval to create a temporal data flow fingerprint of the specified IoT device 102a. The temporal data flow fingerprint of a device is a temporary snapshot of the behavior of the device in a defined time interval, which is characterized by quantitative attributes of data units sent and / or received by the device in the defined time interval. The AI module 126 utilizes a machine learning model to derive device identification information of the specified IoT device 102a from the temporal data traffic fingerprint of the specified IoT device 102a. The term "derive" as used herein means to logically derive or derive (e.g., based on defined rules) from input data including the temporal data traffic fingerprint by a trained machine learning model.
[0024] The term "data unit" or protocol data unit (PDU) refers to a single unit of information transmitted between entities of a network in the form of an electrical signal. A data unit consists of protocol-specific control information and a payload. For example: at the data link layer, a data unit refers to a frame; at the network layer, a data unit refers to a packet; at the transport layer, a data unit refers to a segment; and so on. Therefore, the term "data unit" used in this specification and claims is understood in the broadest sense to refer to any PDU.
[0025] The quantity flow collection point 122 is a point for collecting data flow to or from a specified IoT device and / or a counter and / or other quantitative means related to the flow. The quantity flow collection point 122 for the specified IoT device can be established at one or more of several possible points. In a first embodiment, the quantity flow collection point 122 for the specified IoT device 102a can be the device 102a itself. In a second embodiment, the quantity flow collection point 122 for the specified IoT device 102a can be a node of the network 110, through which the data unit transmitted to or from the device 102a is passed. Such nodes may include, for example, network components such as routers, switches, gateways, etc. In the illustrated embodiment, the quantity flow collection point 122 for the specified IoT device 102a can be established at a router switch 104 or a gateway 116. In a third embodiment, the quantity flow collection point 122 for the specified IoT device 102a can simply be a port of a network component linked to the device 102a, such as port 108a of a router switch 104. Quantity flow collection points for other IoT devices connected to the network can be established similarly.
[0026] The data traffic collection point 122 may be configured to filter out data traffic transmitted to or from a given IoT device 102a in a variety of ways.
[0027] In a first example of data traffic filtering, the data traffic collection point 122 can collect data units transmitted to or from a specified IoT device 102a based on the device address of the specified IoT device 102a. The device address may include, for example, a MAC address or an IP address of the device 102a. In various embodiments, where the specified IoT device 102a belongs to a virtual local area network (VLAN), the device address used to filter the data traffic may also include a VLAN identifier that may be mapped to a MAC address or an IP address. In other embodiments, the filtering of data traffic may be based on device addresses / identifiers encapsulated by various technologies, such as Ethernet VPN, VXLAN, MAC on MAC, and the like. This data traffic filtering by device address / identifier may be implemented, for example, by a router switch 104 or a gateway 116.
[0028] In a second example of data traffic filtering, the data traffic collection point 122 can collect data units transmitted via a physical port (e.g., port 108a) linked to a specified IoT device 102a. In some embodiments, such as but not limited to a VLAN, a physical port can have one or more virtual (or logical) interfaces configured thereon. In such a case, only data traffic passing through the virtual interface linked to the specified IoT device can be filtered out.
[0029] In a third example of data traffic filtering, the data traffic collection point 122 may collect data units transmitted via a circuit including a combination of a physical port and / or a virtual interface linked to a specified IoT device. Figure 1 In the configuration shown, if the gateway 116 is used as a data traffic collection point for the IoT device 102a, data traffic transmitted via the following circuit can be filtered out, where the circuit includes the interconnection between the following items: port 130 of the gateway 116 (or a virtual interface configured on port 130), port 128 of the router switch 104 (or a virtual interface configured on port 128), and port 108a of the router switch 104 (or a virtual interface configured on port 108a).
[0030] The data processing module 124 is implemented by a device that converts the quantitative properties of the data units sent / received by the specified IoT device 102a at the defined time interval into a time data traffic fingerprint of the specified IoT device 102a. The data processing module 124 can be implemented at the device that collects the data traffic, or at a different processing device, for example, by mirroring the data traffic. Figure 1 In the example shown, the data processing module 124 resides in the network manager 120.
[0031] In an exemplary embodiment, the data processing module 124 may be configured to process the quantitative information to create statistical information about data units sent and / or received by a specified IoT device over a defined time interval. The statistical information may relate to, for example, the communication protocol used (which may include protocols at the data link layer and / or the network layer), data priority categories, and other characteristics such as the length or selected content of the data unit.
[0032] The statistical information may include, for example, the number of data units per time interval for each protocol or for each selected set of protocols. In one embodiment, information about the protocol of Ethernet frames (data units in the data link or Ethernet layer) sent / received by a specified IoT device can be obtained from the EtherType value. Figure 2 As shown, EtherType is a two-octet field in Ethernet frame 200 whose value indicates which protocol is encapsulated in the payload field (e.g., IPv4, IPv6, a routing protocol such as OSPF, ISIS, etc.). Plotting the number of Ethernet frames per time interval for each EtherType value provides a technique for conveniently creating statistics about a specific communication protocol that can characterize the behavior of a given IoT device.
[0033] Alternatively or in addition, the statistical information may include the number of data units per time interval for each data priority class. An example of a data priority class is a Quality of Service (QoS) class according to IEEE 802.1p, such as Figure 2 As shown, it may be encapsulated in a specific Ethernet frame format that includes a VLAN tag between the source MAC address and the EtherType field (according to IEEE 802.1Q). Another example of a data prioritization class includes a Differentiated Services Code Point (DSCP) tag, which is specified in a network layer data packet.
[0034] Although specific examples are described only in conjunction with encapsulation protocols, data prioritization categories, etc., it should be understood that the described methods can be applied to a variety of other encapsulation technologies, such as Ethernet VPN, VXLAN, MAC on MAC, etc.
[0035] Alternatively, or in addition to the above, the statistical information may relate to characteristics of data units sent / received by the specified IoT device, such as the distribution of data unit lengths for each time interval and / or statistics of selected contents of the payload of the data units for each time interval (e.g., a data segment containing a portion of a video stream, an application layer protocol such as HTTP, HTTPS, etc.), which may clearly characterize the specified IoT device.
[0036] The AI module 126 receives statistical information from the data processing module 124 as input. The statistical information, which may be pre-processed before being forwarded to the AI module 126, represents a time data traffic fingerprint of a specified IoT device 102a. The AI module 126 employs a trained machine learning model. The machine learning model may be trained on prior data to learn data traffic fingerprints that characterize the behavior of a large number of device types / device type variants. In one embodiment, based on the training data, the machine learning model may be configured to define rules for how to derive device identification information of the IoT device 102a from the input time data traffic fingerprint of the IoT device 102a. Although the AI module 126 is shown as residing in the network manager 120, it should be understood that the AI module 126 may reside in other devices, including but not limited to the router switch 104. In some embodiments, the AI module 126 may be distributed across multiple computing devices.
[0037] The device identification information derived by the AI module 126 may include information about the device type. For example, the device identification information may determine whether the specified IoT device 102a is a sensor, an actuator, a controller, a switch, a router, a server, etc. Depending on the specificity of the temporal data traffic fingerprint of the IoT device 102a, the device identification information may also include information about device type variants. For example, the device identification information may specify whether the specified IoT device 102a is a temperature sensor or a pressure sensor, a file server or a web server, a device that communicates using a specific IoT protocol (such as PROFINET, GOOSE, etc.), a device release, a release date, etc.
[0038] The described techniques may be combined, for example, with analyzing a source MAC address of a data unit received from the IoT device 102a or a destination MAC address of a data unit sent to the IoT device 102a to identify the manufacturer of the IoT device 102a based on the first 6 digits of the MAC address and a lookup in a corresponding table.
[0039] In a specific embodiment of the above technology, the data processing module 124 can create a graphical image from quantitative information about data units sent / received by a specified IoT device 102a over a defined time interval. A graphical image is a digital image, which is typically but not necessarily limited to a two-dimensional image. In various embodiments, a graphical image representing a temporal data traffic fingerprint of a specified IoT device 102a can be created based on statistical information, which can, for example, be related to one or more of the following: the protocol used, the data priority, the data unit length distribution, the selected content of the data unit, etc.
[0040] Figure 3A simplified example of a two-dimensional image 300 created from statistical information associated with an EtherType value is shown. Here, all possible EtherType values are plotted along the x-axis 302, first for transmitted data frames and then for received data frames. The number of transmitted / received data frames per second for a particular EtherType value is plotted along the y-axis 304. The resulting two-dimensional image of the histogram represents the temporal data traffic fingerprint of the specified IoT device 102a. The image 300 may be pre-processed and forwarded to the AI module 126. The AI module 126 classifies the graphic image 300 representing the temporal data traffic fingerprint into one of many learned data traffic fingerprint types to derive device identification information for the specified IoT device 102a. The graphic image 300 is temporal and is expected to change at different time intervals. The statistical parameters are selected so that the characteristics of the graphic image are still strong enough to clearly associate the behavior of the specified IoT device 102a with the learned device type (similar to AI-based recognition of a cat or dog). To this end, the AI module 126 may employ a classification type machine learning model that is implemented using an artificial neural network and is implemented using a machine learning tool, including, for example, open source software such as TensorFlow or commercial tools that may or may not be enhanced by AI acceleration hardware.
[0041] In other embodiments, as an alternative or in addition to creating statistical information, other techniques may be used to process quantitative information from data units sent / received over defined time intervals to create a time data traffic fingerprint for a specified IoT device. As an example, the data processing module 124 may be configured to create a time series from data units sent and / or received by a specified IoT device 102a over a defined time interval. The time series may be based on, for example, data points related to the communication protocol used, which may include protocols at the data link layer and / or network layer, data priority categories, and other characteristics such as the length or selected content of the data unit. The time series may be used to create a data traffic fingerprint based on the dynamic behavior of the specified IoT device 102a (e.g., a first type of data traffic burst followed by a second type of data traffic burst, etc.). The machine learning model in the AI module 126 may be fed data during the training phase, for example, including a large number of graphs (graphic images) and / or digital data describing snapshots of device behavior over time in terms of sending and / or receiving data traffic.
[0042] The technology described herein can be applied, for example, in network troubleshooting. In one application, the technology described can be integrated into network maintenance and management tools. In another application, the technology can be used as an aid in expanding an existing network.
[0043] The described techniques achieve a number of practical benefits. For example, the described techniques do not require specific knowledge of the operator of the device identification system. Furthermore, the rules for correct identification do not need to be manually designed or thought out. The use of machine learning methods avoids the need to manually predefine complex decision rules. On the device side, the described techniques are non-intrusive and do not require specific adjustments or capabilities, whereby the techniques can also be used for third-party devices. Furthermore, the described techniques are flexible in terms of which communication attributes to look at, how many parameters to use, etc., and can also be combined with non-communication parameters and / or attributes.
[0044] Figure 4 A computer-implemented method 400 for obtaining device information according to an aspect of the present disclosure is illustrated. The method may be performed by a computing device such as a network manager 120, a router switch 104, or may be performed on multiple computing devices. In one embodiment, the method 400 may be performed as a cloud service. Figure 4 It is not intended to indicate that the operation blocks of method 400 will be performed in any particular order, or that all blocks of method 400 will be included in every case. Additionally, method 400 may include any suitable number of additional operations.
[0045] Block 402 of method 400 involves receiving a collection of data units sent and / or received by a designated IoT device among a plurality of IoT devices connected to a network. In one embodiment, block 402 may involve receiving data units sent and / or received by the designated IoT device from a data traffic collection point via mirroring the data traffic. As described above, the data traffic collection point may be, for example, the designated IoT device itself, or a network node through which data units transmitted to or from the designated IoT device are passed, or a port of a network component linked to the designated IoT device.
[0046] Block 404 of method 400 involves processing quantitative information related to data units collected over defined time intervals to generate a time data traffic fingerprint for a specified IoT device. As described above, processing the quantitative information may include, for example, creating statistical information related to the data units collected over defined time intervals. The statistical information may include one or more of the following items: the number of data units per time interval for each protocol or each selected set of protocols, the number of data units per time interval for each data priority category, the distribution of data unit lengths per time interval, and statistics of selected contents of the payload of the data units per time interval. As an alternative or in addition to creating statistical information, other techniques (e.g., creating a time series) may be used to process quantitative information from data units collected over defined time intervals to create a time data traffic fingerprint for a specified IoT device. In one embodiment, block 404 may involve creating a graphical image from the quantitative information that represents the time data traffic fingerprint for a specified IoT device.
[0047] Block 406 of method 400 involves using a machine learning model to derive device identification information of a specified IoT device based on a temporal data traffic fingerprint of the specified IoT device. The machine learning model can be trained based on prior data to define rules on how to derive device identification information from an input temporal data traffic fingerprint of an IoT device. In one embodiment, block 406 may involve using the machine learning model to classify a graphical image representing a temporal data traffic fingerprint into one of a plurality of learned data traffic fingerprint types to derive device identification information of the specified IoT device.
[0048] Figure 5 An exemplary computing environment 500 including a computer system 502 is illustrated in which aspects of the present disclosure may be implemented. The computing system 502 may be configured, for example, Figure 1 The network manager 120 shown. Computers and computing environments such as computer system 502 and computing environment 500 are known to those skilled in the art and are therefore briefly described herein. It should be noted that in the case of implementing embodiments of the present disclosure in a cloud environment, some items (e.g., keyboard 536) included in the exemplary computing environment 500 may not be applicable.
[0049] like Figure 5 As shown, computer system 502 may include a communication mechanism such as a system bus 504 or other communication mechanism for transferring information within computer system 502. Computer system 502 also includes one or more processors 506 coupled to system bus 504 for processing information. Processor 506 may include one or more central processing units (CPUs), graphics processing units (GPUs), or any other processors known in the art.
[0050] The computer system 502 also includes a system memory 508 coupled to the system bus 504 for storing information and instructions to be executed by the processor 506. The system memory 508 may include computer-readable storage media in the form of volatile and / or non-volatile memory, such as read-only memory (ROM) 510 and / or random access memory (RAM) 512. The system memory RAM 512 may include other dynamic storage devices (e.g., dynamic RAM, static RAM, and synchronous DRAM). The system memory ROM 510 may include other static storage devices (e.g., programmable ROM, erasable PROM, and electrically erasable PROM). In addition, the system memory 508 may be used to store temporary variables or other intermediate information during the execution of instructions by the processor 506. A basic input / output system 514 (BIOS) containing basic routines that help to transfer information between elements within the computer system 502, such as during startup, may be stored in the system memory ROM 510. The system memory RAM 512 may include data and / or program modules that are immediately accessible to and / or currently being operated on by the processor 506. System memory 508 may additionally include, for example, operating system 516 , application programs 518 , other program modules 520 , and program data 522 .
[0051] The computer system 502 also includes a disk controller 524 that is coupled to the system bus 504 to control one or more storage devices for storing information and instructions, such as a magnetic hard disk 526 and a removable media drive 528 (e.g., a floppy disk drive, an optical drive, a tape drive, and / or a solid-state drive). Storage devices may be added to the computer system 502 using an appropriate device interface (e.g., small computer system interface (SCSI), integrated device electronics (IDE), universal serial bus (USB), or FireWire).
[0052] The computer system 502 may also include a display controller 530 coupled to the system bus 504 to control a display 532 (such as a cathode ray tube (CRT) or a liquid crystal display (LCD), etc.) for displaying information to a computer user. The computer system 502 includes a user input interface 534 and one or more input devices (such as a keyboard 536 and a pointing device 538) for interacting with a computer user and providing information to the one or more processors 506. The pointing device 538 may be, for example, a mouse, a light pen, a trackball, or a click stick for communicating directional information and command selections to the one or more processors 506 and for controlling cursor movement on the display 532. The display 532 may provide a touch screen interface that allows input to supplement or replace the communication of directional information and command selections by the pointing device 538.
[0053] The computer system 502 may perform some or all of the processing steps of embodiments of the present disclosure in response to one or more processors 506 executing one or more sequences of one or more instructions contained in a memory such as a system memory 508. Such instructions may be read into the system memory 508 from another computer-readable storage medium, such as a magnetic hard disk 526 or a removable media drive 528. The hard disk 526 may contain one or more data memories and data files used by embodiments of the present disclosure. The data storage contents and data files may be encrypted to increase security. The processor 506 may also be used in a multi-processing device to execute one or more instruction sequences contained in the system memory 508. In alternative embodiments, hard-wired circuits may be used in place of or in conjunction with software instructions. Therefore, the embodiments are not limited to any particular combination of hardware circuits and software.
[0054] The computer system 502 may include at least one computer-readable storage medium or memory for storing instructions programmed according to embodiments of the present disclosure and for containing data structures, tables, records, or other data described herein. As used herein, the term "computer-readable storage medium" refers to any medium that participates in providing instructions to one or more processors 506 for execution. Computer-readable storage media can take many forms, including but not limited to non-temporary, non-volatile media, volatile media, and transmission media. Non-limiting examples of non-volatile media include optical disks, solid-state drives, magnetic disks, and magneto-optical disks, such as magnetic hard disks 526 or removable media drives 528. Non-limiting examples of volatile media include dynamic memory, such as system memory 508. Non-limiting examples of transmission media include coaxial cables, copper wires, and optical fibers, including lines that constitute the system bus 504. Transmission media can also take the form of sound waves or light waves, such as those generated during radio wave and infrared data communications.
[0055] The computing environment 500 may also include a computer system 502 operating in a networked environment using logical connections to one or more remote computers, such as a remote computing device 544. The remote computing device 544 may be a personal computer (laptop or desktop), a mobile device, a server, a router, a network PC, a peer device, or other common network node, and typically includes many or all of the elements described above with respect to the computer system 502. When used in a networked environment, the computer system 502 may include a modem 542 for establishing communications over a network 540, such as the Internet. The modem 542 may be connected to the system bus 504 via a network interface 546 or via other appropriate mechanism.
[0056] The network 540 can be any network or system known in the art, including the Internet, an intranet, a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a direct connection or a series of connections, a cellular telephone network, or any other network or medium that can facilitate communication between the computer system 502 and other computers (e.g., remote computing devices 544). The network 540 can be wired, wireless, or a combination thereof. Wired connections can be implemented using Ethernet, a universal serial bus (USB), RJ-6, or any other wired connection known in the art. Wireless connections can be implemented using Wi-Fi, WiMAX, and Bluetooth, infrared, cellular networks, satellites, or any other wireless connection methods known in the art. In addition, several networks can work alone or communicate with each other to facilitate communication in the network 540.
[0057] Embodiments of the present disclosure may be implemented with any combination of hardware and software. In addition, embodiments of the present disclosure may be included in an article (e.g., one or more computer program products) having, for example, a non-transitory computer-readable storage medium. The computer-readable storage medium includes, for example, computer-readable program instructions for providing and facilitating the mechanisms of embodiments of the present disclosure. The article may be included as part of a computer system or sold separately.
[0058] A computer-readable storage medium may include a tangible device that can retain and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a corresponding computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network.
[0059] Although various aspects and embodiments have been disclosed herein, other aspects and embodiments will be apparent to those skilled in the art. The various aspects and embodiments disclosed herein are for purposes of illustration and not limitation, with the true scope and spirit being indicated by the appended claims.
Claims
1. A system for obtaining information about an IoT device connected to a network, the system include: a data traffic collection point configured to collect data units transmitted to or from a designated IoT device among a plurality of IoT devices connected to the network, a data processing module configured to process quantitative information about the data units transmitted to or from the specified IoT device collected over a defined time interval to create a temporal data traffic fingerprint of the specified IoT device, wherein the data processing module is further configured to process the quantitative information to create statistical information about the data units transmitted to or from the specified IoT device over the defined time interval, wherein the statistical information includes the number of data units per time interval for each protocol or each selected set of protocols, wherein the data processing module is further configured to create a graphical image from the quantitative information, the graphical image representing the temporal data traffic fingerprint of the specified IoT device, and An artificial intelligence module is configured to utilize a machine learning model to derive device identification information of the specified IoT device from the time data traffic fingerprint of the specified IoT device, wherein the artificial intelligence module is further configured to classify the graphic image representing the time data traffic fingerprint into one of a plurality of learned data traffic fingerprint types to derive the device identification information of the specified IoT device.
2. The system according to claim 1, in, The data traffic collection point is configured to collect data units transmitted to or from the designated IoT device based on a device address or identifier of the designated IoT device.
3. The system according to claim 1, in, The data traffic collection point is configured to collect data units transmitted via a physical port linked to the designated IoT device, or via a virtual interface linked to the designated IoT device, or via a circuit including a combination of a physical port and / or a virtual interface linked to the designated IoT device.
4. The system according to any one of claims 1 to 3, in, The quantity traffic collection point is selected from the group consisting of: the designated IoT device, a network node through which a data unit transmitted to or from the designated IoT device is passed, and a port of a network component linked to the designated IoT device.
5. The system according to any one of claims 1 to 3, in: The data unit comprises an Ethernet frame, and The statistical information includes the number of Ethernet frames in each time interval for each EtherType value, where the EtherType value indicates an encapsulation protocol.
6. The system according to any one of claims 1 to 3, in, The statistical information includes the number of data units per time interval for each data priority.
7. The system according to any one of claims 1 to 3, in, The statistical information includes the data unit length distribution in each time interval.
8. The system according to any one of claims 1 to 3, in, The statistical information includes statistics regarding selected contents of the payload of the data unit for each time interval.
9. The system according to any one of claims 1 to 3, in, The device identification information includes information about the type of device.
10. The system according to claim 9, in, The device identification information also includes information about device type variants.
11. The system according to any one of claims 1 to 3, in, The system is embedded in a router switch.
12. The system according to any one of claims 1 to 3, in, The system is distributed across multiple computing devices.
13. A computer-implemented method for obtaining information about an Internet of Things (IoT) device connected to a network, the method include: receiving a collection of data units transmitted to or from a specified IoT device among a plurality of IoT devices connected to the network, processing the quantitative information about the data units transmitted to or from the specified IoT device collected during a defined time interval to create a temporal data traffic fingerprint of the specified IoT device, wherein the quantitative information is processed to create statistical information about the data units transmitted to or from the specified IoT device over the defined time interval, wherein the statistical information includes the number of data units per time interval for each protocol or each selected set of protocols, wherein a graphical image is created from the quantitative information, the graphical image representing the temporal data traffic fingerprint of the specified IoT device, and A machine learning model is used to derive device identification information of the specified IoT device from the time data traffic fingerprint of the specified IoT device, wherein the graphic image representing the time data traffic fingerprint is classified into one of multiple learned data traffic fingerprint types to derive the device identification information of the specified IoT device.
14. The method according to claim 13, in, The statistical information includes one or more of the following: The number of data units per time interval for each protocol or for each selected set of protocols, The number of data units per time interval for each data prioritization category, The distribution of data unit lengths per time interval, and Statistics about selected contents of the payload of the data units for each time interval.
15. The method according to any one of claims 13 to 14, in, The device identification information includes information about the type of device.
16. A non-transitory computer-readable storage medium comprising instructions which, when processed by a computer, configure the computer to perform the method according to any one of claims 13 to 15.
17. A computing system, include: one or more processors; and a memory storing instructions that, when executed by the one or more processors, configure the one or more processors to: receiving a collection of data units transmitted to or from a specified IoT device among a plurality of IoT devices connected to the network, processing quantitative information about the data units transmitted to or from the specified IoT device collected over defined time intervals to create a temporal data traffic fingerprint of the specified IoT device, wherein the quantitative information is processed to create statistical information about the data units transmitted to or from the specified IoT device over the defined time intervals, wherein the statistical information includes the number of data units per time interval for each protocol or each selected set of protocols, wherein a graphical image is created from the quantitative information, the graphical image representing the temporal data traffic fingerprint of the specified IoT device, and A machine learning model is used to derive device identification information of the specified IoT device from the time data traffic fingerprint of the specified IoT device, wherein the graphic image representing the time data traffic fingerprint is classified into one of a plurality of learned data traffic fingerprint types to derive the device identification information of the specified IoT device.
