Devices, systems, and methods for detecting proximity-based mobile propagation

[0001] 1. Field of the Subject Disclosure

[0002] The present subject disclosure relates to mobile malware. More specifically, the present subject disclosure relates to detecting proximity-based mobile malware propagation.

[0003] 2. Background of the Subject Disclosure

[0004] Mobile communication devices, such as cellular telephones, have become a common tool of everyday life. Cellular telephones are no longer used simply to place telephone calls. With the number of available features rapidly increasing, cellular telephones are now used for storing addresses, keeping a calendar, reading e-mails, drafting documents, etc. These devices are small enough that they can be carried in a pocket or purse all day, allowing a user to stay in contact almost anywhere. Recent devices have become highly functional, providing applications useful to business professionals as well as the casual user.

[0005] Proximity-based Mobile Malware Propagation (PMMP) is a category of malware that propagates through proximal connectivity such as WiFi, Bluetooth and infrared. The target victims are any communication device that has a WiFi, Bluetooth, IR, or any other module for proximal communication. These modules are now included in default configurations from many manufacturers. More risk exists for devices that are in “discoverable” mode, which broadcasts connection availability to all nearby devices. Also at risk are devices with either no password or PIN protection, or a weak one. Although slower than propagation schemes such as network-based instant messages and emails, proximity-based malware is compelling in its unique advantage that it is unobservable by the service provider network. Thus, it is substantially more challenging to detect proximity than network-based malware propagation.

[0006] Proximity-based propagation, by establishing short range wireless connection with victims, is a preferred method for mobile malware. Detection for proximity-based malware is still an open issue due to the fact that such malware has two main advantages compared to the network-based propagation. First, proximity-based propagation is difficult to detect since the communication between the attacker and the victims bypasses network-based security inspection. The provider network cannot observe any traffic or signals since such attacks launch locally. Second, proximity-based propagation is more likely to succeed due to the weak security in local connectivity technologies. Consequently, there is an increasing amount of mobile malware that propagates through proximity-based WiFi and Bluetooth connections. Well-known mobile malware that utilizes such vulnerabilities includes Lasco, Locknut, Cabir, ComWar, PBStealer, and Skuller. Given sufficient time, a Bluetooth malware can infect all susceptible devices in the network. Therefore, it is important to detect such activities locally and quickly.

[0007] Malware that uses PMMP can execute in three different ways. One way is through an established connection. If the victim device has already established connections with other devices, the attacker can utilize these established connections to infect other victim devices. Another way is to scan-connect. The attacker can actively scan and search for all the devices within the proximity. Then the malware will attempt to connect to these newly discovered devices and request to establish new connections. If these devices have no or weak passwords and PIN numbers, or if users acknowledge these connection requests, then these devices will be infected. One other way is to re-connect. If the victim device caches previously established connection settings, including security cookies such as password/PIN, then the attacker can avoid security challenges to establish connections and execute the propagation.

[0008] The victim device is a device that has been infected and controlled by the attacker to propagate the malware to other benign devices in proximity. Malware propagation through an established connection is difficult to detect. But the impact of such propagation is restricted in small local areas due to two aspects. First, assume that the mobile devices move frequently, and that the average duration of an established connection is short. The probability that the victim device has a live connection when it is infected is very low. Second, the number of devices being infected through the established connection is also small, usually one device at a time. Thus, such propagation most likely will die out and impact only a few devices.

[0009] What is needed is a method of tracking PMMP in cases where the mobile device is infected without an active connection.

[0010] The present subject disclosure solves the above problems with novel devices, systems, and methods that leverage an agent that resides in a mobile communication device to detect PMMP. The agent injects one or several trigger network connections in the candidate connection list. These connections appear as legitimate networks and devices. However, the triggers connect to an agent server on a service provider's network. Essentially, the method is based on the assumption that malware lacks the intelligence to differentiate the trigger network connection from a normal one. Therefore, by attempting to connect through the trigger network connection, the malware reveals itself.

[0011] Unlike other detection methods, which need either malware signatures or traces, the proposed system, devices, and methods do not rely upon collecting such statistics and patterns. For this purpose, the light-weight technology can be massively deployed on mobile communication devices easily. The system collects local malware information that is unreachable by the ISP core networks. Whenever a trigger connection is attempted, the agent collects new malware signatures. The proposed proximity-based propagation detection exposes mobile malware quickly since these agents are deployed on mobile communication devices at different geographic locations. The proximity-based propagation detection is a passive detection system, which does not use any of the radio resources when the network is attack free. By tracing back the telephone numbers, the infected mobile communication devices are identified, which assists deploying mitigation plans, such as disinfection patches for customers, etc. This trigger system can be extended in many areas, by inserting trigger resources such as links, directories, devices, etc., in “sensitive” places, where risks reside.

[0012] In one exemplary embodiment, the present subject disclosure is a mobile communication device. The mobile communication device includes a processor, a memory in communication with the processor, a transceiver in communication with the processor, a malware on the memory, and an agent logic on the memory for discovering a plurality of devices in a proximity, compiling a list of discovered devices in the proximity, inserting a trigger into the list of discovered devices, receiving a request to connect to the trigger from the malware, and reporting a malware activity to an agent server on a network. The trigger appears to the malware to be a discovered device.

[0013] In another exemplary embodiment, the present subject disclosure is a system for tracking proximity-based mobile malware propagation. The system includes a network, a mobile communication device in communication with the network, an agent server in communication with the network, a malware on the mobile communication device, and an agent logic on the mobile communication device for discovering a plurality of devices in a proximity, compiling a list of discovered devices in the proximity, inserting a trigger into the list of discovered devices, receiving a request to connect to the trigger, and reporting a malware activity to the agent server. The trigger appears to the malware to be a discovered device.

[0014] In yet another exemplary embodiment, the present subject disclosure is a method for tracking proximity-based mobile malware propagation. The method includes discovering a plurality of devices in a proximity, compiling a list of discovered devices in the proximity, inserting a trigger into the list of discovered devices, receiving a request to connect to the trigger, and reporting a malware activity to an agent server on a network. The trigger appears to a malware to be a discovered device.

[0023] The present subject disclosure solves the above problems with novel devices, systems, and methods that leverage an agent that resides in a mobile communication device to detect PMMP. The agent injects one or several trigger network connections in the candidate connection list. These connections appear as legitimate networks and devices. The triggers, however, connect to an agent server on a service provider's network. The devices, systems, and methods can operate based on the assumption that malware lacks the intelligence to differentiate the trigger network connection from a normal one. Therefore, by attempting to connect through the trigger network connection, the malware reveals itself.

[0024] The total number of deployed malware detection agents depends on the usage of such a system. If it is for the purpose of collecting new PMMP based malware signatures, then a number of selected mobile devices in the network are recruited as agents at different locations. If the purpose is to identify all devices that have been infected by the malware, then triggers should be deployed on each device.

[0025] In principal, the proposed idea is extended to detect any attacks that rely on the discover-infect propagation model. In other words, if the malware searches for system resources or devices in its discovery phase, a trigger device is created for the malware to find. For example, a malware attempts to attack flash memory of mobile phones (such as the Cardblock attack in 2005) by copying the malware to the victim phone, then a bogus flash drive is deployed, which can be utilized as a trigger system for alarms.

[0026] “Mobile communication device,” as used herein and throughout this disclosure, refers to any electronic device capable of wirelessly sending and receiving data. A mobile communication device may have a processor, a memory, a transceiver, an input, and an output. Examples of such devices include cellular telephones, personal digital assistants (PDAs), portable computers, etc. The memory stores applications, software, or logic. Examples of processors are computer processors (processing units), microprocessors, digital signal processors, controllers and microcontrollers, etc. Examples of device memories that may comprise logic include RAM (random access memory), flash memories, ROMS (read-only memories), EPROMS (erasable programmable read-only memories), and EEPROMS (electrically erasable programmable read-only memories).

[0027] “Logic,” as used herein and throughout this disclosure, refers to any information having the form of instruction signals and/or data that may be applied to direct the operation of a processor. Logic may be formed from signals stored in a device memory. Software is one example of such logic. Logic may also be comprised by digital and/or analog hardware circuits, for example, hardware circuits comprising logical AND, OR, XOR, NAND, NOR, and other logical operations. Logic may be formed from combinations of software and hardware. On a telecommunication network, logic may be programmed on a server, or a complex of servers. A particular logic unit is not limited to a single logical location on the telecommunication network.

[0028] Mobile communication devices communicate with each other and with other elements via a network, for instance, a wireless network, or a wireline network. A “network” can include broadband wide-area networks such as cellular networks, local-area networks (LAN), and personal area networks, such as near-field communication (NFC) networks including BLUETOOTH®. Communication across a network is preferably packet-based; however, radio and frequency/amplitude modulations networks can enable communication between mobile communication devices using appropriate analog-digital-analog converters and other elements. Communication is enabled by hardware elements called “transceivers.” Mobile communication devices may have more than one transceiver, capable of communicating over different networks. For example, a cellular telephone can include a cellular transceiver for communicating with a cellular base station, a Wi-Fi transceiver for communicating with a Wi-Fi network, and a BLUETOOTH® transceiver for communicating with a BLUETOOTH® device. A network typically includes a plurality of elements that host logic for performing tasks on the network.

[0029] For the following description, it can be assumed that most correspondingly labeled structures across the figures (e.g., 115 and 215, etc.) possess the same characteristics and are subject to the same structure and function. If there is a difference between correspondingly labeled elements that is not pointed out, and this difference results in a non-corresponding structure or function of an element for a particular embodiment, then that conflicting description given for that particular embodiment shall govern.

[0030] FIG. 1 shows a proximity-based mobile malware propagation model, according to an exemplary embodiment of the present subject disclosure. The model includes a source mobile communication device 100 including a malware 115, a plurality of victim mobile communication devices 120-124, each including a malware 115, a wireless access point 126, and an internet 128. In this model, source mobile communication device 100 communicates with victim mobile communication devices 120 and 121 through a wireless peer-to-peer (P2P) communication protocol in order to spread malware 115. Before communicating with victim mobile communication devices 120 and 121, malware 115 must use source mobile communication device 100 to discover them. Upon command from malware 115, source mobile communication device 100 broadcasts a discover signal. Victim mobile communication devices 120 and 121 respond to source mobile communication device 100. Then, source mobile communication device 100 sends each of victim mobile communication devices 120 and 121 malware 115. Source mobile communication device 100 is also in communication with wireless access point 126. Victim mobile communication devices 122, 123, and 124 are connected to wireless access point 126 in order to communicate with internet 128. Source mobile communication device 100 sends malware 115 to each of victim mobile communication devices 122-124 through wireless access point 126. Malware 115 is capable of repeating these same actions on victim mobile communication devices 120-124 to send malware 115 to more mobile communication devices through P2P communication or through other wireless access points.

[0031] This model shows the potential problem in the current local network. Local malware propagation via wireless access points and P2P communication is stealthy due to lack of security deployment on wireless access points and mobile communication devices. The P2P communication used by these mobile communication devices can be any of BLUETOOTH, Near Field Communication (NFC), etc. Wireless access points use any WiFi standard including 802.11a, b, g, i, n, etc.

[0032] FIG. 2A shows a mobile communication device 200, according to an exemplary embodiment of the present subject disclosure. Mobile communication device 200 includes a display 202, a keypad 204, a microphone 206, and an antenna 208. Display 202 is a liquid crystal display (LCD) that serves as a visual output for the user. Keypad 204 is an input for entering information and commands to mobile communication device 200. Microphone 206 accepts aural input and allows mobile communication device 200 to deliver voice communication to the network and other mobile communication devices. Antenna 208 sends and receives wireless radiofrequency (RF) signals to and from wireless networks and other wireless devices.

[0033] FIG. 2B shows the inner components of a mobile communication device 200, according to an exemplary embodiment of the present subject disclosure. The inner components of mobile communication device 200 include a processor 210, a memory 212 including an agent logic 213 and a malware 215, a transceiver 214, a removable memory card 216, and a battery 218. Processor 210 receives input and issues commands to deliver output through the other components. Memory 212 holds information for enabling processor 210 to operate the other components of mobile communication device 200, and contains agent logic 213 and malware 215. Agent logic 213 intercepts discovery requests, such as from malware 215, and inserts triggers into discovered device lists. Triggers appear to a user of mobile communication device 200 to be unknown and/or unattractive options for communication. However, triggers appear to malware 215 to be substantially similar to every other option for communication. When a trigger is selected for communication, such as by malware 215, agent logic 213 is used to report malware activity to an agent server on a network. Transceiver 214 converts wireless signals received by antenna 208 to information capable of processing by processor 210, and vice-versa. Transceiver 214 can use one or more wireless protocols, including cellular RF, WiFi, BLUETOOTH, etc., to communicate with the network and other mobile communication devices. Battery 218 powers mobile communication device 200.

[0034] There are many embodiments of a mobile communication device that are capable of being equipped with the present subject disclosure. For instance, many legacy model cellular telephones are capable of executing the agent logic described above. In other embodiments of the mobile communication device, other displays are used, such as an LED display, OLED display, etc. In some embodiments, the display is used as a touch-sensitive input device, i.e. a touch screen. A touch screen allows the user to view output on the display as well as use the display to provide input. In some touch screen embodiments, the mobile communication device may not have a physical keypad for input. Instead, a virtual keypad is displayed on the touch screen and the user provides input by touching the virtual keys. Other forms of input such as full keyboards, accelerometers, motion sensors, etc., can be utilized in the mobile communication device. The memory can be a non-removable internal memory, or a removable memory such as in a subscriber identity module (SIM) card or a memory card inserted into a memory card reader. Many mobile communication devices have more than one transceiver or a transceiver that supports more than one protocol. For instance, it is not uncommon for a mobile communication device to support cellular radio frequency (RF), WiFi, and BLUETOOTH® protocols.

[0035] FIG. 3 shows a system for tracking proximity-based mobile malware propagation, according to an exemplary embodiment of the present subject disclosure. The system includes a source mobile communication device 300 including an agent logic 313 and a malware 315, a plurality of victim mobile communication devices 320-324, each including a malware 315, a wireless access point 326, a network 328, and an agent server 330. In this system, source mobile communication device 300 communicates with victim mobile communication devices 320 and 321 through wireless P2P communication in order to spread malware 315. Before communicating with victim mobile communication devices 320 and 321, malware 315 must use source mobile communication device 300 to discover the victim mobile communication devices 320 and 321. Upon command from malware 315, source mobile communication device 300 broadcasts a discover signal. However, agent logic 313 intercepts this discovery request. When victim mobile communication devices 320 and 321 respond to source mobile communication device 300, agent logic 313 compiles a list including victim mobile communication devices 320, 321, and at least one trigger. Triggers appear to a user of mobile communication device 300 to be unknown and/or unattractive options in the list. However, triggers appear to malware 315 to be substantially similar to every other option in the list. When a trigger is selected for communication, such as by malware 315, agent logic 313 reports malware activity to agent server 330 through network 328. When malware 315 instructs source mobile communication device 300 to send each of victim mobile communication devices 320 and 321 malware 315, source mobile communication device 300 necessarily sends malware 315 to agent server 330. Agent server 330 is not affected by malware 315, and does not send malware 315 to other mobile communication devices as do victim mobile communication devices 320-324. Agent server 330 records the reception of malware 315 as well as malware from any other mobile communication device having agent logic 313 and malware. Source mobile communication device 300 is also in communication with wireless access point 326. Victim mobile communication devices 322, 323, and 324 are connected to wireless access point 326. Source mobile communication device 300 must first discover victim mobile communication devices 322-324 through wireless access point 326 by sending a discovery request as with P2P communication. Agent logic 313 intercepts this discovery request and inserts at least one trigger in a discovered device list. When source mobile communication device 300 sends malware 315 to each device in the discovered device list, source mobile communication device 300 sends malware 315 to agent server 330 through network 328.

[0036] FIG. 3 shows the overall architecture of the system, and how it works. The key innovation is the use of the trigger, which is a fake network or device created by the agent. Not all malware distribute themselves to each and every device, but may distribute to a random selection of devices. In this case, malware is not sent to the agent server 330 every time, but still has a possibility. If the agent server 330 assumes this fact, then the agent server 330 can estimate the total number of victim mobile communication devices on the network based on the number of triggers that are activated.

[0037] FIG. 4 shows a flow chart for detecting and reporting mobile malware, according to an exemplary embodiment of the present subject disclosure. The steps in this flow chart are split by a dotted line. The steps on the left are performed by a malware while the steps on the right are performed by an agent logic. Both the malware and the agent logic are stored and run from a mobile communication device. First, the malware submits a discovery request S440 to the operating system of a mobile communication device. The agent logic intercepts the discovery request and performs the discovery S441 to search for nearby mobile communication devices. A plurality of mobile communication devices respond to the search, the responses received by the agent logic S442, and the agent logic compiles a list of nearby mobile communication devices. As part of this compilation, the agent logic inserts one or more triggers S443 into the list of nearby mobile communication devices. Once the triggers have been inserted, the list is returned to the malware S444. When the malware receives the list of nearby mobile communication devices, the malware selects a mobile communication device with which to connect S445. The agent logic receives this selection and determines whether the selection is a mobile communication device or a trigger S446. If the selection is a mobile communication device, then the selection is passed on to the operating system and a connection is made. If the selection is a trigger, then the agent reports the malware activity to the agent server S447.

[0038] In other embodiments, the steps performed on the right side of FIG. 4 may be split between the agent logic and an operating system of the mobile communication device. In further embodiments, the agent logic is an integral part of the operating system. However, whether the step is performed by the agent logic or the operating system is not important to the functionality of the method in FIG. 4.

[0039] Scan-connecting is a very general way to search for susceptible devices and adopted by many malware. By the steps explained above, the agent detects the malware inevitably by detecting connection requests to the trigger. Moreover, since most users do not connect to unknown devices, the chance of a user selecting to connect to the trigger(s) is much lower than the malware. This makes the malware easily detected. When the malware on the victim device attempts to scan and make new connections, an exemplary embodiment of the proposed detection scheme works as follows: The user application or the malware requests to scan the proximity for “discoverable” Bluetooth/WiFi devices. Next, the operating system performs network/device discovery. Available networks/devices will respond to the discovery request, and a candidate list will be generated accordingly. The agent logic inserts trigger network(s)/device(s) into the candidate list. The responding networks/devices plus the trigger will be reported to the application or malware.

[0040] When compared to scan-connect cases, re-connect cases are more “friendly” to attackers, and the reasons are twofold. First, once the other device is within the proximity, the connection is surely established since the configurations, including the security settings (e.g. password/pin), are cached on the device. Second, the operating system will attempt to re-establish the previous successful connections before attempting to scan for new networks and devices. In order to detect reconnection based malware propagation, a similar technique is used. In this scheme, the agent logic inserts a trigger connection into the cached list of pre-established networks. When a normal user application attempts to re-establish a connection, the user typically knows the last successful used or paired network and device. The user is generally more likely to choose those trusted connections and ignore the trigger. By contrast, the malware does not have such knowledge and inevitably selects the trigger connection upon selecting to connect to all the devices in the cached list.

[0041] In summary, for both cases, the agent logic injects trigger(s) into either the returned network/device list for new discovered connections, or in the cached list for pre-established connections. When users choose the desired network/device to establish connections, they are more likely to connect to acknowledged or trusted networks/devices. Meanwhile, the malware blindly chooses a portion or a full list of candidate networks/devices to establish connections. Therefore, under these assumptions, the PMMP-based detection system raises alarms with a relatively low false detection ratio. The false alarms come from legitimate users blindly choosing networks and devices to make connections, which show the same behavior patterns as the victim devices controlled by the malware.

[0042] The method is based on the assumption that the user device will not automatically attempt to re-establish connections with the cached networks/devices if these networks/devices are not in proximity. In other words, user devices will first discover whether the cached networks/devices are in proximity if they support automatic network connections without human interaction. Devices violating this assumption will make the normal device behave like a malware since it may also attempt to connect to the trigger. Such devices make detection difficult.

[0043] FIG. 5 shows a flow chart for reporting mobile malware, according to an exemplary embodiment of the present subject disclosure. Malware is reported by an agent logic on a mobile communication device. Prior to the process, malware attempts to send a message to a device it has discovered. First, the agent logic receives the message S550, which is addressed to a trigger. Next, the message is converted S551 into a format readable by an agent server on a network. Once converted, the message is forwarded to the agent server S552 on the network. Depending on the communication protocol, the malware may expect to see a confirmation of a sent message. The agent logic first determines if the communication protocol warrants a confirmation S553 to be sent to the malware. If a confirmation is warranted, then the agent logic submits a confirmation to the malware, which falsely confirms S554 that the message was sent to the device intended by the malware, even though the message was actually forwarded to the agent server.

[0044] FIG. 6 shows a program 660 for connecting to a device, according to an exemplary embodiment of the present subject disclosure. Connection program 660 is shown on a mobile communication device 600 having an agent logic. When a user wants to connect to a device, a screen, as shown in this figure, is displayed for the user using display 602. Connection program 660 includes a discovered device list 662 including a trigger 663, a connect button 664, and a cancel button 666. When the user browses through discovered device list 662, trigger 663 appears as an unfamiliar and/or undesirable option. Another option, “Matt's Laptop,” has been highlighted for connection because it is familiar to the user. To connect to “Matt's Laptop,” the user highlights the device, as shown, and activates connect button 664. If the user does not wish to connect to any of the devices in discovered device list 662, then the user activates cancel button 666. In this embodiment, activation of buttons is performed by using keypad 604.

[0045] If the scan is a legitimate user application request, the user will most likely ignore the trigger and only choose networks/devices that look benign or well-known to the user, and establish the new connection. But, if the scan is from a malware, then due to the lack of user interaction, the malware is unable to distinguish a trigger from the responded networks/devices. Thus, there is a chance that the malware will select the trigger 663 instead of the legitimate networks/devices. As such PMMP-based scanning increases, the probability of the malware choosing at least one trigger will approach 100% (given an event P, if the probability of P occurring is not strictly 0, it is best to assume that P will occur, since it will almost surely happen, provided that enough time be granted). The agent logic sends the malware reports to the agent server, which reports the malware signatures and victim identities to the service provider for the network.

[0046] Other embodiments of the mobile communication device feature a touch screen that performs as a display and an input. The connect button and cancel button can be activated by touching an area of the touch screen in these embodiments.

[0047] FIG. 7 shows a connection program 770 for accessing a resource, according to an exemplary embodiment of the present subject disclosure. Connection program 770 is shown on a mobile communication device 700 having an agent logic. When a user wants to access a resource, a screen, as shown in this figure, is displayed for the user using display 702. Connection program 770 includes a discovered device list 772 including a trigger 773, an open button 774, and a cancel button 776. When the user browses through the discovered device list 772, trigger 773 appears as an unfamiliar and/or undesirable option. Another option, “SIM Card,” has been highlighted for access because it is familiar to the user. To access “SIM Card,” the user highlights the resource, as shown, and activates open button 774. If the user does not wish to access any of the devices in discovered device list 772, then the user activates cancel button 776. In this embodiment, activation of buttons is performed by using keypad 704.

[0048] Other embodiments of the mobile communication device feature a touch screen that performs as a display and an input. The connect button and cancel button can be activated by touching an area of the touch screen in these embodiments.

[0049] The foregoing disclosure of the exemplary embodiments of the present subject disclosure has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the subject disclosure to the precise forms disclosed. Many variations and modifications of the embodiments described herein will be apparent to one of ordinary skill in the art in light of the above disclosure. The scope of the subject disclosure is to be defined only by the claims appended hereto, and by their equivalents.

[0050] Further, in describing representative embodiments of the present subject disclosure, the specification may have presented the method and/or process of the present subject disclosure as a particular sequence of steps. However, to the extent that the method or process does not rely on the particular order of steps set forth herein, the method or process should not be limited to the particular sequence of steps described. As one of ordinary skill in the art would appreciate, other sequences of steps may be possible. Therefore, the particular order of the steps set forth in the specification should not be construed as limitations on the claims. In addition, the claims directed to the method and/or process of the present subject disclosure should not be limited to the performance of their steps in the order written, and one skilled in the art can readily appreciate that the sequences may be varied and still remain within the spirit and scope of the present subject disclosure.