Open-loop wake-up radio based on transmitter fingerprinting

Abstract

A device (e.g., an IoT device) includes a first radio (126) and a memory device (227) accessible to the first radio (126). The memory device is configured to store a fingerprinting feature (229) for a particular transmitter device. A second radio (122) and a processor (210) are also included. The processor (210) is coupled to the first and second radios (126, 122). The first radio (126) is configured to extract a fingerprinting feature of a first received wireless signal, determine that the extracted feature matches a fingerprinting feature stored in a storage device, and, in response to determining that the extracted feature matches the feature stored in the storage device (227), transition the second radio (122) from a low-power state to a higher-power operating state and continue to receive an input signal.

Description

[Technical Field]

[0001] One challenge in wireless networks that include battery-powered devices is battery life. One type of wireless device is an Internet-of-Things (IoT) device. IoT devices often have sensors that can be used to monitor environmental conditions (e.g., temperature), machine operating conditions, or other types of conditions. IoT devices are generally "headless," meaning they have no direct user input / output capabilities (e.g., no keyboard, no display, etc.). IoT devices are often battery-powered and may be installed in an environment or machine and not directly accessed by a user. Many applications for the use of IoT devices benefit from IoT device batteries that last for a long time (e.g., several years). Summary of the Invention

[0002] In at least one example, a device includes a first radio and a memory device accessible to the first radio. The memory device is configured to store characteristics for a particular transmitter device. A second radio and a processor are also included. The processor is coupled to the first and second radios. The first radio is configured to extract characteristics of the first received wireless signal, determine that the extracted characteristics match characteristics stored in the storage device, and, in response to determining that the extracted characteristics match the characteristics stored in the storage device, transition the second radio from a lower power state to a higher power state of operation.

[0003] For a detailed description of various examples, reference will now be made to the accompanying drawings. [Brief explanation of the drawings]

[0004] [Figure 1] 1 illustrates a wireless network including IoT devices in some examples.

[0005] [Figure 2] 1 illustrates an example implementation of IoT with wake-up radio.

[0006] [Figure 3] 1 illustrates an example method for an IoT device to verify an access point based on features extracted from a wireless signal. DETAILED DESCRIPTION OF THE INVENTION

[0007] Some battery-powered wireless devices include a "primary" radio and a "wake-up" radio. The primary radio is used to transmit and / or receive data during runtime operation of the device. The primary radio can enter a low-power state (e.g., sleep, hibernate, etc.) during periods of non-use. The wake-up radio receives wireless signals from transmitters in the wireless network to determine when to wake up the primary radio. The wake-up radio may operate in an "open-loop" or "closed-loop" configuration. The closed-loop wake-up radio is pre-configured, for example, to recognize a specific sequence of symbols from the transmitter or to negotiate with the transmitter for a transmitter-specific sequence of symbols. The closed-loop wake-up radio and transmitter may follow a specific wireless protocol to determine and / or negotiate the wake-up signal. The wake-up signal negotiation is additional to data and message signaling.

[0008] Although wireless devices operate according to applicable standard protocols (e.g., IEEE 802.11, Bluetooth Low Energy, etc.), wireless transmitters within a given wireless network can be distinguished from other wireless transmitters due to imperfections in the transmitter's analog components. Such imperfections can result from randomness introduced during the manufacturing of transmitter components (e.g., digital-to-analog converters, filters, frequency mixers, power amplifiers, etc.). For example, the threshold voltage or on-resistance of a metal-oxide-semiconductor field-effect transistor (MOSFET) can vary slightly from transistor to transistor, even if fabricated using the same process step. Such nonlinear effects result in each transmitter having a unique "fingerprint." Therefore, radio frequency (RF) fingerprinting can be used by a receiver to identify a particular transmitter among other possible transmitters, thereby activating the primary radio. Activating the primary radio based on RF fingerprinting is an "open-loop" process in that negotiation of a specific set of symbols between the transmitter and receiver is not required. Thus, a wake-up signal (generated internally to the wireless device) for activating the primary radio in response to a particular transmitter aspect (the extracted "feature") is generated based on the transmitter's standard transmissions, without adding a specific / integrated wake-up signal to the normal data transmission protocol. The receiver determines an RF fingerprint of the transmission that uniquely identifies the particular transmitter with which the receiver should associate, and then uses the extracted feature to activate only the primary radio when a valid feature is detected (e.g., the extracted feature matches a feature stored in the radio's memory). The fingerprinted transmission may also include an identifier of the IoT device that the transmitter intends to activate. Fingerprinting a transmission with the IoT device's unique identifier may cause only the particular IoT device to activate its primary radio when a future transmission using the associated fingerprint is detected.

[0009] Some examples are directed to a battery-powered Internet-of-Things (IoT) device that includes a primary radio and a wake-up radio. The primary radio is used by the IoT device to transmit and / or receive data during runtime operation. In one example, a battery-powered IoT device may have one or more integrated or external sensors, and the IoT device's primary radio is used to transmit sensor data or event information to a wireless network. To conserve battery power, the primary radio transitions to a low-power state (e.g., a sleep or hibernate state). While in the low-power state, the primary radio cannot be used to transmit or receive wireless signals. Instead, the primary radio must be activated for that purpose. The wake-up radio uses RF fingerprinting (i.e., a transient "signature") to detect when a valid transmitter (e.g., an access point) is attempting to communicate with an IoT device that includes a wake-up radio. In this context, a valid transmitter is a transmitter with which the IoT device is paired and should communicate. When the wake-up radio detects a valid fingerprint, it wakes up the primary radio (i.e., transitions from a lower power state to a higher power state) to continue decoding the received wireless signals, thereby enabling run-time operations such as transmitting sensor data and receiving wireless communications from the transmitter. As such, the wake-up radio described herein uses RF fingerprinting (e.g., comparing newly extracted features with one or more features stored in memory) to wake up the primary radio. Because wake-up events are also generated while receiving data in the course of normal operation (i.e., there is no dedicated wake-up signal), negotiation of a specific set of dedicated wake-up symbols between the transmitter and receiver according to a specific protocol is not required for the open-loop wake-up radio described herein.

[0010] 1 illustrates an example of a wireless network 100 including access points 110 and 111 and IoT devices 120 and 130. Any number of IoT devices (one or more) may be connected to the network. ) may be included in the wireless network 100. In this example, each IoT device includes a battery, a primary radio, and a fingerprinting wake-up radio (WUR). IoT device 120 includes a battery 121, a primary radio 122, and a fingerprinting wake-up radio 126. Similarly, IoT device 130 includes a battery 131, a primary radio 132, and a fingerprinting wake-up radio 136. During long periods of non-use, the primary radios 122, 132 of IoT devices 120, 130 are turned off to conserve battery power. However, the fingerprinting wake-up radios 122, 132 remain powered on continuously. In the configuration of FIG. 1, the primary radio 132 in IoT device 130 is turned off. The wake-up radio 126 in the IoT device 120 detects a valid fingerprint from the wireless signal transmitted by the access point 110 (e.g., extracts a feature from the wireless signal and determines that the feature matches a feature stored in the radio's memory). In response to detecting the valid fingerprint, the wake-up radio 126 causes the primary radio 122 in the IoT device 120 to wake up (e.g., be powered on or otherwise transitioned to a higher power, fully operational state) to complete reception on the incoming wireless signal. The wake-up radio 136 in the IoT device 130 is not configured to recognize the same RF fingerprint as the wake-up radio 126 (e.g., the feature extracted by the wake-up radio 136 does not match any feature stored in the radio's memory), so the primary radio 132 in the IoT device 130 remains in a low power state (e.g., sleep, hibernation, etc.). Instead, the wake-up radio 136 of the IoT device 130 may be configured to recognize an RF fingerprint derived from the access point 111.

[0011] While FIG. 2 illustrates additional details regarding the implementation of IoT device 120, the same architecture may apply to IoT device 130 and / or other IoT devices within wireless network 100. In this example, IoT device 120 includes a battery 121, a primary radio 122, a fingerprinting wake-up radio 126, a processor 210, sensors 220, and a memory device 227. The primary radio 122, fingerprinting wake-up radio 126, and sensors 220 are coupled to processor 210. Battery 121 provides operating power to some or all of the active components within IoT device 120. While one processor 210 is shown in this example, other implementations may include two or more processors. Similarly, two or more sensors 220 may be provided. Sensors 220 are application specific. Examples of sensors 220 include temperature sensors, current sensors, voltage sensors, etc.

[0012] Each radio 122, 126 is coupled to an antenna. The primary radio 122 is coupled to antenna 225, and the wake-up radio 126 is coupled to antenna 235. Thus, each radio may be connected to its own antenna. In other examples, one antenna or antenna array is shared between the two radios 122, 126. As mentioned above, the primary radio 122 is used for purposes other than the fingerprinting wake-up radio. The primary radio 122 is used to exchange (transmit and / or receive) wireless signals with an access point during device runtime. For example, the primary radio 122 may be used to receive requests from an access point (e.g., access point 110) for sensor readings, to send data and / or signals from the sensor 220 to the access point (e.g., access point 110), etc. In implementations in which the IoT device 120 responds to requests received from an access point, the primary radio 122 in the IoT device may be powered off after transmitting / receiving information to the access point while the IoT device waits for another request from the access point. Alternatively, the primary radio 122 may be powered down following a predetermined period of non-use (eg, 30 seconds, 2 minutes, etc.).

[0013] The fingerprinting wake-up radio 126, in at least some implementations, remains continuously powered on and operational and is used to detect a valid fingerprint from the access point's standard wireless signals. In response to detecting a valid RF fingerprint, the primary radio 122 is transitioned from a low power state to a higher power state to receive incoming signals.

[0014] RF fingerprinting can be performed based on the following example categories: transient-based RF fingerprinting generation and steady-state-based RF fingerprinting generation. In transient-based RF fingerprinting generation, a transmitter transmitting from an off state to an on state triggers a unique transient feature in the transmitted wireless signal that appears before the actual transmission of a data packet. In steady-state-based RF fingerprinting generation, a unique feature is present in the transmitter's wireless signal during the modulation phase. In this case, the fingerprinting wake-up radio generates a fingerprint from at least one received symbol. Any of a number of different types of RF fingerprinting techniques can be implemented by an IoT device to verify a transmitter. Verifying a transmitter means that the IoT device verifies whether the wireless signal it receives is from a transmitter with which the IoT device is associated (e.g., paired) and whether the extracted feature matches a feature already stored in the device's memory.

[0015] One example of transient-based RF fingerprinting involves determining the power spectral density (PSD) of a preamble, such as that in an IEEE 802.11a preamble. In this particular RF fingerprinting technique, the PSD can be calculated as follows: It is characterized by the PSD coefficients. TIFF0007795258000001.tif722Here, X(k) is the coefficient of the discrete Fourier transform of the input signal x(m) and is given by the following equation. TIFF0007795258000002.tif834

[0016] The PSD of a wireless signal received from a transmitter can be used to uniquely identify that transmitter. Thus, the PSD varies between transmitters and is generally repeatable for a given transmitter. The fingerprinting wake-up radio described herein can be used to determine the PSD for an incoming wireless signal. The PSD for one or more transmitters with which the IoT device is associated is stored in memory 227 within the IoT device as a fingerprint 229. The determined PSD for a given wireless signal can be compared to the PSD stored in memory within the IoT device to determine whether a valid transmitter is attempting to communicate with the IoT device. If the PSD calculated by the IoT device matches the PSD stored in memory of the IoT device, the primary radio is transitioned from a lower power state to a higher power state (i.e., woken up).

[0017] The fingerprints 229 stored in memory 227 may be provided to or determined by the IoT device 120 in any suitable manner. In one example, a user device 211 may be coupled to the processor 210 and used to indicate to the processor 210 that the processor 210 is about to enter a training mode in which the processor 210 determines fingerprints of wireless signals it receives, and that the processor 210 is about to receive and store the fingerprints in memory 227 for subsequent use to enable the primary radio 122. In another example, a user may program one or more fingerprints 229 via a graphical user interface implemented on a computer system external to the IoT device 120 and have the external computer system transmit the fingerprints to the IoT device for storage in memory 227.

[0018] 3 illustrates an example of a method implemented by an IoT device (e.g., IoT device 120) to verify a transmitter using RF fingerprinting. The IoT device communicates with other transmitters at 306 and determines and stores the transmitter's fingerprint at 308. Fingerprinting techniques such as those described above can be used. The IoT device's primary radio 122 can be awake during steps 306 and 308 and can be used to provide signals from the transmitters to the IoT device's processor 210 for fingerprint determination. The processor 210 can store the fingerprint in memory 227.

[0019] At 310, the primary radio 122 is transitioned to a lower power state (e.g., sleep, hibernate, etc.). In one example, the processor 210 sends a signal to the primary radio 122 to transition to the lower power state following use of the primary radio to respond to a request received from a transmitter. In another example, the processor 210 sends a signal to the primary radio 122 to transition to the lower power state when a timer times out during a period when the primary radio 122 is not being used.

[0020] At 320, the fingerprinting wake-up radio (which remains on and operational) begins receiving a wireless signal. The received wireless signal may be from a valid or invalid transmitter. If the wireless signal is from a valid transmitter, the primary radio 122 should transition to a higher power (operational) state, but if the wireless signal is not from a valid transmitter, the primary radio 122 should not transition to a higher power state and should therefore remain in a low power state. As described above, a valid transmitter is a transmitter with which the IoT device is paired and with which the IoT device should communicate. The wireless signal received at 320 may include receipt of an IEEE 802.11 message preamble. Transitioning the primary radio 122 to a higher power state may include one or more of turning on power to the primary radio, increasing the operating voltage to the primary radio, clocking the primary radio at a higher frequency, etc.

[0021] At 330, the method includes extracting features from the received wireless signal. In one example, the extracted features include calculating the PSD of the received wireless signal, as described above. The fingerprinting wake-up radio 126 may calculate the PSD of the received wireless signal.

[0022] At 340, the method includes determining whether the extracted features match any features stored in memory 229 within the IoT device 120. In one implementation, the fingerprinting wake-up radio 126 makes this determination. In another example, the fingerprinting wake-up radio 126 provides the extracted features to the processor 210, which compares the extracted features to features stored in memory 227. In either case, a comparison is made between the newly extracted features and any features previously stored in memory 227. The extracted features and the features stored in memory 227 may include, for example, the PSD of the preamble of the wireless packet.

[0023] At 350, if the extracted feature does not match any feature stored in memory 227, the power state of primary radio 122 remains in the low power state. Thus, primary radio 122 is not activated.

[0024] At 360, if the extracted feature matches at least one feature stored in memory 227, the primary radio is woken up and continues decoding the received signal. In one example, the fingerprinting wake-up radio 126 determines a match and sends a signal to the processor 210 to wake up the primary radio 122. In another example, the fingerprinting wake-up radio extracts a feature from the wireless signal at 330 and provides the feature to the processor 210, which determines that a match exists and commands the primary radio 122 to transition to a higher power state (e.g., by providing an enable signal to the primary radio 122). Once the primary radio 122 transitions to the higher power state, the primary radio continues to receive incoming wireless signals and provides such signals to the processor 210 for further processing. While in the higher power state, the primary radio 122 may also be used to transmit data (e.g., sensor data).

[0025] The term "couple" is used throughout this specification. This term may encompass a connection, communication, or signal path that enables a functional relationship consistent with this description. For example, in a first example, device A is coupled to device B if device A generates a signal to control device B to perform a certain action, or in a second example, device A is coupled to device B via an intervening component C such that device B is controlled by device A via a control signal generated by device A, where intervening component C does not substantially change the functional relationship between device A and device B.

[0026] Modifications may be made to the exemplary embodiments described, and other embodiments are possible, within the scope of the claims of the invention.

Claims

1. A device, a storage device configured to store characteristics for a particular transmitter device; a first radio; a second radio, Extracting a feature of the first received wireless signal; determining that the extracted features match features stored on the storage device; transitioning the first radio from a low power state to a high power state of operation, causing the first radio to decode the first received wireless signal, in response to determining that the extracted feature matches a feature stored in the storage device. the second radio configured to Including, The device, wherein the extracted features are based on a power spectral density of the first received wireless signal and the stored features on the storage device are based on a power spectral density associated with the particular transmitter device.

2. 10. The device of claim 1, The device further includes a sensor coupled to the second radio.

3. 3. The device of claim 2, The device, wherein the second radio is further configured to use the first radio to transmit wireless data of the sensor.

4. 10. The device of claim 1, The device further comprising a processor configured to store the extracted features in the storage device.

5. 10. The device of claim 1, The second radio extracting a characteristic of the second received wireless signal; determining that the extracted feature of the second received wireless signal does not match any feature stored on the storage device; The device may be further configured to:

6. 6. The device of claim 5, the second radio is further configured to not transition the first radio from the lower power state to the higher power state of operation in response to determining that the extracted feature of the second received wireless signal does not match any feature stored in the storage device.

7. 6. The device of claim 5, the first radio configured to receive an enable signal, the enable signal being asserted to transition the first radio from the low power state to the high power state; In response to determining that the extracted characteristics of the second received wireless signal do not match any characteristics stored in the storage device, the first radio does not receive the enable signal.

8. 10. The device of claim 1, The device further includes a processor coupled to the first radio and the second radio.

9. The device of claim 1, A device wherein the features stored in the storage device are extracted from previously received wireless signals from the particular transmitter device.

10. The device of claim 1, A device that is pre-programmed with features stored in said storage device.

11. a first network device, a first wireless receiver, transitions between low and high power states; receiving a first wireless signal containing data from a second network device while in the high power state; the first wireless receiver configured a second wireless receiver configured to receive a second wireless signal while the first wireless receiver is in the low power state; a processor coupled to the first wireless receiver and the second wireless receiver, determining a first fingerprint associated with the second network device based on the first wireless signal; determining a second fingerprint associated with the second wireless signal; comparing the second fingerprint to the first fingerprint to determine if the second wireless signal is associated with the second network device; transitioning the first wireless receiver from the low power state to the high power state based on the second wireless signal being associated with the second network device; the processor configured to Including, A first network device, wherein the first fingerprint is based on a power spectral density of the first wireless signal and the second fingerprint is based on a power spectral density of the second wireless signal.

12. 12. The first network device of claim 11, the first wireless signal includes an IEEE 802.11a preamble; A first network device, wherein the first fingerprint is based on a power spectral density of an IEEE 802.11a preamble of the first wireless signal.

13. 12. The first network device of claim 11, the first wireless receiver is further configured to be coupled to a first antenna; The first network device, wherein the second wireless receiver is further configured to be coupled to a second antenna different from the first antenna.

14. 12. The first network device of claim 11, A first network device, wherein the processor is further configured to compare the second fingerprint with the first fingerprint associated with the second network device based on the second network device being paired with the first network device.

15. 12. The first network device of claim 11, A first network device further comprising: a memory coupled to the processor, the memory configured to store the first fingerprint.

16. 12. The first network device of claim 11, The first network device further includes a sensor coupled to the processor, the sensor configured to provide a set of sensor data.

17. 1. A method comprising: extracting a characteristic of the first received wireless signal by the first radio while the second radio is in a low power state; determining that the extracted features match features stored on a storage device; transitioning the second radio from the lower power state to a higher power state in response to determining that the extracted characteristics match characteristics stored on the storage device; Including, The method, wherein the extracted features are based on a power spectral density of the first received wireless signal and the stored features on the storage device are based on a power spectral density associated with a particular transmitter.

18. 18. The method of claim 17, extracting a characteristic of a second received wireless signal while the second radio is in the low power state; determining that the extracted feature of the second received wireless signal does not match any feature stored in the storage device; The method further comprises:

19. 20. The method of claim 18, responsive to determining that the extracted feature of the second received wireless signal does not match any feature stored in the storage device, maintaining the second radio in the low power state and not transitioning the second radio to the high power state.

20. 18. The method of claim 17, The method further comprising using the second radio to transmit sensor data.

Images