DETECTING DROP TYPE SURFACE

1 . A method for determining whether a fall of an electronic device is on a hard surface or a soft surface, the method comprising:

collecting N samples of acceleration data after detecting a free-fall event;

calculating a number of crossing events from the N samples of acceleration data, the number of crossing events corresponding to a number of times an acceleration norm calculation for each sample of the N samples exceeds a first threshold;

determining that the fall is on the hard surface in response to the number of crossing events being greater than a second threshold; and

determining that the fall is on the soft surface in response to the number of crossing events being less than the second threshold.

2 . The method of claim 1 , further comprising detecting the free-fall event by a finite state machine circuit of a sensor of the electronic device.

3 . The method of claim 2 , where the determining that the fall is on the hard surface or the soft surface comprises determining by a machine learning core of the sensor, the method further comprising communicating surface type of the fall to a processor of the electronic device.

4 . The method of claim 3 , wherein the finite state machine circuit and the machine learning core are always ON, wherein the finite state machine circuit communicates the free-fall event detection to the machine learning core, and, in response, the machine learning core determines whether the fall is on the hard surface or the soft surface.

5 . The method of claim 3 , wherein the finite state machine circuit is always ON and the machine learning core is initially in low-power mode, wherein the finite state machine circuit communicates the free-fall event detection to the processor, and, in response, the electronic device enables the machine learning core and the machine learning core determines whether the fall is on the hard surface or the soft surface.

6 . The method of claim 3 , wherein the sensor can change device configurations independently from the processor, wherein the finite state machine circuit is always ON and the machine learning core is initially in low-power mode, wherein, in response to the finite state machine circuit detecting the free-fall event, the finite state machine circuit enables the machine learning core and the machine learning core determines whether the fall is on the hard surface or the soft surface.

7 . The method of claim 1 , wherein the first threshold and the second threshold are stored in a memory storage of the electronic device, the first threshold and the second threshold being configurable threshold values.

8 . A method for determining whether a fall of an electronic device is on a hard surface or a soft surface, the method comprising:

collecting N samples of acceleration data after detecting a free-fall event;

applying a high-pass filter on the N samples of acceleration data;

calculating a variance from the N samples of acceleration data after applying the high-pass filter on the N samples of acceleration data;

determining that the fall is on the hard surface in response to the variance from the N samples of acceleration data after applying the high-pass filter on the N samples of acceleration data being greater than a threshold; and

determining that the fall is on the soft surface in response to the variance from the N samples of acceleration data after applying the high-pass filter on the N samples of acceleration data being less than the threshold.

9 . The method of claim 8 , further comprising detecting the free-fall event by a finite state machine circuit of a sensor of the electronic device.

10 . The method of claim 9 , where the determining that the fall is on the hard surface or the soft surface comprises determining by a machine learning core of the sensor, the method further comprising communicating surface type of the fall to a processor of the electronic device.

11 . The method of claim 10 , wherein the finite state machine circuit and the machine learning core are always ON, wherein the finite state machine circuit communicates the free-fall event detection to the machine learning core, and, in response, the machine learning core determines whether the fall is on the hard surface or the soft surface.

12 . The method of claim 10 , wherein the finite state machine circuit is always ON and the machine learning core is initially in low-power mode, wherein the finite state machine circuit communicates the free-fall event detection to the processor, and, in response, the electronic device enables the machine learning core and the machine learning core determines whether the fall is on the hard surface or the soft surface.

13 . The method of claim 10 , wherein the sensor can change device configurations independently from the processor, wherein the finite state machine circuit is always ON and the machine learning core is initially in low-power mode, wherein, in response to the finite state machine circuit detecting the free-fall event, the finite state machine circuit enables the machine learning core and the machine learning core determines whether the fall is on the hard surface or the soft surface.

14 . The method of claim 8 , wherein the threshold is stored in a memory storage of the electronic device, the threshold being configurable.

15 . A sensor of an electronic device, the sensor comprising:

an accelerometer configured to collect acceleration data;

a finite state machine circuit configured to receive the acceleration data from the accelerometer and detect a free-fall event; and

a machine learning core, in response to detecting the free-fall event, configured to:

calculate a number of crossing events from N samples of acceleration data collected after the free-fall event, the number of crossing events corresponding to a number of times an acceleration norm calculation for each sample of the N samples exceeds a first threshold,

determine that a fall of the electronic device is on a hard surface in response to the number of crossing events being greater than a second threshold, and

determine that the fall is on a soft surface in response to the number of crossing events being less than the second threshold.

16 . The sensor of claim 15 , wherein the machine learning core is further configured to communicate surface type of the fall to a processor of the electronic device.

17 . The sensor of claim 16 , wherein the finite state machine circuit and the machine learning core are always ON, wherein the finite state machine circuit communicates the free-fall event detection to the machine learning core, and, in response, the machine learning core determines whether the fall is on the hard surface or the soft surface.

18 . The sensor of claim 16 , wherein the finite state machine circuit is always ON and the machine learning core is initially in low-power mode, wherein the finite state machine circuit communicates the free-fall event detection to the processor, and, in response, the electronic device enables the machine learning core and the machine learning core determines whether the fall is on the hard surface or the soft surface.

19 . The sensor of claim 16 , wherein the sensor can change device configurations independently from the processor, wherein the finite state machine circuit is always ON and the machine learning core is initially in low-power mode, wherein, in response to the finite state machine circuit detecting the free-fall event, the finite state machine circuit enables the machine learning core and the machine learning core determines whether the fall is on the hard surface or the soft surface.

20 . The sensor of claim 15 , wherein the first threshold and the second threshold are stored in a memory storage of the electronic device, the first threshold and the second threshold being configurable threshold values.