Patent Application
20060149541
1. Field of the Invention
The present invention relates generally to acoustic sensing systems, and more particularly to acoustic sensing systems for gunfire detection.
2. Related Art
Acoustic sensing systems for gunfire detection use multiple acoustic sensors to detect the supersonic shock cone from a ballistic projectile. Conventional solutions for detection of supersonic shock waves have been expanded, over the last several years, to include detection of acoustic characteristics from subsonic projectiles as well.
The performance of existing art acoustic gunfire detection systems is generally less than the level of performance desired in order to be truly effective in field environments. One of the principal drawbacks to conventional systems is their inability to deal with the high levels of acoustic interference that are often present in tactical environments. Such systems do not enable acoustic gunfire detection systems to deal effectively with background noise, and thereby achieve optimum detection sensitivity.
Exemplary embodiments of the invention provide a system and method for implementing real-time adaptive threshold triggering. An exemplary embodiment of the invention may provide a device to generate a trigger signal based on a real-time adaptive threshold. The device may include a noise state estimator to monitor an audio signal, estimate a level of background noise in the audio signal, and output the estimate of the level of background noise. The device may also include a static offset generator coupled to the noise state estimator. The static offset generator may receive the estimate of the level of background noise, apply an offset from the estimate of the level of background noise to form a static threshold, and output the static threshold. The device may further include a dynamic offset generator to generate a dynamic offset from a post-trigger level of background noise and output the dynamic offset from the post-trigger level of background noise; circuitry coupled to the static offset generator and the dynamic offset generator to receive the static threshold and the dynamic offset from the post-trigger level of background noise, apply the dynamic offset from the post-trigger level of background noise to the static threshold to form an adaptive threshold, and output the adaptive threshold; and a comparator to compare the audio signal to the adaptive threshold and generate a trigger signal if a magnitude of the audio signal is greater than a magnitude of the adaptive threshold.
A further exemplary embodiment of the invention may provide a method for generating a trigger signal based on a real-time adaptive threshold. The method may include monitoring an audio signal, estimating a level of background noise in the audio signal, generating a static threshold by applying an offset from the estimate of the level of background noise, generating a dynamic offset from the a post-trigger level of background noise, generating an adaptive threshold by applying the dynamic offset to the static threshold, comparing a magnitude of the adaptive threshold to a magnitude of the audio signal, and generating a trigger signal if the magnitude of the audio signal is greater than the magnitude of the adaptive threshold.
Still a further exemplary embodiment of the present invention may provide a system for implementing real-time, adaptive threshold triggering. Such a system may include a microphone to receive an audio signal, a device to generate a trigger signal based on a real-time adaptive threshold coupled to the microphone to form an adaptive threshold and generate a trigger signal if a magnitude of the audio signal is greater than a magnitude of the adaptive threshold. The system may also include a waveform capture module coupled to the microphone to receive the audio signal and convert the audio signal into a series of waveform packets and a waveform analysis processor to extract characteristics from the waveform packets.
In yet a further exemplary embodiment of the invention, a vehicle may incorporate a system for implementing real-time, adaptive threshold triggering. Such a vehicle may be a combat vehicle.
Further objectives and advantages, as well as the structure and function of preferred embodiments will become apparent from a consideration of the description, drawings, and examples.
The foregoing and other features and advantages of the invention will be apparent from the following, more particular description of a preferred embodiment of the invention, as illustrated in the accompanying drawings wherein like reference numbers generally indicate identical, functionally similar, and/or structurally similar elements.
Embodiments of the invention are discussed in detail below. In describing embodiments, specific terminology is employed for the sake of clarity. However, the invention is not intended to be limited to the specific terminology so selected. While specific exemplary embodiments are discussed, it should be understood that this is done for illustration purposes only. A person skilled in the relevant art will recognize that other components and configurations can be used without parting from the spirit and scope of the invention. All references cited herein are incorporated by reference as if each had been individually incorporated.
This continuous data stream of the digitized audio signal 4 may be converted to a series of waveform packets 7 by waveform capture circuitry 5. The waveform packets may be discrete time-slices of the continuous data stream, suitable for subsequent processing by a waveform analysis processor 8, for example. The waveform analysis processor can analyze the waveform packets 7 to extract specific characteristics from the audio waveform. In an exemplary embodiment of the invention, this data may be analyzed with respect to signatures and timing, and is correlated to measurements obtained from other channels, in order to compute the location of gunfire origin (i.e., a shooter's position).
Trigger generator circuitry 9 may activate the waveform capture circuitry 5 by means of a trigger 6, for example. In an exemplary embodiment of the invention, a trigger may be generated whenever the magnitude of the digitized audio signal 4 crosses a defined threshold, indicating the probable presence of a feature of interest within the data stream. The waveform capture circuitry 5 may respond to the trigger 6 by capturing a discrete time-slice of the digitized audio signal 4, and forwarding the resulting waveform packet 7 to the waveform analysis processor 8 for further processing.
The threshold at which trigger generator circuitry 9 produces a trigger 6 may be critical to overall system performance. For example, if the threshold is too low, triggers can be continuously generated, and the waveform analysis processor will be flooded with waveform packets that contain nothing but background noise. This condition is likely to produce false reports of detected gunfire (i.e., false alarms). If, however, the threshold is set too high, system sensitivity can be reduced. Legitimate gunfire events may fail to trip the threshold, and thus go undetected. The problem is complicated by the fact that the level of background noise can be widely variable. In such an environment where the level of background noise is widely variable, it may be very difficult to set a static threshold that does not result in a significant compromise in system performance.
A dynamic offset generator 14 may produce the dynamic offset signal 15, in units of dB. This signal may be referred to as the Proportional Sensitivity Time Control (PSTC) discussed in
The real-time adaptive threshold 16 may be constantly adjusting, in units of dB for example, to follow the quasi-static changes in background noise level, and to apply a dynamic PSTC after each trigger transient. In this manner, the system sensitivity is fully optimized, providing maximum permissible sensitivity in any given noise environment, without allowing undue risk of false alarms.
The output of the latch may be an up/down flag 22 that may select either of two inputs of a multiplexer (MUX) 23, in order to implement a switched gain 24. The switched may gain assume, for example, the “Up” value (1.125, or +1 dB) if no pre-trigger signal was detected in the preceding 1 second measurement interval, and the “Down” value (0.875, or −1 dB) if at least 1 pre-trigger signal was detected. The switched gain 24 may then multiply the estimated noise level 11 to form a revised estimate 25, which may be clocked into the register 26 at a 1 Hz Rate, for example.
Thus, the noise state estimator 10 may operate by measuring the digitized audio signal 4 against the estimated noise level 11 over a one second interval. If the digitized audio signal is greater than the estimated noise level at any time during the 1-second measurement interval, the revised estimate will be 1 dB higher. If however, the digitized audio signal remains below the estimated noise level for the entire 1-second measurement interval, the revised estimate will be 1 dB lower. In this manner, the estimated noise level may continuously increase or decrease in 1 dB increments, tracking changes in background noise level at slew rates of 1 dB/second.
The dynamic offset generator 14 may be implemented by storing, for example, Sensitivity Time Control (STC) data in a Read Only Memory (ROM) 32, and clocking through the ROM addresses with a 6-Bit Counter 31, at a 200 Hz clock rate. The counter 31 may be configured to count through all states whenever it is triggered, then hold at terminal count. In the hold state, the counter therefore addresses ROM location 63, which is programmed with zero offset. This state may be held until a trigger signal 6 is received, for example.
In an exemplary embodiment of the invention, when a trigger 6 occurs, and the PSTC ReTrigger Enable 28 is valid, a PSTC Trigger 29 may be accepted. The PSTC Trigger 29 may then reset the 6-Bit counter 31, and enable it to begin counting though the ROM 32 addresses. The PSTC trigger may simultaneously capture the peak amplitude of the trigger transient that occurred on the digitized audio signal 4, by latching this value into the peak amp detect (PSTC) register 30, for example. As the counter 31 clocks through the ROM 32 addresses, the stored data at each address may be multiplied 33 by the amplitude value captured in the peak amp detect (PSTC) register 30. This may scale the STC profile stored in the ROM 32 to be proportional to the peak amplitude stored in the peak amp detect (PSTC) register 30. The resulting Proportional STC (PSTC) profile is the dynamic offset 15 term.
Since a trigger 6 may occur at any time, it is possible that the counter 31 may already be part of the way through generating the PSTC profile from a previous trigger. A decision should be made, therefore, as to whether to restart the PSTC profile (i.e., Re-Trigger), or simply allow the already running PSTC profile to run to completion. In an exemplary embodiment of the invention, this decision may be made by a comparator 37, which may enable a Re-Trigger 28 if the New PSTC Level 38 is greater than the present value of dynamic offset 15. If the New PSTC Level is less than the current dynamic offset, the PSTC Re-Trigger may not be enabled, and the current PSTC profile may be allowed to run to completion. The New PSTC Level 38 may be determined by capturing the peak amplitude of the trigger in a register 34 and multiplying it 36 with a register 35 that contains the first value in the STC profile (i.e., the data in ROM 32 Location 0).
In an exemplary embodiment of the invention, the static offset generator 12 may be implemented as a variable offset 27, in decibels, from the estimated noise level 11. The output of the variable offset 27 may be the quasi-static threshold 13, in decibels, as noted above.
The estimated noise level 11 may divided by two 49 before it is routed to the high-side input of the half-scale multiplexer 45, and again by two 50 before being routed to the high-side input of the quarter-scale multiplexer 46. The adder 51 may then sum the estimated noise level 11 with the outputs of the half-scale multiplexer 45 and the quarter-scale multiplexer 46 to produce the adder output 52. Depending upon the state of control Bits CB141 and CB040, the adder output can thus assume the values 1.0, 1.25, 1.5, or 1.75 times the Estimated Noise Level 11. In such an embodiment, this equates to an offset of 0 dB, +1.94 dB, +3.52 dB, or +4.86 dB over the Estimated Noise Level.
In an exemplary embodiment of the invention, the adder output 52 may drive the low-side input of the double-scale Multiplexer 47, and twice this value 53 may drive the high side input. The output of the double-scale multiplexer 47 may drive the low-side input of the quad-scale multiplexer, and four times 54 the output drives the high-side input. Depending upon the state of control bits CB242 and CB343, the quasi-static threshold can thus assume the values 1, 2, 4, or 8 times the adder output 52.
The combined effect of the four control bits (CB3, CB2, CB1, and CB0) is to enable the quasi-static threshold 13 to be offset from the estimated noise level 11 by between 0 dB and 23 dB, in increments of approximately 1.5 dB. Table 1 below shows the offset, in dB, resulting from each possible state of the four control bits.
In an exemplary embodiment of the invention, the value in the peak amplitude detect register 30 may be loaded into a 16-bit shift register 55 shift logic 59 clocks the shift register 55 to left shift the amplitude data until the MSB 60 is set (This left justifies the data). The number of shifts, N, required to left justify the amplitude data may be stored for later processing.
When the amplitude data is left justified, the three most significant bits (MSB 60, MSB-161, and MSB-262), each control one of the multiplexers 56, 57, 58. Each multiplexer can select between a low-side input of zero, or a high-side input.
In such an embodiment, the 16-Bit value in the ROM 32 may drive the high-side input of the unity multiplexer 56. This same data, divided by two 63, may drive the high-side input of the half-scale Multiplexer 57 and, when divided by two again 64, may drive the high-side input of the quarter-scale multiplexer 58.
The adder 65 may then sum the outputs of the unity multiplexer 56, half-scale multiplexer 57 and the quarter-scale multiplexer 58, to produce the adder output 66.
Depending upon the state of the shift register's three MSBs 60, 61, 62, the adder output can thus assume the values 0, 0.25, 0.5, 0.75, 1.0, 1.25, 1.5, or 1.75 times the 16-Bit value in the ROM 32. This resolution is sufficient to ensure accuracy to 1 part in 8 (1 dB). However, in an exemplary embodiment of the invention, the adder output 66 may be normalized to full scale, and may be right-justified to properly scale the answer. The shift register 67, and shift control logic 59 right-justify the answer by right shifting by N (i.e., the same number of times the original amplitude data was left shifted).
This process may produce a dynamic offset 15 that is proportional to the product of the data in the peak amp detect register 30 and the values stored in ROM 32, accurate to 1 dB.
The embodiments illustrated and discussed in this specification are intended only to teach those skilled in the art the best way known to the inventors to make and use the invention. Nothing in this specification should be considered as limiting the scope of the present invention. All examples presented are representative and non-limiting. The above-described embodiments of the invention may be modified or varied, without departing from the invention, as appreciated by those skilled in the art in light of the above teachings. It is therefore to be understood that, within the scope of the claims and their equivalents, the invention may be practiced otherwise than as specifically described.
