WO2020257855A1

WO2020257855A1 - Device and method for indicating an emergency exit

Info

Publication number: WO2020257855A1
Application number: PCT/AU2020/050643
Authority: WO
Inventors: Travis MACILWAIN; Ian WAKELING
Original assignee: Wat Export Import Pty Limited
Priority date: 2019-06-28
Filing date: 2020-06-25
Publication date: 2020-12-30

Abstract

An evacuation device comprising a microphone to record an audible signal, for conversion to a visual signal, in the form of a flashing light to aide evacuation from a smoke filled environment. In the preferred embodiment, the unit senses and evaluates an audible signal consistent with the combined frequency and decibel characteristics of a working smoke alarm activating lights. The method involves detecting an audible signal and evaluating its specific frequency level and wave pattern, in combination with the decibel level, against the threshold required for a working smoke alarm. The method includes converting sound to a digital signal to activate an electrical circuit that emits a visual signal. The emergency light may be associated with a location where the unit's flashing lights can be viewed beneath descending smoke thereby indicating the exit point.

Description

DEVICE AND METHOD FOR INDICATING AN EMERGENCY EXIT

Field

[0001] The present invention generally relates to a device and method for use in indicating an emergency exit.

Background

[0002] Smoke alarm systems are used for detecting smoke or fire in buildings and other structures, and upon detection a smoke alarm system typically produces a loud, audible alarm to alert the occupants of the building, so that they can be evacuated in an early stage of the emergency.

[0003] Generally, if a fire emergency happens during the day, people have more chance to successfully escape from the building, not only because they are more likely to smell, see, or taste the smoke's presence before they need further assistance to escape, but also with the daylight it is easier for occupants to locate the safety exit for evacuation.

[0004] However, at night, if occupants are woken from sleep, or caught by surprise and encounter built up smoke, then the audible noise may be of little further assistance to exit the danger.

[0005] Some fire alarm systems have been developed to light up lamps or lights in an emergency situation in addition to activating the audible alarm. However, upgrading existing fire alarm systems to these newly developed integrated systems is usually expensive and inconvenient. Further, if the existing system covers the whole building, the time and amount of work required for replacing the entire system can be exorbitant. The systems may also provide little assistance to those seeking to exit a building or structure. In other words, smoke alarms only tell occupants they have a problem, but not how to solve it.

[0006] It is desired to address or ameliorate one or more disadvantages or limitations associated with the prior art, or to at least provide a useful alternative.

Summary

[0007] According to some embodiments, the present invention provides a device for indicating an emergency exit, including: a sound sensor for detecting an audio signal; a light emitting module for producing a light signal indicative of a location of or direction to an emergency exit; and a controlling module configured to:

determine whether the detected audio signal includes an audible alarm generated by an external smoke alarm system; and if the detected audio signal includes an audible alarm, cause the light emitting module to produce the light signal.

[0008] An embodiment of the device is self-contained, in that it is in a housing separate and not connected to the smoke alarm system. The device seeks to address, in use, the deficiency of most smoke alarms, that is to direct a person to a safe escape, i.e. the emergency exit.

[0009] According to some embodiments, the present invention provides a method for indicating an emergency exit, including:

detecting an audio signal; determining whether the detected audio signal includes an audible alarm generated by

an external smoke alarm system; if the detected audio signal includes an audible alarm, causing a light signal to be

produced for indicating a location of or direction to an emergency exit.

Brief Description of the Drawings

[0010] Embodiments of the present invention are hereinafter described, by way of example only, with reference to the accompanying drawings, wherein:

[0011] Fig. 1 illustrates an exemplary structure of a device for indicating a safety exit in an emergency situation;

[0012] Fig. 2A shows an example of the device;

[0013] Fig. 2B shows an example of the device in use;

[0014] Fig. 3 illustrates an example of a method executed by the device for indicating the safety exit in an emergency situation;

[0015] Fig. 4 shows an example of hardware components of the device; [0016] Figs. 5 - 9 are exemplary circuit diagrams of some hardware components of the device;

[0017] Figs. 10 - 12 show examples of signals before, during and after processed by an envelope generator of the device;

[0018] Figs. 13 - 19 are exemplary circuit diagrams of some other hardware components of the device;

[0019] Fig. 20 illustrates a controlling process implemented by a microcontroller of the device;

[0020] Fig. 21 illustrates an example of a beep discrimination algorithm implemented by the microcontroller;

[0021] Fig. 22A and Fig. 22B show an example of a digital signal being converted into a binary signal by the microcontroller;

[0022] Figs. 23 A - 23E show some examples of applying the beep discrimination algorithm to different signals; and

[0023] Figs. 24 - 27 show an exemplary design of the device.

Description

[0024] Described herein is a device for use in indicating the destination, location of or direction to a safety exit in an emergency situation. The device detects the audible alarm produced by a smoke detector and turns on a light emitting module to provide illumination and direction for someone trying to exit a room, building or structure. The device compensates for the inability of existing smoke alarms to ensure safe escape for occupants.

[0025] The device can be located at or adjacent to the emergency exit, preferably near the floor, such that that the occupants can see, and in particular see the source of light produced by the device, when they are trying to escape a smoke-filled room or building by crawling. The device may be used for either residential or commercial premises. One significant advantage of the device is that it emits a light stimuli/signal that captures the occupant's attention and allows them to focus just on it alone to arrive a safe exit destination. It is envisaged that as the smoke fills from the ceiling downwards the escaping occupant might be crawling under the smoke level and thus cannot see the exit sign affixed above the doorway. Hence the flashing light provides the missing stimuli for the occupant to concentrate on when crawling or making their escape to a safe exit obscured by smoke. The flashing light may be white light to provide sufficient visibility. Alternatively, it may be of any other suitable colour.

[0026] Effectively the device illuminates the destination of or route to safety, by converting an audible indication produced by an existing smoke alarm system to a visual indication.

[0027] The device 100, as shown in Figure 1 , includes a sound sensor 102, a light emitting module 104, and a controlling module 106.

[0028] The sound sensor 102 is adapted to detect an audio signal. The detected audio signal is processed by the controlling module 106 to determine whether the audio signal includes an audible alarm generated by an external smoke alarm system. If so, the controlling module 106 activates the light emitting module 104 to produce a light signal for indicating a location of or direction to an emergency exit.

[0029] The controlling module 106 determines that the detected audio signal includes an audible alarm generated by an external smoke alarm system if one or more predetermined criteria are met. The specific criteria may be selected based on international, national or regional standards, regulations or rules for smoke alarm signals, for example, one or more of the following standards:

• Australian Standard AS 3786:2014 "Smoke alarms using scattered light, transmitted light or ionization" published by Standards Australia;

• ISO8201 "Acoustics— Audible emergency evacuation signal";

• ISO 7731 "Ergonomics - Danger signals for public and work areas - Auditory danger signals";

• ISO 12239 "Smoke alarms using scattered light, transmitted light or ionization".

[0030] The criteria may include, for instance, requirements on frequency, loudness, and/or signal pattern. For example, the standard smoke alarm signal in Australia has a fundamental frequency of 3,100 Hz and a temporal pattern, and the loudness is prescribed to be at least 85 dBA but not more than 105 dBA at 3 metres from the source. Accordingly, the controlling module 106 may be configured to recognise signals having some or all of these characteristics.

[0031] Alternatively, the smoke alarm signal may be defined as having other suitable frequency, pattern and/or loudness. For example, Section 4.2.1.3 of Australian Standard AS 3786:2015 sets out an optional type of smoke alarm signal which has a fundamental frequency of 520 Hz with odd harmonics to approximate a square wave. A low-frequency alarm is considered to have better performance at waking certain user groups, e.g., elderly people, deep sleeping young adults, people with hearing impairment, and people affected by alcohol, than the standard high-frequency alarm signal.

[0032] Alternatively, the criteria for recognising the smoke alarm signal may be selected based on corresponding industrial standards in other nations or regions.

[0033] One the inherent difficulties encountered in producing the device 100 is ensuring the control module 106 only activates the light emitting module 104 when the audible alarm is generated and not in response to other similar sounds emitted by other devices. The device 100 is a self-contained unit used solely to guide a person to an exit and the device 100 only reacts to the frequencies or particular pulse rate of the sound emitted by a smoke detector or separate alarm system.

[0034] The sound sensor 102 may be a commercially available microphone.

[0035] The light emitting module 104 may include, for example, one or more LED lights. The light signal for indicating the emergency exit may be, e.g., flashes or strobes. Preferably, the LED lights are white LED lights, which provide sufficient illumination that can be seen in a smoke-filled area. Alternatively or additionally, the light emitting module 104 may include other suitable type of light emitting device.

[0036] The controlling module 106 determines whether the detected audio signal includes an audible alarm generated by an external smoke alarm system.

[0037] Optionally, the device 100 may further include a switch (now shown in Fig. 1) for turning the device 100 ON/OFF, and/or one or more user input devices for receiving user inputs e.g., a reset button 108 as shown in Fig. 1. The reset button 108 may be used by a user to stop or pause the light signal after the light signal is triggered, e.g., after the user confirms that the smoke alarm is a false alarm, or in smoke alarm testing situations.

[0038] Fig. 2A shows an example of the device 100. In this example, the device 100 has an elongated case 110 for housing the controlling module 106. The light emitting module 104 includes a plurality of LED lights, provided in a horizontal row on the front side of the case 110. Near the LED lights is the sound sensor 102 taking the form of a microphone. The reset button 108 is provided proximate to one end of the elongated case 1 10.

[0039] Fig. 2B shows the device 100 in use. Preferably, the device 100 is located at or adjacent to an emergency exit. For example, the device 100 may be affixed to a lower portion of an emergency exit door 200, but slightly elevated from the floor (for example, 10cm to 15cm off the floor). This allows the device 100 to be easily seen by a person trying to escape a smoke- filled room by crawling. Alternatively, the device 100 may be affixed to the floor board next to the emergency exit door 200.

[0040] In some implementations, more than one device 100 may be used for indicating the location of the emergency exit. For example, in addition to the front side of the emergency exit door, a second device 100 may be affixed to the back of the emergency exit door, preferably slightly elevated from the floor (e.g., 10-15cm off the floor).

[0041] Fig. 3 illustrates a method 300 executed by the device 100 for indicating the location of or direction to a safety exit in an emergency situation.

[0042] At step 310, the device 100 detects an audio signal through the sound sensor 102.

[0043] At step 320, the device 100 determines, using the controlling module 106, whether the detected audio signal includes an audible alarm generated by an external smoke alarm system.

[0044] If so, at step 330, the device 100 causes the light emitting module 104 to produce the light signal for indicating the location of or direction to the emergency exit.

Exemplary Hardware Architecture

[0045] Fig. 4 shows an example of hardware components of the device 100.

[0046] In this example, the device 100 includes:

• the sound sensor in the form of a microphone (402);□ the controlling module 106, including:

o an amplifier (404); o a band-pass filter (406); o an

envelope generator (408); o a comparator (410); and

o a microcontroller (412);

• the light emitting module 104 in the form of four white LEDs and an associated driver (414); • the reset button (416); and

• power supply (not shown):

o a 3V low dropout linear regulator (LDO) power

supply for accommodating

3xAAA batteries; and

o a DC bias generator.

[0047] An exemplary design of the device 100 of Fig. 4 is shown in Figs. 24 - 27.

Exemplary Signal Flow

[0048] The audio signal detected by the microphone 402 is amplified by the amplifier 404, resulting in an amplified signal that is large enough for subsequent processing.

[0049] The amplifier 404 is a two-stage amplifier. With a two-stage amplifier, sufficient gain can be obtained using much less power compared to a single-stage amplifier. This reduces the power consumption of the device 100.

[0050] Alternatively, the amplifier 404 may be any other type of amplifier that provides sufficient amplification of the input audio signal. For example, the amplifier 404 may include three or more stages, or only one stage.

[0051] After the amplification, the band-pass filter 406 restricts the frequency range of the amplified signal to a predetermined range. The predetermined range is determined based on the typical frequency ranges of the audible alarms generated by commercially available smoke detectors. The band-pass filter 406 includes a high-pass filter section and a low-pass filter section, for filtering out low frequencies and high frequencies outside the predetermined range, respectively.

[0052] Next, the envelope generator 408 (which may also be referred to as the "envelope follower") generates a voltage envelope of the signal output from the band-pass filter 406. This reduces the sampling rate required for the microcontroller 412. In other words, the required operation speed of the microcontroller 412 can be slowed down, which facilitates reducing power consumption of the device 100.

[0053] The envelope signal output by the envelope generator 408 is sent to the comparator 410, which determines whether the peak signal level of the envelope is above a predetermined threshold. If so, the comparator 410 triggers the microcontroller 412 to wake up from a sleep mode to analyse the envelope signal. This ensures that only signals that are strong enough will be analysed by the microcontroller 412, thereby further reducing the power consumption of the device 100.

[0054] The microcontroller 412 has an active mode and a sleep mode (the sleep mode may also be referred to as a "low power mode"). In the sleep mode, the microcontroller 412 suspends at least some of its signal processing functions and accordingly reduces its power consumption.

[0055] The microcontroller 412 is put in the sleep mode until a change in the signal output by the comparator 410 is detected, which triggers an interrupt and wakes up the microcontroller 412. The microcontroller 412 then switches to the active mode, in which it samples the analogue input signal to convert it to a digital signal, e.g., by using an integrated or separate analogue-to-digital converter (ADC). The microcontroller 412 then performs the alarm detection algorithm on the digital signal, as described in further detail below, to determine whether the digital signal includes a pattern associated with an audible alarm generated by an external or separate smoke alarm system (the audible alarm may also be referred to as a "beep", and the associated signal pattern may be referred to as a "beep pattern").

[0056] If the microcontroller 412 detects a beep pattern, it causes the LED lights 414 to flash by activating the driver transistor.

[0057] The flashing of the LED lights may be stopped after a predetermined period of time has lapsed, and the period determined ensures that the occupants are given sufficient time to escape through the safety exit.

[0058] In some other embodiments, an analogue-to-digital (AD) conversion may be performed before the detected audio signal is processed by the band-pass filter 406 or the envelope generator 408, rather than after the comparator 410. However, performing the AD conversion after the comparator 410 can lower the required sampling rate, and thus reduce power consumption and increase battery life of the device 100.

Microphone and Amplifiers

[0059] Figs. 5 - 7 illustrate exemplary circuit diagrams of the microphone 402 and the two stages of the amplifier 404, respectively. [0060] The microphone 402 shown in Fig. 5 is a silicon based microphone that provides low power consumption.

[0061] The output signal of the microphone 402 is AC-coupled to the first stage of the amplifier 404. Accordingly, DC bias is applied before the signal is input into the amplifier 404, to bring the voltage into the range that the amplifier 404 can process.

[0062] The two stages of the amplifier 404 are both low quiescent current devices, so that the amplifier 404 has alow power consumption. As described above, the two stages allow a higher gain to be achieved at a given frequency while maintaining low power consumption. In the example shown in Fig. 6, the gain of the first stage of the amplifier 404 is (1 + (680K/120K.)) = 6.6. In the example shown in Fig. 7, the gain of the second stage (inverting amplifier) of the amplifier 404 is (680K/120K) = 5.6. This results in a cascaded gain of 36.96.

[0063] In addition, as described in further detail below, the band-pass filter 406 also provides an additional gain.

[0064] As shown in Figs. 6 and 7, the capacitors across R12 and R11 limit the high frequency output and form a first-order low-pass filter. The low-pass threshold is (1/(2 x p x R12 x C4)) = 4980Hz, such that it is above the highest expected frequency of the audible alarm generated by the smoke detector(s) to be used together with the device 100.

[0065] The capacitor in series with R13 limits the low frequency and forms a first-order high- pass filter. The high-pass threshold is (1/(2 x p x R13 x C6)) = 2822Hz, such that it is below the lowest expected frequency of the audible alarm generated by the smoke detector(s) to be used together with the device 100.

[0066] As shown in Fig. 7, the output signal of the second stage of the amplifiers 404 is referred to as AMP2, which is the amplified signal.

Filter

[0067] Fig. 8 illustrates an exemplary circuit diagram of the filter 406.

[0068] In this example, the filter 406 is an inverting active band-pass filter with inverting gain of R33 / R32 = 5.66, for a total gain throughout of 36.96 x 5.66 = 209.19. [0069] As described above, the standard smoke alarm signal in Australia currently has a fundamental frequency of 3,100 Hz and a temporal pattern. Accordingly, a detected alarm signal that is output by a smoke alarm system is expected to have a frequency range around 3,100Hz.

[0070] The high-pass threshold of the filter 406 is (1/(2 x p x R32 x Cl 5)) = 2822Hz, which is below the expected smoke alarm frequency range.

[0071] The low-pass threshold of the filter 406 is (1/(2 x p x R33 x 08)) = 4980Hz, which is above the expected smoke alarm frequency range.

[0072] DC bias is applied to the high-pass input stage to bring the operating voltage to a range that the filter 406 can process.

[0073] Working in conjunction with the first-order filters in the amplifiers 404, the filter 406 removes signals of frequency ranges that are not part of the smoke alarm.

Envelope Generator and Comparator

[0074] Fig. 9 illustrates an exemplary circuit diagram of the envelope generator 408 and the comparator 410.

[0075] The input signal of the envelope generator 408 is AC coupled through the capacitor C3.

[0076] The circuit after this is a precision rectifier which produces a fully rectified waveform. The capacitor C7 on the output acts as a low-pass filter to remove all the high frequency content from the signal, thereby generating the envelope of the signal. An example of this process is shown in Figs. 10 - 12.

[0077] Although there is some reduction in overall amplitude due to the effect of the filter capacitor C7, this reduction is not significant and does not reduce the overall performance of the device 100 in detecting alarms.

[0078] The output envelope signal contains only the low frequency content of the input signal, and thus can be processed at a low sampling rate. [0079] The comparator 410 is used to compare the peak of the envelope signal with a predetermined threshold. If the peak is above the threshold, the output COMP OUT signal will go from high to low.

The microcontroller 412 is configured to monitor this COMP OUT signal, and wake up from the sleep mode if a change of signal level from high to low is detected. This triggers the microcontroller 412 to run a beep discrimination algorithm for checking for a beep pattern. The beep discrimination algorithm reduces the false positive result caused by sounds other than a smoke alarm, e.g., background noise, music, phone rings.

Microcontroller

[0080] Fig. 13 illustrates an exemplary circuit diagram of the microcontroller 412.

[0081] The microcontroller 412 runs the system software, and is responsible for:

• detecting the change in the comparator output signal COMP OUT, and waking up to process the input audio signal;

• sampling the input audio signal at a suitable rate;

• running the beep discrimination algorithm, as described in further detail below; and□ causing the LEDs to flash if beeps are detected.

[0082] Preferably, the microcontroller 412 is selected such that it has a low operating current, which further reduces the power consumption of the device 100.

LEDs and Driver

[0083] Fig. 14 illustrates an exemplary circuit diagram of the LEDs and an associated driver.

[0084] When the microcontroller 412 detects a beep pattern, it triggers the signal LED ON goes high to turn MOSFET Q1 on, which causes all the four white LEDs to turn on together.

[0085] The four LEDs have separate current limiting resistors, so that any variation in the LEDs on threshold voltage does not lead to a difference in individual LED brightness. [0086] The current limiting resistors value can be changed to alter the brightness of the LEDs. The brightness of the LEDs is also affected by the battery voltage VBAT, which powers the LEDs.

Reset button

[0087] Optionally, the device 100 may be provided with a reset button 416 for user interaction. An example of the reset button 416 is shown in Fig. 15.

[0088] As described above, the reset button 416 may be used for stopping the flashing of the LED lights if the user confirms that the smoke alarm is a false alarm. It can also be used in smoke alarm tests.

Power Supply and Power Management

[0089] The device 100 is powered by batteries. As shown in Fig. Three AAA batteries (BT1, BT2, BT3) are arranged in series to provide an operating voltage of 4.5VDC when new and 3.9VDC when nearly flat.

[0090] The LEDs require a voltage of 3.6V and thus can run directly from the batteries.

[0091] The microcontroller 412 cannot run from voltages above 3.6VDC. Accordingly, a low dropout linear regulator (LDO) as shown in Fig. 17 is used to reduce the voltage to 3.0V DC. This reduced voltage is also used to power the microphone 402, the amplifier 404, the bandpass filter 406, the envelope generator 408, the comparator 410 and the DC bias. Preferably, the LDO is selected so that it has low quiescent current, which may further improve the battery life. For the same reason, the resistors in the LDO are preferably selected to have high resistances.

[0092] The DC bias circuit as shown in Fig. 18 is used to bias the operating voltage of the two stages of the amplifier 404 and the filter sections of the filter 406 to half the available supply rail. This allows these components to operate within their designed voltage range.

[0093] In some embodiments, the battery voltage VBAT may be divided down and provided to an ADC (as shown in Fig. 19), such that the battery voltage can be monitored, e.g., by the microcontroller 412. [0094] In some embodiments, the device 100 does not have an ON/OFF switch, such that once deployed, it runs continuously. A power down can be achieved by removing the batteries from the device 100. The initial power on control can be achieved either by a user inserting the batteries into the device 100, or removing a plastic tab that prevents the electric contact between the batteries and the device 100.

[0095] Alternatively, the device 100 may include an ON/OFF switch for turning the device on or off.

Controlling Process

10096] As described above, upon detection of the change in the signal COMP OUT output by the comparator 410, the microcontroller 412 wakes up from the sleep mode to process the input signal, and controls the LEDs accordingly.

[0097] Fig. 20 illustrates a controlling process implemented by the microcontroller 412 of the device 100.

[0098] At step 2010, the microcontroller 412 converts the input analogue signal into a digital signal by using an internal ADC module (this may alternatively be achieved by using a ADC component separate from the microcontroller 412).

[0099] At step 2020, the microcontroller 412 determines whether the digital signal is associated with beeps generated by a smoke alarm system, by running the beep discrimination algorithm.

[00100] If beeps are detected, at step 2030 the microcontroller 412 causes the LEDs to flash.

[00101] Otherwise, the microcontroller 412 checks whether the reset button 416 is pressed (step 2040).

[00102] If the reset button 416 is pressed, the microcontroller 412 stops the LEDs from flashing (step 2050).

[00103] Otherwise, the microcontroller 412 returns to step 2010. [00104] The controlling process may include other suitable alternative or additional steps, and may be implemented in any suitable programming language.

[00105] For example, the following code is an example of the controlling process implemented using the C programming language:

Beep Discrimination Algorithm

[00106] As describe above, the microcontroller 412 determines whether the sampled signal is associated with beeps generated by a smoke alarm system by running the beep discrimination algorithm.

[00107] Preferably, this is determined based on one or more of the following criteria:

• whether the signal is periodic;

• whether the duty cycle of the signal is within a predetermined duty cycle range; and□ whether the noise level is below a predetermined threshold. [00108] Preferably, the microcontroller 412 determines that a beep pattern is detected when all these three criteria are met.

[00109] An example of the beep discrimination algorithm performed by the microcontroller 412 is illustrated in Fig. 21.

[00110] At step 21 10, the microcontroller 412 converts the digital signal into a binary signal. An example of the signals before and after the conversion is shown in Figs. 22A and 22B.

[00111] Fig. 22A shows a digital signal after the A-D conversion. The signal is sampled at 50Hz. As shown in Fig. 22B, an average of the signal of this digital signal is calculated, and a binary signal is generated by comparing the read value (the value in Fig. 22A) with the average, wherein high level represents a positive comparison result and low level (zero) represents a negative comparison result.

[00112] The binary signal of Fig. 22B can be tracked for a predetermined number of cycles. The number of cycles tracked may be determined based on the frequency of the signal. For example, high frequency signals may be tracked for more cycles than low frequency signals.

[00113] For instance, for high frequency signals (e.g., signals with periods of no more than 20 counts), 4 cycles are tracked; for low frequency signals (e.g., signals with periods of more than 20 counts), 2 cycles are tracked.

[00114] For each cycle, a number of signal characteristics are extracted by the microcontroller 412, for example:

(i) the duration of the high signal (also referred to as "high count");

(ii) the period (also referred to as "period count");

(iii) the duty cycle; and

(iv) the noise level.

[00115] Based on (i), the deviation (or "error") in the duration of the high signal in the tracked cycles can be calculated. Based on (ii), the deviation (or "error") in the period in the tracked cycles can also be calculated. [00116] For example, for the signal shown in Fig. 2B, the periods are more than 20 counts. Accordingly, two cycles of the signal are tracked, based on which the following signal characteristics are obtained:

First Cycle High count = 21 ;

First Cycle Period count = 40 ;

First Cycle Duty Cycle = 51.5% ;

Second Cycle High count = 22 ;

Second Cycle Period = 42 ;

Second Cycle Duty Cycle = 52.4% ; and

Noise Average = 7.

[00117] Accordingly, in the two tracked cycles, the error in the high count is 4.5%, and the error in period is 4.7%.

[00118] At step 2120, the microcontroller 412 determines whether the signal is periodic. This may be conducted by checking whether the frequency or period of the signal is stable, e.g., by comparing the error in period with a predetermined threshold, e.g., 20%. Accordingly, if the error in period for two consecutive cycles is less than 20%, it is determined that the signal is periodic.

[00119] In some other embodiments, whether the signal is periodic may be determined further based on the error in the duration of the high signal (i.e., error in the high count). For example, the error in the high count of each cycle may be compared with a predetermined threshold (e.g., 20%), and only if the error in the period and the error in the high count are both below the corresponding threshold, the signal is taken to be periodic.

[00120] For example, for the signal shown in Fig. 2B, the error in the high count and the error in period are 4.5% and 4.7%, respectively. Accordingly, this signal is deemed periodic by the microcontroller 412.

[00121] If the signal is periodic, the microcontroller 412 proceeds to step 2130 to determine whether the duty cycle of the signal is within the predetermined duty cycle range. [00122] The duty cycle for each cycle is defined as the percentage of the duration of the high signal in that cycle in relation to the period of the cycle.

[00123] The predetermined duty cycle range can be selected based on the typical beeps of smoke alarm systems.

[00124] Preferably, different duty cycle range may be selected for signals with different frequencies. This allows more accurate detection of the beep pattern.

[00125] For example, in some embodiments, for high frequency signals (e.g., signals with periods of no more than 20 counts), the range may be between 45% and 80%; for low frequency signals (e.g., signals with periods of more than 20 counts), the range may be between 35% and 65%.

[00126] For example, for the signal shown in Fig. 2B (which is considered to be a low frequency signal as its periods are more than 20 counts), the duty cycle in the two tracked cycles are 51.5% and 52.4%, respectively, both fall within the range of 35% to 65%. Accordingly, the duty cycle of this signal meets the requirement.

[00127] If the duty cycle meets the requirement, the microcontroller 412 proceeds to step 2140 to determine whether the noise level is below a predetermined threshold.

[00128] The noise level can be quantified by measuring and averaging the signal differential (the signal differential being the difference between the signals before and after the binarization step shown in Fig. 22B). As beeps produced by smoke alarm systems typically contain low noise level, this step may facilitate filtering out irrelevant audio signals, e.g., phone rings.

[00129] The noise threshold can be selected based on the typical noise level on the site where the device 100 is expected to be used.

[00130] For example, in some embodiments, the noise threshold is selected to be 16.7%. Accordingly, for the signal shown in Fig. 2B, the average noise level in the two tracked cycles is 7, which is equivalent to 13% of the high signal level. This noise level is below the 16.7% threshold, and thus meets the requirement. [00131] If the noise level meets the requirement, the microcontroller 412 proceeds to step 2150 to conclude that a beep pattern is detected.

[00132] Otherwise, the microcontroller 412 concludes at step 2160 that the signal does not include beeps generated by a smoke alarm system.

[00133] By running this beep discrimination algorithm, the microcontroller 412 can determine whether the digital signal is associated with beeps generated by a smoke alarm system, and can at least partially filter out irrelevant audio signal, e.g., claps, bass, bumps and bells, music, phone rings, and random noise.

Application of the Beep Discrimination Algorithm

[00134] Figs. 23 A - 23E show some further examples of applying the above-described beep discrimination process to different signals.

[00135] In these examples, the signals input to the microcontroller 412 are sampled at 50Hz, and the following criteria are used for beep discrimination: a) the deviation (or "error") in the signal period is: (i) less than 20% or (ii) no more than 1 count; b) for high frequency signals (signals with periods of no more than 20 counts), 4 cycles need to be recognized and the duty cycle are between 45% and 80%; c) for low frequency signals (signals with periods of more than 20 counts), 2 cycles need to be recognized and the duty cycle are between 35% and 65%; and d) noise level is less than 16.7%.

[First Example]

[00136] In a first example, the converted binary signal is shown in Fig. 23A. [00137] For this signal, the periods are more than 20 counts. Accordingly, two cycles of the signal are tracked, and based on which the following signal characteristics are obtained:

• High Counts are 13 and 12, with error of 7.9% ;

• Period Counts are 31 and 31, with error of 0% ;

• Duty Cycles are 41.93% and 38.71% ; and □ Noise Average is 9.4, corresponding to 7% .

[001381 Accordingly, it can be recognized that:

the period error is less than 20%;

the duty cycles are between 35% and 65%; and

the noise level is less than 16.7%.

[001391 As a result, it can be recognised that this signal contains beeps.

[Second Example]

[001401 In a second example, the converted binary signal is shown in Fig. 23B.

[00141] For this signal, the periods are less than 20 counts. Accordingly, four cycles of the signal are tracked, and based on which the following signal characteristics are obtained:

• High Counts are 3,4,4 and 4, with error of 1 count ;

• Period Counts are 6,6,7 and 6, with error of 1 count ;

• Duty Cycles varies from 50% and 66% ; and

Noise Average is 3, corresponding to 15% .

[00142] Accordingly, it can be recognised that: the period error is no more than 1 count;□ the duty cycles are between 45% and 80%; and

• the noise level is less than 16.7%.

[00143] As a result, it can be determined that this signal also contains beeps.

[Third Example]

[00144] In a third example, the converted binary signal is shown in Fig. 23C.

[00145] For this signal, the period error exceeds the 20% threshold. Accordingly, the microcontroller 412 recognises that this not a periodical signal, and thus does not contain beeps.

[Fourth Example]

[00146] In a fourth example, the converted binary signal is shown in Fig. 23D.

[00147] For this signal, the periods of the first two cycles are less than 20 counts. Accordingly, four cycles of the signal are tracked, and based on which the following signal characteristics are obtained:

• High Counts are 5,5,3 and 4;

• Period Counts are 18,20, 19 and more than 20, with error of 2, corresponding to 10% ; and

• Duty Cycles varies from 15% and 28%.

[00148] Accordingly, it can be recognised that although the period error is less than 20% (i.e., the signal is periodic), the duty cycles are lower than 35% and thus not within the predetermined range.

[00149] As a result, it can be determined that this signal does not contain beeps.

[Fifth Example] [00150] In a fifth example, the converted binary signal is shown in Fig. 23E.

[00151] For this signal, the periods are more than 20 counts. Accordingly, two cycles of the signal are tracked, and based on which the following signal characteristics are obtained:

• High Counts are 23 and 25, with error of 8% ;

• Period count is 45 and 42, with error of 8% ;

• Duty Cycles are 51.1 1 % and 51.02% ; and

• Noise Average is 28, corresponds to 23% .

[00152] Accordingly, it can be recognised that the period error is less than 20% (i.e., the signal is periodic), and the duty cycles are within the range of 35% to 65%. However, the average noise level exceeds the 16.7% threshold, and thus fails the noise criteria.

[00153] As a result, it can be determined that this signal does not contains beeps.

Alarm Management

[00154] Once a valid beep pattern has been detected, the microcontroller 412 causes the LEDs to start flashing on and off. This may continue until the beep pattern has no longer been detected for a predetermined period of time (e.g., 5 seconds), or the reset button 416 is pressed. In some embodiments, a timer may be used to ensure that the LEDs continues flashing for a predetermined period (e.g., at least 10 seconds) before it can be stopped by pressing the reset button.

Battery Management

[00155] In some embodiments, the battery voltage may be monitored by the microcontroller 412, such that when the battery power is low (e.g., below 3.9V), a visual indication may be provided to inform the user that the battery needs to be replaced. For example, if the battery voltage is below 3.9V, the microcontroller 412 may control the LEDs to flash once every 10 seconds as a signal of low battery power. [00156] According to at least some embodiments, the device 100 described herein can provide a visual indication of an exit in an emergency situation upon detection of an audible alarm, which may facilitate occupants of a building or structure escaping from a dangerous area. The device 100 can be affixed near ground level adjacent or on an exit door to guide a person to the exit when smoke obscures vision.

[00157] The device described herein is a self-contained device that can be used together with various existing smoke alarm systems, without requiring modification or upgrading of the existing systems. The operation of the device is independent from the smoke alarm system and does not require direct signal or data communication with the smoke alarm system.

[00158] Further, according to at least some embodiments, the beep discrimination process or algorithm run by the described device allows accurate detection of a beep pattern associated with an audible alarm generated by an external smoke alarm system. It also allows fdtering out irrelevant audio signals, and thus can reduce false detection. This not only improves the reliability of the device 100, but also reduces power consumption and achieves longer battery life.

00159] Further, according to at least some embodiments, the hardware components of the device require low operating current, which further reduces the power consumption of the device.

[00160] The device described herein can be used not only with smoke alarm systems, but also with other types of emergency detection systems that produces audible alarm sounds.

[00161] The reference in this specification to any prior publication (or information derived from it), or to any matter which is known, is not, and should not be taken as an acknowledgment or admission or any form of suggestion that that prior publication (or information derived from it) or known matter forms part of the common general knowledge in the field of endeavour to which this specification relates.

[00162] Many modifications will be apparent to those skilled in the art without departing from the scope of the present invention as hereinbefore described with reference to the accompanying drawings.

Claims

DEVICE AND METHOD FOR INDICATING AN EMERGENCY EXIT THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS:

1. A device for indicating an emergency exit, including:

a sound sensor for detecting an audio signal; a light emitting module for producing a light signal indicative of a location of or direction to an emergency exit; and a controlling module configured to:

2. The device of claim 1, wherein whether the detected audio signal includes an audible alarm is determined based at least partially on:

whether the audio signal is periodic.

3. The device of claim 1, wherein whether the detected audio signal includes an audible alarm is determined based at least partially on:

whether a duty cycle of the signal is within a predetermined duty cycle range.

4. The device of claim 3, wherein the predetermined duty cycle range is different for audio signals of different frequency ranges.

5. The device of claim 1, wherein whether the detected audio signal includes an audible alarm is determined based at least partially on:

whether a noise level of the signal is below a predetermined noise level threshold.

6. The device of claim 1, wherein determining whether the detected audio signal

includes an audible alarm includes: converting the detected audio signal into a binary signal; and determining, based at least partially on the binary signal, one or more of the following signal characteristics:

a duration of high

signal; a period; a duty

cycle; and a noise level.

7. The device of claim 6, wherein: the signal characteristics are determined by tracking a predetermined number of

consecutive cycles of the binary signal; and for audio signals of a higher frequency

range, more consecutive cycles are tracked

than audio signals of a lower frequency range.

8. The device of claim 1, wherein the controlling module includes: an amplifier for amplifying the detected audio signal; a frequency filter

for restricting the frequency of the amplified signal to a

predetermined range; and an envelope generator for generating a voltage envelope

of the filtered signal.

9. The device of claim 8, wherein the controlling module further includes:

a microcontroller for determining whether the voltage envelope is associated with an audible alarm generated by an external smoke alarm system.

10. The device of claim 9, wherein the controlling module further includes: a comparator for determining whether a peak signal level of the voltage envelope is above a predetermined peak threshold; wherein the microcontroller is configured to

perform the determination only if the comparator determines that the peak signal level of the voltage envelope is above the predetermined peak threshold.

1 1. The device of any one of the preceding claims including a housing affixed to or at least adjacent to the exit, and the housing comprises the sound sensor, controlling module and light emitting module.

12. A method for indicating an emergency exit, including:

produced for indicating a location of or direction to an emergency exit.

13. A self-contained device for performing the method of claim 12 and the device is in use affixed to or at least adjacent to the exit near ground level.