Integrated sensor system

Abstract

The present invention relates to a method and apparatus for providing noise and/or vibration monitoring. Specifically, the present invention relates to the provision of a distributable and/or configurable method and/or apparatus for noise and/or vibration monitoring. According to a first aspect, there is provided a method of monitoring ambient sound in an area for use with a device for recording ambient sound, comprising the steps of: receiving ambient sound data from the device; determining when a parameter related to the ambient sound data exceeds a predetermined threshold; and storing ambient sound data when the predetermined threshold is exceeded.

Claims

1. A method for storing ambient sound recorded in an area using a microphone of a device, comprising: recording a signal of ambient sound using the microphone of the device; using a processor for executing a code for: applying a signal analysis on the signal for determining an ambient sound parameter; when the ambient sound parameter exceeds a predetermined threshold storing a recording of the ambient sound in a memory location; monitoring a length of time during which the ambient sound parameter exceeds the predetermined threshold; and only upon determining that when the ambient sound parameter has exceeded the predetermined threshold for longer than a duration of a predetermined time period, making the recording available to a user.

2. The method according to claim 1, wherein the ambient sound parameter is a measure of ambient noise.

3. The method according to claim 1, wherein the ambient sound is continually recorded and stored until the ambient sound parameter falls below the predetermined threshold.

4. The method according to claim 1, comprising the further steps of: recording a sample of the ambient sound into a local data buffer; overwriting contents of a local data buffer with the recorded sample; and storing the contents of the local data buffer when the predetermined threshold is exceeded.

5. The method according to claim 4, comprising the further step of: assembling the stored contents of the local data buffer and the recording of the ambient sound into a single audio file; wherein the audio file is continuous.

6. The method according to claim 4, wherein the length of the sample is between 1 and 30 seconds.

7. The method according to claim 1, wherein the processor is of the device.

8. The method according to claim 1; wherein the processor is of server external to the device.

9. The method according to claim 1, wherein the recording is stored on a memory of a server external to the device.

10. The method according to claim 1, wherein recording is stored on a data storage of the device; wherein the method further comprising: receiving a request to transmit one or more items of the stored recording, and transmitting the one or more items of stored recording to an external server in response to the request; wherein the device is in communication with the external server.

11. The method according to claim 1, wherein the making the contents of the local storage available to a user comprises transmitting one or more items of stored recording to an external server upon receipt of a request for said one or more items of stored recording from the external server, wherein the device is in communication with the external server.

12. The method according to claim 1, comprising the further step of: deleting the stored recording where the further predetermined threshold is not exceeded.

13. The method according to claim 1, wherein the stored recording of the ambient sound is time-stamped.

14. The method according to claim 1, comprising the further step of: transmitting an alert signal when the predetermined threshold is exceeded.

15. Apparatus for recording ambient sound, operable to carry out the method of claim 1.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

(1) Embodiments will now be described, by way of example only and with reference to the accompanying drawings having like-reference numerals, in which:

(2) FIG. 1 shows a schematic diagram of the physical components of a device according to an embodiment;

(3) FIGS. 2a, 2b, 2c, 2d and 2e show circuit diagrams illustrating various possible configurations for components of a power circuit of an embodiment of the device;

(4) FIG. 3 shows a schematic view of the physical components of an analyser board of an embodiment of the device, with the connections between components shown;

(5) FIGS. 4a, 4b, 4c and 4d show circuit diagrams illustrating various possible configurations for electronic components of an audio analyser of an embodiment of the device;

(6) FIGS. 5a and 5b show circuit diagrams illustrating various possible configurations for electronic components of a vibration analyser of an embodiment of the device;

(7) FIG. 6 shows electrical connections between various boards of an embodiment of the device;

(8) FIG. 7 shows a flow diagram illustrating an example of the audio recording functionality of the device;

(9) FIG. 8 shows a flow diagram illustrating a further example of the audio recording functionality of the device;

(10) FIG. 9 shows a flow diagram illustrating how a plurality of devices may be operated based on configuration data from an external server;

(11) FIG. 10 shows a flow diagram illustrating how the device downloads configuration data;

(12) FIG. 11 shows a schematic view of a network of devices monitoring a noise/vibration source, indicating how data is transmitted; and

(13) FIG. 12 shows a system architecture diagram of an exemplary network containing a device.

. DESCRIPTION OF SPECIFIC EMBODIMENTS OF THE INVENTION

(14) Referring now to FIG. 1, there is shown a schematic diagram of the internal physical components of a device 100 according to an embodiment, which will now be described.

(15) The device 100 comprises a casing 101, a battery 103, and a number of printed circuit boards (PCBs), each PCB having a distinctive function. In the embodiment shown in FIG. 1, 5 PCBs are provided, referred to as a radio board 115 for holding communication modules, an analyser board 113 for holding componentry for analysing captured data, a processor board 111, a power board 117 and a socket board 119 arranged to hold components allowing a physical connection to be made with the device. An advantage of providing multiple PCBs which perform discrete functions is that an entire PCB (with associated componentry) can easily be replaced if necessary, as opposed to the difficulties associated with replacing individual components towards the end of their lives.

(16) The device 100 provided is configured to measure and process desired parameters related to noise and vibration. It will be appreciated that the two types of sensor to be used with the device 100 are a microphone 20, to measure noise, and an accelerometer 30, to measure vibration. Advantageously, the device 100 may therefore be used to monitor both noise and vibration, although only one sensor of each type may be used if an application demands that only noise or vibration needs to be monitored. The device 100 may monitor both noise and vibration at the same type. Alternatively, only one or noise or vibration may be measured even if a microphone 20 and accelerometer 30 are both connected to the device 100.

(17) These sensors are provided external to the device 100 and are preferably connected to the device 100 via a TNC connector 407 for the microphone 20 and a LEMO connector 409 for the accelerometer 30. These connectors are preferably located directly on the socket board 119, as will be described later. The device is configured to use any class 1 microphone 20 with a sensitivity of 50 mV/Pa and a Bias Voltage between 8-30V, (DC), and any type of triaxial accelerometer 30 with a sensitivity of 1000 mV/g. Preferably, a Bias Voltage of 24V (DC) is used. Both sensors are preferably TEDS-compliant. In the case where a TEDS sensor is used, the device 100 may be able to recognise the sensor and read the sensor's specific parameters, including calibration factors, from the sensor's memory. When required, the device 100 may also be able to save information about sensor parameters onto this memory.

(18) It will be appreciated that sensors with different specifications may also be used with the device 100; however, the analyser board 113 may need to be replaced with an alternative board to meet the sensors' specification. The device 100 is arranged to be modular, such that easy replacement of the analyser board 113 and other components is possible.

(19) The device 100 is designed to be used outdoors, so the casing 101 of the device 100 should preferably be resistant to water and dust ingress so as to achieve at least IP54 rating. The casing 101 is preferably made of aluminium with a composite top cover attached permanently to the aluminium enclosure. The socket board 119 is preferably provided towards the bottom of the device 100, where the casing 101 provides holes to allow all sockets to be accessed from the exterior of the device 100, as is shown schematically in FIG. 1. The casing 101 is preferably arranged to provide access to the device's internal componentry through the bottom of the device 100 via a removable section (not shown).

(20) The battery 103 is internally-housed and preferably is a rechargeable lithium-polymer battery. The battery 103 may be housed around the power board 117. A fully charged battery 103 is preferably able to power the device 100 for at least 10-12 hours of continuous operation, and more preferably for at least 24 hours of continuous operation. The device 100 may be operable for use in a low power mode, where certain functions are shut down in order to improve the battery life of the device 100. To ensure continuous operation over a longer period, the device 100 may be connected to any external power source providing DC voltage between 12-50V (DC) via a power socket 411. Those sources can include an external power supply, an external battery pack, solar panels or a wind turbine. This external power input may be used to charge the battery 103 and/or directly power the device 100. Where no external power supply is available or where the power supply is interrupted, the device 100 may switch to being powered by the internal battery 103 to provide uninterrupted measurements.

(21) The processor board 111, the analyser board 113, power board 117, radio board 115 and socket board 119 can be physically arranged one on top of the other, separated by a distance of approximately 1 cm.

(22) As shown in FIG. 1, the processor board 111 comprises one or more processors 121, operational RAM memory 123, NAND FLASH memory 125 for local storage (which may be provided as a removable Micro USB card or be attached directly to the processor board 11), SERIAL memory 127 for system and configuration files, and various other connection and transmission-related components. In some embodiments, three processors 121 are used. In some embodiments, 512 MB of operational RAM memory 123 and 64 MB of SERIAL memory 127 are provided.

(23) The processor 121 used may be an AT32UC3 Audio series processor, for example. Its functions will be described later on with reference to how measurements are taken and analysed. The processor board 111 may further comprise a real time clock 129 powered by a battery, such as a CR3022 battery, to properly mark each sample for analysis with a timestamp. In some embodiments, the real time clock 129 has an accuracy of at least 2 ppm.

(24) The processor board 111 in some embodiments can also comprise a memory card 131, such as an SD card, which can be used to store data which is later exported. The processor board 111 may optionally further comprise one or more USB connections 133, which may be used to allow the device to interface with a computer, for example, via USB sockets 403, 404 located on the socket board 119. The processor board 111 may optionally also comprise means 135 for communicating over LAN networks and the Ethernet physical layer via an Ethernet connection port 405 located on the socket board 119.

(25) The processor board 111 may also comprise an additional accelerometer 409, configured to collect data related to movement of the device 100. This accelerometer 409, however, may not be suitable for use in vibration monitoring.

(26) The socket board 119 comprises all external connections to the device. The socket board 119 comprises TNC connector 407 and LEMO connector 409, as mentioned earlier. The socket board 119 may also optionally comprise one or more USB OTG sockets 403 for file transfers, a USB RS socket 404 for device configuration and calibration, an ETHERNET socket 405 for LAN connections and a signal output connector 401 for the optional provision of an external signalisator. Alternatively, a multiport 415 may be provided on the socket board 119, where this multiport 415 is arranged to accept a microphone 20 and/or accelerometer 30 cable and/or act as a power input. In some embodiments, an on/off switch 413 is also provided next to the aforementioned sockets on the exterior of the device 100, where this switch 413 is configured to communicate with the processor and power boards 111, 117 to start or shut down the device 100 when switched.

(27) The radio board 115 may comprise means 301 for communication using a GSM cellular network and a subscriber identification module (SIM). The means 301 for communication preferably uses GPRS/HSDPA/3G/4G networks. The radio board 115 may also comprise means for communicating over wireless local area networks (WLAN) (such as a WLAN client 303) and means for communicating over Bluetooth 4.0 networks (such as a Bluetooth 4.0 controller 305). The Bluetooth network may be used to communicate with Bluetooth-enabled user devices, such as computers and mobile devices, and, optionally, may be used to transfer current readings and/or download stored data, and/or read information related to the status of the device 100.

(28) The radio board 115 may also comprise a GPS receiver and associated componentry (or ‘GPS chip’ 307), which allows gathered results to be associated with the location at which they were gathered.

(29) FIGS. 2a to 2e show circuit diagrams illustrating various possible configurations for components of the power board 117. The power board 117 is arranged to transmit electrical power from the battery 103 to the electronic components of the device 100 and as such may comprise means for controlling power supplied to the processor 121 (FIG. 2a), means for receiving an input from the battery and switching the connection to the battery on and off (FIG. 2b), means for regulating power from the battery (FIG. 2c), means for regulating the voltage supplied to various other components of the main board 111 (FIG. 2d), and means for charging the battery using an external power source via the power input 411 (FIG. 2e).

(30) FIG. 3 shows a schematic view of the components of the analyser board 113, with the connections between components shown. The analyser board 113 comprises of 2 subsections: a noise analyser 211 and a vibration analyser 213. The noise analyser 211 is used to take noise related measurements and perform initial processing on these measurements. The noise analyser 211 comprises a connection from a microphone 20 (via the socket board 119), a circuit 201 arranged to supply a bias voltage to the microphone (using a phantom power input when required), a relay circuit 203, an electrical amplifier 205, an electrical filter 207, preferably an anti-aliasing filter, and an analogue to digital convertor 209. In use, analogue electrical signals output by the microphone 20 pass through the relay circuit 203 which selects the most accurate signal path depending on the microphone used and signal strength. The signal may then be filtered and/or amplified before being converted to digital signals and sent to the processor 121.

(31) FIG. 4a shows a circuit diagram illustrating a possible configuration of the circuit 201 arranged to supply a bias voltage to the microphone 20, along with the connections to and from the microphone 20. Similarly, FIG. 4b shows a circuit diagram illustrating a possible configuration of the relay circuit 203. FIG. 4c shows a circuit diagram illustrating a possible configuration of the antialiasing filter 207, which is provided in this embodiment as an 8 order Butterworth antialiasing filter which is set for an input signal with a frequency of 20 kHz. FIG. 4d shows a circuit diagram illustrating a possible configuration of the analogue to digital convertor 209, which is arranged in this embodiment to provide a maximum of 24 bit output at 96 kSPS (samples per second).

(32) Similarly to the noise analyser 211, the vibration analyser 213 is used to take vibration-related measurements and perform initial processing on these measurements. The vibration analyser 213 comprises a connection from an external accelerometer 30 (via the socket board 119), one or more antialiasing circuits 215, one or more electrical amplifiers 217, one or more analogue to digital convertor (ADCs) circuits 219 (which also acts as low-pass filters), and a connection to the processor 121 through SPI BUS 221. Several types of accelerometer 30 may be used, such as piezoresistive or piezoelectric sensors, strain sensors, or MEMS sensors. Whichever type of accelerometer 30 is used, the ADC circuit 219 is used to detect a change of a parameter, typically resistance, and cause an output of an electrical signal which is proportional to the measured acceleration. Preferably, acceleration is measured in three axes, necessitating the use of three antialiasing circuits 215, amplifiers 217 and ADC circuits 219. Typically, one tri-axial accelerometer 30 can be used to measure all three axes, as is shown in FIG. 3. The output of the antialiasing circuits 215 is amplified by the electrical amplifiers 217 and then converted into a single digital signal by the analogue to digital convertor(s) 219. The signal is then sent to processor through SPI BUS 221.

(33) FIG. 5a shows a circuit diagram illustrating a possible configuration of the analogue to digital convertor 219 for one axis of the vibration-monitoring functions of the device. FIG. 5b shows a circuit which may be used to provide a reference input voltage into this analogue to digital convertor 219.

(34) FIG. 6 shows a possible embodiment of the electrical connections between the various boards. It will, however, be appreciated that the boards may be connected in many different ways.

(35) Optionally, a digital accelerometer may be used instead of the analogue accelerometer 30 described. Such an accelerometer 30 would require different signal processing functions, so as such it is envisaged that these functions are provided using components provided on a separate, removable board which may replace the conventional analyser board 113.

(36) Further details on the parameters will now be described. The device 100 is arranged to measure a variety of user-selectable statistical parameters related to noise and vibration. The parameters may relate to displacement, octaves, or statistics. Preferably, the device calculates only those parameters selected by the user during configuration, as will be described later on. The device 100 may be arranged to perform spectral analysis on the input signal to calculate one or more parameters. A set of calculated parameters is preferably generated at the end of a user-set time period (referred to as the ‘integration period’), rather than being continuously calculated.

(37) For noise, the parameters may include sound pressure level, highest value measured over a time period (L.sub.max), lowest value measured over a time period (L.sub.min), equivalent continuous noise level (L.sub.eq), sound exposure level (SEL), peak sound measurement (L.sub.peak), day-evening-night equivalent sound level (L.sub.den), and statistical noise levels (L.sub.n), which refers to the noise level exceeded for n % of the time. Frequency values recorded may be grouped as single octave bands or as one third octave bands, depending on the level of detail required. ‘A’, ‘C’ or ‘Z’ frequency weightings may be used in the electronic filters provided in the device. ‘A’ weighting is most useful to model the response of the human ear to noise, ‘C’ weighting is most useful for the measurement of peak sound pressure level, and ‘Z’ weighting provides a ‘flat’ weighting for some other applications. A fast, slow, or impulse time constant may be selected by the user. The integration time period may also be specified, as described later.

(38) The total dynamic range (i.e. the ratio of the largest measurable signal to the smallest measurable signal) of the noise-monitoring function of the device 100 is preferably 16 dB(A) RMS-140 dB(A) Peak. Within this dynamic range, the detected signal from the microphone 20 is preferably linear within at least the range 26 dB(A) RMS-140 dB(A) Peak. The device input frequency range for noise is preferably 20 Hz-20 kHz (although this will depend to some extent on the microphone 20 used). The device's electroacoustic performance, in combination with the microphone 20 used, is in some embodiments such that it conforms to the specifications of IEC 61672-1 for a class 1 sound level meter.

(39) Similarly, a variety of vibration-related parameters may be measured and/or calculated. These parameters may include RMS, MAX, and PEAK vibration measured over a time period, which are useful parameters for measuring machine vibration; peak particle velocity (PPV) which is a measure of ground vibration; and vibration dose value (VDV), which is a cumulative measurement of the vibration level received over a long time period. The dominant discrete frequency of the highest PPV recorded within the integration period may also be calculated. These parameters may be calculated both as total values and in each of the three axes individually. As with noise measurements, frequency values recorded may be grouped as single octave bands or as one third octave bands, depending on the level of detail required.

(40) The dynamic range of the vibration-monitoring function of the device 100 is in some embodiments around 0.0005 m/s.sup.2 RMS to 50 m/s.sup.2 PEAK, although this will depend on the accelerometer 30 used. The device input frequency range for vibration is preferably from 1 Hz-10 kHz. The device's vibration monitoring performance, in combination with the accelerometer 30 used, is preferably such that it conforms to ISO 10816-1 when used to monitor machine vibration.

(41) FIG. 7 shows a flow diagram illustrating an example of the audio recording functionality of the device 100. The device 100 preferably has the function to record audio data to allow a user to recognise a source of noise during playback at a later time. This audio recording function may be implemented using the processor board 111 and/or the analyser board 113 as described, where the processor board 111 and/or the analyser board 113 may be modified to incorporate means for audio recording.

(42) As shown in the Figure, the ambient sound data may be received (step 701) using microphone 20, for example. In step 702, a parameter (such as one of the parameters described above) related to the ambient sound is determined to have exceeded a predetermined threshold. Ambient sound data may then be stored, as shown in step 703.

(43) The audio data is not saved continuously, as very high volumes of data would then need to be saved and/or transmitted. Additionally, saving continuous audio may allow users of the device 100 to be capable of intrusively listening in on audio of the device's surroundings, such as conversations between passers-by, which may lead to public disquiet and may breach local privacy laws. To mitigate such problems, the device 100 only records specific audio samples during periods where predetermined noise limits are exceeded, and for only long enough to allow the source of noise to be identified during later playback.

(44) FIG. 8 shows a flow diagram illustrating a further example of the audio recording functionality of the device 100. In step 801, short samples of audio may be continuously recorded into a buffer in local memory, such as in memory card 131 and/or NAND FLASH memory 125. The contents of the buffer may be continuously overwritten by new audio samples. The samples may be, for example, 10 seconds long. The contents of the buffer are not available for the user to access. In step 802, if measured noise exceeds a predetermined threshold sound pressure level (or equivalent parameter determined as described above), the contents of the buffer may be saved into local memory (step 803), for example by being saved to a different location on memory card 131 and/or NAND FLASH memory 125. In an example, the processor 121 may be used to determine whether the threshold is exceeded. The threshold may be, for example, 70 dB(a)L.sub.eq. Additionally, live audio may be recorded and saved into local memory (step 803), as described with reference to FIG. 7. The contents of the buffer and the recorded live audio may be assembled into a single audio file by the processor 121 prior to being saved to local memory, such that there are no gaps in audio playback. The use of a buffer allows for the source of noise to be recorded before the noise level goes above the threshold, which may make it easier for the user to identify the source of noise.

(45) The device is arranged to continue recording while comparing the measured sound pressure level against the threshold (step 804). If the sound pressure level drops below the threshold for a predetermined short period of time (step 805), such as 2s, the device 100 will stop recording and the recording may be saved to a temporary location (step 807). If the sound pressure level exceeds the threshold within the predetermined time period following it dropping below the threshold, recording will continue. In addition, if it is determined that the noise level has remained above the threshold for more than a predetermined period (step 806), which may be 30 s, for example, the device 100 will stop recording and the recording may be saved to a temporary location (step 807). The size of the audio file is thereby minimised. With the example given, the maximum length of an audio recording may therefore be 42 s. In alternative, the total file length may be compared against a predetermined maximum (for example, 42 s) to determine whether recording should be stopped due to the file being too long. If the threshold is exceeded within the length of a sample after recording has stopped (i.e. within 10 s of recording being stopped), the shortened audio sample present in the buffer will be saved into local memory and recording of live audio will begin as normal.

(46) This recording process is carried out throughout an integration period. At the end of an integration period, an overall sound pressure level or equivalent measurement for the period may be determined. If this overall sound pressure level for the period exceeds the threshold (step 809), the recorded audio file(s) are made available to the user (step 810), for example, by being exported to an external server 150. If the overall threshold is not exceeded, the recorded audio file(s) are not made available to the user, and may be deleted (step 810). This means that only significant sources of noise are reported to the user. The recorded audio file(s) may then be deleted from local storage following transmission to the external server 150.

(47) Optionally, the recorded audio file(s) are only exported to the external server 150 upon request from a user, in order to reduce the costs and/or increase the efficiency of data transmission. The recorded audio files may be sent in an uncompressed format, such as .wav, or a compressed format, such as .mp3. Recorded audio files may be timestamped to assist in the identification of sources of noise. If the maximum local storage of the device 100 is exceeded, older recorded audio files may be overwritten with newer files.

(48) Further details about the processing will now be described. The device 100 performs all processing on collected raw data on the device 100, providing the advantage of easy access to relevant calculated parameters that may readily be analysed. The device 100 may be able to collect data and act on it in near-real time. The integration period may range from a minimum of 1 second to a maximum of 24 hours, for example, depending on the level of detail required. Raw data collected within the integration period is analysed in portions and the desired parameters are calculated at the end of each integration period. For many parameters, data must be integrated over the integration period in order to accumulate data readings into a single value. Different integration periods and parameters are suitable for different situations, and so the high programmability and customisability of the device 100 allows it to be used in a wider range of situations than current equipment.

(49) The set of calculated parameters, referred to as results, are then assembled into a single file with associated time (from the real time clock 129) and/or location (from the GPS chip 307) data. Optionally, the results file may also comprise an identifier of the device, such as a unique number, as well as any audio files recorded during the integration period. The results file may also comprise data and/or metadata to assist in the later processing of the results file, such as data relating to the project with which the device is presently associated. The result file is then transmitted to an external reporting server 150. The functionality of the server 150 will be detailed later on. In case of connectivity issues all results will be stored on the device 100, and when communication is regained all unsent results may be transmitted to the server 150. All results from one day (or over any other period longer than the user-defined time period described above) may be combined into a single file, if desired. The results file is preferably in the form of a Message-Queueing Telemetry Transport (MQTT) data packet. All other transmissions to and from the device 100 are similarly preferably also in the form of MQTT data packets. MQTT is preferred as the protocol for use with the device 100 because it is extremely lightweight, making it suitable for use for applications which must transmit data from remote locations where memory usage should be minimised, such as the device 100.

(50) Optionally, the device 100 may be configured to calculate a number of parameters per time interval, where time intervals are short periods of time such as 10s, alongside calculating parameters at the end of every integration period as described above. These parameters are assembled into a results file and transmitted as described above, providing near-real time measurements of noise and vibration. The ‘near-real time’ parameters may differ from with those calculated at the end of an integration period, or alternatively may overlap. The ‘near-real time’ parameters may include current SPL and L.sub.EQ since the start of the integration period for noise, along with RMS vibration over the period and current PPV for vibration.

(51) The device's configurations settings are configured externally and sent to the device 100, as the device itself has no buttons or screen. The device 100 gathers and processes data based on the configuration settings. The configuration settings can be sent to the device 100 through a network via an external server 150, uploaded via the USB connection (if present) or through Bluetooth connection via a mobile application installed on a user device, as will be described later on. Where the device is configured via the USB connection or Bluetooth connection, the configuration settings are synchronised with the server 150. The configuration settings may comprise the desired parameters to be saved (if not all measured parameters are desired), the user-defined time period, modes of operation, such as the times of day when the device should be active (if continuous measurement is not desired), contact details and any alert thresholds, as will be described later on. The configuration settings may be assembled into a file, which may be referred to as a ‘configuration file’, which may be downloaded by the device from the external server 150 and may be stored in the memory card 131 and/or NAND FLASH memory 125, for example. The external server may optionally be configured to push the configuration file to the device 100 each time a new connection is established between a device 100 and the external server 150. The configuration file or specific data related to the configuration settings may be imaged to each of the processor board 111, analyser board 113, power board 117, radio board 115 and socket board 119. The configuration data may comprise a firmware or software update, which may be pushed out to the device 100 from the external server 150. Where firmware and/or software is updated, the device 100 may be configured to download the update, check the new firmware and/or software for errors, and, if there are no errors, restart itself in order to implement the update.

(52) FIG. 9 shows a flow diagram illustrating how a plurality of devices 100 may be operated based on configuration data from an external server 150. In step 901, configuration data may be sent from the external server 150 to the device 100. In step 902, the devices may monitor noise and/or vibration as described. In step 903, parameters may be calculated as described, where those parameters are specified by the configuration data. The device 100 therefore calculates only those parameters selected by the user during configuration.

(53) FIG. 10 shows a flow diagram illustrating how the device 100 downloads configuration data. In step 1001, the device 100 searches for new configuration data available via a communication network (using the radio board 115). Typically, the device 100 will attempt to connect to the external server 150. In step 1002, existing configuration data present on the device 100 is overwritten by new configuration data. The device 100 may then continue to monitor noise and/or vibration on the basis of the new configuration data. The device 100 may be configured to check for (and download, if possible) new configuration data after a given interval, such as every 30 minutes. This process will also take place when the device is started up, including when the device is started up following a firmware update and/or software update, and when the device is started up following a power failure. The device may also follow the same process when the connection to the external server 150 is re-established after being lost.

(54) The device may be configured so as to minimise disruption in the event of the device losing power. Where the device has switched off due to no external power source being connected and where the internal battery is depleted, the device may be arranged to automatically start up upon the connection of the external power source. When the device starts up, the device may attempt to establish a connection with the external server 150, search for and download a new configuration file as described, and then begin measurement of noise and/or vibration according to the configuration file already present on the device or that has just been retrieved from the server.

(55) The device requires calibration in order to accurately gather data. Calibration may be initiated remotely from the external server 150 or via USB or Bluetooth connection from a user device. Optionally, configuration files may comprise calibration data, such as calibration factors, or may cause calibration to be initiated. Calculated calibration factors may be saved for subsequent reference during processing.

(56) For noise measurement, calibration is performed by initiating calibration, applying a reference signal produced by an external calibrator, and by measuring the applied signal for a fixed time period (for example, 5 seconds) and calculating a final sound pressure level value. A calibration factor may then be determined based on the calculated value and the real sound pressure level of the reference signal (which may be selected by the user at a user device, for example). A user may then accept the calibration factor, whereupon the calibration factor is saved and synchronised between the external server 150 and the device 100.

(57) For vibration measurement, manual calibration may be required. To do this, sensitivity values provided by the manufacturer of the accelerometer may be entered into the device using the external server 150 or a user device connected through a USB connection or Bluetooth. Similarly, a calibration factor may be saved and synchronised between the external server 150 and the device 100.

(58) Alternatively, if the microphone and/or accelerometer used is TEDS compliant, calibration values may be read from the microphone and/or accelerometer, saved to the device 100, and synchronised between the device 100 and an external server 150. In a further alternative, calculated calibration factors may be saved directly to TEDS compliant microphones and/or accelerometers.

(59) The device 100 may, optionally, be arranged to generate alerts when a user-defined threshold for one or more noise or vibration parameters are exceeded. This functionality makes the device 100 suitable for use as an emergency warning system while acting as a passive monitor. Optionally, the threshold used for audio may be the same threshold used to initiate audio recording. Alerts may also be generated in dependence on the condition of the device 100 (for example, when the battery 103 is on low voltage, when the external power is connected or disconnected, when a measurement error is detected, when a sensor is connected or disconnected, which may be detected by measuring the current through the sensor, or when the device 100 is moved, tilted, or dropped, as detected by the internal accelerometer 409 or changes in the location data sent in messages to the server 150), and may be set to different thresholds and/or parameters depending on the time of day or day of the week, for example. The alerts are preferably transmitted to an external server 150 over a data connection, which may transmit the alerts as text messages, notifications within a mobile application or web portal, or e-mails. Alternatively, alerts may be directly transmitted by text message, using the device's GSM cellular networking capability, or by e-mail, using a GSM network or WLAN, for example. In a further alternative, in case of lack of data connectivity, the device may send a text message related to the alerts to an external server 150, which may then transmit the alerts as described. A list of relevant contact details may be user-defined. The device 100 may use different contact details depending on the type and urgency of the alert; for example, different contact details could be set for ‘users’ and ‘administrators,’ where both users and administrators receive alerts about thresholds being exceeded but only administrators receive information about the condition of the device 100. Optionally, an externally provided signalisator (not shown) may be connected via the signal output connector 401 to provide further alert functionality by means of a flashing LED or a tone, for example.

(60) The device may be arranged to shut down sequentially as the batteries are depleted to extend the battery life and to allow the device to record measurements for longer. For example, components unrelated to measurement or calculation of parameters (such as the radio board 115) may be shut down to allow the device to record measurements for longer.

(61) Further details about the networking will now be described in relation to FIG. 11, which shows a schematic view of a network of devices 100 monitoring a noise/vibration source 200, indicating how data is transmitted in a network. As described above, an external server 150 may receive transmissions from the device 100, where the transmissions comprise results files. The server 150 may compile a database of results, which may then be accessed by users. Users may access results by logging on to a web portal 153 or by using a mobile application, for example. The results transmission may be either ‘pushed’ to the server 150 or ‘pulled’ from the device by the server 150, depending on the connection quality and the required time between refreshes of the data. The communication between the device 100 and the server 150 may take place using a variety of media, depending on availability. In the ideal case, a WLAN connection is used, but if this is not available a wired local area network connection could be used, or a GSM network. Where no such communication with the server 150 is possible, for example, in remote areas where wired connections are impractical for reasons of range, data saved on the device's memory card 123 may be uploaded to the server 150 at a later date, or downloaded via Bluetooth connection.

(62) The server 150 may comprise a cluster of servers and services known as a cloud server 151, as shown in FIG. 4. The purpose of the server 150 and/or server cloud 151 is to process the data from the one or more devices 100 (at a ‘back end’) and display the processed data to the user (at a ‘front end’), thereby compiling the results files from one or more devices 100 into useful reporting data for the user. It should be noted that the compilation of data from the devices 100 and the processing of said data to produce useful results may be provided separately, such as on separate servers.

(63) It is envisaged that many devices 100 according to aspects or embodiments described herein form a network with such a server 150 to gather data from a variety of different locations, as shown in FIG. 4, which can then be compiled and presented to the user by the server 150, enabling a large area to be effectively monitored. The envisaged low cost of the device 100, geolocation functionality (via GPS) and the results file assembly (incorporating a variety of user-set parameters relating to noise and vibration) make the device 100 well suited for use in such a system, allowing an integrated, ‘big data’ approach which is impractical and costly to achieve with current noise and vibration monitoring equipment. In such a system, the device 100 acts as a data logger and data processor which gathers and processes all data, generates alerts and records audio data when required, and sends data to the server 150. The server 150 acts to combine and present data to the user and allows the device 100 to be configured and calibrated.

(64) The web portal 153 serves as the ‘front end’ for the server 150, and may allow users to register, maintain and edit an account which the user may use to log in to the web portal 153. A user may be able to view the accounts of other users and set privileges for those accounts, which is useful for large projects where many users with different responsibilities are involved. The user may also create and modify projects, and assign or unassign devices to these projects.

(65) As mentioned above, the web portal 153 preferably allows the user to configure all devices in a network that are registered to the user's account, as long as their account privileges allow it. This configuration data may then be assembled into a configuration file and remotely uploaded to the device via any of the communication means between the server 150 and the device 100 mentioned above. The new configuration file can also be uploaded to the device directly from a computer or a mobile device via Bluetooth connection. In this case, the device 100 will upload said new configuration file to the server 150, in order to synchronise the configuration data. The user may also configure the operating times of the device and alerts, included the recipients of said alerts, as described above. This data may be incorporated into the configuration file.

(66) The web portal 153 is also used to report data from the server 150 to the user via a protocol. This reporting functionality preferably incorporates graphical presentations, including those that integrate the GPS readings from each device in a network with available mapping data, for example, such that each device is shown as a marker on the map. The user may be able to view selected parameters and look at gathered data over a specific time range, and compare it with historic data. The data may be provided in both graphical format (i.e. showing the time-history of noise and/or vibration) and tabulated format. All data may be downloadable and may be exported in a number of formats, such as .csv, .xls, and .pdf. It will be appreciated that the data collected by a network of devices may be presented to the user in a wide variety of useful ways.

(67) It will be further appreciated that the web portal 153 may also be provided as a mobile application or an installed program for a computer, for example. The mobile application may also allow a user device to configure and calibrate the device 100, read parameters calculated by the device, and download recorded audio files and/or results files via a USB or Bluetooth connection

(68) FIG. 12 shows a system architecture diagram of an exemplary network containing a device 100, illustrating how an exemplary network might transmit data from the device 100 to the user. The network preferably uses the MQTT protocol, where all outputs from the device 100 (such as the results files) are in the form of MQTT data packets and the server 150 comprises a MQTT broker. Data is received at the device via microphone 20 and accelerometer 30. The device 100 may also receive data from an external power supply and GPS satellites.

(69) Alerts and/or a status of the device may be indicated to the user on site via an LED and/or a signalisator, as described. The results files are transmitted to a service provider 161 as MQTT data packets via either GPRS or WiFi. Where WiFi is used, the data packets are transmitted via a router 163. The data packets are then transmitted to a MQTT broker 152 via a load balancer 154 and a Domain Name System (DNS) server 156. The MQTT broker 152 may comprise a parser. The MQTT broker 152 may then interface with message broker software 158 (such as Rabbit MQ) to serve the data to the web portal 153, where it may accessed by the user. The ‘raw’ data from the devices may optionally be saved into data store 165, and the web portal 153 may query data store 165 and access processing resources (not shown) to present useful processed results to the user. These processed results may be saved into further data store 167. The DNS server 156, load balancer 154, MQTT broker 152, and message broker software 158 are shown as part of the external server 150, but it will be appreciated that this is merely a schematic representation and different components may be provided on different servers. Similarly, components that are not shown as part of the external server 150 may be provided as part of the external server 150.

(70) In an alternative embodiment, the device 100 may have a reduced number of user-selectable parameters. For example, the device may not perform spectral analysis on input signals, and may instead calculate only overall noise and/or vibration levels.

(71) It will be understood that aspects and embodiments have been described above purely by way of example, and modifications of detail can be made within the scope of the invention.

(72) Each feature disclosed in the description, and (where appropriate) the claims and drawings may be provided independently or in any appropriate combination.

(73) Reference numerals appearing in the claims are by way of illustration only and shall have no limiting effect on the scope of the claims.