WO2013179202A2 - Environmental parameter determination device and method using acoustics - Google Patents

Abstract

Properties of the air such as temperature, pressure, humidity and air flow can modify the propagation pathway of an acoustic signal. By making acoustic measurements of a known acoustic path it is possible to estimate at least one of the temperature, the pressure, the humidity and the air flow using perturbations present in the measured signal. Moreover, by combining data from several acoustic pathways topographical reconstruction of profiles of at least one of the temperature, the pressure, the humidity and air flow can be obtained. Profiling these data over an extended volume such as an office or green house is particularly important for local control.

Description

ENVIRONMENTAL PARAMETER DETERMINATION DEVICE AND METHOD USING ACOUSTICS

FIELD OF THE INVENTION

The present invention relates to a system, apparatus, device, method and computer program product for heating, ventilation and air conditioning (HVAC) applications, especially to the field of environment energy management, where it is necessary to control the temperatures within specific tolerances and/or to provide a fire warning system.

BACKGROUND OF THE INVENTION

In modern commercial premises automatic control of environmental parameters such as temperature, humidity and air flow can provide huge benefits in terms of energy reduction and productivity. For example in horticulture improving the homogeneity of temperature profiles can dramatically increase the crop yield. To provide effective automatic control, sensing of environmental parameters must be included into the feedback loop of the control system. Most effective control can be achieved when environmental parameters are sensed on a local scale and local actuators are used to attain the optimum environmental parameters for a given function at a given time and location. The smaller the local environment which can be controlled, the more flexible and productive a given floor space can become. Thus, it is advantageous to have an extensive network of control sensors and actuators to maximise productivity.

Modern semiconductor temperature and humidity sensors can be constructed in one compact package, but measurements of air flow require different mechanisms. As a result separate sensors are required to test for these parameters where air flow, temperature and humidity data is important. Therefore, where measurements of temperature, humidity and air flow are required at least two sensors are deployed. These sensors measure the local conditions as spot measurements. If measurements at multiple locations are required, more sensors must be placed at each required location, or grid point. Although the local temperature, humidity and fluid flow at each location can be measured, the properties of the air at intermediate positions between these grid points is unknown. It is possible to interpolate the data measured at each grid point to estimate the parameters of the air at intermediate positions; however, it is possible that very large local perturbations in the environment are either missed by this interpolation, or strongly affect it. For example a temperature sensor installed next to a lighting element will read a much higher temperature when the lighting element is switched on due to local heating by the lamp, however, the temperature as a whole throughout the space may be only marginally affected. In order to minimise this risk, the grid spacing of the sensor layout must be made very small, requiring a large number of sensors, greatly increasing the cost and complexity of installation, commissioning and long term running costs. Further, to measure the parameters of the air at the central position of a room would require the installation of a series of sensors at that point. This can be very inconvenient, as it can impact upon the usage of the space.

It has been disclosed in US 6,681,635 that acoustic tomography can be used to determine specific parameters of the air using acoustic means. Here the Time of Flight (ToF) - the time taken for an acoustic signal to travel from an acoustic transmitter to an acoustic receiver - was used to estimate the speed of sound of the air. As the speed of sound in air varies with the temperature and humidity of the air, it is possible to infer these parameters from the speed estimates, and therefore from the ToF measurements. Further any perturbation of the air due to drafts or convection currents also introduces variations in the time of flight measurements. Thus if sufficient data is collected it is possible to estimate all these parameters via acoustical means.

As the speed of sound varies with the local conditions of the air as it travels from a transmitter to a receiver, the speed of sound represents the integral over all perturbations along the acoustic path. By combining data from different transmitter receiver pairs, the local perturbations can be calculated using computer tomography to reconstruct the temperature, humidity and air flow profiles in three dimensions. An advantage of this approach is that three dimensional profiles of the parameters of the air can be measured remotely with a two dimensional array of sensors located on a surface. No sensors need to be installed at awkward locations, as would be the case with conventional sensors, allowing the internal space to remain uncluttered. However, a disadvantage of a ToF based system is the timing accuracy required to resolve the small time differences caused by a slight change in the speed of sound as the parameters of the air are varied. This problem is exacerbated when a large number of receivers and transmitters are installed to profile the temperature in three dimensions (3D). To maximize the efficacy of the system, the time instants of reception and transmission at each node need to be synchronized. This requires very fast

communication between the nodes, as well as a dedicated network and protocol system making such systems very expensive when scaled to many nodes.

Another disadvantage of this approach is that ToF data yields only one data point for a given transmitter and receiver. Given that multiple parameters affect the speed of sound, one cannot use a single ToF measurement to isolate all of these parameters. In order to profile temperature, humidity and drafts many transmitter receiver pairs must be used together to generate sufficient data redundancy that numerical optimization schemes can estimate the individual parameters at any given point in the measurement space.

Thus, prior art systems concerning acoustic tomography offer a solution to conveniently monitor profiles of significant parameters of the air - such as temperature, humidity, air flow and aerostatic pressure - in three dimensions, but requires a significant installation base to be fully operational, which can be exceedingly costly. SUMMARY OF THE INVENTION

It is an object of the present invention to provide an improved environmental air parameter determination system and method with low installation and maintenance costs. Also, it is an object of the present invention to provide an environmental air parameter determination system which enables the determination of at least one environmental parameter at locations where no conventional sensors can be installed. Another object of the present invention is to provide an environmental parameter determination system and method with the ability of creating a profile map - through means of acoustic tomography - in order to control heating, ventilation and/or air conditioning, especially within a building, without the drawbacks of prior art acoustic tomography approaches. Further, an object of the present invention is to provide a system which enables the reduction of the required quantity of detectors for an environmental parameter. At least one of these objects is achieved by an apparatus as claimed in claim 1, an air parameter determination device as claimed in claim 3, a system as claimed in claim 7, and a method as claimed in claim 4.

Accordingly, several environmental air parameters can be determined by a single transmitter and a single receiving apparatus, wherein the environmental air parameter determined within a specific measurement area is characteristic of an average value of the parameters along the distance between a signal generator and a receiver, especially between a respective emitting point and a respective receiving point. Thereby, it is possible to install the signal generator or transmitter resp. emitting apparatus and the receiver resp. receiving apparatus at locations which are easily accessible, e.g. a side wall of a room, but to measure the several air parameter variations integrated over the acoustic path, which includes the parameter variations in the middle of the room. This is especially of interest in case the room is quite high, e.g. a room in the form of a gallery or studio with a ceiling height of 3, 4, 5 or even more meters, as the case may be with balcony- like mezzanine, so that e.g. the temperature or another environmental parameter in the middle of the room can be determined, e.g. in front of a window front, without the need of placing a temperature sensor in the middle of the room. Thus, the acoustic (ultrasonic) air parameter determination can provide a highly accurate measurement over great distances exceeding 10 m.

Moreover, temperature, humidity and flow sensors can be combined into one unit, greatly reducing the cost of ownership. This new sensor unit operates by transmitting an acoustic signal through the air to a receiver. The acoustic path taken by the transmitted signal is modified by changes in the parameters of the air. Thus, any modification to the transmitted signal can be used to extract information about the temperature, humidity and air flow between the transmitter and the sensor. As the perturbations in the signal are generated while the signal is on route, the system does not need to make point measurements, but measures the average 'state' of the air lying along acoustic path. As such, an 'average' measurement between the transmitter and receiver may be provided, which has recorded the influence of the air at each intermediate point. By combining the data from several such sensors, with preferably overlapping acoustic pathways, a calculation process - computer tomography - can be used to accurately profile the measureable quantities over space. The interpretation of such a topological profile can be more meaningful than simple interpolation techniques used with grid measurements and is based on real data points averaged over intermediate positions

Furthermore as each transmitter can transmit a signal which is received at multiple receivers, a large number of measurements can be made with relatively few sensor units. Thus, the proposed solution can achieve better or comparable resolution to a conventional spot measurement technique with a much lower hardware installation base.

By integrating signal generators, e.g. loudspeakers, and receivers, e.g.

microphones, into lighting devices, the positions and distances to each other of the loudspeakers and microphones can be determined very accurately via a process of acoustic localization. An acoustic localization infrastructure may be installed for the particular purpose of determining the distance between devices of a super-system, such as a lighting system, or building automation system, where the devices may be lighting elements, switches or control panels.. The distances between emitting and receiving apparatuses can be derived from an automatic commissioning process. Once the identification of all the lights, switches or sensors has been established (which is, somehow or other, generally done in context with the installation of lights and switches), accurate distances can be extracted, e.g., from a building plan. Thereby, the system can determine temperatures or other environmental parameters based on these identifications to calibrate the acoustic system. An important variable in acoustic air parameter determination is accurately calculating the distance between a respective transmitter and receiver. However, since the main components of an acoustic localization system, and an acoustic air parameter determination system are essentially the same, a single system can provide dual functionality.

The acoustic air parameter determination system according to embodiments of the present invention may be based upon measuring the effects certain parameters of the air have upon the acoustic propagation of a sound wave. More explicitly the proposed solution is based upon frequency domain analysis of a sound wave to determine attenuation of a test signal at each frequency of interest, after travelling along an acoustic path from an acoustic emitter to an acoustic receiver. The frequency dependent absorption of sound is deterministically dependent upon the temperature, humidity and absolute pressure of the air. Thereby, performing acoustic measurements along a given acoustic path between a transmitter and receiver and comparing the relative frequency response of the received signal to a mathematical model of the process, allows the average air temperature, pressure and humidity of the air lying along the acoustic path to be estimated in a single measurement. Known methods (as described e.g. in US 6,681,635) based on time of flight and speed estimation are unable to separate the influence of temperature and humidity on the relative change in the speed of sound. Thus, such ToF based methods are unable to estimate temperature and humidity independently in one measurement.

By measuring the influence of the air on the frequency response of the received signal between each transmitter-receiver pair, it is possible to create a detailed profile of air parameters over the whole measuring space (e.g. office space), even in three dimensions (3D), enabling advanced HVAC applications. As an example, the use of three different loudspeakers and three different microphones provided in different locations can provide a quantity of at least nine temperature and other parameter values, as each loudspeaker can emit at least one acoustic signal to each microphone. The device can provide acoustic signals only in such directions or to such receiving apparatus that the minimum distance is kept, in order to ensure good accuracy. Thus, in an office

environment, where there are many lights and switches, many measurements can be made over the whole office to create a temperature and other parameter's profile of the whole environment. In contrast thereto, with individual sensors, only punctiform, local temperature or parameter values in direct proximity to a respective sensor can be determined. Using a parameter profile derived from frequency response alteration of an acoustic signal, more intelligence can be used to facilitate more advanced air conditioning, such that local variations in temperature or other environmental parameters can be evened out. Thereby, it is possible to deliberately vary the temperature or other parameter's profile in different areas of an environment. The temperature or other parameter's profile could also be used for other applications, for instance to detect the build-up of heat in an office to provide early warning of a fire, i.e. to provide a fire alarm. Thereby, the temperature or other parameter's profile resp. temperature or other parameter determination system is sensitive to such an extent it that can give an early warning of the build up of heat in an area of a building, enabling a warning to be sent before a fire has started.

Further, integrating one or several reflectors to the arrangement can provide for further measurement points. In particular, it can be shown that the resolution of the temperature and humidity estimation improves as the length of the acoustic path increases. Therefore the parameters of the air can be calculated more accurately over distances exceeding 10 m, rather than over 1 to 2 m. Thus, e.g. based on a building or lighting plan, it can be chosen which devices the measurements should be derived from in order to provide good accuracy. Thereby, a detailed parameter profile enables advanced HVAC control, which makes it possible to realize e.g. a varying temperature profile, especially over large offices. For example, HVAC outlets can be controlled such that a correction of draughts is provided, wherein heat can be moved, e.g. from a hot boiler room area to a cold reception area.

The environmental parameter determination system can be applied to any environment where speakers and microphones are employed, and it is not confined to the ceiling, as switches and sensors and/or lights can be placed anywhere in the environment as can speakers and microphones. Typical indoor environments for which an application of the temperature or other parameter determination system is possible are e.g. offices, theatres, cinemas, exhibition halls, stations, airport terminals, shopping malls, warehouses and greenhouses for horticulture applications etc. Also, early detection of a fire can be ensured, by detecting a change in heat, or a hotspot in the temperature profile. This is increasingly important in context with a high amount of unsupervised electrical and electronic equipment in buildings.

Although the main embodiment is for air, the technique can also be applied to underwater environments, like swimming pools and fluid containers in factories, especially based on the respective composition resp. the sound characteristics of the liquid. The transmission of sound through air varies with temperature, pressure, humidity, concentration of C0₂ and other gases and air flow. If a test signal is transmitted from a sender to a receiver, the received sound will be modified depending upon the properties of the air through which it has travelled and the frequency content of the transmitted signal. By recording the received signal and comparing it to the signal sent by the transmitter, the effects of the air on the transmission can be calculated and related to the various parameters such as temperature, humidity, air-flow and gas composition.

The distance dependent attenuation of air follows a distinct relationship at a given frequency, humidity, temperature and pressure. Based upon comparison of a measured test signal transmitted from an emitter to a receiver, to that of a model of the acoustic path, it is possible to estimate the quantities of distance, humidity, temperature and pressure. If, however, the distance is known and assumed fixed, the number of unknown variables can be reduced resulting in simplified calculation of the temperature humidity and pressure from the measured relative signal attenuation at a set number of distinct frequencies. Ideally, a frequency sweep can be used to probe a wide frequency range and to provide more data points for the air parameter calculations. It is, however, possible to make measurements as several discrete frequencies, and use this information to derive the necessary data for air parameter determination.

Air flow also affects the acoustic pathway between a transmitter and a receiver. Therefore the influence of air flow will perturb properties of the sound transmission and represent measureable quantities which can be used to estimate the air flow.

According to a first aspect, an acoustical parameter determination device can be provided, wherein the emitting apparatus itself or the receiving apparatus itself can comprise a parameter detector arranged to receive values from a frequency response analysis of the received sound in order to determine environmental parameters within the measurement area independently from any parameter detector of the emitting apparatus or the receiving apparatus, and arranged to provide parameter data with respect to specific points within the measurement area to a processor device for creating a parameter profile map of environmental parameters, respectively, determined within the measurement area. Thereby, basically, average parameter values are determined, the average relating to the environmental parameters along the measuring section, but also other parameter values at specific points can be estimated. In particular, a grid of measuring sections can be established and visualized, each measuring section corresponding to a kind of acoustic beam, especially beams crossing each other. Thereby, an estimation of parameter values at the vertices of such a grid can be made, i.e., it is not required to provide a sensor or a sensor node at a respective vertex. In other words, although an average value is provided by each measuring section, spot values can be determined with respect to specific points. The spot values are to be determined in such a way that they provide, as a result, the respective average value. The system may not need crossing paths, but paths that lie close to each other is enough. The actual process for reconstruction of the profile may be based on computer tomography (CT), similar techniques as used in making 3D x-ray images with CT scanners.

Further, the emitting apparatus can comprise a second signal generator and/or the receiving apparatus can comprise a second receiver arranged in a second distance to each other, so that between respective first and second generator resp. receiver, acoustic signals can be provided with another frequency or at a different instants of time, in order to provide further temperature or other parameter data based on merely two apparatus, e.g. in case the apparatus are provided in illuminating devices with large dimensions. Thereby, the first and second signal generator can also be arranged to provide acoustic signals in specific directions, especially within a narrow sector, so that a specific receiver can be addressed. In particular, the specific points (pi, p₂; pr, p₂') within the measurement area respectively correspond to the centre points of the distances between a respective signal generator and receiver. Further, the device can be arranged to provide acoustic signals in different planes and directions, so that an environmental parameter's profile map can be created in 3D.

Also, the device can further comprise a reflector in order to provide further parameter values with respect to locations different than the temperatures or other environmental parameters determined using the direct acoustic path from the signal generator to the receiver.

A respective measurement point can be arranged along the distance between a respective emission point and a respective reception point, and its exact location can be chosen e.g. based on heuristics, in order to allocate the determined parameter to an appropriate point along the distance such that the determined average parameter value is also representative of the specific parameter in proximity to the measurement point. For example, in a room which is provided with glass panels only at one side, in summer, when heating is provided by solar radiation, on a measurement section (at least approximately) orthogonal to the glass panels, the measurement point can be provided closer to the glass panel side than to the opposite side. The amount of dislocation of the measurement point with respect to the half of the absolute distance between emission point and reception point might depend on the effective temperature or other parameter gradient. Thereby, the effective temperature or other parameter gradient might be determined e.g. by providing two further measurement sections, one in proximity to the glass panels (at least

approximately) parallel to the glass panel side, and the other in proximity to the opposite side (at least approximately) parallel to the glass panel side.

Further, the method can be carried out such that a first acoustic signal and/or a second acoustic signal emitted to at least two receivers in order to determine at least two parameter values (T_ls T₂, T₃), especially with respect to at least two measurement points (pi; p_2; p₃), and wherein the steps for determining a temperature or other

environmental parameter based on the first acoustic signal resp. the steps for determining a temperature or other environmental parameter based on the second acoustic signal are repeated in order to determine an averaged temperature or other parameter value, especially with respect to time, based on temperatures or other environmental parameters determined based on the first acoustic signal resp. based on temperatures or other environmental parameters determined based on the second acoustic signal. Thereby, the locations of a signal generator and a receiver and the respective distances between a respective signal generator and receiver can be derived from a lighting or building plan. Further, an entry of the determined temperatures or other environmental parameters can be provided in a temperature or other parameter's profile map. In other words, basically, for each measurement of the relative frequency dependent absorption of sound, an average temperature or other parameter of the air between two points can be derived. During a first measurement, parameter values for several measurement sections can be determined, e.g. three sections between a first emitting point (ei) and respective receiving points (r_ls r₂, r₃), providing first average parameter values (T_{l s} T₂, T₃). During a subsequent measurement, e.g. based on a second acoustic signal, again environmental parameter values for several measurement sections can be determined, corresponding to the previous measurement sections or differing from these previous measurement sections, e.g. three sections between a second emitting point (e₂) and respective receiving points (r_{l s} r₂, r₃), providing second average environmental parameter values (TV, Τ_2', Τ_3').

According to a second aspect which can be combined with the above first aspect, any system for acoustical determination of environmental parameters in heating, ventilation and air conditioning (HVAC) applications can be provided, wherein a controller device arranged to communicate with an HVAC system for controlling the HVAC system in order to provide cooling and/or heating based on the temperature or other parameter's profile map is provided in order to even out local variations in temperature or another environmental parameter, especially arranged to provide cooling and/or heating to specific regions of the measurement area which is covered by the system for acoustical

determination of temperatures or other environmental parameters.

Thereby, the system can also be provided with a thermostat or thermistor temperature sensor providing a single reading at a specific measuring spot. The sensor can serve for comparing its reading with a temperature value of the acoustic profile determined by the acoustic temperature determination technique of the emitting resp. receiving apparatus. Thereby, the local temperature sensor can enable a calibration of the acoustic temperature determination system, wherein depending on the resolution, preciseness resp. accuracy of the sensor, any effects which may rely on varying C02 level or barometric (air pressure) can be accounted for. Thus, the quality of air could be determined in conjunction with temperature determination. In particular, such a thermostat or thermistor temperature sensor can be incorporated in retrofit situations where this sensor already exists.

Further, the system can be provided with a second signal generator for acoustic emission for generating a second acoustic signal. Several signal generators in conjunction with several receivers can provide for the determination of a multitude of different parameter values, so that a temperature or other parameter's profile map with a high resolution can be generated. Further, the system can make use of an available microphone and loudspeaker in an acoustic based automatic commissioning system. In other words, the system is arranged to communicate with an HVAC system in order to direct cooling or heating in specific regions of the measurement area, to provide a consistent temperature or other environmental parameter over the measurement area or to provide specific temperatures or other environmental parameters in different regions of the measurement area.

Of course other measuring options can be used. For example, the arrangement of a sound analysis controller or parameter detector can be chosen such that for a specific environment, installation costs and/or required amount of inter- communication in order to create a temperature or other parameter's profile map can be minimized.

The above apparatus may be implemented as a hardware circuit integrated on a single chip or chip set, or wired on a circuit board. As an alternative, at least parts of the apparatus may be implemented as a software program or routine controlling a processor or computer device.

It shall be understood that the apparatus of claim 1, the device of claim 3, the method of claim 4 and the system of claim 7 may have similar and/or identical preferred embodiments, in particular, as defined in the dependent claims.

It shall be understood that a preferred embodiment of the invention can also be any combination of the dependent claims with the respective independent claim.

These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

In the following drawings:

Fig. 1 shows a schematic drawing of an arrangement for acoustic

measurements;

Fig. 2 shows a schematic drawing of an acoustical air parameter

determination device according to a first embodiment, especially provided with one acoustical emitting apparatus and a multitude of acoustical receiving apparatus arranged within lighting devices;

Fig. 3 shows a more detailed schematic drawing of an acoustical air

parameter determination device and system according to a second embodiment, especially provided with one acoustical emitting apparatus and two acoustical receiving apparatus, wherein two temperature values are determined;

Fig. 4 shows a detailed schematic drawing of an acoustical air parameter determination device and system according to a third embodiment, especially provided with one acoustical emitting apparatus and three acoustical receiving apparatus, wherein three temperature values are determined;

Fig. 5 shows a detailed schematic drawing of an acoustical air parameter determination device and system according to a fourth embodiment, especially provided with two acoustical emitting apparatus and three acoustical receiving apparatus, wherein six temperature values are determined;

Fig. 6 shows a two dimensional temperature profile which can be provided by an acoustical parameter determination device according to the embodiments shown in Fig. 3, 4 and 5;

Fig. 7 shows a schematic block diagram of an acoustical parameter

determination device and an acoustical parameter determination system according to the invention;

Fig. 8 shows a schematic diagram illustrating the effect of air flow on an acoustic pathway; Fig. 9 shows a basic system configuration of according to a fifth

embodiment;

Fig. 10 shows a schematic block diagram of an estimation scheme according to a sixth embodiment;

Fig. 11 shows a schematic diagram illustrating how air flow direction and magnitude can be estimated using a directional receiver;

Fig. 12 shows a schematic diagram illustrating how profiles of air

parameters can be created;

Fig. 13 shows a schematic diagram of a dual system according to a seventh embodiment;

Fig. 14 shows a schematic diagram of an arrangement according to an

eighth embodiment for measuring concentration of gases; and

Fig. 15 shows a schematic diagram of a transceiver based sensor according to a ninth embodiment.

DETAILED DESCRIPTION OF EMBODIMENTS

In the following embodiments, an enhanced environmental air parameter determination system, especially for HVAC applications, is proposed where at least two temperature values can be generated in order to create an environmental parameter profile map.

According to the embodiments, the average temperature, humidity and relative magnitude of the air flow with respect to a distance between respective emitting apparatus and receiving apparatus can be determined. Hence, over great distances, these parameters can be controlled without the need of placing sensors exactly at the locations which are to be controlled. The further implementation will mainly depend on which technology is used for inter-communication and for providing switches, lights or the like in a building.

In the following, ten embodiments using acoustic parameter determination are described.

Fig. 1 shows a schematic drawing of an arrangement for acoustic measurements, wherein an acoustical emitter 10A is provided in a specific distance d to an acoustical receiver 20A, so that the modification of the sound signal transmitted by emitter 10A can be recorded at receiver 20A to deduce the frequency dependent absorption of the air in comparison to a mathematical model of the sound transmission. In other words, it is principally shown how according to the invention, it is possible to make use of

microphones and loudspeakers for determination of certain parameters of the air, e.g. in an acoustic based automatic commissioning system, especially based on the parameter d and the relative frequency content of the transmitted and received acoustic signals. Thereby, it has been found that this state of the art arrangement can be used for determination of parameters of the air, such as temperature, in context with HVAC applications also.

Fig. 2 shows a schematic drawing of an acoustical air parameter determination device according to a first embodiment, especially provided with one acoustical emitting apparatus 10 and a multitude of acoustical receiving apparatus 20 arranged within lighting devices 20a, wherein the arrangement of the lighting devices 20a is shown in two dimensions, but an arrangement in three dimensions is likewise possible. Thereby, each receiving apparatus 20 can be conceived as lighting devices 20a with embedded microphones. A speaker 11 provided in the emitting apparatus 10 emits an acoustic signal which is received by each of the microphones 21 provided in the receiving apparatus 20 and a comparison can be made to determine the relative acoustic absorption of the air as a function of frequency and there from determine the parameters of the air, especially by a parameter detector 45 in communication with the emitting apparatus 10 and receiving apparatus 20. With this technology, an air parameter determination made from the relative frequency spectrum of the received signal after transmission over a known distance is possible (wherein exemplarily, two distances di, d₂ are indicated). The determined parameters are characteristic of the average of those parameters in the medium, e.g. air or water, between the two devices. Since the locations of the devices 10, 20 can be defined, e.g. based on a lighting plan, an entry in a parameter profile - such as a temperature profile - can be provided, especially with respect to the area between these devices. Thus, parameter determinations can be made between several different points, in the embodiment shown especially based on six parameter values.

In one specific embodiment, in an office environment, any device (e.g. PIR presence detector) is fitted with a loudspeaker resp. signal generator 11 and sited at the centre of a tile mounted on the office ceiling. The nearest lighting device is fitted with a microphone 21 at a defined distance from the sensor, wherein the distance can be derived from e.g. the ceiling plan. Since the sound emitted from the signal generator 11 can be received by any nearby microphone 21 and microphones 21 can be placed in every lighting device 20a, one sound can be used to create a series of parameter determinations extending over a wide area of the office, not necessarily restricted to the ceiling. Further

measurements from other sensors can be used to calculate additional points, e.g. repeating measurements over certain areas to obtain a high accuracy, until the whole environment can be profiled. Measurements made from ceiling to floor or desk level can also be made, especially in order to create a 3D plot of the temperature profile. That is to say, with an acoustic signal generator 11 and at least two acoustic receivers 21, a measurement plane can be spanned, and with an acoustic signal generator 11 and at least three acoustic receivers 21, a measurement space can be spanned, likewise with at least two acoustic signal generators 11 and at least two acoustic receivers 21. This enables multiple measuring points with relatively few sensors 11, 21.

Fig. 3 shows a more detailed schematic drawing of an acoustical parameter determination device and system according to a second embodiment, especially provided with one acoustical emitting apparatus 10 and two acoustical receiving apparatus 20, wherein multiple parameter values are determined using two acoustic paths with reference to the reading points pi and p₂, especially provided within an office environment 60 indicated by the frame shown in Fig. 3. Both emitting apparatus 10 and receiving apparatus 20 are provided within standard switches or sensors of a building, e.g. within lighting devices 20a, respectively. Apart from the type of emitting apparatus 10 and receiving apparatus 20 and the parameter detector 45 shown in Fig. 2 and a processor device 46. 45.

With this technology, several parameter values may be determined at two locations based on a first and second comparison of the relative frequency spectrum of the received sound, wherein the points pi and p₂ are provided at least approximately in the middle of the respective distance di and d₂. The frequency dependant absorption of the air may be measured between a first emission point ei corresponding to a point from which acoustic signals are emitted from a signal generator 11 , and first and second receiving points ri, r₂, at which acoustic signals are received by a respective receiver 21. Fig. 4 shows a detailed schematic drawing of an acoustical temperature determination device and system according to a third embodiment, especially provided with one acoustical emitting apparatus 10 and three acoustical receiving apparatus 20, wherein parameter values can be established at three locations with respect to the measurement point pi, p₂, p₃, especially in two different planes, providing a 3D temperature map. The system is provided within an indoor environment 60 indicated by the frame shown in Fig. 4, e.g. an office. Apart from the type of devices shown in Fig. 3, the system further comprises a third receiver 21 with a third receiving point r₃, and a reflector 25. With this technology, parameters of the air can be determined based on three different values for the frequency dependant acoustic absorption of the air, wherein the points pi, p₂ and p₃ are provided at least approximately in the middle of the respective distance di and d₂ Further, by making use of the reflector 25, parameters of the air can be determined at three other positions by determining the frequency dependant absorption of the sound along the reflected paths. All these temperature values can be determined by means of a first acoustic signal, and by selecting an appropriate position of the reflector 25, the three other temperatures can be determined with respect to measuring points which are substantially different than the points pi, p₂ and p₃. Providing a second or third acoustic signal can provide further six temperature values respectively, so that by timely spacing of the signals, not only the temperature profile can be determined but also a temperature profile can be correlated to time, providing temperatures as a function of time resp. temperature gradation. The emitting apparatus 10 can emit the acoustic signals in specific directions, so that e.g. a first acoustic signal is specifically directed to the receiving apparatus 20 and a second acoustic signal is specifically directed to the reflector. Alternatively or in addition, the emitting apparatus 10 can emit the acoustic signals all around, i.e. in a sector encompassing 360°C.

Fig. 5 shows a detailed schematic drawing of an acoustical air parameter determination device and system according to a fourth embodiment, especially provided with two acoustical emitting apparatus 10 and three acoustical receiving apparatus 20, wherein air temperature parameters may be determined at six positions, which can be allocated, e.g., to the measurement point pi, p₂, p₃, and pr, ρ₂', p₃', especially in different planes, providing a 3D temperature map. The system is provided within an indoor environment 60 indicated by the frame shown in Fig. 4, e.g. an office. Apart from the type of devices shown in Fig. 3, the system further comprises a third receiver 21 with a third receiving point r₂, and a second emitting apparatus 10 with a second emitting point e₂. One of the emitting apparatus 10 emits a first acoustic signal in order to determine first parameter values, especially with respect to the measurement points pi, p₂, p₃ serving as a basis for creating the temperature profile map, and after this measurement, the other of the emitting apparatus 10 emit a second acoustic signal in order to determine second parameter values, especially with respect to the measurement point pr, p₂',

In other words, basically, for each acoustic measurement, an average parameter value - be that temperature, humidity or air flow - between two points can be derived. During a first measurement, parameter values for several measurement sections can be determined, e.g. three sections between emitting point ei and respective receiving points r_ls r₂, r₃, providing average parameter values ΤΊ, T₂, T₃. During a subsequent measurement, e.g. based on a second acoustic signal, again parameter values for several measurement sections can be determined, corresponding to the previous measurement sections or differing from these previous measurement sections, e.g. three sections between emitting point e₂ and respective receiving points r_{l s} r₂, r₃, providing average parameter values TV, Τ_2', Τ_3'. The processor device 46 can be in communication with the respective emitting apparatus 10 in order to prevent interference between successive signals by multiplexing the emission of respective signals in time. In addition to the system elements shown in Fig. 4, the system of Fig. 5 is further provided with a controller device 47 which is in communication with the processor device 46 in order to control a HVAC system, especially based on temperature profile map data provided by the processor device 46. Likewise as in context with the embodiments shown in Fig. 2, 3 and 4, any connections or communication paths may be wired or wireless, and wireless communication can be effected in local area networks (LAN) and/or wide area networks (WAN) and can be based on bluetooth, infrared, radio communication or other standardized and well known communication techniques.

Fig. 6 shows a two dimensional temperature profile which can be provided by an acoustical temperature determination device according to the embodiments shown in Fig. 3, 4 and 5. Of course other parameter profiles can also be depicted in this way, for example a profile of the relative magnitude of air flow or humidity. Depending on the arrangement of emitting apparatus and receiving apparatus and reflectors, temperature data provided by a temperature detector to the processor can be used for generating a three dimensional temperature profile. Thereby, a determined temperature profile serves as a basis for creating a temperature profile map 50, which can be used for displaying the temperature distribution to a service engineer or for indicating a danger point resp.

dangerous spot in view of fire. The resulting temperature profile map 50 can be used to identify hotspots 51 and cool spots where a HVAC system arranged for processing the temperature data representing a specific temperature profile can direct the appropriate cooling to specific areas (e.g. hotspots 51) in order to provide a more consistent temperature over the whole environment. Alternatively, the resulting temperature profile may enable a more targeted HVAC system where different temperatures can be supported by the HVAC system over the width and length of the environment meeting the individual needs of the occupants or those of the room.

Fig. 7 shows a schematic block diagram of an acoustical parameter determination system according to fifth embodiment. Two emitting apparatus 10 are respectively provided with a signal generator 11, a parameter detector 45, and two receiving apparatus 20 are respectively provided with a receiver 21 and a parameter detector 45, wherein the redundancy of the parameter detectors 45 is not necessarily required, i.e., parameter detectors 45 can also be provided independently of emitting apparatus 10 or receiving apparatus 20 in communication with the emitting apparatus 10 and the receiving apparatus 20. Emitting apparatus 10 and receiving apparatus 20 are in communication with the parameter detector 45 which is also in communication with a processor device 46 via a second connection 42, in order to generate a temperature profile map, and the processor device 46 is in communication with a controller device 47 via a third connection 43, in order to control heating and/or ventilation, and the controller device 47 is in communication with a HVAC system 48 via a fourth connection 44, in order to actively implement control instructions in e.g. deviation of air, opening of windows, or energy management in general.

In the following embodiments, a transmitter receiver combination is described, which monitors temperature, humidity, pressure and flow by analysing the modifications to an acoustic signal occurring as a consequence of its propagation through the air. By combining data sets from multiple receivers or sensors acoustic tomography can be employed to reconstruct all of these features for use in air as a feedback and control system for HVAC applications and monitoring of environmental parameters in

greenhouses and industrial premises.

The atmospheric attenuation of sound is also dependent on the frequency, absolute pressure, temperature and humidity. Furthermore, the speed of sound is dependent on several factors including the gas composition, the humidity and the temperature of the air. By performing acoustic measurements of the relative attenuation of sound propagation, the major parameters affecting the acoustic transmission of sound can be estimated. It is also possible to include speed of sound information to enhance the robustness and accuracy of the parameter estimates, although it is not strictly necessary. Speed of sound data can be determined using time of flight methods, or using methods which determine relative changes in the group delay of a signal, and relating this to changes in the speed of sound. In (2), the atmospheric attenuation of sound is given by the formula

where and f is the frequency, Po is the reference pressure (1 arm), P is the actual pressure under measurement conditions, To is the reference temperature (293.15 K), T is the actual temperature in Kelvin under measurement conditions, F_ro is the frequency of maximum absorption for the vibrational relaxation for atmospheric Oxygen and F_r is the frequency of maximum absorption for the vibrational relaxation for atmospheric Nitrogen. Formula for calculating F_ro and F_r are given in Bass, H. E., et al. : "Atompheric absorption of sound: Further developments", 1994, Acoust. Soc. Am., pp. 680-683.

Thus, to estimate the parameters of the air such as temperature, humidity and pressure, measurements of the frequency dependent absorption of the sound as it propagates through the air can be matched to a mathematical model based on the presented formula. A numerical optimisation scheme can be employed to determine the temperature, pressure and humidity values which match closest to the measured data. The distance between the transmitter and the reciever is an important parameter, which can also be estimated from the recorded data using the numerical optimisation, or can be measured in a manual process and input to the model as a fixed variable. To make absolute statements of the parameter values it is important that the frequency response of the transducer is also known in advance, since the signal transmitted by the transducer is affected by the transducer's own frequency response. It is, however, possible to base any temperature, humidity or pressure estimates on relative changes in the frequency spectrum. In this situation either external calibration may be used to arrive at absolute parameter values, or many data points must be collected for sufficient data redundency to make estimates of the absolute values of the parameters of the air. This process is particularly suited to individual sensors, or a low number of sensors deployed within the same space.

However, if many sensors are used, rather than attempt a piecemeal to determine the individual parameters measured by each individual acoustic transmitter and reciever pair, the data from all sensors can be combined into a single set and an acoustic computer tomography may be used to generate a 3D profile of the parameters of the air directly. Tomography makes use of a technique called filtered back projection. Essentialy a mathematical model attempts to build parameter profiles of the temperature, humidity pressure and airflow, which best match the perturbations seen in the measured data.

Typically this method may be based on the Radon transform. Further details on such tomopgraphy can be gathered from e.g. W. Munk, P. Worchester and C. Wunsch "Ocean Acoustic Tomography" Cambridge University Press, 1995, ISBN 978-0-521-11536-0, P. Toft "The Radon Transform: Theory and implementation" Ph.D Thesis 1996 (Technical discussion of the Radon transform and its applications to solving computer tomography problems, in particular related to x-ray imaging), and A. Abubakar "Three-Dimensional Nonlinear Inversion of Electrical Conductivity", PhD Thesis 2000. Thus, in isolation one process can be used to determine the average temperature, humidity and pressure along an air column lying between an acoustic transmitter and reciever. If many sensors are available, with crossing acoustic paths, or paths which lie close to each other, computer tomogrpahy is a preferred method of extracting a 3D prameter profile.

To make continuous measurements in an environment without causing discomfort to occupants of the space, the transmission should be restricted to the frequency range above human audibility. This also incurs the advantages that the effects of attenuation are particularly strong, as the atmospheric absorption scales approximately with the square of the frequency f². The relative attenuation is a function of frequency for different relative humidity levels (0-100%) with a clear trend towards higher attenuation as the humidity increases, but only at higher frequencies. At lower frequencies a fold over occurs. This means that it is not possible to rely on single frequency measurements due to ambiguities introduced by the fold over. Instead a spectrum of measurements will be required. To get the best resolution and humidity discrimination, frequency measurements would be made over the entire range from 20 kHz to 100 kHz. However it is clear that a fore-shortened frequency range may be used to estimate the temperature pressure and humidity from the relative frequency spectrum of a measured signal. It may be that this frequency range is provided by a single transducer, or that multiple transducers will be used to cover as much of the frequency spectrum as possible, or key frequencies to maximize the resolution yet minimizing the hardware cost. Sound is attenuated by a 1/r relationship due to spreading of the wave energy as it propagates outwards from its source. The 1/r attenuation relationship has a much greater effect upon the overall measured attenuation than either the humidity or the temperature, however, the effects of this 1/r relationship are frequency independent. This means that this single attenuation parameter does not influence the frequency dependent attenuation of the air, and thus the frequency dependent and frequency independent effects can be separated. It is important to account for the 1/r attenuation otherwise the results of the numerical optimization and curve fitting can be modified significantly. The most simple method of dealing with measurement uncertainties is to include the distance between the transmitter and the microphone in the numerical optimization, although care is needed to ensure that the optimization parameters do not overly favour one optimization parameter (degree of freedom) over all others. As the distance between the transmitter and the receiver increases, the relative attenuation as higher frequencies increases, improving the signal to noise ratio and allowing a more accurate fit. Thus if the distance between the acoustic emitter and acoustic receiver is sufficiently large and if the frequency response of the transducer is accurately calibrated, only one measurement is required to accurately establish an estimate of the temperature and humidity of the air column lying between the transmitter and receiver. By making multiple measurements between the same transmitter receiver pair we can estimate the amount of noise in the signal, and use this to determine the relative strength of any drafts.

It is also possible to estimate the parameters of the air without accurately calibrating the frequency response of the acoustic emitter and acoustic receiver. Multiple measurements are required over an extended period of time. Each measurement can be compared with the previous measurements. Any change in the parameters of the air will cause a change in the relative frequency response between respective measurements.

Numerical optimization procedures can be implemented to estimate the system parameters, including the distance from the emitter and receiver, air temperature, humidity and pressure. Since these quantities vary in a deterministic way, if sufficient data is collected, the relative changes can be fitted to a model to provide the best estimates of the absolute parameters for the temperature, humidity and air pressure.

Another measureable quantity is the time of flight (ToF) of the acoustic path, which can be used to calculate the speed of sound and which can be combined with the above frequency domain approach for improved accuracy. Absolute values of the ToF can be difficult to obtain directly, complicated by the phase response of the transducers, typically limited bandwidth, and the necessity for synchronization of the receiving and transmitting clocks.

On the other-hand, determining the relative timing difference between multiple measurements is simple and can be highly accurate. Using relative Time of Arrival (ToA) calculations complicated absolute determination of the speed of sound, therefore it is easier to monitor relative changes in the parameters of the air, and as a result, a calibration step could be beneficial to give absolute parameter estimations. However, by combining the results of multiple measurements over time an external calibration system may no longer be required, as shown below.

A relative timing system can be created by transmitting a tonal signal from the transmitter, for example at 40 kHz, and measuring the phase of the 40 kHz tone at a microphone. If the speed of sound changes then the wave will travel faster or slower, and the phase of the signal at the microphone will change accordingly. This simple illustrative example shows how a change in relative phase can allow calculation the relative change in temperature. Given a temperature change AT the relative change in the speed of sound is approximately given by:

AC = C₂ - C₁ = 0.606(Γ₂ - Τ ) = 0.606ΔΓ

Given that the frequency of the test tone and distance to the microphone remain the same the change in phase is due to a modification of the wave vector:

Δφ = x(k₂ - k ) = xAk

which is related to the relative change in the speed of sound via: c₂ c_l

Given that Ci = c₀ the equilibrium speed of sound at the starting temperature an expression for the temperature change as a function of the measured phase change can be derived.

A key advantage of the relative ToA approach is that highly accurate clock

synchronization between devices is not necessary. This can greatly simplify the data transmission and underlying communication network supporting the distributed acoustic parameter determination system, when compared to a system based on absolute ToF. It is possible that clock drift will result in a systematic drift in the ToA, or phase angle at a particular frequency. Since drifting clocks are likely to drift at the same rate, a systematic drift in time can easily be corrected for in processing step.

Of course the speed of sound does not just vary with temperature. Cramer, O.: "The variation of the specific heat ration and the speed of sound in air with temperature, pressure, humidity, and C02 concentration", Acoustical Society of America, 1993, Vol. 95 gives the speed of sound as:

₂ RT„ 2pB^

c; = y— (l + ^-)

⁰ M RT

where γ is the ratio of the specific heats of the air (a real gas) which are given by:

C¹— C°———

p ^{~ p} M RT dT 2 >

and C' = C' -— 1+ ^(7^)

M \ RT dT

where R denotes the universal gas constant, T denotes the temperature, M denotes the molecular weight, p denotes the atmospheric pressure and B dentoes the second virial coefficient of state. Methods for calculating B are provided in Cramer, O.: "The variation of the specific heat ration and the speed of sound in air with temperature, pressure, humidity, and C02 concentration", Acoustical Society of America, 1993, Vol. 95, along with references to the necessary data sources.

The main consequences of the above formula is that the speed of sound is essentially independent of the absolute pressure over the frequency and pressure ranges of interest. The C0₂ concentration and humidity both influence the speed of sound, however, their relative effects are much smaller than the influence of the temperature. Given the very small influence on the speed of sound due to C0₂ it is unlikely that it will be possible to resolve the concentration of this gas under normal conditions (although not impossible). The effects of humidity though subtle can influence any speed estimate. By combining the temperature and humidity estimation using frequency domain analysis with the formula above one obtains a more accurate and robust method of estimating the parameters of the air.

Although numerical methods can be used to estimate the absolute parameters of the air, a calibration phase of the acoustic tomography system can be beneficial for ensuring accuracy, and preventing potential issues such as numerical instabilities or convergence to local minimums which are well known problems in numerical optimization schemes. Thus where accuracy and reliability is paramount it can be important to perform a calibration step. It is possible to perform calibration using 'gold standard' external calibration devices to provide a reference level.

On the other hand, as already indicated earlier, because the parameters of the air vary in a deterministic way, it is also possible to perform an absolute calibration by combining data recorded over a long duration and from multiple sensors. For example if the temperature rises and the relative humidity increases the relative frequency content at the microphone will be modified, as will the time or arrival. Taken in isolation these changes are only relative to an unknown base line. However since the starting conditions affect the magnitude of the relative changes, if sufficient data is available a numerical optimization scheme can be used to match all relative changes back to the initial starting conditions in an absolute sense. This calibration procedure would improve over time as more data became available and would also benefit from the combination of data from multiple transmitter receiver pairs.

Fig. 8 shows a schematic diagram illustrating the effect of air flow on an acoustic pathway between an ultrasound transducer 70 (acoustical emitting apparatus) and a microphone as acoustical receiver 21. In the left part of Fig. 8, a straight arrow indicates the direct acoustic path with no cross flow of air, while in the right part of Fig. 8 a cross flow CF of air influences the acoustic path so that it is bent and thus becomes longer. Therefore, the influence of the air flow will perturb properties of (ultra)sound transmission and can be used to estimate the air flow.

Fig. 9 shows a schematic diagram of a basic configuration according to a sixth embodiment which comprises at least one acoustic transducer (ultrasonic transducer) 70, at least one receiver (e.g. microphone) 21, provided at a known distance d from the transducer 70, and a processor 46. A signal is transmitted from the transducer 70 and arrives at the receiver 21. The microphone signal is then analysed at the processor 46 using signal processing techniques and compared to the known transmitted signal. The analysis may determine the relative frequency content of the received signal compared to the known transmitted signal, and the time taken for the signal to travel the distance d between the transmitting apparatus (transducer 70) and the receiver 21. The system may also analyse the short term variation in the signal, and the longer term trends. Inputting this data into mathematical models it is possible to estimate the average pressure, temperature and humidity of the air column between the transducer 70 and the receiver 20 and to estimate the magnitude of the air flow.

Humidity and pressure and temperature detection can be achieved by having the microphone signal processed using digital signal processing such that the data provides the instantaneous phase delay at the receiver 21 for each frequency of interest, and the instantaneous attenuation at each frequency of interest. These variables are then passed to the processer 46 for interpretation.

Fig. 10 shows a schematic block diagram of an estimation scheme according to a seventh embodiment where instantaneous data Μ[φ_ί, A_f] calculated at a processor 46 is filtered according to the parameters to be estimated. The instantaneous data Μ[ Ε¾ A_f] includes instantaneous phase data <i>_f and amplitude data A_f at each frequency f of interest. The instantaneous values contain a lot of rapid variations due to system noise, external noise and factors such as air flow near the device. To remove the influence of this noise a low-pass filter 84 can be applied to each data stream for each frequency of interest. This process provides a longer term average which allows stable estimates of the environmental parameters to be made. The time characteristic for the low-pass filter 84 may depend on the time resolution of the system. The longer the time characteristic, the less the noise affects the estimates. However, the system becomes slower in responding to changes in the

environment. Typically temperature, humidity and pressure sensing does not need to be reported very frequently and the time constant can be chosen to offer a good compromise between noise rejection and update speed. Therefore, a first estimation unit 85 for estimating temperature, pressure and humidity is connected to the output of the low-pass filter 84.

Alternatively, multiple filters could be used, one for a noisy, but fast update, and one for a stable long term update. The advantage of this system would be the ability to detect fast build ups of heat due to the break out of fire, for example. To determine the pressure humidity and temperature at the first estimation unit 85, only the relative attenuation at each frequency may be required. This can be input into models such as disclosed in Bass et al: "Atompheric absorption of sound: Further developments", Acoust. Soc. Am. , 1994, pages 680-683, to provide estimates for these parameters. It is however advantageous to measure the phase information as well, as this can be used to determine the group delay of the signal. The group delay can be used to calculate the time of flight from the transducer 70 to the receiver 20 (e.g. microphone). Since the speed of sound is directly affected by the temperature of the air, the time of flight can be used to directly estimate the speed of sound, and therefore the temperature of the air. This data can then be combined with the mathematical models to increase the accuracy of the humidity and pressure estimation.

Steady state air flow can affect the measured speed as well as the measured attenuation at different frequencies. By combining the phase and attenuation data, the additional signal perturbations introduced by steady state flows can be accounted for and partially or fully removed. More specifically, air flow can affect the transmission time and the relative attenuation of the acoustic signal. Drafts and random air flows will create short term fluctuations with zero mean. By contrast steady state flows will influence the transmission path over the long term, and different methods will be required to detect and estimate the influence of steady state flow.

To measure the influence of small non-steady air flows and drafts it is possible to examine the variability of the instantaneous values for the attenuation or amplitude data A^and phase data <i>_f. The influence of drafts causes short term fluctuations in these instantaneous values about a constant average. A high-pass filter 82 can be applied to the instantaneous data Μ[ Ε¾ A_f]_at each frequency of interest. This high-pass filter 82 removes the influence of long term trends, due to temperature, pressure and humidity, leaving only short term fluctuations due to noise and the influence of drafts. Strong drafts will create larger fluctuations in the data. Therefore, the magnitude of the drafts (air flow) can be estimated in a subsequent second estimation unit 83 by comparing the measured variance of the instantaneous data with look-up tables.

Thus, in the seventh embodiment, the acoustic measurement system consists of a transducer (not shown in Fig. 10) and a microphone (not shown in Fig. 10) at a known fixed separation and a controller or processor 46 are used to measure the temperature, pressure and humidity of the air, and also determine the relative movement of the air. Here, the sensor may consist of a broad band transducer capable of transmitting ultrasound over a wide range of frequencies (such as an electrostatic or electrets type). Alternatively, the transducer may be an array of tuned elements, where each element may be tuned to a different frequency (this may be a piezo ceramic, electrostatic or electret transducer). It is desirable that a broad spectrum of ultrasound be available for sampling to maximise the resolution for humidity determination based on a mathematical model of the frequency dependant attenuation of air. A frequency range of 20 kHz to 100 kHz would provide a useful range of reference points for humidity estimation, however reduced frequency ranges may also work well.

Fig. 11 shows a schematic diagram illustrating how air flow direction and magnitude can be estimated using a directional receiver 21. To determine the direction of steady state air flows, directional sensors are needed to determine the direction of arrival DoA of the sound received at the receiver 21 (e.g. microphone). By comparing the measured direction of arrival DoA with the known direction of origin (DoO) or actual location of the transducer 70 an estimate of steady state air flow (e.g. cross flow CF) can be made.

Furthermore, a directional microphone array can be used as the receiver of the acoustic sensor or receiver 21. The directional microphone array can be constructed from omni-directional microphones, and signal processing can be used to enhance the signal strength in an arbitrary direction. The microphone array can be used to establish the direction of arrival DoA of a test signal. Comparing this information to the known location or direction of origin DoO of the transmitter, the directionality and magnitude of air flow may be determined.

In some cases, such as when the distance between the transducer and microphone is unknown, establishing absolute values for the temperature, pressure, humidity and draft can be based entirely on a numerical model. If data is noisy, the relative accuracy of the estimates will be poor. To improve the estimates, 'gold standard' sensors, (sensors with known verified accuracy) can be used to provide input to the processor. If for example the absolute pressure is known, it can be input as a constant into the estimation procedure (e.g. based on a mathematical model), thereby removing a variable from the estimation procedure. This will help improve the accuracy of estimates of the remaining parameters. This would be particularly useful if many cheap acoustic sensors were deployed to provide a fine grid of measurements, and only one expensive 'gold standard' sensor was used to provide a background reference for calibration.

A spot measurement is capable of measuring local conditions only. For example, if a temperature sensor is embedded into a lighting infrastructure, the local ambient temperature will be dependent on whether the lamp is on or off. The advantage of the acoustic sensor is that the measurements reflect the average parameters over an extended column of air. Thus, the influences of local parameter fluctuations are averaged over a larger measurement domain. An acoustic sensor system would be less influenced by lamp than a local spot measurement. The fact that the obtained data is an average over an extended measurement can also be exploited to make detailed and accurate profiles of the parameters of the air over a volumetric distribution. This is because the measurement contains information about the entire acoustic path, not just the local conditions at the sensor. If any change occurs at any point along this path, the sensor will detect this change, although it will be averaged over the entire path. If two sensors with intersecting paths are used, any change common to both paths must therefore be local to the intersection point.

Fig. 12 shows a schematic diagram illustrating how profiles of air parameters can be created by a plurality of receiving apparatuses or sensors SI to S7 and a plurality of respective transducers 70 with intersecting acoustic paths. A homogeneous environment with constant ambient temperature, humidity and pressure is assumed. A strong draft present at the marked location D of Fig. 12 will cause sensors SI and S6 to be jointly influenced, while all other sensors will measure only the ambient conditions. This mutual interaction can be used to profile the location as well as strength of the draft. This is a simple explanation of a tomographical reconstruction of the parameters of the air over a distributed volume. Much more involved processing (e.g. mathematical calculations) is required to correctly reconstruct profiles of all parameters of interest in a 3D space using arrays of acoustic sensors. The processing may be similar to other fields such a computer tomography of X-ray images and oceanography, where acoustic signals in water are used to build tomographical models of the ocean currents and temperatures. An illustrative example of a temperature profile was already shown in Fig. 6 above.

Furthermore, intersecting cross-paths between all transmitters and receivers may be used to create a detailed 3D profile of the parameters of the air over the whole measurement volume. To maximise the number of intersecting cross-paths it is

advantageous to use non-directive ultrasonic transducers whose acoustic radiation is distributed evenly in every angular direction. This ensures that every receiver receives a signal from each transducer with a clear line of sight. To enhance signal to noise and reduce interference between transmitters code division multiplexing (CDM), frequency division multiplexing (FDM) or time division multiplexing (TDM) may be used. Non- directive transducers can be created using a transducer which is acoustically small compared to the wavelength of sound to be reproduced. At 40 kHz, the wavelength is approximately 8 mm. An acoustically small transducer at 40 kHz would therefore be required to be physically smaller than 8 mm. Alternatively arrays of small transducers could be used to actively steer an acoustic beam (using acoustic beam forming) of sound into a particular direction. By raster scanning the beam over all directions, each receiver would receive a signal from each transmitter, maximising the number of intersecting cross- paths. Other means of generating omni-directional sound sources exist, such as the use of acoustic lenses etc.

Thus, measurement data from multiple sensors can be combined and fed as input to a computer where acoustic tomography is performed to build up detailed profiles of temperature, humidity, pressure and air flow over a distributed volume. Again, the acoustic sensors may be combined with data from 'gold standard' sensors to calibrate and refine the estimates of the desired parameters of the air.

Fig. 13 shows a schematic diagram of a dual system according to an eigth embodiment where two acoustic sensors are used in parallel to improve estimates of local properties of the air. By shielding one transducer/receiver pair in a porous tube 90 the influence of drafts on the measurement can be eliminated, thus allowing improved accuracy of the parameter estimation. The porous tube 90 will ensure the air inside the tube is in equilibrium with the ambient air, but shelter it from the influence of drafts and steady state flow.

Fig. 14 shows a schematic diagram of an arrangement according to a ninth embodiment where sound is used to measure the composition of air. Sealed tubes of the same dimensions are filled with different gases. For the measurement of air (A), the major constituents are used: pure oxygen (0₂), carbon dioxide (C0₂) and Nitrogen (N₂), as shown in Fig. 14. Since these tubes are co-located they will experience the same temperature and gases inside will be shielded from drafts. Any difference in the speed of sound can be associated with the speed of sound in each gas. A tube containing an ideal concentration of gas for the environment may also be used as a reference to provide calibration data. Fig. 15 shows a schematic diagram of a transceiver based sensor according to a tenth embodiment. Here, the same transducer (i.e. transceiver 75) is used for transmitting and receiving acoustic signals. By aiming or directing the transceiver 75 at a reflective surface 100, a single sensing or measuring device can transmit a signal, then enter a receiving mode and wait for the acoustic test signal to return to the device, as shown in Fig. 15. The acoustic path is twice the distance from the transceiver 75 to the reflecting surface. Transceivers can also be used in multiples so that one transceiver first transmits to all other transceivers. Then, it later enters into a receiving mode as a different transceiver takes over the transmitting mode.

In the above embodiments, the effect of temperature on the acoustic measurement may be eliminated, leaving air flow as the most significant parameter affecting the acoustic measurement. In applications such as greenhouses it is common place to use powered fans to create draughts and air flows to achieve an even temperature distribution throughout the growing space. However, it is desirable to reduce the running speed of the fans to the minimum setting that achieves the desired even temperature distribution, in order to save energy. It is not a simple matter to determine what this minimum setting should be for each of the fans throughout the greenhouse, given the physical effects on the air flow of the supporting structures, watering and drainage installations and the varying size of the plants themselves. By briefly running the fans at their maximum setting, it can be assumed that an initial even starting temperature can be achieved. The challenge is now to determine by how much the speed of the individual fans can be reduced, for energy economy, while still being able to maintain an even temperature throughout the growing space.

The following procedure can be used, with the advantage that the measurement procedure is somewhat simplified compared with other embodiments. After running the fans at their maximum setting for a short period, the fan speeds are now reduced to the setting that is believed to be adequate to maintain and even temperature throughout the growing space and a special set of acoustic measurements is made. For this set measurements made at this time it can be assumed that it is the air flow that is the major influence of the observed readings, because the temperature is believed to be even throughout the growing space.

Unlike measurements made at other times, this set of measurements will mainly be affected by air flow, rather than temperature, because any temperature effects have been averaged out. This allows any local areas of inadequate or excessive air flow to be identified, and the fan speeds can adjusted according to experimentation or experience to compensate. This procedure could be repeated periodically, to recalibrate the effect of individual fan speeds on various air flows, given the changing size and shape of the plants being grown

Typical indoor environments considered for this invention are: offices, greenhouses, theatres, cinemas, exhibition halls, stations, airport terminals, shopping malls, warehouses etc. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In particular, variations including but not limited to any combination of speaker and microphone in e.g. lights, switches, ventilation grills, sensors, presence detectors, C02 sensors, or daylight sensors can be effected.

In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality.

A single processor, sensing unit or other unit may fulfil the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

It is noted that the proposed solution according to the above embodiments can be implemented at least partially in software modules at the relevant functional blocks. The resulting computer program product may comprise code means for causing a computer to carry out the steps of the above procedures of functions of the embodiments. Hence, the procedural steps are produced by the computer program product when run on the computer.

A computer program may be stored/distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems.

Any reference signs in the claims should not be construed as limiting the scope thereof.

In summary, properties of the air such as temperature, pressure, humidity and air flow can modify the propagation pathway of an acoustic signal. By making acoustic measurements of a known acoustic path it is possible to estimate at least one of the temperature, the pressure, the humidity and the air flow using perturbations present in the measured signal. Moreover, by combining data from several acoustic pathways topographical reconstruction of profiles of at least one of the temperature, the pressure, the humidity and air flow can be obtained. Profiling these data over an extended volume such as an office or green house is particularly important for local control.

Claims

CLAIMS:

1. A receiving apparatus (20) for acoustical determination of an environmental parameter which influences transmission of sound through air, said apparatus comprising:

a. a receiver (21) for acoustic reception of an acoustic signal transmitted by a transmitter (10; 70) to enable determination of effects of the air on the transmission of said acoustic signal; and

b. a detector device (45, 46; 83, 85) for estimating said at least one environmental parameter by comparing the received acoustic signal to a known transmitted acoustic signal by means of frequency domain analysis.

2. The apparatus according to claim 1, wherein said detector device (45, 46; 83, 85) is adapted to determine a relative acoustic absorption of the air as a function of frequency and there from determine said environmental parameter.

3. An acoustical parameter determination device comprising at least one emitting apparatus (10) and at least one receiving apparatus (20) according to claim 1, further comprising:

a. an acoustic receiver (21) provided in the receiving apparatus (20);

b. a parameter detector (45) for determining a relative acoustic absorption of the air as a function of frequency and there from determine said parameter;

c. wherein a signal generator (11) of said emitting apparatus (10) and the acoustic receiver (21) of said receiving apparatus (20) are arranged in a first defined distance (di) to each other.

4. A method for acoustically determining at least one environmental parameter which influences transmission of sound through air, said method comprising:

a. receiving an acoustic signal at a reception point to enable determination of effects of the air on the transmission of said acoustic signal; and b. estimating said at least one environmental parameter by comparing the received acoustic signal to a known transmitted acoustic signal by means of frequency domain analysis.

5. The method according to claim 4, wherein measurement data from a plurality of receivers (21) is combined and processed to build up a profile of at least one of temperature, humidity, pressure and air flow over a distributed area or volume.

6. The method according to claim 4, further comprising using measurement data of said acoustic signal obtained from a receiver of a high quality standard to calibrate or refine estimates.

7. A system for acoustical determination of at least one environmental parameter, said system comprising a receiving apparatus according to claim 1 and a transmitter (10; 70) for transmitting said acoustic signal.

8. The system according to claim 7, wherein said transmitter (70) comprises a transducer array of tuned transducer elements, each element being tuned to a different frequency.

9. The system according to claim 7, comprising first and second pairs of transmitters (70) and receivers (21) of acoustic signals, wherein one of said pairs is shielded by a porous tube (90).

10. The system according to claim 7, comprising a plurality of pairs of transmitters (70) and receivers (21) of acoustic signals, wherein said pairs are arranged in respective tubes filled with different gases for use as references to provide calibration data for the estimation.

11. The system according to claim 7, wherein said receiving apparatus comprises a directional microphone (21) for establishing a direction of arrival of said acoustic signal, wherein said system is adapted to compare said direction of arrival to a known location of said transmitter (70) to estimate a magnitude of air flow.

12. The system according to claim 7, wherein said receiving apparatus and said transmitter are integrated in a single transceiver (75) which transmits said acoustic signal towards a reflective surface (100).

13. The system according to claim 7, further comprising a processor device (46) for processing determined temperatures (T_ls T₂, T₃; Tr, Τ₂', T₃ ) provided by a temperature detector (45) to generate a temperature profile map (50) by setting the determined temperatures (T_ls T₂, T₃; Tr, Τ₂', T₃ ) in relation to specific points (pi, p₂, p₃; pr, ρ₂', p₃') within the measurement area, especially based on a lighting or building plan; and a controller device (47) arranged to communicate with a heating, ventilation and air conditioning (HVAC) system for controlling the HVAC system to provide cooling and/or heating based on the temperature profile map (50), especially to specific regions of the measurement area which is covered by said system for acoustical determination of temperatures.