MONITORING AND/OR CONTROL SYSTEM
The present invention relates to a monitoring and/or control system.
Conventional in-building monitoring and/or control systems (such as HVAC - Heating, Ventilation and Air Conditioning - networks) typically comprise a central control unit which communicates with sensors (e.g. thermometers mounted on interior walls or partitions) and actuators (e.g. radiator valves, window controls etc.) distributed throughout the building.
A significant problem with conventional systems is a lack of system flexibility. In particular if the sensors/actuators need to be moved (for instance when the partitioning arrangement within a building is being changed) then the HVAC network also needs to be completely reconfigured.
In accordance with the present invention we provide a building monitoring and/or control system comprising one or more nodes, each node communicating with one or more associated remote units by wireless communication, each remote unit being adapted to monitor and/or control one or more parameters or conditions associated with the building.
The present invention recognises the problem presented by hard wiring in a conventional system. The remote units are easily installed and/or reconfigured since no hard wiring is used to connect them to the nodes.
The nodes and remote units may communicate using rf
(radio frequency) links or infrared links. Preferably, however, the nodes communicate with the remote units by near field inductive communication. The features and advantages of inductive communication are discussed below.
Alternating current (AC) flowing in a wire or coil generates an AC magnetic field H. Maxwell's equations for this situation show that the magnetic field generated has two dominant regions: near field and far field.
The near and far field components have equal strength at λ/2π, where λ is the (free) space wavelength defined by
speed of light c divided by the frequency f of the AC current, c/f.
For far field radiation, the magnetic field is part of a travelling EM (electromagnetic) wave, and this is the normal mode of propagation for radio, TV etc. In this region, field strength is inversely proportional to range r and hence power is inversely proportional to the square of the range.
For near field, the magnetic field strength falls inversely proportional to the cube of range and power decreases at the -6th power.
Inductive communication uses the magnetic near field generated, for example, by an AC current flowing through a coil of wire. It has the advantage over rf (radio frequency) based systems that power drops faster: as the
6th power of distance rather than the 2nd. This makes inductive communication most suitable for a cellular structure where there is a high density of remote units in a 3-dimensional structure such as a building.
Typically the system comprises a commercial building control system, utility business communication system, or domestic environmental control system.
For example, the domestic environmental control system may comprise a HVAC system. The utility business communication system typically comprises a system for monitoring and controlling the supply of gas, water or electricity to a building.
In its simplest form the system may comprise a single node. For example, in a utility business communication system a single node may communicate with a single remote unit which may be a gas, water or electricity flow meter.
Typically, however, the system comprises a linear or cellular arrangement of a plurality of nodes.
Typically, the nodes each communicate with a plurality of remote units. The remote units typically comprise sensors, displays, control units and/or actuators used in systems of the type described above.
In the case where one or more of the remote units are sensors, the sensors may be of any suitable type, for instance infrared proximity sensors (for intruder detection), temperature sensors such as room, duct or water-temperature sensors, differential pressure sensors, inlet/outlet air flow sensors, window contact sensors, air quality or humidity sensors, smoke/fire detectors, light switch detectors, blinds control switch sensors or frost protection sensors.
In the case where one or more of the remote units are displays, the displays typically indicate status or temperature, humidity, air flow values etc. The displays may receive display data from a node and/or may receive display data directly from nearby sensors (by inductive communication).
In the case where one or more of the remote units are control units, the control units are typically used to control a particular parameter. For instance the control units may be temperature setting dials or blind control switches etc.
In the case where one or more of the remote units are actuators, the actuators may be radiator valves, window openers, blinds motors or light switching relays etc.
In the case of a utility business communication system, the node may be an outside interrogator such as a portable reader carried by an inspector, or a fixed reader.
The reader communicates inductively with gas, water and/or electricity meters either distributed throughout a building
(in the case of a portable reader) or within the range of the fixed reader. Alternatively, the node may be a prepaid or credit card operated unit which communicates with the metering units for water, gas and/or electricity and with an actuator for control of the supply of water, gas or electricity. Depending on the state of the card the node signals the remote unit to open or close the actuator for control of the supply of water, gas and/or electricity.
Instead of having the card reader formed integrally with
the node, a separate card reader may be provided as a remote unit which communicates via a wireless link with the node.
Typically, the remote units are battery or solar cell powered. However, the remote units (in particular actuators such as radiator valves which may have a high power requirement) may also be mains powered.
The remote units may be fixed or portable.
Typically the system is part of an integrated monitoring and/or control system which comprises a central processor connected to the nodes. The nodes may also be wireless units which communicate data to the central processing unit and/or other nodes using wireless links. Preferably, however, the nodes are powered by hard wiring and are connected for data transmission to the central processor and/or other nodes by hard wiring. The nodes may also be linked to the central processor by rf links (such as Ultra High Frequency - UHF).
Typically the remote units comprise one or more antennae which generate and/or detect magnetic fields.
The remote units may each comprise separate transmitter and detector antennae, individually optimised.
However, each remote unit preferably comprises a dual function antenna which both transmits and detects magnetic fields.
The antennae may be air cored or ferrite cored. Ferrite cored antennae achieve lower cost for the same performance and allow less wire to be used for the same magnetic dipole produced. Typically, the remote unit antennae are single-axis antennae although they may be multiple axis.
Typically the nodes also comprise one or more antennae: either separate or dual function transmitter/detectors. The node antennae may be single-axis or alternatively may be multiple (for example figure-of-eight) or large antennae.
The nodes may comprise three antennae having axes substantially perpendicular to one another and the signals to or from each of the antennae are phase shifted to direct the emitted signal and/or the most sensitive receiving direction to a specific one of the remote units.
Conveniently, the system further comprises means to measure the direction and level of signals received by each node and thereby determine the position of the or each remote unit. This allows the system to detect the position of the remote units.
The central processor may periodically make a request to each of the remote units to send out a signature signal which allows the central processor to identify the specific remote unit which it has located, typically within a sphere having a diameter of a meter or less. The central processor contains a true picture of the location of all remote units in operation, even if they are removed only temporarily from the intended place. This is particularly important since the remote units are so easily shifted from one place to another. Further, the physical tracking and verification of the location of the remote units is very costly and so substantial savings arise while maintaining the system.
Typically, the nodes are mounted to ceilings within a building, and transmit/sense signals covering a volume in the same and/or adjacent floors typically containing a number of remote units.
The spacing of nodes within a building will be dependent upon a number of parameters. These parameters include compatibility with building lay-out, range and coverage of a single link between a remote unit and node, interference from the next transmitting link, number of remote units per node, and bandwidth available. The design also depends upon the composition of the building, particularly whether or not the magnetic signals can be transmitted/ sensed through the floor.
The nodes may be arranged in a close-packed structure such as hexagonal close packed (ABAB) or face centred cubic (ABC). Alternatively the nodes may be arranged in a square lattice.
Successive layers of nodes will typically be placed every two or three floors in a building (where signals can penetrate through the floor).
Communication between nodes and remote units may be carried out by frequency division multiplexing, but is preferably carried out by time division multiplexing. The time division multiplexing protocol typically involves two types of access to the network:
(i) regular "polled" pre-programmed accesses, in which sensors such as thermometers are polled, e.g. every five minutes; and
(ii) intermittent "alarm" accesses or carrier sense multiplexing, which are necessary for example when an actuator is switched on by a light switch sensor signal.
The system allocates a portion of its time (e.g. 25% to 80%) to regular polling of sensors etc., and the remaining portion (i.e. 75% to 20%) to listening for random signals from actuators, control units etc. Preferably the regular polling may use up to 80% of the system time and the listening time is evenly dispersed between the regular pollings.
Typically, the system frequency for the time division multiplexing system will be between 9kHz and 135kHz.
Preferably the frequency is between 10kHz and 15kHz. It has been found that this frequency range gives the greatest immunity from noise sources.
Transmission from the nodes to the remote units may or may not be at the same frequency as transmission from the remote units to the nodes.
Typically the remote unit to node communication is carried out by bi-phase modulation and the node to sensor communication is amplitude modulated. Phase modulation may
be used for communication between the node and the sensor, although this results in a higher system cost.
In one example data is communicated between the node and remote unit using either ASK, BPSK, FSK or QSK modulation.
In accordance with a second aspect of the present invention, there is provided apparatus adapted to interact with a building monitoring and/or control system, the building monitoring and/or control system comprising one or more nodes each node communicating with one or more remote units by wireless links, each remote unit being adapted to monitor and/or control one or more parameters or conditions associated with the building, the apparatus comprising a portable processor adapted to communicate with one or more of the remote units by wireless communication to check their functions, and/or adapted to communicate with one or more nodes by wireless communication to check the response of the system.
The wireless communication may be rf, infrared or preferably near field inductive communication.
The portable processor (such as a lap-top or PC) typically comprises an interface and antenna to interact with the system according to the first aspect of the invention. It may act as a node and interrogate a remote unit to check its functions (settings, status or measured values, initiate switching of lighting, operate apparatus) or act as one of the remote units to check the system response.
Typically the monitoring and/or control system further comprises a central processor connected to the nodes, and the apparatus is adapted to communicate with the central processor via the nodes. The processor can then transfer programme settings (parameters or a new release of the programme) to the central processor. The "lap-top" tool is of importance to the engineers setting up or maintaining a system according to the first aspect of the invention.
In accordance with a third aspect of the present invention, there is provided a monitoring and/or control system comprising a cellular structure of nodes which communicate with one or more remote units by near-field inductive communication.
Whilst the first and second aspects of the invention relate to in-building applications, the third aspect of the invention can also be applied to other non-building related monitoring and/or control applications.
A number of embodiments of the invention will now be described with reference to the accompanying figures, in which:-
Figure 1 is a cross-section of a first building containing a communication system according to the present invention;
Figure 2 is a block diagram of one of the sensors;
Figure 3 is a schematic diagram illustrating the geometry of a remote unit antenna;
Figure 4 is a functional block diagram of one of the nodes;
Figure 5 is an implementation block diagram of one of the nodes;
Figure 6 is a schematic diagram illustrating the geometry of a node antenna;
Figure 7 illustrates a time division protocol for the system;
Figure 8 is a schematic view of part of the system shown in Figure 1 illustrating the principle of localisation of remote units;
Figure 9 is a cross-section of a second building containing a communication system according to the invention;
Figure 10 is a first example of a utility business communication system according to the present invention; and
Figure 11 is a second example of a utility business communication system according to the present invention.
Figure 1 is a schematic cross-section of a building 1 having seven floors 2-8, and incorporating a communication system according to the present invention. Nodes are arranged in a square lattice on every other floor. Three example nodes are labelled 9-11. The spacing 50 between nodes is typically in the region of 10-llm. Each node is attached to the ceiling and covers a cubic volume surrounding the node. For instance, the node 9 transmits signals to, and receives signals from remote wireless units 12-18 which are distributed as required within its area of coverage.
Each node carries out two-way communication with the distributed remote units every 30 seconds. Each node comprises a 3-axis, electronically steered air cored antenna. Each node transmits on a single frequency by time division multiplexing. The operating frequency is approximately 12.5KHz, with a band width of 500Hz.
The system of Figure 1 assumes that signals can be transmitted through the floors.
Each node 9-11 etc. is hard wired by suitable wiring (not shown) to a central processor 19, which receives data signals from the nodes, and transmits control data to the nodes accordingly. For instance, the remote units 12-15 may be temperature sensors, and the remote units 16-18 may be radiator valves which are actuated in response to signals from the temperature sensors 13-15.
One of the remote units 12 is illustrated in Figure 2. The remote unit 12 has an antenna 52, the physical geometry of which is illustrated in Figure 3. Antenna 52 has a single coil comprising approximately 1000 turns of 0.4mm copper wire, wound on a 10cm diameter former. The total copper mass is 0.25kg. The antenna 52 is switched between transmit and receive modes by a signal on TX/RX select line 53 from 8-bit microprocessor 51. The TX/RX select signal causes the antenna coil to be series tuned in transmit mode
(low impedance) and parallel tuned in receive mode (high impedance). In transmit mode the microprocessor directly generates a transmit RF modulated waveform which is output on line 54. The receive circuit consists of a high impedance front end low noise amplifier (not shown), a six pole bessel filter 55 and a single chip FSK demodulator 56. The microprocessor 51 performs the UART function and frames the data messages. The microprocessor 51 provides an interface to desired functions, in this case a 7 segment display 57 and push button inputs 58. In addition the microprocessor 51 receives temperature data signals from temperature sensor 59. The sensor 12 is powered by a 5V battery 60.
The design of one of the nodes is illustrated in Figures 4-6. Figure 4 is a block diagram showing the software functionality of the node system, as implemented in the TMS 320 digital signal processor and 80186 microprocessor shown in Figure 5. Figure 5 is an implementation block diagram of the node software illustrated functionally in Figure 4. Referring first to Figure 4, each node comprises a 3-axis antenna which provides baseband digitised receive x, y and z signals on input channels 100, 101, 102. Function 103 combines the received signals from the 3-axis antenna using direction (cosine) vector input signal 104. The output is the optimal signal 105. The process to compute the direction vector signal 104 is defined in Appendix A. This signal is low pass filtered by the digital filter 106. Functions 107-110 implement data demodulation, thresholding and framing. The data modulation scheme in this example is BPSK. Function 107 computes a running average of the mean phase error, which is used to control the VCO shown in Figure 5. This ensures that the VCO is locked to the received carrier phase. Function 109 is a matched filter to the data and an integrate and dump process is used. The integrator gating (sample) is controlled from a Data Phase locked loop 108 which is locked to the data rate. Data thresholding and framing is performed by function 111.
Function 111 outputs the demodulated data stream, and a feedback control signal to the data PLL function 108.
Referring to Figure 5, inputs 100-102 are input to low noise amplifiers 112-114 which amplify the received signals which are input in turn to mixers 115-117. Item 118 is a voltage controlled oscillator nominally at 12.5kHz the carrier frequency of the received signal. The VCO 118 is controlled by the data demodulation function (software) so that the VCO 118 remains phase locked with the average carrier frequency (phase rate). Mixers 115,116,117 provide baseband I&Q (in-phase and quadrature) signals. The digitised signals are multiplexed into the software function provided by TMS 320 DSP 121 and microprocessor 122. The outputs of the software function are: demodulated and framed data 123 and VCO control signal 124. The data 123 is input to a wire backbone which is fed to the central microprocessor 19. The DSP processor 121 functions as a front end processor, and the microprocessor 122 performs "housekeeping" functions.
In the case of the node 9, the microprocessor 122 is programmed to carry out the following:
• phase-lock to the data edges of the signal received from the remote units 12-18
• measure signal amplitude in all three axes (coils, channels) and compare these to presignal noise in each channel
• calculate the apparent signal direction from the ratio of the three axis amplitudes
• calculate the optimum signal weighting combination (based on S/N) of the three axes
("optimum antenna synthesis")
• decode the incoming data using the optimum antenna synthesis
• log the number of messages per 300 (repeat) messages that have 0, 1, ...10,>10 bit-errors per message.
The node also comprises transmitter circuitry not shown in Figures 4 and 5. The transmitter circuitry is similar to the remote unit circuitry illustrated in Figure
2, and generates transmit signals which are transmitted by the node antenna to remote units in its area.
Two possible designs for the node antenna are:
1. multipole (e.g. Figure of 8) configurations. These have the feature that the ratio of signal level between that on close objects to that from far away objects is larger than for small, dipole antennae. They would provide significant advantages where man-made noise sources are further from the node than the remote units. This is unlikely to be the case in this application. Multipole antennas also have the disadvantage of lower coverage because their pattern has null points.
2. Large antennae. Large antennae reduce the signal ratio between remote units that are far away from the node and those that are near. This is not of significant advantage, and increases interference from neighbouring nodes.
Figure 6 schematically illustrates the form of a node antenna 130. The antenna 130 comprises three antennae 131, 132, 133 which are identical to the remote unit antenna illustrated in Figure 3 and which are arranged orthogonally. The outputs of each of the antennae 131-133 constitute the channels 100-102 illustrated in Figures 4 and 5.
An example of a suitable single frequency time-multiplexed communication protocol between the nodes and remote units is illustrated in Figure 7. The network array of nodes is divided into N time channels. In the case of Figure 7 N=8. That is, each one of the nodes transmits to all of its associated sensors and receives their signals in one of the eight time slots 70,71 etc.
The third time slot 70 is illustrated in detail. Each node has M associated sensors. The node associated with time slot 70 has 6 associated sensors. The time slot 70 is divided into 6 time slots 72,73 etc. In each time slot 72,73 etc. the sensor transmits 100 bits of data to the node in a first portion 74 of the time slot and receives 100 bits of data from the node in a second portion 75 of
the time slot. No data is transmitted in a central portion 76 of the time slot.
Assuming a maximum of 20 sensors per node, fully packed time slots, and a square cell array with N=8, the required data rate is 1000 bits per second. Different bit rates may be appropriate in different directions. The sensors keep time to within 0.5 bits to know when to transmit. This requires a 30ppm crystal of the low cost watch type.
For synchronisation the node can send data telling the sensor when to transmit data next.
Where the number of time slots N is large then the interval between time slots associated with interfering nodes ("interfering" nodes being nodes which are located close to each other and are therefore liable to interference) is large. The large interval results in lower interference. However, large N means that the number of time slots must be large and the information must be transferred faster with a larger bandwidth. Inductive links are appropriate for such an approach in a 3-dimensional building because:
(a) interference fall-off with distance is fast. An infinite 3-dimensional inductive communications network could operate, whereas a radio based system would be infinitely noisy;
(b) Rayleigh fading is not an issue, so allowable signal to interference ratio is typically higher;
(c) detection angle may be varied electronically in the nodes top trim out interfering emitters; and (d) the nodes can co-operate to help sensors in trouble with interference. Typically this can be done by optimising the orientation of their transmitted dipole.
Figure 8 is a schematic view of Figure 1, illustrating the use of three nodes 9,10,11 to determine the position of (i.e. "localise") remote unit 17.
Three nodes 9,10,11 have a range indicated by the arrows 27,28,29. The remote units 12,15 and 17 are within the range of the three nodes 9,10,11. The remote unit 18
will be localised by the node 11 and two other ones here not shown. The remote unit 17 is being localised by the three nodes 9,10 and 11 which are determining the directions 30,31 and 32 respectively and the received signal strengths. The processor 19 calculates from the received signal strengths and directions the point in 3-dimensions which is common to all three directions 30,31 and 32 and therefore localises the remote unit 17 within the building. After localising all remote units, the processor 19 can display an actualised layout of the system.
The node receive antenna comprises three orthogonal antennae sensitive to the field in the x, y and z directions. The background noise received is not thermally limited and it has been found that in most instances the "noise" is from interference sources close to the receive antenna, and hence the "noise" can vary significantly in the x, y and z direction receive channels. In a thermally noise limited situation noise would be equally distributed in the x, y and z receive channels.
The transmitter and receiver positions are fixed and do not vary with time. Therefore, it is possible to synthesise the node receive antenna direction. It has been found in practice that this direction may not be the direct line of sight path.
As the interference is not uniform in the x, y and z receiver channels the synthesised receive pointing direction can vary significantly from the geometrical line of sight direction. The optimum direction will depend on the background noise vector in the x, y and z channels.
A routine in the node microprocessor 122 first measures the background noise power in the x, y and z directions, then measures the signal plus noise power in the x, y and z directions and computes the optimum antenna pointing direction. The optimum antenna direction consists of a set of 3 direction cosines which "weights" the x, y and z direction receive signals.
The "weights" are computed from the background noise measurements and noise plus signal movements. For
reference the "C" programme which comprises these weights is set out in the attached Appendix A.
A portable lap-top computer 40 according to the second aspect of the invention is also shown in Figure 1. The lap-top computer 40 can communicate with remote wireless units or nodes within its range and can be used to interrogate sensors, initiate actuators etc., receive signals from nodes to check the system response or transfer parameters to the central processor 19.
A second embodiment is shown in Figure 9. In this case, the nodes 20 are mounted between floor level 21 and the ceiling level 22. The range of each node 20 is indicated schematically by dotted lines 23. The nodes 20 communicate with ceiling mounted remote units 24 and wall mounted remote units 25. The central processor 19 is omitted for the purpose of clarity.
Examples of a utility business communication system are illustrated in Figures 10 and 11.
Figure 10 illustrates a fixed node 80 with a display 81. The node 80 communicates via inductive communication with sensors 82-84 which are distributed in a building within range of the node 80. The sensor 82 monitors the flow of water through water supply network 85. The sensor 83 monitors the flow of gas through gas supply network 86. The sensor 84 monitors the flow of electricity through electricity supply network 87. Data received from the sensors is used to display the number of consumed units of gas/water/electricity on display 81, which can be read by an inspector. Alternatively the node 80 and display 81 may be a single portable unit which is carried by a meter inspector.
Figure 11 illustrates an alternative utility business communication system in which the same reference numerals are used for equivalent items from Figure 8. In the embodiment of Figure 9 the node 80 and display 81 are replaced by a node 90 which incorporates a card reader for reading a prepaid card 91. The node receives data from sensors 82-84. The prepaid card is programmed to allow a certain number of gas/water/electricity units to be
consumed. The node transmits a signal to an actuator 92-94 which controls the supply of gas, water or electricity. When the prepaid card runs out of units, the node transmits a signal to actuator 92, 93 or 94 which causes the actuator to cut off the related supply.