GB2513733B - Emergency system - Google Patents

Description

EMERGENCY SYSTEM

This invention relates to an emergency system, and particularly, but not exclusively to awireless fire alarm system. The system is adapted for use in any suitable environment,but is particularly suitable for use in the construction industry.

It is known that construction sites can be potentially dangerous and therefore it is importantthat all necessary safety standards are met. It can be particularly important to have areliable emergency system that will, for example warn people on the construction site ornearby, of emergencies such as fire, through the detection of fire, heat and/or smoke. Itis also important that any emergency system can provide a means for manually raising analarm by manually activating call points forming part of the system.

Also such systems should be easily adapted to, for example, the changing size of abuilding under construction. This means that it is important that emergency systems foruse in the construction industry have the facility to be extended or to be reduced in size asnecessary to meet the changing requirements of the construction site. A known emergency system comprises a plurality of devices that are adapted tocommunicate with one another wirelessly.

The devices are able to communicate unidirectionally on a strict timeslot basis dependingon the number of the unit in question. In such a known system, an activated unit will sendout a radio message after it has been triggered by a call point press or detection headactivation. The radio message indicates the start of the transmission timing, with eachdevice relaying the radio message sequentially in turn. There is no acknowledgemechanism for this communication method. Instead, each message is sent three timesconsecutively and it is assumed that at least one of the three messages has been correctlyreceived. In addition, a predetermined length of time after the original transmission, theradio communication is repeated again in the same sequence.

Typically, there will be a restriction on the number of units forming such a system due tothe particular communication protocol being used and the fact that that the protocol willdepend upon the unit numbers in the system.

According to a first aspect of the present invention there is provided a wireless emergencysystem comprising a plurality of units, each unit comprising a transceiver adapted to transmit and receive data to/from other units in the system, each unit comprising a testercomprising a transmitter adapted to transmit a silent message to other units in the system,and an indicator adapted to indicate to a user that a unit has received the silent messageand that the strength of the signal received is above a predetermined level, wherein eachunit further comprises a delayer for delaying transmission of data by a predeterminedrandomised delay time, and a clear channel assessment for checking that no other unitwithin range of the unit in question is transmitting at that time wherein the delayer isoperatively coupled to both the transmitter and the clear channel assessment and the clearchannel assessment is operatively coupled to both the transmitter and the delayer,whereby if the clear channel assessment determines that the channel is not clear, thedelayer will delay transmission of data by a predetermined randomised delay time suchthat the transmitter will transmit only if the channel is clear.

The purpose of the tester is to exercise the radio communication part of the units andprovide confirmation that each unit is able to communicate successfully with other units onthe same site.

The message sent by the tester may be sent without any audible signal also being sent.In other words the signal may be sent without a siren or other audible alarm also beingsent. For this reason the test carried out may be described as a silent test, and may beregarded as a way of testing that the system is in a state ready to react to a situationrequiring an alarm to be raised.

The silent tester may be initiated from any unit in the system.

The silent tester may comprise a silent test button. The silent test button may be pressedby a user in order to activate the silent tester.

Activation of the silent tester causes the unit associated with the silent tester to send outa silent message activation, which activation causes the indicator on each other unit toindicate that the unit has received the silent message.

The indicator may provide any convenient type of indication that a particular unit hasreceived the silent message. In some embodiments, the indicator comprises a visualindicator such as an LED indicator and/or display. This allows for convenient inspectionof the system to take place in order to check that each and every unit has received thesilent message.

The network comprises a (logical) network layer, a physical (PHY) layer and anapplication layer.

The delayer in each of the units reduces the probability that more than one unit willtransmit at the same time.

Each unit also comprises a clear channel assessment. This component will check thatno other units within range of the unit in question is transmitting or receiving at that time.This reduces the possibility that there will be interference with other units in the same, orother systems. A system according to the first aspect of the invention allows each unit to communicatewith other units in the system using a repeated broadcast radio message. This ensuresthat the radio message is propagated across the network with each unit acting as a relayin a mesh-like structure.

In embodiments of the invention, the system adopts a decentralised mesh topology. Insuch a system there is no system map and the units in the system do not know whatother units exist in the system.

As long as every unit in the system is connected to the mesh-like structure by at leastone sufficiently strong link, the system will function correctly.

In some embodiments each unit will be connected to more than one other unit. Each unitcan thus be regarded as having a plurality of links to other units within the system. Thisadds to the robustness of the system through path diversity.

Any given unit will be able to communicate with other units within its range. A unit thatreceives data from another unit may then in turn transmit the data to yet another unitwithin range of that unit. Whenever a unit receives a transmission it may then in turntransmit the data received to another unit within range of the unit in question. Each stagein this mesh-like structure can be regarded as a “hop”.

The system could, in theory, have an indefinite number of units with many relay hopsbeing required in order to ensure that all units are in communication with all other units.However, in order to ensure timely communication and satisfy latency and powerconsumption requirements, in some embodiments of the invention a limit of communication hops has been placed on the system. In many embodiments the limit tothe number of hops is three. Beyond three hops the system will still work but the time topropagate messages across the network may be long and cannot be guaranteed.

In any given system, data may be transmitted repeatedly with the interval betweenrepeated transmissions being a random interval of time. When a unit receives atransmitted message it may relay the message a plurality of times also with a randominterval of time between repeated transmissions.

In one embodiment of the invention, when a unit in the system wishes to communicatean event, the data may be transmitted five times at random intervals of time. When aunit hears the message it relays the message three times also at random intervals oftime. In other embodiments, the data may be transmitted and relayed any convenientnumber of times.

For every transmission, short random delays are introduced by the delayer of typically afew milliseconds. In this way the probability of a relay clash is reduced. A unit will therefore typically hear the same message several times.

Each distinct event is identified by a combination of the originating unit identifier and amessage sequence number.

The network layer may maintain a history buffer detailing which distinct event(s) it hasheard in a given period of time, known as the history buffer period, and does not passrepeats up to higher level application logic. The history buffer is defined as beingbetween some PHY active and 30 seconds after there has been no PHY activity. Everytime a relayed packet is received by the network layer it checks to see whether it hasalready relayed the packet and passed it up to higher level application logic. If it has, itupdates its received signal strength indication (RSSI) entry for this message and doesnothing else. If it has not, it passes the message up to higher level application logic,stores a new entry in the history buffer and relays the packet. 30 seconds after there hasbeen no PHY activity, the history buffer is cleared.

The system may operate on any suitable frequency but in many embodiments thesystem will operate within the radio frequencies. However, it is to be understood that the system may operate within different bandwidths and may for example operate in thevisible range of frequencies.

In some embodiments, each transceiver will operate in a frequency band from 868.0MHz to 868.6 MHz.

In some embodiments, each transceiver may comprise a CC1125 Category 1 integratedhalf duplex radio transceiver operating at a power of 25 mW and at a frequency of 868.3MHz. In such an embodiment each unit may have a duty cycle of less than 1 %.

It is to be understood however that the transceivers could operate on differentfrequencies within or outside of the radio bandwidth, and could also operate at differentpowers.

The data may be sent in any convenient format but in many embodiments will be sent inpackets. Each packet may contain an address unique to the site at which the system isinstalled. It may also contain a unit number to identify a particular unit, a message typeto identify the type of message, a command and a 16 bit check sum or cyclic redundancycode to ensure that the received packets have not been corrupted. A suitable data packet format is set out below.

The data packets may however have some other configuration.

When operating at radio frequencies, it is preferable for over the air data to be random.This results in the smoothest power distribution over the occupied bandwidth. To assistwith this, the PHY layer implements a simple data whitening algorithm. This works onthe assumption that fields of consecutive 1’s and fields of consecutive 0’s are statisticallymore likely in the data packets used in systems according to the present invention thanfields of alternating 1 ’s and 0’s.

Units in the systems according to the invention will look for a sync word to detect thestart of a valid data packet. This can be an effective mechanism only if the sync worddoes not occur anywhere else in the data system.

To ensure that this is the case, a token removal algorithm is implemented.

Each data packet includes a synchronisation word, known as a token, which is constantfor all data packets and allows the start of each frame to be identified, this word, ortoken, must therefore not appear anywhere else in the data packet. To avoid this tokenappearing within the data each instance of AB is replaced by ~A~B~AB, where A is thefirst byte of the token and B is the second byte of the token and ~A is the bitwise inverseof A.

Wherever the token has been removed it needs to be replaced after the data packet hasbeen received, this is performed by the following single algorithm, replace ~A~BX1X2with ~X1X2.

For the one step replacement algorithm to work at the receiving end any instances of theinverse of the token need also to be replaced as they are a marker to indicate whichbytes have been replaced. For instances in the data where the inverse of the token ~A~Boccur these are replaced with ~A~BA~B which can be returned to the original ~A~Busing the same algorithm above, i.e. Replace ~A~BX1X2 with ~X1X2.

This can be further explained by an example where the sync word or token is chosen tobe 0x930B (A represents 0x93 and B represents OxOB). (In binary 0x930B = 1001 0011 0000 1011. The inverse of this (where 1 becomes 0 andvice versa) is 0110 1100 1111 0100 = 0x6CF4)

Any instance of AB (0x930B) in the packet data is replaced, according to the algorithm~A~B~AB with 0x6CF46C0B.

Any instance of ~A~B (0x6CF4) in the packet data is replaced, according to the secondalgorithm ~A~BA~B, with 0x6CF493F4.

These bytes will then be replaced during the receive process using the single algorithm,replace ~A~BX1X2 with -X1X2. e.g. for 0x6CF46C0B, the receiver finds the inverse token 0x6CF4 and replaces it with0X930B. e.g. for 0x6CF493F4, the receiver finds the inverse token 0x6CF4 and replaces it with0X6CF4. A 16 bit CCITT CRC (X16+X12+X5+1) is used to detect packet corruption. In the receivedirection the CRC check is performed by the transceiver. In the transmit direction,efficient CRC generation is performed by the PHY layer.

In some embodiments of the invention the transceiver is powered for some of the timeonly. In such embodiments, the transceiver may comprise a Wake On Radiocomponent. This enables the power consumption of the transceiver to be optimisedwhilst permitting low latency communication. This in turn increases the battery life of theunit.

In embodiments of the invention, the units forming a system according to embodimentsof the present invention may be battery powered. Each unit must be ready to reactimmediately to any given event. This means that a low power sleep mode in which a unitis nevertheless still capable of receiving data is advantageous.

This can be difficult to achieve since the unit requires significant power in order to listenfor incoming data. In a particular embodiment in which each unit comprises a CC1125,transceiver a method for automatically polling for data activity may be used.

This mode of operation may not always be sufficient however in order to guaranteereception of data, because in order to guarantee receipt of data, a data packet preamblelonger than the maximum allowed by the CC1125 must be achieved.

The inventors have nevertheless realised that this may be achieved using CC1125transceivers, through appropriate configuration of data packets. In particular, theinventors have realised that if the sync word, length field, address field, and some of thedata payload are programmed to be the preamble data word (repeated), and then thesesame fields are instead manually (not automatically by the transceiver) inserted after thisextra preamble data, a preamble longer than the maximum supported by the transceiver(according to its datasheet) can be realized and thus a low power sleep mode may beachieved.

In order for this to function correctly, digital data processing normally performed in thetransceiver must be performed manually in the PHY layer.

In units according to embodiments of the invention, the PHY layer has different transmitand receive configurations.

In the transmit direction, the PHY provides a fully processed packet with a non-standardextra long preamble which the unit transmits transparently. In the receive direction, theadvanced features of the transceiver in the unit are leveraged to enable optimum sleeplevel and duration.

Because it is necessary in some embodiments to configure the transceiver of the unit inbetween receiving and transmitting data, the CCA (Clear Channel Assessment) isperformed instead by firmware provided in the PHY layer.

In such embodiments therefore the PHY layer listens to ascertain whether the radiochannel is clear. If it is it proceeds with transmission. If it is not it waits for a randomback off of typically a few milliseconds and then tries again.

This listen and back off cycle is repeated for up to 10 seconds (although other timeperiods could be used) at which point the PHY will transmit regardless of the channeltraffic.

In a particular embodiment of the invention each transceiver will draw an average currentof approximately 60μΑ without an additional receive amplifier, and approximately 100uAwith an additional amplifier to provide increased range.

Each unit may comprise a microprocessor for controlling operation of the unit.

The delayer and clear channel assessment may form part of the microprocessor.

The microprocessor may determine whether any received data packets have a siteaddress which matches the site address of the system. If a received data packet has anaddress matching the full site address, the data within the data packet will be processedand actions will be performed by the microprocessor.

The microprocessor may remain in a low power sleep state until a data packet isreceived which matches part of the site code and has passed a cyclic redundancy codecheck indicating that there is a high probability that the packet has been received errorfree.

The unit may comprise a monitor for monitoring the signal strength of transmitted andreceived data. This is to ensure that at least one data packet from a unit is above arequired attenuation reserve threshold. The monitor may form part of themicroprocessor.

In particular, if a unit receives radio messages from peer units within the history bufferperiod, but none of the signals within that period is above a reliable signal strength, a lowradio warning will be raised.

In parallel, if a period of PHY activity is started by a transmit operation originating at alocal unit, the network will expect to have heard at least one relay of this message from a neighbouring unit by the end of the history buffer period. If no relay is heard, a separateradio absent warning will be raised.

Each unit may comprise an attenuation reserve threshold.

The attenuation reserve threshold determines whether a signal has sufficient signalstrength. The purpose of the attenuation reserve threshold is to allow for variations in thereceived signal strength due to external environmental changes which may includeatmospheric conditions, unit positioning, movement of people, furniture, equipment etc.The attenuation reserve artificially raises the minimum required received signal strengthto provide a margin of safety to help ensure radio messages are robustly received andnot operating near to the point where a radio signal might not be correctly received. Thesystem uses the attenuation reserve threshold during silent tests such that units whichcorrectly receive a signal which is above the threshold will successfully enter the silenttest mode, units which correctly receive a signal which is below the threshold will fail thesilent test and will indicate a “low signal” warning instead.

For fire alarms however, the attenuation reserve threshold will be disregarded and allunits which correctly receive a radio message will enter the fire alarm state.

The monitor also monitors the PHY layer, so if a PHY operation takes too long, the PHYis reset. Whenever the PHY is requested by the network layer to perform an action (e.g.to send a packet), the network layer starts a timer. If the timer expires before the PHYcalls the network layer back to signal successful completion, the network layer assumesthat there is something wrong with the PHY layer (or the transceiver it abstracts from)and performs a PHY reset sequence to return to functional state.

Each unit may comprise a fault warning unit for transmitting a fault signal in the eventthat the signal strength of the data packets falls below the attenuation reserve threshold.Under such circumstances communications will continue but a signal from the unitindicating the fault will be transmitted and displayed. The fault warning may form part ofthe microprocessor of each unit. A unit which initiates (not relays) a radio message listens for the radio message beingrelayed onwards by other units as confirmation of correct receipt of the message. If this‘radio echo’ is not received then the unit indicates a fault.

The units in the system may comprise any suitable units, and when the system is afirepoint system, at least one of the units comprises a firepoint device. In such a systemat least one of the other units comprises a detector unit adapted to detect smoke/heat,although it is not necessary to have such units.

In such a system, alarm messages are acted upon by the one or more firepoint units.

The system may also comprise a base unit comprising an identifier adapted to identifywhich of the one or more units of the system has been activated. The base unit maythen emit an alarm message in response to identifying one or more units having beenactivated. The identifier may form part of the microprocessor.

It is not necessary for a system according to the first aspect of the present invention tohave a base unit, but some embodiments will have such a unit which preferably may bepositioned inside a building or an enclosure. There is however no requirement for thebase station unit to be positioned indoors it is typical for it to be in a site cabin but couldequally be located outside.

In all systems whether with or without a base unit it is possible to identify which unitshave been activated by visual inspection of each unit.

In some embodiments of the invention, each detection unit will have a visual indicator ofan alarm state. In some embodiments of the invention this visual indicator comprises anLED display.

The period of time for which a detection unit may emit an alarm signal may be anyconvenient time, and in some systems is 30 minutes, unless the fire alarm is cancelledbefore the lapse of that time period.

In some embodiments of the invention firepoints which have initiated an alarm by beingactivated may indicate the alarm state by means of a mechanical flag which is visible toan observer.

In some embodiments of the invention, if any of the units within the system is manuallyactivated, an activation alarm may be emitted by that unit for as long the unit remainsmanually pressed.

The signal that is emitted may comprise an audible signal.

The audible signal may comprise a siren chirp.

In some embodiments of the invention, a unit may be manually activated by a userpressing a button to activate the unit. In such embodiments, the activation alarm may beemitted for as long as the button remains pressed.

In some embodiments the base station unit comprises an internal GSM modem andantenna.

The base station may comprise a display unit. The display unit is adapted to indicate toa user which of the units in the system has been activated. The base station may alsobe adapted to display information relating to units having low battery power, units thathave been tampered with, or those having a poor radio signal.

The units forming a system according to embodiments of the invention are sealed at themanufacturer with a battery pack connected within the unit to avoid the need to open andpossibly damage the unit during installation.

In order to conserve battery power and to permit safe transportation of each unit by van,boat, aeroplane etc, each unit may be placed into a low power mode, known as a transitmode. In this mode the transceiver is disabled to prevent transmission or reception ofany messages and all electronic circuitry is placed into the lowest power state. Pressinga call point or activating a detector head in transit mode for example, will have noconsequence.

In an embodiment of the invention, a unit may be put into transit mode by holding a pairbutton and then pressing a test button three times in quick succession. The unitindicates it is entering transit mode by, for example illuminating a plurality LEDs.Additionally if the unit has already joined a site network it will send a message to informthe base station (if fitted) that a unit has been put in transit mode and is therefore nolonger part of the installed system. A unit is brought out of transit mode by following the same button press sequence ofholding pair button and triple pressing test button. Units being brought out of transit mode have their site code details returned to default and are ready to be installed bypairing with another unit. Unit numbering is unaffected by this process.

In other embodiments, different processes may be followed.

The units may comprise a back tamper component. The back tamper component isadapted to indicate when a unit has been removed from either a wall or a ceiling, forexample on which it has been mounted. Typically, a firepoint unit will be mounted on awall whereas a smoke/heat detector may be mounted on a ceiling.

The back tamper component is adapted to be resettable by reaffixing the unit to awall/ceiling as appropriate.

Once the unit has been reaffixed it will confirm that it can communicate with at least oneother component in the system.

The back tamper unit is adapted to perform a radio check when the unit has beenreinstalled.

Each unit may comprise a first housing component adapted to receive and containelectronics and/or firmware associated with the unit, and a second housing componentcomprising a first connector for connecting to the electronics and/or firmware containedin the first unit.

The electronics may be in the form of an electronics board.

The first connector may be used to form connections with different electroniccomponents in the first housing depending on the behaviour of the detection unit that isrequired.

The first housing component may comprise a second connector operatively connected tothe electronics and/or firmware. When the electronics comprises an electronics board,the second connector may be mounted on the electronics board.

The first connector is adapted to engage with the second connector.

The first connector may be designed so that only certain connections are made when thefirst connector and the second connector engage with one another.

For example, in a system comprising firepoint units, heat detectors and smoke detectors,each of these types of detectors may be formed from a first housing comprisingsubstantially identical electronics and/or firmware. However, when the first connectorengages with the second connector different combinations of electronics components areconnected to form part of the particular unit. The particular electroniccomponents/firmware which are included in the overall circuitry will determine how theunit will behave and therefore whether it is a firepoint unit, a heat detector or a smokedetector, for example.

In other words, each unit within the system comprises a substantially identical firsthousing containing substantially identical electronics and/or firmware. The behaviour ofthe unit is determined when the connection is made between the first housing componentand the second housing component.

The second housing component may comprise a fascia adapted to engage with the firsthousing component to form a closed housing.

The fascia may have different features depending on the type of unit that is to be formed.For example, all firepoint units may look the same whereas all heat detectors may lookdifferent to the firepoint units, and may all be similar to one another.

The first connector may comprise a membrane connector comprising a flat ribbon cable,although any other type of connector may also be suitable.

If a flat ribbon cable is used, the cable may comprise connections printed onto a flexiblesubstrate.

The cable may comprise a terminal connector which is adapted to engage with thesecond connector. The second connector may comprise a connector mount operativelyconnected to the electronics/firmware.

In one embodiment of the invention, the first connector comprises four connections,although in other embodiments a different number of connections may be present.

In a connector having four connections, up to 16 different combinations of connection arepossible. In other words, by determining which one or more of the connections areconnected in use with corresponding connections in the first unit, 16 different types of unitbehaviours may be provided. A logical zero is produced by not connecting that particular connection, and a logical oneis produced by forming a connection between the connector and a connector to a particularelectronics component.

All connections may be kept close to the terminal end of the fascia in order to reduce thepossibility of electro-magnetic incompatibility problems and to obviate the need for electro-static discharge protection.

Looping the connections to the far end of the fascia brings the lines close to the unitenclosure which could increase the susceptibility of electrical noise and electro-staticdischarge.

All of the units forming the system may have substantially the same first housing holdingthe same electronics for all units in the system. The connector of the fascia will determinewhether the unit is, for example, a firepoint, a smoke/heat detector, or a base unit.

This is an important feature of the invention as claimed since it means that the firstcomponent is common to all devices forming the system.

According to a second aspect of the present invention there is provided a method of testinga wireless emergency system according to the first aspect of the invention comprising thesteps of: causing one of the units to transmit a message to other units in the system;checking that all other units in the system have received the message and checking thatthe strength of each signal received is above a predetermined level.

The method may comprise the further step of setting an attenuation reserve threshold foreach unit; and checking that the signal received by each unit is above the attenuationreserve threshold.

The invention will now be further described by way of example only with reference to theaccompanying drawings in which:

Figure 1 is a schematic representation of an emergency system according to anembodiment of the first aspect of the present invention;

Figure 2 is a flow diagram showing how data is transmitted around the system ofFigure 1;

Figure 3 is a schematic representation of a device in the form of a firepoint formingpart of the system in Figure 1;

Figure 4 is a block diagram of the firepoint device shown in Figure 3;

Figure 5 is a schematic representation of a device in the form of a smoke/heatdetector forming part of the system of Figure 1;

Figure 6 is a block diagram of the detector shown in Figure 5; and

Figure 7 is a schematic representation of a device in the form a detector formingpart of the system of Figure 1;

Figure 8 is a block diagram of the base station shown in Figure 7;

Figure 9 is a block diagram illustrating the test feature of an embodiment of theinvention;

Figure 10 is a flow chart showing how a test procedure would be carried out in anembodiment of the invention;

Figure 11 is a schematic representation showing the inside of a housing suitablefor forming any of the devices illustrated and described hereinabove;

Figure 12 is a schematic representation showing the first connector forming partof the fascia which is adapted to engage with a second connector on a main printedcircuit board contained within the housing of the devices; and

Figure 13 is a detailed representation of the first connector shown in Figure 10;

Figures 14 to 17 are schematic representations showing different types ofconnectors that could be used in a system according to embodiments of the invention.

Referring initially to Figure 1, an emergency system according to an embodiment of thefirst aspect of the invention is designated generally by the reference numeral 2.

The system 2 comprises a plurality of units 4 which are positioned at various locations ona construction site. However it is to be understood that the units could be installed in anydesirable location.

In the illustrated system 2 there are four units identified by the reference numerals 6, 8,Wand 12.

Each of the units comprises a delayer 12 (shown in Figure 2) for setting a randomiseddelay time between either activation of the unit or receipt of data by the unit from anotherunit.

The system 2 adopts a decentralised mesh topology. As such there is no system mapand none of the units 4 know which other units 4 exist in the system.

As long as every unit in the system is connected by at least one sufficiently strong link,the system will function correctly.

It can be seen that some units, in this case unit 8 are connected to more than one otherunit. Unit 8 is connected to two other units 6, 10 by a sufficiently strong link, and to athird unit, 12, by a weaker link illustrated by a broken line.

Due to the plurality of links the system has a degree of robustness.

Turning now to Figure 2, a flow chart shows how data is transmitted around the system2. If one on the units 4 wishes to transmit data regarding an event, the delayer 12forming part of the unit will generate a randomised delay which in this example isbetween 7 and 70 milliseconds in one millisecond steps. A clear channel assessment 14 will then check to see that no other units within range aretransmitting.

If the channel is clear, the data will be transmitted.

In the illustrated embodiment, data will be transmitted five times at random intervals.This is achieved by means of the incremental transmit counts 16, 18. Once the data hasbeen transmitted five times, the transmit count will be cleared at the transmissionprocedure will have finished.

If the clear channel assessment shows that the channel is not clear, the delayer 12 willcause a randomised delay of between 7 and 70 milliseconds in one millisecond steps,and will repeat this process until the clear channel assessment shows that the channel isclear.

Turning now to Figures 3 and 4 a firepoint unit, is illustrated schematically.

The firepoint unit shown in Figure 3 and is illustrated schematically by the referencenumeral 30, the device comprises an antenna 32 fixable inside the housing 20. Thehousing is held together by screws 34. Inside the housing are electronics and firmwareshown in more detail in Figure 4. The firepoint device 30 comprises a pair button 36, asilent test button 38, an alarm LED 40, a fault warning LED 42, a unit activate LED 44, acall point 46 and sounder 48. These components will be described in more detailhereinbelow. The housing 20 also comprises wall mounting points 50 which enable thehousing 20 to be mounted on a wall.

The housing comprises a first housing component 90 (shown in Figure 9) in which theelectronic/firmware is housed, and a second housing component 92 in the form of afascia which is attachable to the first housing unit. In this example, the second housingcomponent (the fascia) is attached to the first housing component by the screws 34.

Turning now specifically to Figure 4, the electronics contained within the first housingcomponent 90 are shown in more detail. The unit comprises a microprocessor 505controlling operation of the unit 30. The microprocessor contains within it the delayer 12and clear channel assessment 14, amongst other components. Other parts of theelectronics will be described herein below with reference also to the electronics formingother units within the system.

Turning now to Figures 5 and 6 a smoke/heat detector is designated generally by thereference numeral 300. Parts of the detector correspond to parts of the firepoint device30 have been given corresponding reference numerals for ease of understanding. Aswell as the common components identified in Figure 5, the detector 300 furthercomprises a smoke/heat detector unit 310. A base station 400 is illustrated schematically in Figures 7 and 8. Parts of the basestation 400 corresponding to parts of the firepoint 30 and the smoke/heat detector 300have been given corresponding reference numerals for ease of understanding.

In addition to the components that are common to one or both of the other devices 30,300 the base station further comprises a LCD display 410, and navigation buttons 420.

Turning now particularly to Figures 4, 6 and 8, it can be seen that each of the three typesof devices 30, 300 and 400 have the same electronics which are designated generally bythe reference numeral 500, and which are held within housing 20.

Set out below are details of the components shown particularly in Figures 4, 6 and 8.

All three types of device also have the same microprocessor 505, battery pack 510, aback tamper 520, a radio transceiver daughterboard 530 as well as antenna 32.

The battery pack 510 comprises a 12 alkaline C-cell battery pack. Due to the intermittentpowering required because the sleep modes of the units, each unit will have a battery lifeof approximately 3 years and will have consistent performance over the entire batteryvoltage range.

This will be achieved by utilising a number of techniques as set out below: 1) Power Supplies - The units use switched mode power supplies instead of linearregulators to improve the efficiency of power conversion from the nominal 6 volt batterypack to provide 3.3 and 24 volt power supply outputs which are constant over the fullbattery range. The 24 voltage power supply output is reduced to 16V for detection unitsand is controlled by firmware.

2) Sleep Modes - Wherever possible the microprocessor and other integrated circuitsare kept in the lowest power modes possible utilising built-in sleep and hibernationmodes as appropriate. 3) Low Power Design - The electronics have been designed to ensure that the normal,quiescent state, of signals are the lowest power state with any alarm or error statesdrawing higher power.

Within the main board electronics that are common to all units within the system there isan internal tamper switch 540, and a detection head tamper (not shown). The internaltamper switch 540 together with the detection head tamper and the back tamper 520which is also common to all units helps to ensure that if any units are tampered with, analarm is sounded.

The internal tamper will be activated when the unit enclosure is opened. This tamper isindicated by a flashing LED on the unit facia and by a radio message to a base station (ifpresent in the system configuration). The internal tamper can be reset only by anauthorised user, and a log of the tamper reset is kept indicating date, time and reset keyfor warranty and auditing purposes.

The tamper reset key is electronic and consists of a pushbutton, potential divider (formedby a resistor network), printed circuit board and a connector. To reset the internaltamper the “key” is mated with a connector on the printed circuit board then thepushbutton pressed. The main unit detects the voltage from the potential divider, if thisvoltage is correct then the unit can be closed and the tamper cleared in the next 30seconds. Two different potential divider outputs have been catered for to allow forsealing by manufacturer or sealing by an end customer to allow for warranty claims to becorrectly processed.

The back tamper 520 detects when a unit has been removed from either a wall (forfirepoints and base stations) or a ceiling (a smoke/heat detectors). This tamper isindicated by a flashing LED on the unit fascia and by a radio message to the base station(if present in the system configuration).

The back tamper can be reset by reaffixing the unit to a wall/ceiling as appropriate.When reaffixed the unit will confirm that it can communicate with at least one othercomponent in the system. If successful the unit will then enter its normal mode of operation. If unsuccessful the unit will indicate that it has a problem with communicationby flashing its fault LED in a defined pattern.

The detection head tamper enables smoke and heat detector units to detect when adetector head has been removed from its mounting. This will be indicated by a flashingLED on the unit fascia and by a message to the base station if present. This tamper canbe reset by replacing a smoke or heat detector in its mounting.

The unit may be placed in a “transit mode” when units are being shipped by key presseson the pair and test buttons e.g. holding down the pair button and pressing the testbutton three times. This ensures that power from the battery is not used up when notrequired. When it is required to activate a unit after shipping all that is required is toholding down the pair button and pressing the test button three times.

The “transit” mode will also disable radio communications allowing the units to be safelytransported by aircraft if required. A unit may be placed again in the transit mode if it is required to return the unit or shipthe unit to another site.

Each unit will be able to indicate visually that it is powered and functioning in order toallow a user to distinguish between a working unit and dead unit. In some embodimentsof the invention each unit will have a green LED light, known as the “Alive” LED, that willflash periodically to indicate that the unit is functioning.

Each firepoint unit comprises a latching call point 550. The latching call point will latchinto a pressed condition and visually indicate that it has been pressed. A latching callpoint can be reset only using a special tool. Any activated call point which has not beenmechanically reset after the fire alarm has ended will enter a ‘chirp’ mode in which thesiren of that particular unit will periodically sound. This is to remind system users that thecall point(s) in question needs to be reset before it can be used to manually raise the firealarm.

Unit numbers may be set over a radio link from a base station. This allows greatersystem flexibility and reusability and obviates the need to open the unit for access.

Each unit will comprise local logs which may be used for debug and warranty purposes.

Base units will retain event logs for all units. Each unit will have sufficient non-volatilememory to allow logging data for at least a complete year to be retained. This will allowfalse alarms to be investigated and also provide details applicable to warranty claims asit will be possible to identify when a unit has been opened, activated or reconfigured.After an internal tamper it will be possible to identify who, or at least which user has resetthe internal tamper.

In systems comprising a base station unit, it will be possible to allow a user to entertelephone numbers via a navigation button keypad in order that text messages can besent from the base station to any of a number of telephone numbers without the need toactivate a fire alarm.

Each unit may be adapted to meet all relevant legislation in the region in which thesystem is being used. At present, all units meet all relevant UK and European legislationincluding: BS EN54-3, 5, 7, 11, Wand 25; BS EN 50130-4;

Set out below is further information relating to the setting up and usage of a system 2illustrated in Figure 1.

Starting and joining a site network

Starting a network

Each unit has a guaranteed unique 32bit number. When the first two units are paired,which is initiated via button presses, the units arbitrate to decide which of the two uniquenumbers will become the site code.

Joining a network

Subsequent units will be added to the network by holding / pressing the PAIR buttons onthe units. When pairing with an existing firepoint or detector unit the new unit added tothe system will take both the unique site code and unit number of the unit it is beingpaired with.

Base station units have additional functionality to allow the unit number to be adjustedprior to pairing.

Joining messages are not relayed to other units in the network and may also use lowertransmit powers to limit the transmission range.

Raising the alarm

Firepoint

Pressing any installed call point in the system will activate the local Sounder andbroadcast a radio message to be picked up by other installed units on the site. There isno delay between activating the call point and the radio message being transmitted,other than that imposed by the communication method outlined above.

Heat / Smoke Detectors

When a heat or smoke alarm is triggered the unit will itself not sound. It will transmit aradio message to all other units in the same way as the firepoint above.

Cancelling a fire alarm

After activation the firepoints will sound for 30 minutes before automatically resetting.

Fire alarms will also be able to be cancelled from any call point which has already beenpressed by resetting the call point with its key. The alarm will not be resettable from anon-activated call point.

Any units with activated call points after either automatic reset or manual reset willperiodically sound (e.g. 1 seconds every 10 minutes) to alert users the call point buttonsneed resetting.

It will not be possible to reset a fire alarm from any detector (smoke or heat).

The alarm can be reset by the base station via a PIN code protected function.

Tampers

As mentioned above, there are three different types of tamper, internal, back anddetector.

Internal tamper

This is activated when the unit housing is opened and as an internal tamper isconsidered to potentially prevent the unit working and can only be reset using anelectronic key on a tamper reset header or tablet PC via USB by an authorised person.

The tamper reset key could allow the ID of the person who has sealed the unit to bestored in the unit. Following reset the unit will provide a 30 second grace period to allowthe unit lid to be correctly closed and secured before re-arming.

In the tamper state the unit will indicate a tamper fault and send a radio message to aGSM base station, if fitted as part of the system.

Back tamper

The back tamper is intended to indicate when a unit has been removed from thewall/ceiling it was fixed to. Removing a unit from the wall/ceiling is not expected to be anissue from a unit integrity viewpoint but could change its ability to communicate to otherunits as it may be moved out of range. The unit will indicate a tamper fault on its ownindicator and send a radio message to a GSM base station, if fitted as part of the system.

When the unit is fixed in place again the unit will check it can still satisfactorilycommunicate with another system component. If successful it will clear the tamper, if notit will indicate a radio link problem.

Detector tamper

The presence of the detector head (smoke or heat) will be monitored. If the head isremoved the unit will indicate a tamper fault on its own indicator and send a radiomessage to a GSM base station, if fitted as part of the system. The tamper fault will becleared when the head is replaced.

Unit numbering

The unit number is only relevant to users who have an optional ‘base station’ unit whichindicates which unit(s) has been activated and which can also send alerts via SMS textmessage to a number of designated people. All units will have a default unit number of 1.

The system will allow a number of units with the same unit number to allow end userflexibility. For example, all units in one area could be numbered the same allowing a fireincident to highlight a particular area of a building(s); equally all units can be numbereddifferently to allow identification of the exact unit which raised the alarm.

Programming the unit number incorrectly will not prevent an alarm being raised butwould make locating the triggering unit difficult.

Silent Tests

Turning now to Figures 9 and 10, the silent test function will be described in more detail.

Figure 9 illustrates a unit 4 forming part of an emergency system 20 and comprising amicroprocessor 505 comprising an A to D converter. The unit 4 may comprise any oneof the devices 30, 300 or 400, for example. If the silent test mode is activated in aparticular unit 4, then a transceiver 910 will send and receive silent test messages, andon reception of signals will measure incoming signal strength.

The microcontroller 505 is operatively connected to tamper switches 915 and the unit'sbattery 920.

The device further comprises a display 925 which in this embodiment comprises aplurality of LEDs.

Turning now to Figure 10, the silent test procedure is shown schematically. Step 1 isindicated by box 1001. The first stage is for a user to press and hold the silent testbutton 900 on any particular unit in order to initiate the silent test procedure.

Next, as shown in box 1002, a silent test message is transmitted to other units 4 in thesystem 2 using a similar radio mechanism to that used when an alarm message is sentout.

As shown in box 1003, if a unit has not heard the silent test message it will continue toindicate a quiescent state as indicated by the LED display 925.

On the other hand, as shown at box 1004, if a unit does hear the silent test message, itwill check the strength of the received signal.

If the received signal strength is not sufficiently above a predetermined level such as anoise floor or an attenuation reserve threshold, the silent test message is discarded, andthe LED display will remain in the quiescent state as shown in box 1005.

On the other hand, if the received signal strength is sufficiently above the predeterminedlevel, then the unit receiving the signal will check its warning inputs to, for example,check that there has been no tampering. This is shown in box 1006.

If no warning exists on the particular unit, it will indicate a silent test pass on the LEDdisplay 925.

On the other hand, if one or more warnings exist on a unit it will indicate on the LEDdisplay 925 that the silent test has failed.

Once the silent test has been carried out, a user may walk around all the units in thesystem and check the indication shown on each of the LED display 925, as shown in box1009.

If all units show that the silent test has been passed, then the system as a whole isconsidered to have passed the silent test, as shown in box 1010.

On the other hand, if any of the units are showing anything other than silent pass on theirrespective LED display 925, the system as a whole is considered to have failed the silenttest as shown at box 1015.

As described above, a silent test function is incorporated into each unit whose purpose itis to exercise the radio communication part of the units and provide visual confirmationthat the unit is able to communicate successfully with other units on the same site. Thesilent test is initiated from any unit via a press on the silent test button. The unit sendsout a silent message (i.e. no Sounder) activation which flashes the alarm LED on eachunit which has successfully received and relayed the radio message. The site can thenbe inspected to check that each and every unit has received the message and thatseparate islands of units have not been inadvertently created by moving units or by thechanging nature of construction sites.

The unit battery capacity has been sized to permit a silent test per week although theactual frequency of this will be determined by the construction site fire officer.

System Components

Firepoint (FP1)

In this embodiment a firepoint unit is wall mounted, battery powered unit consisting ofsounder and call point communicating to other components in the same system via radiointended for indoor and outdoor use within a construction environment. The intended useof the Firepoint is to raise a fire alarm locally and on other radio linked firepoints withinthe same system when its call point is activated.

Battery pack

The unit will utilise a battery pack made up of twelve alkaline C cell batteries in a 4series, 3 parallel combination to provide a 6V, 23Ah battery pack with a polarisedconnector. The connector is polarised and additionally the electronics directly after thebattery will protect against reverse polarity causing damage, no error will be indicated inthe reverse polarity condition.

When not in an alarm condition the battery voltage will be checked periodically. If theremaining battery capacity is less than 30 days of normal operation an alarm will beindicated on the unit by flashing the 'low battery’ LED and transmitting a low batterymessage to be indicated by a base station if it is part of the system. Battery monitoringmay need to be temperature compensated to give a more accurate capacity indicationover the full unit temperature range.

Sounder

The Sounder will be driven from a switched mode power supply to ensure its powersupply is consistent over the complete range of anticipated battery voltages.

Call point

The firepoint will utilise a call point, rated to at least IP33C, with a two wire output froman electrical pushbutton within the unit. The output is normally open and when the callpoint is pushed to raise a fire alarm the contacts are then closed. The call point itself is adirect action, type A call point, which locks into the pressed state which is indicated by aflag in the call point window. The call point can be reset using a plastic key.

Antenna

The antenna for radio communication will extend from the top surface of the unit. Theantenna is fixed from within the enclosure and cannot be removed from the outside. Theantenna will be flexible to reduce the possibility of damage. LED Indicators

The unit will contain three LED indicators, red, amber and green. Due to the nature of theunit these will be visible when directly in front of the unit but may not be visible from allangles.

The LEDS are labelled with their primary functionality when installed on site and in use,they are used for other purposes when pairing and silent testing where the labelling will beinconsistent.

The green LED, may be labelled ‘Active’, and may periodically flash to indicate that theunit is powered and functioning. To conserve power this LED will illuminate for 10milliseconds every 5 seconds.

The red LED, may be labelled ‘Alarm’, and may indicate a fire alarm. The same red LEDmay be used for silent tests and will be on for 10 seconds every second to indicate a silenttest is in progress.

The amber LED, labelled ‘Fault’ will be used to indicate unit faults such as low battery, lowsignal and tamper. These faults may have the following pattern (10 ms on 250 ms off x n)with a 2 seconds gap between groups. This will give a single, double, triple, quadruple etcflash group to indicate which particular type of fault is active. Where two faults aresimultaneously active the highest priority fault will be displayed. ‘Unit tampered’ isconsidered the highest priority, then ‘no radio signal’, ‘low radio signal’ and ‘low battery’.

Buttons

There will two membrane pushbuttons, labelled ‘A’ and ‘B’ on the front face of the unit.These are used to add units to the system, initiate a silent test and also together to entera ‘shipping mode.’

Enclosure

The enclosure will provide at least IP33C protection and will contain all of the unitcomponents. There will be a tamper switch on the inside of the unit to indicate that the unithas been opened by an unauthorised user.

The enclosure lid will be secured to the main body of the unit with 4 Pozidriv® /Phillips®screws.

The enclosure will be wall mounted using two M4 pan head screws, one on either side ofthe unit. These will be exposed when the unit is installed.

Component mounting

The Firepoint should be installed so that its call point is 1.4m above finished floor level.As the Firepoint is a combined unit with a sounder this means that the sounder will be ata level of approximately 1.5m above finished floor level, there does not appear to alocation requirement for sounders.

Smoke / Heat Detector (SD1/HD1)

This may comprise a battery powered unit. The main unit will power and monitor thedetector. The unit is intended for indoor use only.

The unit needs to conform to EN54-7/5 and EN54-25.

Battery pack

The same battery pack as the Firepoint will be used.

Antenna

The antenna for radio communication will be fixed from the inside of the unit and extenddownwards from the unit when mounted on the ceiling, it cannot be removed from theoutside. The antenna will be flexible to reduce the possibility of damage. LED Indication LED indication will be the same as for the Firepoint.

Buttons

Same as Firepoint

Enclosure

The enclosure will provide at least IP22C protection and will contain all of the unitcomponents.

There will be a tamper switch on the inside of the unit to indicate that the unit has beenopened by an unauthorised user.

Component mounting

The unit will be ceiling mounted using 2 M4 pan head screws through the integralmounting points.

Base Station (BS1)

The base station is an optional indoor only unit used to identify which unit(s) has beenactivated and to send out a fire alarm message via SMS (internal GSM modem andantenna) to up to six separate mobile telephone numbers.

Battery pack

The same battery pack as the Firepoint will be used.

Audible alert

The base station will give an audible alert to notify users that it has received a fire alarm.Where multiple alarms are received within quick succession the user will be able to scrollup and down through a list of recently activated alarms.

The base station will also show alarms from units which are indicating other warningssuch as low battery, unit tamper, or poor radio signal. Units with no radio signal will notbe indicated.

Antenna

The antenna for radio communication will extend from the top surface of the unit. Theantenna is fixed from within the enclosure and cannot be removed from the outside. LED Indicators

These will be the same as for the Firepoint and Detector units.

Buttons

Same as the Firepoint and Detector Units with the addition of a 5 button navigation padfor accessing information and changing settings.

Display

The base station will incorporate a 4 line LCD display. This will only be active when analarm or other message has been received. The rest of the time the display will be blankto conserve battery power.

The fascia will provide some impact protection for the display.

Enclosure

The enclosure will provide at least IP22C protection and will contain all of the unitcomponents. There will be a tamper switch on the inside of the unit to indicate that theunit has been opened. (The tamper message may be relayed to a base station if one ispresent).

The enclosure will be wall mounted using two M4 pan head screws, one on either side ofthe unit.

Component mounting

The unit will be wall mounted. There appears to be no specified mounting locations orheights for this unit.

As has been mentioned hereinabove, an important feature of the invention as claimed inembodiments of the invention is that each unit in the system may be housed withinsubstantially the same first housing component which is housed particularly theelectronics and/or firmware. Each component may also have the sameelectronics/firmware, but the behaviour of the unit may be determined by means of aconnection between the fascia and the first housing component. A further advantage of this is that the inside of each first housing component will looksubstantially the same regardless of whether the unit is to be a base station, firepoint etc.

Figure 11 shows a typical layout for the inside of a unit forming part of the system 2.

Figure 11 shows the inside of the first housing component 90. It can be seen that thefirst housing component 90, which has been described hereinabove.

In particular, the first housing component 90 houses the battery pack 510 which is held inposition by battery straps 91.

The electronic components are arranged on a main electronics board 94. The mainelectronics board 94 comprises the second connector 96 which connects to the firstconnector formed in the fascia (second housing component). A connector 98 connectsthe battery back 510 to the main electronics board 94.

The unit also comprises a radio daughterboard 100 having an antenna connection 110for connecting to the antenna 32. The antenna is connected to the first housingcomponent 90 by means of a securing nut 120. The housing component 90 alsocomprises an antenna ground plane 130.

The unit also comprises a back tamper switch cable and connector 140, a tamper resetconnection 150, an internal tamper switch 160, a USB connector 170 and a connector forthe base station electronics 180.

Turning now to Figures 12 to 17, further details of how the first and second connectorsengage with one another is shown in more detail.

In a particular embodiment of the invention, the fascia 92 comprises a first connector inthe form of a ribbon cable connector 200 having a number of individual connectorsnumbered 20 to 21 in Figure 12. Each of these connectors is connectable with acorresponding connector forming part of the second connector 96. The number ofconnectors present on the first connector may be varied so that not all of the connectorsof the second connector engage with a connector on the first connector. This means thatthe electronics components that are active will vary depending on the connections made.

It is thus possible to determine the behaviour/performance of a unit by ensuringappropriate connections.

Although the requirements of various British Standards such as EN54 have beenreferred to within this specification, it is to be understood that the Standards to which thesystem may have to adhere may change from time to time. In addition, in countriesoutside the UK different Standards may have to be met.

Claims (19)

1. A wireless emergency system comprising a plurality of units, each unit comprisinga transceiver adapted to transmit and receive data to/from other units in the system, eachunit comprising a tester comprising a transmitter adapted to transmit a silent message toother units in the system, and an indicator adapted to indicate to a user that a unit hasreceived the silent message and that the strength of the signal received is above apredetermined level, wherein each unit further comprises a delayer for delayingtransmission of data by a predetermined randomised delay time, and a clear channelassessment for checking that no other unit within range of the unit in question istransmitting at that time wherein the delayer is operatively coupled to both the transmitterand the clear channel assessment and the clear channel assessment is operativelycoupled to both the transmitter and the delayer, whereby if the clear channel assessmentdetermines that the channel is not clear, the delayer will delay transmission of data by apredetermined randomised delay time such that the transmitter will transmit only if thechannel is clear.

2. A system according to Claim 1 operating in the radio frequency bandwidth.

3. A system according to any one of the preceding claims wherein each unitcomprises a history buffer.

4. A system according to any one of the preceding claims wherein the transceivercomprises a Wake On Radio component.

5. A system according to any one of the preceding claims wherein each unitcomprises a microprocessor for controlling operation of the unit.

6. A system according to any one of the preceding claims wherein each unitcomprises a monitor for monitoring the signal strength of transmitted and/or received data.

7. A system according to any one of the preceding claims wherein each unitcomprises an attenuation reserve threshold corresponding to the predetermined level.

8. A system according to any one of the preceding claims wherein each unitcomprises a fault warning unit.

9. A system according to any one of the preceding claims comprising a firepointsystem, wherein at least one unit comprises a firepoint device, and at least one other unitcomprises a detector unit adapted to detect smoke and/or heat.

10. A system according to any one of the preceding claims comprising a base unit,which base unit comprises an identifier adapted to identify which of the one or more unitsforming the system has been activated.

11. A system according to any one of the preceding claims comprising an activationalarm for indicating when a unit has been manually activated.

12. A system according to Claim 11 wherein the activation alarm comprises an audiblesignal.

13. A system according to any one of the preceding claims wherein each unitcomprises a transit mode in which the unit is placed into a low power state.

14. A system according to any one of the preceding claims wherein each unitcomprises a back tamper component.

15. A system according to Claim 10 wherein the base unit comprises an internal GSMmodem and antenna.

16. A system according to any one of the Claims 10 or any claim dependent thereonwherein the base unit comprises a display unit.

17. A system according to any one of the preceding claims wherein each unitcomprises a first housing component adapted to receive and contain electronics and/orfirmware, and a second housing compartment comprising a first connector for connectingto the electronics and/or firmware.

18. A system according to Claim 17 wherein the first housing component of each unitcomprises a second connector operatively connected to the electronics and/or firmware,which second connector is adapted to engage with the first connector.

19. A system according to any one of Claims 17 and 18 wherein the second housingcomponent comprises a fascia adapted to engage with the first housing component to forma closed housing.