WO2013135764A1

WO2013135764A1 - Method for operating a network

Info

Publication number: WO2013135764A1
Application number: PCT/EP2013/055126
Authority: WO
Inventors: Thomas Rickes
Original assignee: Robert Bosch Gmbh
Priority date: 2012-03-14
Filing date: 2013-03-13
Publication date: 2013-09-19
Also published as: DE102012203958A1

Abstract

The invention relates to a method for operating a network (2) which has multiple devices (6, 8, 10). Each device (6, 8, 10) of the network (2) is assigned a serial number (16, 18, 20) which comprises n bit positions (22, 24, 26, 28), and the devices (6, 8, 10) are identified in identification rounds using the device serial number (16, 18, 20). According to the invention, - bit values for m of the n bit positions (22, 24, 26, 28) of the serial numbers (16, 18, 20) of the device (6, 8, 10) are checked for a maximum of m checks in an identification round, wherein during a kth check of a kth bit position, solely the at least one device (6, 8, 10) which was authorized to participate in the kth check during a previous k-1th check participates, - the at least one device (6, 8, 10) authorized to participate in the kth check is then queried as to whether the kth bit position (22, 24, 26, 28) of said device has one of two possible bit values, and - the device (6, 8, 10) which has said one of two possible bit values in the kth bit position (22, 24, 26, 28) is authorized to participate in a subsequent k+1th check.

Description

description

title

Method for operating a network The invention relates to a method and a control element for operating a network

Network.

PRIOR ART Fire alarm systems generally include a fire alarm control panel with a

Control element, which is connected via a network designed as a bus network with at least one fire alarm device, wherein a bus protocol is used for communication. The network is often looped back to the point of origin, starting from the fire panel, for redundancy in the event of a short circuit. Therefore, a network section as

Be called loop.

In a common existing protocol, each fire detector as a device has its own address, usually between 1 and 255, allowing the control element to communicate with each device point-to-point.

It is also contemplated that each device will receive its address at a setup by programming it with a so-called programming tool, each device storing an address in a nonvolatile memory.

The programming involved in the installation is performed on each device before it is used for the first time within the network. If more than 100 devices are used in a loop, programming the devices prior to installation is expensive. Therefore, the programming can be performed automatically considering the serial numbers of the devices. Serial numbers often include production characteristics such as date of manufacture and place of manufacture and can therefore be quite long. Should serial numbers record a dataset, the z. B. encoded in 64 bits, results in a number of 2 ⁶⁴ individual serial numbers. Due to this large number, the problem arises of finding a single device with a particular serial number within a reasonable time frame.

The document DE 102 40 832 A1 describes a bus which has one master and several slaves. In the bus, an addressing of the connected slaves is carried out on the basis of the serial number, with a partial feature being interrogated by the master to the individual slaves. Only if a single slave responds to the master's question does this slave receive an address. If several slaves or no slave answers, the question is refined.

Disclosure of the invention

Against this background, a method and a control element with the features of the independent claims are presented. Further embodiments of the invention will become apparent from the dependent claims and the description.

With the presented method, it is no longer necessary to manually program a communication address for a device, wherein an automatic address mapping or an automatic address assignment is performed.

In one embodiment of the method according to the invention, a time-efficient search algorithm for searching for a separate, individual serial number for a device is performed. Furthermore, after an identification round for a device, an address is automatically assigned to it by a typically central control element and thus assigned. In this case, in an embodiment, each new device that is connected to a network, searched by the control element, being found by the control element in each identification round usually a device. Commands, for example bus commands sent by the central control element, can be understood by all devices on the network formed as a bus, for example. There are commands that are directed to individual devices and those that are directed to groups of devices. The method uses a

Property of networks, which makes it possible to detect in a group inquiry the existence of multiple responses of devices that overlap on the network, even if the individual bit pattern of a device can no longer be discriminated.

In one possible embodiment of the method, bit positions of serial numbers are initially determined, which have bits that match all devices of a loop as a group of devices within the network and can be defined as constant.

In doing so, at least one query is carried out by first erasing or clearing a search mask for serial numbers for each device to be searched for. The search mask comprises in design for each bit position to be checked, on the one hand, the information as to whether a k-th bit position should be taken into account in the search, and, on the other hand, the information with which of the two possible bit values the k-th bit position should be filtered. Each device locally maintains such a search mask that can be modified by commands from the central control element. This will send a command to clear all bits of the search mask. As a result, the search mask for the serial numbers is completely reset for every device that does not yet have a valid bus address. In this case, the search mask is used to check, for each k-th bit position of the serial number to be considered, whether there are devices which have a bit value to be filtered, for example 0 or 1, in their serial number at the specified k-th bit position.

The control element sends the command for setting and querying a bit position of the search mask at a k-th check for a k-th bit position of the serial number, once for the first and once for the second bit value. Any device where the kth bit position of the serial number matches the set value of the requested bit value search form sends a response to that command. Devices in which the bit value of the k-th bit position deviates from the requested bit value does not respond. The content of the answer is irrelevant to the control element. It only has to determine if there is no response or at least one response from at least one device and thus possibly multiple simultaneous responses.

If there are answers in both cases, this bit position is not constant in the quantity of the serial numbers of the devices, usually the devices of the group. If there is at least one answer in one case and not in another case, it can be stated that this bit position is the same and therefore constant for all devices, usually for all devices of the group. If there is no response in both cases, uninitialized devices will no longer be on the bus. Constant bit positions are memorized by the control element and can always be skipped in the further procedure for speeding up the method, since within the set of serial numbers, content-identical bit positions do not provide a distinguishing criterion. The control even knows after these queries the bit value at the constant bit positions.

After checking constant bit positions, the command for resetting or deleting the search masks is sent. Thus, only the property of a single bit position is successively checked. That is, each device only checks the match of a single bit position of its serial number with the requested bit value. A bit position other than the bit position being queried is of no significance.

As part of an identification round, the command for setting and querying a k-th selected bit position is sent in a k-th check and thus query, which in design with the highest, non-constant bit position of the search masks of the devices and the first bit value started who - that can. If no responses are obtained, it is assumed that there are no unassigned devices in a loop of the network whose k-th bit position of the serial number has the first bit value. Thereafter, the command for setting and retrieving the k-th bit position for the second bit value is sent. If no response is received either, there are no unassigned devices in the loop whose kth bit position tion is occupied, so that the search process is already completed at the k-th bit position.

However, if for any one of the possible bit values at the k th bit position check, a response sent from at least one device is received, the search continues at the next k + 1 th bit position and so on until the control receives a response from receives a device for the last mth bit position. When a device receives the command to set and poll for a bit position, that device locks the searched bit position for the bit value searched for. This means that for other search commands others

Bit positions are taken into account with the bit values after which the device was previously requested. This applies as long as the search mask is not completely, usually only partially, reset. The kth bit position is handled as soon as the first response from at least one

Device is obtained. That is, when a response to the request for the first bit value at the k th bit position is obtained, the second bit value is not requested, but immediately advanced to the next k-1 th bit position. Only if there is no response to the query for the first bit value, the second bit value for the k-th bit position is also queried. In case that both

If no response is received, it can be concluded that there are no more devices to process and the search is complete.

That device which gives an answer to the command for setting and querying the kth bit position switches or sets an internal flag and thus a marking. Only such a device is ready to receive the subsequent command for setting a temporary bus address. Due to the procedure described above, only one device is identified in an identification round since each device has a serial number with its own bit pattern.

By running through all bit positions, the number of fixed bit positions within the device-internal search masks is refined. Once all the bit positions to be handled are determined, it is ensured that only the serial number of a single device matches the search mask. When addressing devices, the control element assigns a temporary bus address to a single device if a single device is found during an identification round. By assigning an address, the device no longer belongs to the addressless devices that participate in responses to search commands and ignores the search commands. Thereafter, a re-identification round can be performed, whereby another device is taken from the set of addressless devices. This will continue until all devices have received addresses.

A new device, which can be designed as a sensor, for example for the detection of smoke and / or fire, can be connected to the network while a system is in operation, whereby an address can be assigned to this new device only a few seconds later , Thus, devices can be replaced without much effort. If desired, automatically assigned addresses, for example via a network interface, can be changed later. In the methods for operating a network having multiple devices, it is contemplated that each device is assigned a serial number with n bit positions. The devices are searched and identified one after the other in identification rounds on the basis of their serial numbers, whereby in each of the identification rounds one device is usually identified. In an identification round, bit values for m of the n bit positions of the serial numbers of the devices are checked for a maximum of m checks. In the case of a k-th check of a k-th bit position, only that at least one device that has been authorized to participate in the k th check in a previous k-1-th check participates. At a first check (k = 1) that at least one device participates, which is to be identified.

In a k-th check, the at least one device participating in the k-th checking is asked whether its k-th bit position has a first of two possible bit values. If, for at least one of the devices, the k-th bit position has the first bit value, this becomes at least one device for

Entitled to participate in a subsequent k + 1-th review, and this next k + 1-th review carried out. If none of the devices participating in the kth verification has the kth bit position the first bit value, the at least one device participating in the kth verification is asked if its kth bit position has a second bit value If, for at least one of the devices, the kth bit position has the second bit value, this at least one device is authorized to participate in a subsequent k + 1th check, and the next k + 1th check is performed. The device is identified by its serial number during a mth check. At least one device participating in the kth verification is entitled to participate in a subsequent k + 1th validation if its kth bit position has one of two possible bit values.

Hereinafter, individual steps of a possible embodiment for finding a single device by dicing will be described. The coding of the commands shown here is arbitrary or arbitrary and proves that the commands can be encoded in the specified number of bytes. The instructions each comprise three bytes, the first byte being a command identifier, the second byte being a parameter, and the third byte being a checksum relating to the first two bytes. Not all bits of the parameter are always used for these methods, resulting in further functional possibilities. If the transmission of a command due to a communication failure in a device disturbed, for example, received changed, so this can be discarded.

A reset of the search mask for all devices is performed with the query <0x01> <0xC0> <CS> of the control element. Then the query <0x02> <0xBF> <CS> searches for devices whose k-th bit position of the serial number is 0. Bits 0 to 5 of the parameter comprise the position, bit 6 comprises information on whether a bit value of the search mask to be checked should be set to 0 or 1 (here 0), bit 7 is constant 1.

If a device is found, it replies with <0x00> <0xFF>,

otherwise there is no response and it is checked whether the kt bit position of the serial number has the bit value 1. The bit positions will be decoded until the last, mth bit position is reached. If a response is received, the next k + 1th bit position is checked.

It is envisaged that after the last, mth bit position of the serial number last searched, since then the devices for the command 0x03, the setting of a temporary address, are ready to receive. As a rule, there must be a single device to be found. Thus, all other previously checked bit positions are locked in advance. For temporary or temporary assignment of a new bus address for a device discovered during an identification round, the control element provides the query <0x03> <0xFF> <new bus address> <CS>, whereupon the device with <0xFF> <new bus address> < CS> answers what the control element confirms with <0x04> <CS>. The control element has thus recognized that the device has received the correct address and confirms that this address may now be accepted. From then on, the device has a valid address and will not participate in any further described device identification steps because devices with a valid address ignore the search commands described above.

After performing identification of the devices and assigning the new addresses, the control element may issue further commands using the new ones

Send bus addresses to the devices. However, after shutting down a system comprising the devices and the controller, the temporary bus addresses are lost. Therefore, by addressing via the temporary address in a further step, a permanent address can also be allocated which is saved in a non-volatile memory of the device. In this case, the device automatically resumes this address after restarting the system and therefore no longer participates in a further automatic address assignment. The order of processing the bit positions is for the operation of the

Procedure irrelevant. It is thus possible to determine the sequence of bit positions in any permutation. In an embodiment, however, it is possible to start with a highest, for example, 63rd bit position, and finally to conclude by decrementing k values, a last, zeroth bit position is checked. The method may be performed for a plant comprising a network with the typically centralized control element and spatially distributed devices connected together. The system can be provided as a security system, such as. A fire alarm system for a building. In this case, the control element is designed as a control or control center. The devices can be configured as sensors, for example video cameras or, in the case of a fire alarm system, as fire detectors. Regardless, the method for searching for devices and assigning addresses to devices for various network-type devices may be performed with a control element performing a function of a master and a number of devices performing functions of slaves.

Whether a device or multiple devices respond during a scan is irrelevant to the process. The control only recognizes that at least one device is responding. However, it is possible to measure a response current, the z. B. 20 mA per device is. By monitoring the voltage, the devices can prevent too many devices from responding at once; H. if 200 devices wanted to pull 20 mA, the voltage could break down, causing some devices to cancel their response immediately. This avoids a complete breakdown of the network. In combination, there remains a measurable summation current at which the control element detects the presence of at least one response or no response.

The control element according to the invention is designed to carry out all steps of the presented method. In this case, individual steps of this method can also be carried out by individual components of the control element. Furthermore, functions of the control element or functions of individual components of the control element can be implemented as steps of the method. In addition, it is possible that steps of the method are realized as functions of at least one component of the control element or the entire control element.

Further advantages and embodiments of the invention will become apparent from the description and the accompanying drawings. It is understood that the features mentioned above and those yet to be explained below can be used not only in the particular combination indicated, but also in other combinations or in isolation, without departing from the scope of the present invention.

Short description of the drawing

Figure 1 shows a schematic representation of an embodiment of the control element according to the invention for an example of a network in carrying out an embodiment of the method according to the invention.

Figure 2 shows a first Flussidagramm to the embodiment of the inventive method.

FIG. 3 shows a second flow chart relating to the embodiment of the method according to the invention.

Figure 4 shows a third Flussidagramm to the embodiment of the inventive method.

Embodiment of the invention

The invention is schematically illustrated by means of an embodiment in the drawings and will be described in detail below with reference to the drawings.

FIG. 1 shows, in a schematic representation, a network 2 which comprises an embodiment of a control element 4 according to the invention. Furthermore, the network 2 comprises a plurality of devices 6, 8, 10, wherein only three of the devices 6, 8, 10 are shown in FIG. The control element 4 is here connected via a trained as a bus line connection 12 with the devices 6, 8, 10. In addition, in the example of the network 2 shown here, the devices 6, 8, 10 are connected to one another via connections 12.

The network 2 shown by way of example in FIG. 1 may be loop-shaped, the devices 6, 8, 10 being connected in series. In this case For example, the network 2 shown in FIG. 1 may also be formed as a loop of a larger, superordinate network, the devices 6, 8, 10 in the loop being referred to as a group of devices 6, 8, 10. However, the embodiment of the method according to the invention that can be carried out by the control element 4, which is described with reference to FIG. 1 and the following FIGS. 2, 3 and 4, can also be used for other configurations of networks 2, for example for star-shaped networks, within which devices 6, 8 , 10 with each other and / or with the control element 4 differently, as shown in Figure 1 explicitly shown, are connected, performed.

Independently of the embodiment of the method according to the invention described here, the usually central control element 4 is designed to control and thus to control and / or regulate functions of the devices 6, 8, 10 within the network 2. Accordingly, the control element 4 within the network 2, the function of the so-called master assigned, whereas the devices 6, 8, 10 are assigned within the network 2 functions of so-called slaves.

It is furthermore provided that each of the devices 6, 8, 10 has a memory 14 in which, in addition to other data, a respective serial number 16, 18, 20 of the device 6, 8, 10 is stored. In this case, each serial number 16, 18, 20 of each device 6, 8, 10 has n bit positions 22, 24, 26, 28, wherein in FIG. 1 of each serial number 16, 18, 20 only four bit positions 22, 24, 26, 28 are shown. In this regard, FIG. 1 shows in detail a highest, η-1 th bit position 22, a second highest, η-2 th bit position 24, a third highest η-3 th bit position 26 and a lowest, 0 th bit position 28.

The illustrated in Figure 2 first Flussidagramm to the embodiment of the inventive method represents a flow of an identification round 38 for identifying and / or searching a device 6, 8, 10 by the control element 4. It is provided at a start 40 that a sequence, according to which the n bit positions 22, 24, 26, 28 of the serial numbers 16, 18, 20 of the devices 6, 8, 10 are checked here. When reset 42, search masks of devices 6, 8, 10 are reset. After that, a counter is set to k = 1 during a first check and a first bit value check 44 is carried out for a first bit position 22, 24, 26, 28, wherein each device 6, 8, 10 participating in the identification round 38 is asked whether its first bit position 22, 24, 26, 28 of the serial number 16 to be checked 18, 20 has a first of two possible bit values. If this is the case for at least one of the devices 6, 8, 10 (yes 46), this at least one device 6, 8, 10 will have a subscription authorization 48 for a subsequent, in this case second check, of a second bit position 22, 24, 26, 28 assigned. If none of the participating devices has the first bit value at its first checked bit position (no 50), a second bit value check 52 is performed for all devices 6, 8, 10 for the first bit position to be checked, checking to see if the bit position has a second bit value , Each device 6, 8, 10, which responds positively (yes 46) receives a subscription 48 for checking the next bit position 22, 24, 26, 28. If it should be provided that no device 6, 8, 10, the first Bit position of the serial number 16, 18, 20 has the second bit value (no 50), the identification round 38 is completed (end 54).

After each check carried out with the final allocation of a subscription authorization 48 for at least one device 6, 8, 10, a meter check 56 is carried out. In this case, it is checked whether the performed k-th check for the k-th bit position 22, 24, 26, 28 was already the mth and thus last check for an mth and thus last bit position 22, 24, 26, 28 , Accordingly, it is checked whether the counter is still k <m. If this is the case (yes 46), an incrementation 58 of the counter is performed by 1, with the counter k being increased to k + 1. After that, after the kth check of the k-th bit position

22, 24, 26, 28 the k + 1-th check for the k + 1-th bit position 22, 24, 26, 28 of the serial numbers 16, 18, 20 performed. At a next k + 1-th, here second, checking a k + 1-th or second bit position 22, 24, 26, 28, only that at least one device 6, 8, 10 participates whose previously checked k-th ( first) bit position has either the first bit value or the second bit value and has received a subscription authorization 48 for the k + 1-th verification.

As part of the identification round 38, a total of m of n bit positions 22, 24, 26, 28 are checked by a first bit value check 44 and, if necessary, by a second bit value check 52. Usually, everyone gets to

Participation in a k th review entitled device 6, 8, 10, whereby the sub- is issued at the previous k-1th check, then asked if its kth bit position 22, 24, 26, 28 has one of two possible bit values. In this case, a kth check is followed by a kth + 1st check and so on. After each check carried out with the final assignment of a subscription 48 for at least one device 16, 18, 20, a meter check 56 is carried out. In this case, it is checked whether the performed kth check for the kth bit position 22, 24, 26, 28 was already the mth check for the mth and thus last bit position 22, 24, 26, 28. Accordingly, it is checked if k <m. If this is the case (yes 46), an incrementation 58 of the counter is performed by 1, wherein the counter k is increased to k + 1. Thereafter, after the kth check of the kth bit position 22, 24, 26, 28, the k + 1th check for the k + 1th bit position 22, 24, 26, 28 of the serial numbers 16, 18, 20 is performed , If the counter check 56 proves (no 50) that a final state 60 is reached, which means that k = m, an identification 62 is completed for the device 6, 8, 10 to be identified during the identification round 38 that has been carried out a detection of m to be checked bit positions 22, 24, 26, 28 of the serial number 16, 18, 20 of this device 6, 8, 10 is performed, said device 6, 8, 10 based on its now recognized serial number

16, 18, 20 is identified. Thereafter, the identification round 38 for this device 6, 8, 10 ended (end 54).

It is intended to carry out an identification round 38 for each device 6, 8, 10 to be identified. In this case, only one device 6, 8, 10 is identified on the basis of the serial number 16, 18, 20 in each identification round 38, since each serial number 16, 18, 20 is assigned only once. In each identification round 38, m checks are performed for m bit positions 22, 24, 26, 28 of the serial numbers 16, 18, 20. A maximum of two devices 6, 8, 10 take part in the last, mth check of the mth bit position 22, 24, 26, 28, to which in all previous m-1 checks a right to participate 48 in the respective subsequent check is granted has been. A device which has not been given the right to participate 48 after a k th verification is excluded from all m - k subsequent checks of the identification round 38. Before carrying out a first identification round for a first device 6, 8, 10 to be identified, it can be determined in an embodiment of the described method which c bit positions 22, 24, 26, 28 of the total n bit positions 22, 24, 26, 28 of FIG Serial numbers 16, 18, 20 of all devices 6, 8, 10 are constant. Such determination of constant serial numbers is illustrated by the second flowchart of FIG.

Here, at a start 70, it is determined according to which order the total of n bit positions 22, 24, 26, 28 are to be checked one after the other, wherein a counter is set to k = 1, so that first a first bit position is checked according to the sequence. In this case, when checking the first bit position 22, 24, 26, 28 for all devices 6, 8, 10, a first bit value check 72 is carried out, with all devices 6, 8, 10 being asked whether their first bit position 22, 24, 26 , 28 has the first of two possible bit values. Immediately thereafter, a second bit value check is performed for all devices 6, 8, 10

74 for the first bit position 22, 24, 26, 28 to be checked here, in which all devices 6, 8, 10 are asked whether their first bit position 22, 24, 26, 28 has a second bit value. It is envisaged that this will only be checked, whether at each performed

Bit value check 72, 74, d. H. in the first bit value check 72 and the second bit value check 74 no device 6, 8, 10 or at least one device 6, 8, 10 has responded. In a decision step 76 it is decided whether the checked kth, here first bit position 22, 24, 26, 28, is constant or not. In this case, the checked k-th bit position 22, 24, 26, 28 is constant, if only one of the two bit value checks 72, 74, d. H. either at the first bit value check 72 or only at the second bit value check 74, at least one device 6, 8, 10 has responded and none of the devices 6, 8, 10 has responded to the other of the two bit value checks 72, 74. If at least one device 6, 8, 10 has answered in each of the two bit value checks 72, 74, the checked k th bit position is

22, 24, 26, 28 not constant.

After decision 76, a counter check 78 is performed, checking to see if k <n. Thus, it can be reconstructed whether the last, nth bit position has already been reached. If k = n (no 80), the end 82 is reached, so that all n bit positions 22, 24, 26, 28 of the serial numbers 16, 18, 20 of all 6, 8, 10 have been checked to see if they are constant or not. If the counter check shows 78 (yes 84) that k <n, an incrementation 86 is performed for the counter, the counter being incremented from k to k + 1. Subsequently, after the already checked k-th bit position 22, 24, 26, 28, a k + 1-th bit position 22, 24, 26, 28 is checked to see whether it is constant for all devices 6, 8, 10 or not. Thus, at each check for each of the n bit positions 22, 24, 26, 28, a first bit value check 72 and a second bit value check 74 are performed and then a decision 76 is made.

A general method for operating the network 2 with a plurality of devices 6, 8, 10 is illustrated by the third flow diagram of FIG. In this case, after a start 90 in a constantity check 92, it is checked which c of the total of n bit positions 22, 24, 26, 28 of the serial numbers 16, 18, 20 are constant for all devices 6, 8, 10. Details of this constant checking 92 are shown in FIG. In at least one identification round 94 to be performed, as is also shown in detail with reference to the flow chart from FIG. 2, only those m = n-c bit positions 22, 24, 26, 28 which are not constant are checked. Constant bit positions 22, 24, 26, 28 can be skipped. After carrying out at least one identification round 94 for at least one device 6, 8, 10, wherein a device 6, 8, 10 is identified on the basis of its serial number 16, 18, 20 in each conducted identification round 94, this identified device 6, 8, 10 is included an address 96 associated with an address. Thereafter, a continuation test 98 checks whether further devices 6, 8, 10 are to be identified and addressed. Is that the case

(Yes 100), an identification round 94 is performed for a next device 6, 8, 10 to be identified. If this is not the case (no 102), the end 104 has been reached and an identification and addressing of all devices 6, 8, 10 to be identified and addressed are concluded.

A search mask associated with a device 6, 8, 10 comprises a search value and a search filter, both of which are 64 bits long. The search filter determines that only bit values with a filter bit set to 1 are taken into account during the search. The search value determines with which bit position 22, 24, 26, 28 of the serial number 16, 18, 20 in the search a match must be achieved. A search command specifies the bit position and the bit value. Of the Bit value, for example 0 or 1, comes in the search value at bit position 22, 24, 26, 28 and replaces the previous bit value at this bit position 22, 24, 26, 28. At bit position 22, 24, 26, 28 is written in Search filter sets the filter bit to 1. The search command then returns an answer if (search value & search filter) = (serial number & search filter). "&" here corresponds to a bitwise logical "And". The bit value can thus be set to 0 or 1 in succession, but the search filter is only reset as a whole in the above method. Therefore, in the course of time, more and more bit positions 22, 24, 26, 28 are taken into account in the identification rounds 38, 94 to be performed.

A list of constant bit positions 22, 24, 26, 28 may also be used as bit patterns, e.g. A 1 at each point, which is constant. However, in a comparison of the internal serial number 16, 18, 20, which is read out after addressing, the search mask built up during the search and / or serial number 16, 18, 20 can be used. A comparison can be used to verify that the procedure has gone through properly. Therefore, it is also possible to retrieve the serial number 16, 18, 20 again from the device 6, 8, 10, instead of deriving only from the search mask of the control element 4. If commands are transmitted disturbed, there may be differences. In contrast, a multiple transmission of search commands can be made, but this costs transmission time. If more than one device 6, 8, 10 remains at the end, that is probably what happened. In this case, the identification round 38, 94 can be ignored and repeated. If, despite errors, a single device 6, 8, 10 with an unexpected serial number has been identified, this can also be ignored, since the goal of singling has nevertheless been achieved.

As a rule, at least one device 6, 8, 10, which is to be identified in an identification round 38, 94, is assigned an internal search mask with at least one placeholder before carrying out the checks. In this case, this at least one placeholder after a k-th review on the eligibility 48 is assigned a mark, is specified for this at least one device 6, 8, 10, whether it is entitled to participate in the subsequent k + 1-th review is or not. The at least one placeholder is assigned a positive mark at the k th check on the eligibility 48, if the at least one device 6, 8, 10 is authorized to participate in the subsequent k + 1 th check. The at least one placeholder is assigned a negative mark in the kth check if the at least one device 6, 8, 10 is not entitled to participate in the subsequent k + 1th check, which means that the eligibility 48 is denied.

In a first check of an identification round 38, 94, the highest of the m of n bit positions 22, 24, 26, 28 is checked. For each subsequent, hard check, the bit position 22, 24, 26, 28 is decremented by one. Alternatively, in the first check, the lowest of the m of n bit positions 22, 24, 26, 28 is checked, with the bit position 22, 24, 26, 28 being incremented by one at each subsequent k th check,

Before a, for example, first, identification round 38, 94 to be performed, it is determined which m of n bit positions of the serial numbers 16, 18, 20 of the devices 6, 8, 10 are checked, whereby it is determined which c bit positions 22, 24, 26, 28 for all devices 6, 8, 10 have the same bit values, wherein in the m = n - c checks of an identification round 38, 94 only those m bit positions 22, 24, 26, 28 are checked, which are valid for all devices 6, 8, 10 have different bit values. Thus, in an identification round 38, 94 to be performed constant bit positions 22, 24, 26, 28 are skipped. Such constant checking 92 may be performed before each identification round 38, 94 as needed. A device 6, 8, 10 identified in an identification round 38, 94 is excluded at each subsequent identification round 38, 94.

Claims

A method of operating a network (2) comprising a plurality of devices (6, 8, 10), wherein it is provided that each device (6, 8, 10) of the network has a serial number (16, 18, 20), the n bit positions (22, 24, 26, 28), the devices (6, 8, 10) being identified in identification rounds (38, 94) by their serial numbers (16, 18, 20),

wherein bit values for m of the n bit positions (22, 24, 26, 28) of the serial numbers (16, 18, 20) of the devices (6, 8, 10) are checked in an identification round (38, 94) for a maximum of m checks, wherein at a k-th check of a k-th bit position (22, 24, 26, 28) only that at least one device (6, 8, 10) participates, which in a previous k-1-th check to participate in the k th review was authorized,

wherein the at least one device (6, 8, 10) authorized to participate in the kth check is asked whether its kth bit position (22, 24, 26, 28) has one of two possible bit values,

- whereby that device (6, 8, 10) is authorized to participate in a subsequent k + 1-th verification whose k-th bit position (22, 24, 26, 28) has this one of two possible bit values.

Method according to Claim 1, in which the at least one device (6, 8, 10) authorized to participate in the kth check is asked whether its kth bit position has a first of two possible bit values,

- In the event that for at least one of the devices (6, 8, 10), the k-th bit position (22, 24, 26, 28) has the first bit value, this at least one device (6, 8, 10) The eligibility of (48) to participate in a subsequent k + 1th review and that this next k + 1th review will be carried out;

in the case that for none of the devices (6, 8, 10) participating in the k-th verification, the k-th bit position (22, 24, 26, 28) has the first bit value, which is at least one at the k asking whether the kth bit position (22, 24, 26, 28) has a second bit value, - In the event that for at least one of the devices (6, 8, 10), the k-th bit position (22, 24, 26, 28) has the second bit value, this at least one device (6, 8, 10) Eligibility (48) to participate in a subsequent k + 1 review, with this next k + 1 review being conducted.

Method according to Claim 1 or 2, in each of which one device (6, 8, 10) is identified in one of the identification rounds (38, 94), wherein in a first check of an identification round (38, 94) that at least one device (6, 8, 10) that is to be identified, and in which the at least one device (6, 8, 10) to be identified is identified on the basis of its serial number (16, 18, 20) during a last mth check.

Method according to one of the preceding claims, in which at least one device (6, 8, 10) to be identified in an identification round (38, 94) is assigned an internal search mask with at least one placeholder before performing the checks, provided is to assign to this at least one placeholder, after a k th verification, a mark indicating to that at least one device (6, 8, 10) whether it is entitled to participate in the subsequent k + 1 review or Not.

Method according to Claim 4, in which at least one placeholder is assigned a positive marking in the case of the k-th verification if the at least one device (6, 8, 10) is entitled to participate in the subsequent k + by issuing a subscription authorization (48) 1), and / or in which at least one placeholder is assigned a negative mark in the k-th verification if the at least one device (6, 8, 10) does not participate in the subsequent k + 1 -tested review.

Method according to one of the preceding claims, in which a start (40) of an identification round (38, 94) is decided on an order according to which the m bit positions (22, 24, 26, 28) are checked successively.

7. Method according to one of the preceding claims, in which it is determined before an identification round (38, 94) to be carried out which m of n bit positions (22, 24, 26, 28) of the serial numbers (16, 18, 20) of the devices (6 , 8, 10), wherein it is determined which c bit positions (22, 24, 26, 28) have the same bit values for all devices (6, 8, 10), wherein in the m = n - c checks of an identification round ( 38, 94) only those m bit positions (22, 24, 26, 28) are checked which have different bit values for all devices (6, 8, 10).

A method according to any one of the preceding claims, in which that device (6, 8, 10) identified in an identification round (38, 94) is assigned an address valid within the network (2).

9. The method according to any one of the preceding claims, wherein a device (6, 8, 10), which is identified in an identification round (38, 94) is excluded in subsequent identification rounds (38, 94).

A control element for operating a network (2) which is connected to devices (6, 8, 10) of the network (2) and communicates via the network (2) with these devices (6, 8, 10), wherein the control element (4) is adapted to perform at least one step of a method according to any one of the preceding claims.