GB2401228A

GB2401228A - Data transmission

Info

Publication number: GB2401228A
Application number: GB0309992A
Authority: GB
Inventors: Sean Treen; Brendan Sleight
Original assignee: TRANSP FOR LONDON
Current assignee: TRANSP FOR LONDON
Priority date: 2003-04-30
Filing date: 2003-04-30
Publication date: 2004-11-03
Anticipated expiration: 2023-04-30
Also published as: GB2401228B; GB2401228B8; GB2401228A8

Abstract

Data is communicated between a central control unit that generates a sequence of periodic control messages and at least one traffic signal controller or sensor that is arranged to respond to the sequence of periodic control messages by generating a series of periodic reply messages. A first control message packet is received over the digital packet based network at a first control message arrival time. The packet contains data for a first control message and the data for the first control message is stored. A first periodic control message is output to the or each traffic signal controller or sensor at a delay after the arrival time corresponding to a predetermined buffer period. A second control message packet containing data for a second control message is received over the digital packet based network at a second control message arrival time and the data for the second control message is stored. A second periodic control message is output at a delay after outputting of a first control message based on a predetermined periodic interval.

Description

240 1 228 Data Transmission System and Method The present invention

relates to a system and method for the remote control of a distributed system and, in particular but not exclusively to a system and method for the remote control of traffic signals.

To facilitate the co-ordination of elements of a distributed system, such as traffic controllers or signals, it is advantageous to provide a central control system which controls and co-ordinates each of the individual components of the system. Such systems may be used, for example, to ensure a continuous flow of traffic around an urban area. For example, the system may allow cars travelling along a main road or a busy road to encounter a series of green traffic lights to allow the traffic to keep flowing along the road.

Such control by a central system has been implemented, as described in more detail below, by creating a network of physical connections, for example copper leased line connections, to each element of the distributed system and controlling the system through a series of control signals and response signals transmitted at regular intervals. In a traffic signal control system, for central control of the system to be maintained, messages must be sent and received between the central computer and the outstation equipment at a predefined periodic interval. If control messages are not received at the periodic interval, which is defined by standards in the United Kingdom as a period of one second, then the outstation equipment reverts to local control. The copper leased line connections can provide means for transmitting the control and response signals reliably and at the required interval. However, the system is inflexible since the physical connection must be moved if the control device for the traffic signal is moved. Also a new physical connection must be created for any traffic signal to be added to the system. For many systems, it may also be necessary to rely on a third party to implement and maintain the physical network. In addition, for redundancy to be introduced into the system, a second physical connection must be connected to each traffic signal control site.

For messages to be sent reliably over wireless transmission means using the existing technology, it would be necessary to maintain an always-open connection over the wireless link. However, maintaining such a connection would be inefficient, since it would require a large amount of bandwidth. Packet based networks, either over a physical connection or over a wireless link, require the use of less bandwidth, but are inherently unreliable and the delivery of control and reply messages to the components of the system in sequence and at the required periodic interval cannot be guaranteed or even expected. Hence prior art packet based networks are inherently unsuitable for controlling a system that requires control and reply messages to be transmitted at a predefined periodic interval.

According to a first aspect, there is provided a method of communicating data via a digital packet based network between a central control unit that generates a sequence of periodic control messages and at least one traffic signal controller or sensor that is arranged to respond to the sequence of periodic control messages by generating a series of periodic reply messages, the method comprising: receiving a first control message packet containing data for a first control message over the digital packet based network at a first control message arrival time; storing the data for the first control message; outputting a first periodic control message to the or each traffic signal controller or sensor at a delay after the arrival time corresponding to a predetermined buffer period; receiving a second control message packet containing data for a second control message over the digital packet based network at a second control message arrival time; storing the data for the second control message; and outputting a second periodic control message at a delay after outputting of a first control message based on a predetermined periodic interval.

This may allow messages to be received from a control unit over the digital packet based network, which may not provide reliable transmission of messages at predetermined intervals, buffered and transmitted to the traffic signal controller or sensor at a regular predetermined interval. Storing the first message for a predetermined buffer period may provide some tolerance in the system to allow subsequent messages to be delayed by a greater amount than the first message but still arrive in time to be output at a delay based on a predetermined periodic interval.

The present system may allow messages to start to be transmitted at a delay period based on a predetermined periodic interval immediately after receipt of the first message, since the delay period can synchronise to the receipt of the first message. Hence it may not be necessary to perform statistical analysis of the network or to wait for a number of messages to be received before messages start to be output. As described in more detail below, an adaptive buffer may be used to adjust the delay time if necessary and the delay time may be adjusted based on receipt of subsequent messages.

Synchronising to the first message may allow the system to take into account a typical time for the message to be transmitted over the network. This transmission time is likely to differ for each traffic signal or sensor since the route to each component over the network is different.

Preferably, the method further comprises: receiving subsequent control message packets over the digital packet based network, each subsequent control message packet having a corresponding control message arrival time; storing the data for each subsequent control message; and outputting subsequent periodic control messages at a delay after outputting of the preceding control message based on a predetermined periodic interval.

Hence a series of control messages may be received with varying intervals and may be output periodically.

A further aspect provides a method of communicating data via a digital packet based network between a central control unit that generates a sequence of periodic control messages and at least one traffic signal controller or sensor that is arranged to respond to the sequence of periodic control messages by generating a series of periodic reply messages, the method comprlsmg: receiving a first reply message packet containing data for a first reply message over the digital packet based network at a first reply message arrival time; storing the data for the first reply message; outputting a first periodic reply message to the central control unit at a delay after the arrival time corresponding to a predetermined buffer period; receiving a second reply message packet containing data for a second reply message over the digital packet based network at a second reply message arrival time; storing the data for the second reply message; and outputting a second periodic reply message at a delay after outputting of a first reply message based on a predetermined periodic interval.

Similarly to the first aspect, this may allow reply messages to be sent from the traffic signal controller or sensor and received over the digital packet based network at varying intervals. The buffering may then allow the messages to be output at a delay based on a predetermined periodic interval.

Preferably, the method of the second aspect further comprises: receiving subsequent reply message packets over the digital packet based network, each subsequent reply message packet having a corresponding reply message arrival time; storing the data for each subsequent reply message; and outputting subsequent periodic reply messages at a delay after outputting of the preceding reply message based on a predetermined periodic interval.

Hence, a series of messages may be received from the digital packet based network and output at a predetermined periodic interval.

As described in more detail below, in the embodiments described herein, the central control unit may generate a sequence of control messages, the control message data may be encapsulated within a control message packet, optionally with a control sequence number, and the control message packet may be transmitted across the digital packet based network. A receiver may receive the packets at varying intervals and buffer the message data, allowing a control message to be output to the traffic signal controller or sensor at a predetermined periodic interval as described. The traffic signal controller or sensor may generate a sequence of reply messages that may be formed into a sequence of reply message packets and transmitted across the digital packet based network. On receipt, these messages may also be buffered as described herein and may be passed as a sequence of reply messages to the central control unit.

Preferably, the delay between subsequent messages is substantially constant. Hence a sequence of control messages or reply messages may be transmitted at a substantially constant interval.

Preferably, the predetermined periodic interval is constant. This may be set as a parameter of the system.

According to a highly preferable embodiment, the predetermined periodic interval is equal to the interval at which the central control unit generates periodic control messages, preferably one second. Hence messages may be output at substantially the same rate as the average rate at which they are received. Setting the predetermined periodic interval to one second may allow the present system to integrate with pre-existing systems or to meet standard requirements for the transmission of messages.

Preferably, the delay between subsequent messages is substantially equal to the predetermined periodic interval.

According to a highly preferable embodiment, the delay between subsequent messages remains constant within the tolerance period of the central control unit and the traffic signal or sensor.

Hence the control unit or the signal or sensor may not detect the variation in the delay between subsequent messages and may function as if messages were being output at a constant delay period.

Preferably, the delay between subsequent messages may be varied by a correction time below a predetermined correction threshold. This may allow some correction, for example if the first message was transmitted over the digital packet based network particularly quickly, the buffer period for each message may need to be increased for the system to tolerate messages that arrive less quickly over the network.

Preferably, the predetermined correction threshold is less than 10% of the predetermined periodic interval, preferably less than 5% of the predetermined periodic interval. Hence the correction threshold is small compared to the periodic interval.

Preferably, the correction time is a set correction time, preferably less than about 50% of the tolerance period of the central control unit and the traffic signal controller or sensor.

Preferably, the correction time is a set correction time greater than around 10 milliseconds and less than around 40 milliseconds.

Further preferably, the predetermined correction time is about 25 milliseconds.

Preferably, the delay may be varied based on comparing the arrival time for each message to an expected arrival time for each message. Comparing the arrival time of a message to the expected arrival time may allow the system to determine whether a message has arrived late. If a predetermined number of messages, for example 60 messages, arrive later then expected, for example outside a predetermined range of values around each expected arrival time, then the delay period may be adjusted as described above and each message may be buffered for a longer period of time. Hence the buffer may be adaptive since, if the first message arrives in a shorter or longer time than the average transmission time, the delay period may slowly be adjusted to allow a large proportion of the messages to arrive and be output within the predetermined period. One embodiment of the adjustment of the delay period is described in more detail below/ Preferably, the expected arrival time for each message may be calculated based on the arrival time of the first message combined with an integer number of predetermined periodic intervals.

Preferably, the buffer period for subsequent messages is permitted to vary from message to message. Hence the buffer may adapt as the delay period between subsequent messages is varied.

According to a highly preferable embodiment the predetermined buffer period for the first message is less than a predetermined periodic interval. This may allow a message to be transmitted within the same periodic interval as it is received. This may act to minimise the round trip delay time between control messages being sent out by the central control unit and reply messages being received. This may be advantageous, since the central control unit may expect a reply to a control message that has been sent out within a predefined period.

Preferably, the round trip time for messages is less then 5 seconds, preferably about 4 seconds.

Preferably, the predetermined buffer period for the first message is based on an expected variation of transmission time for the message over the digital packet based network. However, advantageously, it may not be necessary to perform statistical analysis on the network before messages are transmitted. This may allow the system to be put into operation immediately over the network without a delay to the first message.

Preferably, wherein the predetermined buffer period for the first message is based on an expected standard deviation of transmission time for the message over the digital packet based network.

Further preferably, the predetermined buffer period for the first message is at least about one standard deviations from the mean transmission time and, further preferably, is at least about two standard deviations from the mean transmission time. Providing a long predetermined buffer period may allow a large proportion of the messages to be received over the digital packet based network before they are due to be output. Hence, a long buffer period may provide a higher tolerance within the system to network delays. However, this advantage must be balanced with the disadvantage that a long buffer period may delay the round trip time of messages through the system. The buffer period selected may depend on the expected variance in the network transmission time, for example a longer buffer period may be required if the variance is expected to be great.

Preferably, the predetermined buffer period for the first message is less than about 50% of the predetermined periodic interval, preferably less than about 25% of the predetermined periodic interval.

Preferably, the predetermined buffer period for the first message is greater than about 2% of the predetermined periodic interval.

According to a particularly preferable embodiment, the predetermined buffer period for the first message is about 100 milliseconds for a one second periodic interval. This may provide a balance between providing tolerance in the system to network delays and minimising the delay to the round trip time as discussed.

Preferably, the control messages generated by the central control computer are formed into control message packets before they are transmitted over the digital packet based network.

Preferably, the reply messages generated by the at least one traffic signal controller or sensor are formed into reply message packets before they are transmitted over the digital packet based network.

According to one embodiment, each control message packet encapsulates a periodic control message and each reply message packet encapsulates a periodic reply message.

According to a further embodiment, each control message packet comprises data from which a control message may be constructed and wherein each reply message packet comprises data from which a reply message may be constructed.

According to a further highly preferable feature, each control or reply message packet further comprises a control sequence number or a reply sequence number. This may allow the message sequence to be determined if messages arrive over the digital packet based network out of sequence. This may be particularly advantageous if, for example, more than one message packet arrives within the same predetermined periodic interval.

Preferably, subsequent control or reply message packets comprise sequential control or reply sequence numbers.

Preferably, the method further comprises determining the control sequence number of a control message packet on receipt of the control message packet over the digital packet based network.

Preferably, the method further comprises determining whether the control sequence number of a control message packet follows sequentially from the control sequence number of the preceding control message packet.

According to a particularly preferable embodiment, if more than one packet arrives in a single periodic interval, the method comprises outputting only the control message corresponding to the highest control sequence number. The message with the lower sequence number may be discarded, hence only the latest control message may be output. This may avoid confusing or contradictory messages being output.

Preferably, the method further comprises determining the reply sequence number of a reply message packet on receipt of the reply message packet over the digital packet based network.

Further preferably, the method further comprises determining whether the reply sequence number of a reply message packet follows sequentially from the reply sequence number of the preceding reply message packet.

According to a further preferable feature, if more than one packet arrives in a single periodic interval, the method comprises outputting only the reply message corresponding to the highest reply sequence number. This may provide similar advantages to those described above for the control messages.

The sequence number applied to a reply message may not be the same number as the sequence number of the control message to which the reply message is a reply. Hence the present system does not have to keep track of which control message a reply message was sent in response to.

In an alternative embodiment, however, the sequence number appended to the reply message may correspond to the control message in response to which the reply message was sent. In a further alternative embodiment, the sequence number appended to the reply message may be unrelated to the sequence number of the control message.

According to a further feature, in one embodiment, the method further comprises comparing the data for a control message to the stored data for the preceding control message and determining whether the data for the control message differs from the stored data for the preceding control message.

Further, if the data for the control message does not differ from the stored data for the preceding control message, the method may further comprise rejecting the data for the control message.

Hence, according to this feature, control messages may only be transmitted over the digital packet based network if they will act to induce a change of state in the traffic signal controller or sensor. This may allow the digital packet based network to be used more efficiently since fewer control messages may need to be transmitted over the digital packet based network. A control message that is identical to the preceding control message may be output to the traffic signal controller or sensor at the delay period after the preceding message. Hence the traffic signal controller or sensor may still receive a control message every predetermined interval even though a message is not necessarily sent over the digital packet based network each second. As described in more detail below, all reply messages are preferably transmitted to the central control unit to allow the central control unit to detect if there is a line error in transmission to the outstation equipment.

According to one preferable embodiment, messages may be output to the central control computer or to the or each traffic signal controller or sensor via an analogue modem. This may allow messages to be converted from transmission over a digital network to transmission over an analogue network and may allow pre-existing traffic control signals or sensors and pre existing central control units to be integrated with the present system without modification. This may mean that it is not necessary to obtain official approval of new equipment in order to implement the present system. Also the system described may appear to the central control unit and the traffic control signals or sensors to be no different to the existing system.

Messages may further be communicated to the traffic signal controller or sensor or to the to central control unit via a network of physical connections. The network of physical connections is a leased line network of copper connections. Hence the system may be integrated with the present system that operates over a leased line network. In addition, the physical connections may provide a reliable means to transmit messages at the required predetermined periodic interval over the last step to the traffic signals or controller.

According to a highly preferable embodiment, the digital packet based network is a wireless network. Further preferably, the digital packet based network is a General Packet Radio Switched (GPRS) network.

Using a wireless network may allow messages to be sent efficiently between components, particularly over large distances. This may also provide flexibility in the system since it may be possible to move components in the system around without having to re-wire substantial parts of a physical network. Using a wireless connection may also allow the central control computer to be located at a distance from the outstation equipment, for example in a different geographical region. This may allow a single central control computer to control a number of networks of outstation equipment. It may also allow redundancy for each network to be provided by allowing each network to be controlled potentially by one of number of different central control computers. The GPRS network may transmit messages using the Internet Protocol (IP).

Preferably, the packets are transmitted over the digital packet based network using the User Datagram Protocol.

Preferably, the control messages include messages that control the status of traffic signals. The status of individual traffic signals is preferably controlled by a local controller which may control, for example, a set of traffic signals at a junction. The control messages may be sent to the local controller which may then co-ordinate the individual signals. Alternatively, a control message may be sent to each traffic signal individually. Control messages may contain data that indicates which groups of traffic signal should be at green and which groups should be at red.

Preferably, the reply messages include messages that contain information relating to the status of traffic signals. These messages may be generated and sent for each individual traffic signal or an overall status message may be sent by a local controller. Reply messages may comprise signals that indicate commands such as: northbound green, eastbound red.

Preferably, the method further comprises transmitting control messages or receiving information to or from further outstation equipment connected over the digital packet based network. For example, the outstation equipment may be used as a local area network (LAN) hub.

The further outstation equipment may comprise for example: a variable message sign; a traffic volume sensor; a pollution sensor; or a pedestrian or public transport sensor.

According to a further preferable embodiment, wherein a control message packet and a reply message packet each include an address byte, wherein the address byte comprises an identifier of the destination component for the message packet and a sequence number. Hence the sequence number may be inserted into the address byte, which may be efficient since this may not require additional bytes to be transmitted over the digital packet based network.

According to a further aspect, there is provided apparatus for communicating data via a digital packet based network between a central control unit that generates a sequence of periodic control messages and at least one traffic signal controller or sensor that is arranged to respond to the sequence of periodic control messages by generating a series of periodic reply messages, the apparatus comprising: means for receiving a first control message packet containing data for a first control message over the digital packet based network at a first control message arrival time; means for storing the data for the first control message; means for outputting a first periodic control message to the or each traffic signal controller or sensor at a delay after the arrival time corresponding to a predetermined buffer period; means for receiving a second control message packet containing data for a second control message over the digital packet based network at a second control message arrival time; means for storing the data for the second control message; and means for outputting a second periodic control message at a delay after outputting of a first control message based on a predetermined periodic interval.

Preferably, the apparatus further comprises: means for receiving subsequent control message packets over the digital packet based network, each subsequent control message packet having a corresponding control message arrival time; means for storing the data for each subsequent control message; and means for outputting subsequent periodic control messages at a delay after outputting of the preceding control message based on a predetermined periodic interval.

According to a further aspect there is provided apparatus for communicating data via a digital packet based network between a central control unit that generates a sequence of periodic control messages and at least one traffic signal controller or sensor that is arranged to respond to the sequence of periodic control messages by generating a series of periodic reply messages, the apparatus comprising: means for receiving a first reply message packet containing data for a first reply message over the digital packet based network at a first reply message arrival time; means for storing the data for the first reply message; means for outputting a first periodic reply message to the central control unit at a delay after the arrival time corresponding to a predetermined buffer period; means for receiving a second reply message packet containing data for a second reply message over the digital packet based network at a second reply message arrival time; means for storing the data for the second reply message; and means for outputting a second periodic reply message at a delay after outputting of a first reply message based on a predetermined periodic interval.

Preferably, the apparatus further comprises: means for receiving subsequent reply message packets over the digital packet based network, each subsequent reply message packet having a corresponding reply message arrival time; means for storing the data for each subsequent reply message; and means for outputting subsequent periodic reply messages at a delay after outputting of the preceding reply message based on a predetermined periodic interval.

Preferred features of the method aspects may be applied to the apparatus and aspects unless otherwise stated and the apparatus aspects provide corresponding advantages to those of the method aspects outlined above.

According to a further aspect, there is provided a method of configuring a network for communicating data between a central control unit and at leastone traffic signal controller or sensor according to a method disclosed in either of the first two aspects of any of their preferable features.

There is further provided a computer program or computer program product for carrying out a method according to either of the first two aspects or any of their preferable features.

Embodiments of the invention will now be described in more detail with reference to the following drawings in which: Fig. 1 is a schematic diagram of a prior art traffic control system; Fig. 2 is a schematic diagram of an overview of one embodiment of the system described herein; Fig. 3 is an example of timing diagram showing messages sent in one embodiment of the system described herein; Fig. 4 is a graph showing the delay time in milliseconds for a plurality of control messages; Fig. 5 is a schematic diagram showing the format of one embodiment of a control address byte; Fig. 6 is a schematic diagram showing the format of one embodiment of a reply address byte; Fig. 7 is a schematic diagram showing an overview of the message sending process according to one embodiment of the system described herein.

One example of a prior art traffic management system is the Urban Traffic Control (UTC) system, which will now be described with reference to Fig. 1. The UTC system is controlled by a central computer 110, which, in this embodiment, is an Alpha computer running traffic control software. The central computer 110 is connected via an ethernet connection 112 to a Front End Processor (FEP) 114, which controls the transmission of data to and from the distributed traffic control components, such as traffic signals. The FEP 114 is connected via a serial connection 116 to a leased line modem 118, which in turn is connected to a leased line copper network 120, which may be provided by a telecommunications network provider 122.

Typical leased line modems 118 that may be used are analogue 200 or 1200 baud leased line modems. A further copper connection 124 then connects the telecommunications network to a plurality of outstation modems or integrated facilities cards 126, each of which is connected to a traffic control component.

The UTC system and transmission system outlined above is specified by the United Kingdom Department of Transport in documents MCE360C, MCE312C, MCE361A and the control equipment is specified in document MCS215.

In the specified prior art system, the smallest resolution of timing sent and held in the model is one second and control data is sent every second. In the prior art system, a traffic controller, which may control a set of traffic signals, is designed to remain under the control of the central UTC computer 110 provided that UTC control data is received every second on the second. The UTC 110 must then receive into its buffer the reply data by 999ms after sending out the control message. In the embodiment described, the reply message is three messages old by the time it is returned to the UTC 110, due to the round trip time for sending a message to the traffic controller, but the reply message is still tied to the most recent control message. That is, a reply message (that replies to a control message sent out four seconds ago) must be received within 999ms of the most recent control message sent.

An overview of one embodiment of the present invention will now be described with reference to Fig. 2.

As in the prior art system, a central Urban Traffic Control (UTC) computer may be used to monitor, manage and control the traffic signal or other traffic control equipment. As in the prior art system, the UTC computer may also be connected to a Front End Processor (FEP) 214 via an Ethernet connection 212.

The FEP 214 may be used to control the interface to the network over which signals are transmitted to at least one piece of outstation equipment, such as a traffic signal controller.

Control signals for each piece of outstation equipment may be received from the UTC 210 at the FEP 214. In one embodiment, a block of signals may be sent to the FEP 214 from the UTC 210 for a plurality of outstations. The FEP 214 may then perform the mapping between the block format and the outstation equipment. To allow this mapping to be performed, configuration data may be downloaded from the UTC 210 to the FEP 214, for example using UDP/IP or TCP/IP. Routing information to route each message to the correct outstation equipment may be stored by the FEP 214 and information about the availability of each outstation may be sent to the UTC 210 by the FEP 214.

Sequence numbers may also be added to each control signal by the FEP, as described in more detail below, before the control signals are output to the wireless network and to each adapter unit. In this embodiment, the signals are output over a further ethernet connection 216 to a wireless network, for example a General Packet Radio Switched (GPRS) network. The signals may be transmitted via, for example, a Customer Premises Equipment (CPE) Router 218 to a Gateway GPRS Support Node (GGSN) 222. The signals transmitted from the FEP 214 over the wireless network may be received by an adapter unit 224 which may be connected, for example over a copper connection 236 to the outstation equipment 238. In this embodiment, the adapter unit 224 may comprise a GPRS modem 226 connected, for example over a RS-232 or a TLA/EIA-232 connection 228, to a processor 230, which may contain software for processing the signal received. The processor may further be connected to a modem 234, typically a 1200bps or a 200bps analogue modem, over, for example, a TTL link 232, via which the control message may be passed to the outstation equipment 238. The adapter unit 224 may receive the control message as described and may remove the sequence number. It may then buffer the message for the required time, as described in more detail below, before sending messages to the outstation equipment 238 at a predetermined interval.

In response to each control message, the outstation equipment 238 may generate a reply message which may be sent to the adapter unit 224. The adapter unit may receive this message and append a sequence number, which may be the sequence number taken from the most recent control message. The reply message may then be sent over the wireless network to the FEP 214.

The FEP 214 may receive the reply message from the adapter, decode the sequence number and insert this message in to the queue of messages to be sent to the UTC computer 210. In the present embodiment, the queue position may be determined from the sequence number.

Messages may be assembled into blocks to be sent to the UTC 210.

In the present embodiment, a control message is sent by the UTC 210 to the outstation equipment 238 every second and a reply message (replying to a previous control message) is also expected back from the outstation equipment or controller 238 every second.

According to one embodiment, the FEP 214 may have an Internet Protocol (IP) address on the wireless network, for example 10.10.10.10. The IP address for the FEP may be configurable.

The subnet mask for the FEP may further be 255.255.255.0 and a default gateway address may be set, for example 10.10.10.1. Further, a port number such as 10011 may be set for transmissions from the FEP to the outstation equipment. The packets sent from the FEP to the outstation equipment may use any available port number, but preferably they will use the same port number as packets received at the FEP.

Each adapter unit may be assigned an IP address, preferably a dynamic IP address. In one embodiment, this may be controller by RADIUS, the Remote Authentication Dial-In User Service, for example using the Password Authentication Protocol (PAP). In one embodiment, the dynamic IP assigned may be based on the user identification given. An outstation name may be set at the site identification, a typical site identification code may be 08/000052, and the user identification may be set to match the site identification or the outstation name, for example J08000052. The user identification and outstation name preferably start with an alpha character to allow the system to be DNS compatible. For example, a further feature of the system may be that the FEP can interrogate the RADIUS to determine the IP address of each site.

Alternatively, an operator may enter records, for example giving the outstation name and the IP address.

As described in more detail below, components of the network, such as the FEP 214, the adapter unit 224 or both components, may manage one or all of the configuration, buffering, sequencing, timing, interfacing, and GPRS connectivity of the system and messages may be transmitted at predetermined intervals, for example on a second-by-second basis. Preferably, components of the system described herein may be implemented by integrating the components into a pre-existing network or a design for a new network that uses pre-existing components, for example a prior art UTC or prior art outstation equipment. Hence the system may be implemented without requiring modification to this equipment and without obtaining national standards approval for any modified equipment.

In an alternative embodiment, the existing components of the control network may be redesigned to incorporate some or all of the functionality described herein. For example, a GPRS modem may be incorporated into the outstation equipment and the UTC computer may perform some of the functionality of the FEP described herein, for example, buffering, timing and sequencing tasks.

As discussed above, in some systems, such as traffic signal control systems, the timing of control and reply messages sent from and to the central computer 210 may be important in ensuring that the system operates effectively. For example, according to the standards referenced above for controlling traffic signals, to maintain control over the outstation equipment, it is necessary to maintain second-by-second contact with each unit.

In a traffic signal control system, control messages may include messages such as: change to green, stay green, change to red and stay red. Reply messages may include messages such as: northbound red and eastbound green.

In the present embodiment, messages may be sent using User Data Protocol (UDP), since it may not be necessary to ensure that all data sent by the central computer 210 is transmitted to the outstation equipment 238.

One embodiment of a system and method that may be used to control timing and buffering of messages in conjunction with the system described herein will now be described in more detail with reference to Fig. 3.

Control messages may be sent from the UTC 210 at one second intervals 310. As described above, these messages may be transmitted over the wireless network via the PEP 214 and received by an adapter unit 224. In being transmitted over the wireless network, the control messages may each be delayed by different amounts and may not arrive at the adapter 224 at intervals of exactly one second 312 as shown on Fig 3. The control messages may then be entered into a buffer memory in the adapter unit 224 before they are passed to the outstation equipment 238. In this embodiment, the messages are initially buffered for 100ms and this time period is initially sequenced to the arrival of the first control packet. However, the buffer time is preferably varied depending on factors such as the reliability of the network and the distance of the adapter 224 from the UTC computer 210. In addition, the sequencing may be based on a control packet other than the first control packet or may be based on a plurality of control packets.

As illustrated in Fig. 3, buffering the control packets may allow the packets to be passed to the outstation equipment 238 at regular one second intervals 316 by compensating for any delays in the transfer of the control packets over the wireless network.

As described in more detail below, an adaptive buffer may be used. In summary, the relative times each message is received at the adapter unit may be compared to the buffer parameters and recorded. Should the relative time be higher or lower than a preset threshold value over a predetermined number of control messages, the buffer parameters may be adjusted to compensate.

The buffer may have the following parameters which can be configured: À Buffer time (hold time) À Maximum deviation of the delay time before correction of the buffer time À Size of the correction À Sample size (number of messages) Reply packets may be generated by the outstation equipment 238 and may be transmitted over the wireless network back to the PEP 214. Since transmission over the wireless network may have caused further random time delays in the delivery of the packages, the reply packets 318 may be buffered by the PEP 214 or at the UTC 210 and this buffer may be read 320 by the UTC 210 at one second intervals a predetermined time within the one second cycle, for example, in this embodiment, the buffer is read 320 at 999ms into the cycle.

Three timing scenarios will now be discussed in more detail with reference to Fig. 4 which shows the delay time in milliseconds for a plurality of control messages sent from the UTC 210.

In this embodiment, the control message is expected to take less than one second to reach the buffer of the adapter unit 224 and the delay time shown is the time that the packet is delayed.

In this example, the mean delay time for packets reaching the adapter 224 was 313ms and the standard deviation of times was 32ms.

The first control message, to which the buffer may be synchronised in this embodiment, as described above, may be delayed for a mean amount of time, for example around 320ms.

According to the parameters set in the present embodiment, the control message may then be held for lOOms before being passed to the traffic controller or outstation equipment 238. Hence subsequent messages may be passed to the outstation equipment at one second intervals but at a delay of 420ms after the control message should have been received by the adapter. This means that, in this embodiment, the second and subsequent messages must be delayed for less than 420ms in order to be passed to the outstation equipment. Using the standard deviation for this example, the probability of a message suffering a delay of 420ms or less is 99.89% and 1 in 1000 messages will not be delivered to the outstation equipment.

If the first control message suffers a low delay, for example around 250ms, it will also be held in the buffer for l OOms in this embodiment and subsequent messages will have to be delayed by less than 350ms to be delivered to the outstation equipment. The probability of a message suffering a delay of 350ms or less in this example is 87.16%. This means that a number of control messages will not arrive at the buffer in time to be passed to the outstation equipment at the regular one second inters al and these control messages will be lost in this embodiment. In the present embodiment, if more than a predetermined number of subsequent messages, in this case seven subsequent messages, are not delivered to the outstation equipment, then the equipment may revert to local control.

In one embodiment, the outstation equipment may be arranged to return to central control by the UTC 210 automatically on receipt of a further control message. In an alternative embodiment, it may be necessary for an operator to re-initialise UTC control.

According to a further feature of the present embodiment, the buffer in the adapter unit 224 monitors the delay of subsequent messages relative to the first message. If the average offset over a predetermined number of messages, for example 60 messages, is greater than a preset value, for example 25ms, the offset time, or assumed delay time, may be adjusted be a preset offset adjustment amount. The offset adjustment may be configurable and is preferably equal to the average offset threshold value, in this case 25ms.

In the case where the first message suffers a high delay, for example around 450ms, the message will still be held in the buffer for the preset time, for example lOOms and subsequent messages must be delayed by less than 550ms to be passed to the outstation equipment 238. In the present example, the probability of a message suffering a delay of less than 550ms is 99.9999%. Hence a high proportion of control messages will be delivered and there is only a very small chance that the outstation equipment 238 will revert to local control.

As for the previous example, the adapter unit may be arranged to monitor the delay of subsequent messages relative to the first message and, if the average offset time over a predetermined number of control messages, for example 60 control messages, is less than a predetermined value, for example -25ms, the offset time can be adjusted by a predetermined value. The adjustment value is preferably the same as the threshold average offset, for example -25ms in this case.

The values by which the offset time is adjusted may be different for adjustments that increase and adjustments that decrease the time. Preferably, however, the adjustments are the same.

Preferably, the adjustment time is set quite low so that the interval between messages sent to the traffic controller or outstation equipment does not vary significantly from one second. That is, the outstation equipment does not detect that messages are being sent to it at a slightly higher or lower frequency.

In summary, in this embodiment, the time the first control message is received (plus looms in this example) marks the base point for messages to be sent to the controller. That is, the messages are synchronized to the receipt of the first message. If the first message is delayed for an unusually short or long period, in this embodiment, the unit will slowly migrate to the correct offset point. Also, if the network dynamics change, for example, if the messages start being routed via a different path, but are still arriving within the one second frame, the unit may migrate, as described above to the new mean delay time.

The UTC central computer 210 and the outstation equipment 238 may be connected to the wireless network via, for example an internet gateway or a leased line connection. A leased line connection may provide a shorter mean delay time and a smaller deviation to the delay than those given in the example above.

Further considerations that may be taken into account in the development of a network with the features outlined above may be that a lower error rate may be achieved if the buffer time is set high. This may be important since, in some systems, every control message lost acts to degrade the UTC operation.

As a further feature, the buffer length may be configurable via a handset which may interface with each adapter unit to set buffer lengths individually for each adapter. Alternatively, the buffer length for each adapter, or for all adapters, may be configurable centrally.

The system may further be implemented to allow control messages arriving after the buffer expires but within the one second period to be passed directly to the outstation equipment.

The system is preferably arranged to adapt within set boundaries to allow messages to be passed to the traffic controller or outstation equipment as soon as possible. This may provide a minimum timing offset compared with traffic signal controllers or outstation equipment being controlled without buffering.

A further feature that may be implemented in conjunction with the system described herein is sequential verification. There is a chance that control and reply messages transmitted over the wireless network, for example the GPRS network, may be received out of sequence at the FEP 214 or at the outstation equipment 238. The sequential verification feature may allow messages that are received out of sequence to be detected and, in some embodiments, trapped from being forwarded to the outstation equipment or UTC.

In this embodiment, the sequencing is incorporated into the system by using three bits within the address byte. The format of a control address byte according to one embodiment is shown in Fig. 5. The format of a control reply byte according to one embodiment is shown in Fig. 6.

Each byte has four address bits 510, 610 that may be used to identify, for example the UTC computer 210. Each byte also has three bits 512, 612 that may be used for sequencing. In addition, the control address bit has an Even Parity bit 514 and the reply address bit has a Message Transfer bit 614. In the present embodiment, the address bits 510, 610 are set to either 1 or 14, since these numbers are diametrically opposed in binary (0001 and 1110) and hence are least likely to be corrupted or confused.

In the present embodiment, a control address byte is further followed by two control bytes and a reply address byte is followed by seven reply bytes.

Since there are three bits to use for sequencing in this embodiment, the total number of unique combinations of bits is 8. Two examples of how the three bits may be used for sequencing are set out below.

A simple count of 0 to 7 in binary could be used for sequential bits. That is, a sequence of bits lo could be transmitted with the sequencing bits as: 6 17 1 1l 12 13 14 15 6 7 1 111 1 1001 1010 011 100 101 1 110 1 111 If the bits are then returned with the sequencing bits in the order: -6 17 1 1 13 12 14 15 6 17 1 111- 1 000 1 001 1 011 1 010 1 100 1 101 1 110 1 111 -1 then it is clear that two of the packets are out of sequence. One problem that may occur is if the sequencing bits of an out of sequence packet has been corrupted, for example if the "3" bit had been corrupted in the above sequence: -6 17 1 1l 13 12 14 15 6 17 1 110 111 1 000 1 001 1 010 1 010 1 100 1 101 1 110 1 111 then the out of sequence packet would not be detected until the next packet arrived and it is still not possible to determine which of the packets is out of sequence. For most purposes, however, a sequential count is likely to be sufficient.

An alternative sequencing code that could be used is a Gray code. The Gray code below has a Hamming distance of 2, which means that two bits have to change to go from one code in the sequence to the next:

A B IC ID IA B IC D IA

goOO 1011 101 1 110 1 ooo 1 011 I 101 110 1 If a sequencing error occurs, the received sequence may be:

LA IB C I D A C I B I D I A

000 1 011 1 101 1 110 1 000 1 101 1 011 1 110 1 000 If an error or corruption also occurs, the sequence may read:

IA B IC ID IA IC IB ID IA

000 1 011 1 101 1 110 1 000 1 001 1 011 1 110 1 000 1 Hence, using this sequencing code, an out of sequence packet even with an error will still be detected and not mistaken for the next packet in the sequence. One limitation of this sequence, however, is that there are only four different codes used in the sequence.

In the system described herein, the probability of more than four packets being delayed and arriving in the same one second interval is higher than the probability of corruption of the three sequencing bits. Therefore, in the present embodiment, it may be advantageous to use the sequential count of 0 to 7 described above.

In operation, the messages may be received by the adapter unit 224, it may be verified that the sequence number has increased by one and the message may then be buffered and passed to the controller. In the event that a message is delayed, but the messages are still received in the correct order, the late message may be passed directly to the controller or the late message may be ignored and the next message passed to the controller in the following second.

If a message arrives so late that it arrives in the same second as a subsequent message, for example if the message arrives out of sequence, then the system may be arranged so that only the message with the higher sequence number is passed to the controller. Similarly, if packets arrive very late, so that there are multiple out-of-sequence arrivals, only the message with the highest sequence number may be passed to the controller.

As outlined above, if there is a predetermined interval during which no control messages are received, for example seven seconds, then the controller may revert to local control. Control of the controller by the central computer may be regained on receipt of the next control message or may be reinstated by an operator.

According to one embodiment, a reply message may be sent out by the controller only when a valid control message has been received. As described above, the adapter may determine the sequence value. This sequence value may then be used when sending back the reply message in the same second. In this embodiment, messages sent across the wireless network in the same second are attributed the same sequence number. In this embodiment, therefore, a reply to a control message will not bear the same sequence number as the control message to which the reply corresponds.

The system and methods described above can preferably be implemented in conjunction with an existing network. Wireless links may be used to replace some leased line connections so that a wireless network can be used to transmit control messages over at least part of the distance between a central computer and outstation equipment. It is not necessary to replace all physical links with the wireless network, however, and the system and method described herein may be implemented with only some of the outstation equipment connected over wireless links. In addition, in one embodiment, redundancy may be introduced into the system by having both a wireless link and a physical connection between the central computer and the outstation equipment. This may be useful particularly at critical sites, for example busy traffic junctions.

A further use for the system may be as a temporary traffic signal control system which may be used, for example if a physical connection is being moved.

In a system to control traffic signals, additional data may be transmitted using the present system. For example, control signals to control the operation of variable message signs may be transmitted over the system and data may be transmitted, for example from sensors that detect the volume of traffic or the presence of pedestrians, cyclists or public transport or from pollution sensors. The system may also be integrated with further systems, such as a system that indicates the time of the next bus at a particular bus stop or a road user charging system. In this way, the outstation equipment may be used as a hub of a Local Area Network (LAN) which may have its own operating system, and ancillary equipment, for example that described above, may be connected to the LAN ports. Each piece of equipment may have its own unique IP address.

As an additional feature, or in an alternative embodiment, data may be downloaded over the wireless network as a block to a local controller, which may implement the sequence of signals locally.

As described above, a control message is preferably sent to each traffic signal controller every second but, in an alternative embodiment or as an additional feature, a control message may be sent to the controller only when it is necessary to change the state of the signals. This may allow the resources of the network to be used more efficiently.

According to the present standards, however, it is necessary to transmit a control signal to the outstation equipment 238 and to the central computer 210, shown on Fig. 2, every second. One solution to this problem is to implement a system whereby second-by-second communication is maintained between the central computer 210 and the FEP 214 and between the adapter unit 224 and the outstation modem 238 but wherein data is only transferred between the FEP 214 and the adapter unit 224 when there is a change of state.

When a message is sent by the central computer 210 to the FEP 214, the FEP 214 may check to see if a change of state has occurred since the lastmessage. If a change of state has occurred, then the message may be forwarded to the adapter 224. The adapter may receive the message, store it in a buffer and forward the buffer contents to the outstation equipment 238 each second until a further change of state occurs.

In the system described above, where messages are sent between the FEP 214 and the adapter unit 224 every second, messages may be sent using the User Data Protocol (UDP), since it is not critical to ensure that every message is reliably delivered. However, if only a change of state message is sent, it is more important that every message is delivered, hence the Transmission Control Protocol (TCP) may be used to transmit messages.

Since control messages may be sent by the adapter unit 224 to the outstation equipment 238 every second, additional safety features may be provided to ensure that the signals can revert to local control if a change of state message is not received from the FEP 214, for example if there is a failure in the wireless network. Preferably, the system is designed so that a control bit, or force bit cannot exist for more than a predetermined period of time, for example 200 seconds.

After this predetermined period, the adapter 224 may stop sending the control messages to the outstation equipment 238 and may allow the outstation equipment 238 to revert to local control.

The adapter 224 may further be designed to transmit only reply messages that indicate a change of state to the FEP 214. However, the FEP 214 may not be able to detect line errors if it is not receiving real time data on a second by second basis hence it may be preferable to maintain second by second transmission of data for the reply messages.

In the present embodiment, sequencing bits may not be used on the control messages, but they may be added to the reply messages by the adapter 224 as described above.

In any of the embodiments of the system described, the default state for the adapter 224 may be to send its address identifier and all zeros to the FEP 214.

By way of summary, one embodiment of the system will now be described with reference to Fig. 7.

The central control unit 710 may generate a series of periodic control messages 718, and may output the control messages to a Front End Processor (FEP) 712. A series of control message packets 720 may be formed by the FEP, for example by encapsulating the control message data or by using the data to form packets, the packets optionally incorporating a control sequence number, and each control message packet may be transmitted across the digital packet based network. A receiver, for example an adapter 714, may receive the packets 720 at varying intervals and may buffer the message data. There may be one receiver associated with each traffic control signal or sensor or a receiver may receive messages for a plurality of traffic control signals or sensors. The first control message packet 730 may arrive at a first control message arrival time and may be buffered for a predetermined buffer period after receipt. The predetermined buffer period is preferably shorter than the periodic interval between the series of control messages 718 sent by the central control unit 710 so the first control message packet 730 may be output within the same periodic interval in which it was received. The second and subsequent messages may then be received at varying intervals, the intervals depending in part on the variable delay caused to each message in transmission across the digital packet based network. The arrival times of the second 732 and subsequent 734 control message packets are determined and are compared to the arrival time of the first control message packet. The buffer time for the messages may be varied if the second and subsequent message packets take longer than expected to arrive after the first control message packet. The control message packets 720 are preferably converted into a series of periodic control messages 722 and may be output at a periodic interval to a traffic control signal or sensor 716. Preferably, the buffer time is varied only in small steps so that the messages may be output at a constant periodic interval within the tolerance of the traffic control signal or sensor.

The traffic control signal or sensor may generate a reply message in response to each control message it receives and these may be output to an adapter 714 as a series of periodic reply messages 728. The adapter may add a sequence number to each reply message and may format the messages as reply message packets before transmitting the series of reply message packets 726 across the digital packet based network.

The reply message packets may be received at the PEP 712 and an arrival time for each message 732, 734, 736 may be determined as for the control message packets. The messages may be buffered and may be output to the central control unit 710 as a series of periodic control messages 724. The buffering may be adaptive as described above for the control messages.

Modifications of detail are possible within the scope of the invention as defined by the claims and the invention is not to be construed as limited to the embodiments described.

Claims

Claims: 1. A method of communicating data via a digital packet based

network between a central control unit that generates a sequence of periodic control messages and at least one traffic signal controller or sensor that is arranged to respond to the sequence of periodic control messages by generating a series of periodic reply messages, the method comprising: receiving a first control message packet containing data for a first control message over the digital packet based network at a first control message arrival time; storing the data for the first control message; outputting a first periodic control message to the or each traffic signal controller or sensor at a delay after the arrival time corresponding to a predetermined buffer period; receiving a second control message packet containing data for a second control message over the digital packet based network at a second control message arrival time; storing the data for the second control message; and outputting a second periodic control message at a delay after outputting of a first control message based on a predetermined periodic interval.
2. A method according to Claim I wherein the method further comprises: receiving subsequent control message packets over the digital packet based network, each subsequent control message packet having a corresponding control message arrival time; storing the data for each subsequent control message; and outputting subsequent periodic control messages at a delay after outputting of the preceding control message based on a predetermined periodic interval.
3. A method of communicating data via a digital packet based network between a central control unit that generates a sequence of periodic control messages and at least one traffic signal controller or sensor that is arranged to respond to the sequence of periodic control messages by generating a series of periodic reply messages, the method comprising: receiving a first reply message packet containing data for a first reply message over the digital packet based network at a first reply message arrival time; storing the data for the first reply message; outputting a first periodic reply message to the central control unit at a delay after the arrival time corresponding to a predetermined buffer period; receiving a second reply message packet containing data for a second reply message over the digital packet based network at a second reply message arrival time; storing the data for the second reply message; and outputting a second periodic reply message at a delay after outputting of a first reply message based on a predetermined periodic interval.
4. A method according to Claim 3 wherein the method further comprises: receiving subsequent reply message packets over the digital packet based network, each subsequent reply message packet having a corresponding reply message arrival time; storing the data for each subsequent reply message; and outputting subsequent periodic reply messages at a delay after outputting of the preceding reply message based on a predetermined periodic interval.
5. A method according to any preceding claim wherein the delay between subsequent messages is substantially constant.
6. A method according to any preceding claim wherein the predetermined periodic interval is constant.
7. A method according to any preceding claim wherein the predetermined periodic interval is equal to the interval at which the central control unit generates periodic control messages, preferably one second.
8. A method according to any preceding claim wherein the delay between subsequent messages is substantially equal to the predetermined periodic interval.
9. A method according to any preceding claim wherein the delay between subsequent messages remains constant within the tolerance period of the central control unit and the traffic signal or sensor
10. A method according to any preceding claim wherein the delay between subsequent messages may be varied by a correction time below a predetermined correction threshold.
11. A method according to Claim 10 wherein the predetermined correction threshold is less than 10% of the predetermined periodic interval, preferably less than 5% of the predetermined periodic interval.
12. A method according to Claim 10 or 11 wherein the correction time is a set correction time, preferably less than about 50% of the tolerance period of the central control unit and the traffic signal controller or sensor.
13. A method according to any of Claims 10 to 12 wherein the correction time is a set correction time greater than around 10 milliseconds and less than around 40 milliseconds.
14. A method according to any of Claims 10 to 13 wherein the predetermined correction time is about 25 milliseconds.
15. A method according to any preceding claim wherein the delay is varied based on comparing the arrival time for each message to an expected arrival time for each message.
16. A method according to any preceding claim wherein the expected arrival time for each message is calculated based on the arrival time of the first message combined with an integer number of predetermined periodic intervals.
17. A method according to any preceding claim wherein the buffer period for subsequent messages is permitted to vary from message to message.
18. A method according to any preceding claim wherein the predetermined buffer period for the first message is less than a predetermined periodic interval.
19. A method according to any preceding claim wherein the predetermined buffer period for the first message is based on an expected variation of transmission time for the message over the digital packet based network.
20. A method according to any preceding claim wherein the predetermined buffer period for the first message is based on an expected standard deviation of transmission time for the message over the digital packet based network.
21. A method according to Claim 20 wherein the predetermined buffer period for the first message is at least about two standard deviations from the mean transmission time.
22. A method according to any preceding claim wherein the predetermined buffer period for the first message is less than about 50% of the predetermined periodic interval, preferably less than about 25% of the predetermined periodic interval.
23. A method according to any preceding claim wherein the predetermined buffer period for the first message is greater than about 2% of the predetermined periodic interval.
24. A method according to any preceding claim wherein the predetermined buffer period for the first message is about 100 milliseconds for a one second periodic interval.
25. A method according to any preceding claim wherein the control messages generated by the central control computer are formed into control message packets before they are transmitted over the digital packet based network.
26. A method according to any preceding claim wherein the reply messages generated by the at least one traffic signal controller or sensor are formed into reply message packets before they are transmitted over the digital packet based network.
27. A method according to any preceding claim wherein each control message packet encapsulates a periodic control message and each reply message packet encapsulates a periodic reply message.
28. A method according to any of Claims 1 to 25 wherein each control message packet comprises data from which a control message may be constructed and wherein each reply message packet comprises data from which a reply message may be constructed.
29. A method according to any preceding claim wherein each control or reply message packet further comprises a control sequence number or a reply sequence number.
30. A method according to Claim 29 wherein subsequent control or reply message packets comprise sequential control or reply sequence numbers.
31. A method according to Claim 29 or 30 wherein the method further comprises determining the control sequence number of a control message packet on receipt of the control message packet over the digital packet based network.
32. A method according to any of Claims 29 to 31 wherein the method further comprises determining whether the control sequence number of a control message packet follows sequentially from the control sequence number of the preceding control message packet.
33. A method according to Claim 31 or 32 wherein, if more than one packet arrives in a single periodic interval, the method comprises outputting only the control message corresponding to the highest control sequence number.
34. A method according to any of Claims 29 to 33 wherein the method further comprises determining the reply sequence number of a reply message packet on receipt of the reply message packet over the digital packet based network.
35. A method according to Claim 34 wherein the method further comprises determining whether the reply sequence number of a reply message packet follows sequentially from the reply sequence number of the preceding reply message packet.
36. A method according to Claim 34 or 35 wherein, if more than one packet arrives in a single periodic interval, the method comprises outputting only the reply message corresponding to the highest reply sequence number.
37. A method according to any preceding claim wherein the method further comprises comparing the data for a control message to the stored data for the preceding control message and determining whether the data for the control message differs from the stored data for the preceding control message.
38. A method according to Claim 37 wherein, if the data for the control message does not differ from the stored data for the preceding control message, the method further comprises rejecting the data for the control message.
39. A method according to any preceding claim wherein messages are output to the central control computer or to the or each traffic signal controller or sensor via an analogue modem.
40. A method according to any preceding claim wherein the digital packet based network is a wireless network.
41. A method according to any preceding claim wherein the digital packet based network is a General Packet Radio Switched (GPRS) network.
42. A method according to any preceding claim wherein the packets are transmitted over the digital packet based network using the User Datagram Protocol.
43. A method according to any preceding claim wherein the control messages include messages that control the status of traffic signals.
44. A method according to any preceding claim wherein the reply messages include messages that contain information relating to the status of traffic signals.
45. A method according to any preceding claim wherein the method further comprises transmitting control messages or receiving information to or from further outstation equipment connected over the digital packet based network.
46. A method according to Claim 45 wherein the further outstation equipment comprises at least one of: a variable message sign; a traffic volume sensor; a pollution sensor; a pedestrian or public transport sensor.
47. A method according to any preceding claim wherein a control message packet and a reply message packet each include an address byte, wherein the address byte comprises an identifier of the destination component for the message packet and a sequence number.
48. Apparatus for communicating data via a digital packet based network between a central control unit that generates a sequence of periodic control messages and at least one traffic signal controller or sensor that is arranged to respond to the sequence of periodic control messages by generating a series of periodic reply messages, the apparatus comprising: means for receiving a first control message packet containing data for a first control message over the digital packet based network at a first control message arrival time; means for storing the data for the first control message; means for outputting a first periodic control message to the or each traffic signal controller or sensor at a delay after the arrival time corresponding to a predetermined buffer period; means for receiving a second control message packet containing data for a second control message over the digital packet based network at a second control message arrival time; means for storing the data for the second control message; and means for outputting a second periodic control message at a delay after outputting of a first control message based on a predetermined periodic interval.
49. Apparatus according to Claim 48 further comprising: means for receiving subsequent control message packets over the digital packet based network, each subsequent control message packet having a corresponding control message arrival time; means for storing the data for each subsequent control message; and means for outputting subsequent periodic control messages at a delay after outputting of the preceding control message based on a predetermined periodic interval.
50. Apparatus for communicating data via a digital packet based network between a central control unit that generates a sequence of periodic control messages and at least one traffic signal controller or sensor that is arranged to respond to the sequence of periodic control messages by generating a series of periodic reply messages, the apparatus comprising: means for receiving a first reply message packet containing data for a first reply message over the digital packet based network at a first reply message arrival time; means for storing the data for the first reply message; means for outputting a first periodic reply message to the central control unit at a delay after the arrival time corresponding to a predetermined buffer period; means for receiving a second reply message packet containing data for a second reply message over the digital packet based network at a second reply message arrival time; means for storing the data for the second reply message; and means for outputting a second periodic reply message at a delay after outputting of a first reply message based on a predetermined periodic interval.
51. Apparatus according to Claim 50 further comprising: means for receiving subsequent reply message packets over the digital packet based network, each subsequent reply message packet having a corresponding reply message arrival time; means for storing the data for each subsequent reply message; and means for outputting subsequent periodic reply messages at a delay after outputting of the preceding reply message based on a predetermined periodic interval.
52. Apparatus according to any of Claims 48 to 51 wherein the delay between subsequent messages is substantially constant.
53. Apparatus according to any of Claims 48 to 52 wherein the predetermined periodic interval is constant.
54. Apparatus according to any of Claims 48 to 53 wherein the predetermined periodic interval is equal to the interval at which the central control unit generates periodic control messages, preferably one second.
55. Apparatus according to any of Claims 48 to 54 wherein the delay between subsequent messages is substantially equal to the predetermined periodic interval.
56. Apparatus according to any of Claims 48 to 55 wherein the delay between subsequent messages remains constant within the tolerance period of the central control unit and the traffic signal or sensor
57. Apparatus according to any of Claims 48 to 56 wherein the delay between subsequent messages may be varied by a correction time below a predetermined correction threshold.
58. Apparatus according to Claim 57 wherein the predetermined correction threshold is less than 10% of the predetermined periodic interval, preferably less than 5% of the predetermined periodic interval.
59. Apparatus according to Claim 57 or 58 wherein the correction time is a set correction time, preferably less than about 50% of the tolerance period of the central control unit and the traffic signal controller or sensor.
60. Apparatus according to any of Claims 57 to 59 wherein the correction time is a set correction time greater than around 10 milliseconds and less than around 40 milliseconds.
61. Apparatus according to any of Claims 57 to 60 wherein the predetermined correction time is about 25 milliseconds.
62. Apparatus according to any of Claims 48 to 61 wherein the delay is varied based on comparing the arrival time for each message to an expected arrival time for each message.
63. Apparatus according to any of Claims 48 to 62 wherein the expected arrival time for each message is calculated based on the arrival time of the first message combined with an integer number of predetermined periodic intervals.
64. Apparatus according to any of Claims 48 to 63 wherein the buffer period for subsequent messages is permitted to vary from message to message.
65. Apparatus according to any of Claims 48 to 64 wherein the predetermined buffer period for the first message is less than a predetermined periodic interval.
66. Apparatus according to any of Claims 48 to 65 wherein the predetermined buffer period for the first message is based on an expected variation of transmission time for the message over the digital packet based network.
67. Apparatus according to any of Claims 48 to 66 wherein the predetermined buffer period for the first message is based on an expected standard deviation of transmission time for the message over the digital packet based network.
68. Apparatus according to Claim 67 wherein the predetermined buffer period for the first message is at least about two standard deviations from the mean transmission time.
69. Apparatus according to any of Claims 48 to 68 wherein the predetermined buffer period for the first message is less than about 50% of the predetermined periodic interval, preferably less than about 25% of the predetermined periodic interval.
70. Apparatus according to any of Claims 48 to 69 wherein the predetermined buffer period for the first message is greater than about 2% of the predetermined periodic interval.
71. Apparatus according to any of Claims 48 to 70 wherein the predetermined buffer period for the first message is about 100 milliseconds for a one second periodic interval.
72. Apparatus according to any of Claims 48 to 71 wherein the control messages generated by the central control computer are formed into control message packets before they are transmitted over the digital packet based network.
73. Apparatus according to any of Claims 48 to 72 wherein the reply messages generated by the at least one traffic signal controller or sensor are formed into reply message packets before they are transmitted over the digital packet based network.
74. Apparatus according to any of Claims 48 to 73 wherein each control message packet encapsulates a periodic control message and each reply message packet encapsulates a periodic reply message.
75. Apparatus according to any of Claims 48 to 73 wherein each control message packet comprises data from which a control message may be constructed and wherein each reply message packet comprises data from which a reply message may be constructed.
76. Apparatus according to any of Claims 48 to 75 wherein each control or reply message packet further comprises a control sequence number or a reply sequence number.
77. Apparatus according to Claim 76 wherein subsequent control or reply message packets comprise sequential control or reply sequence numbers.
78. Apparatus according to Claim 76 or 77 wherein the apparatus further comprises means for determining the control sequence number of a control message packet on receipt of the control message packet over the digital packet based network.
79. Apparatus according to any of Claims 76 to 78 wherein the apparatus further comprises means for determining whether the control sequence number of a control message packet follows sequentially from the control sequence number of the preceding control message packet.
80. Apparatus according to Claim 78 or 79 wherein, if more than one packet arrives in a single periodic interval, only the control message corresponding to the highest control sequence number is output.
81. Apparatus according to any of Claims 76 to 80 wherein the apparatus further comprises means for determining the reply sequence number of a reply message packet on receipt of the reply message packet over the digital packet based network.
82. Apparatus according to Claim 81 wherein the apparatus further comprises means for determining whether the reply sequence number of a reply message packet follows sequentially from the reply sequence number of the preceding reply message packet.
83. Apparatus according to Claim 81 or 82 wherein, if more than one packet arrives in a single periodic interval, only the reply message corresponding to the highest reply sequence number is output.
84. Apparatus according to any of Claims 48 to 83 wherein the apparatus further comprises means for comparing the data for a control message to the stored data for the preceding control message and means for determining whether the data for the control message differs from the stored data for the preceding control message.
85. Apparatus according to Claim 84 wherein, if the data for the control message does not differ from the stored data for the preceding control message, the data for the control message is rejected.
86. Apparatus according to any of Claims 48 to 85 wherein messages are output to the central control computer or to the or each traffic signal controller or sensor via an analogue modem.
87. Apparatus according to any of Claims 48 to 86 wherein the digital packet based network is a wireless network.
88. Apparatus according to any of Claims 48 to 87 wherein the digital packet based network is a General Packet Radio Switched (GPRS) network.
89. Apparatus according to any of Claims 48 to 88 wherein the packets are transmitted over the digital packet based network using the User Datagram Protocol.
90. Apparatus according to any of Claims 48 to 89 wherein the control messages include messages that control the status of traffic signals.
91. Apparatus according to any of Claims 48 to 90 wherein the reply messages include messages that contain information relating to the status of traffic signals.
92. Apparatus according to any of Claims 48 to 91 wherein the method further comprises transmitting control messages or receiving information to or from further outstation equipment connected over the digital packet based network.
93. Apparatus according to Claim 92 wherein the further outstation equipment comprises at least one of: a variable message sign; a traffic volume sensor; a pollution sensor; a pedestrian or public transport sensor.
94. Apparatus according to any of Claims 48 to 93 wherein a control message packet and a reply message packet each include an address byte, wherein the address byte comprises an identifier of the destination component for the message packet and a sequence number.
95. A method of configuring a network for communicating data between a central control unit and at least one traffic signal controller or sensor according to a method disclosed in any of Claims 1 to 47.
96. A method or apparatus substantially as described herein or as illustrated in any of Figures 2 to 6
97. A computer program or computer program product for carrying out a method according to any of Claims 1 to 47.