Measurement and wireless communication system

Abstract

A frequency hopping wireless communication system is described, comprising a measurement station (10) having a first clock and an interface station (20) having a second clock. The measurement station (10), which may form part of a measurement probe, is configured to transmit measurement information resulting from a measurement event, the measurement information including timing information relating the measurement event to the first clock. The interface station (20) is configured to receive the measurement information from the measurement station (10) and generate a measurement output including timing information defined relative to the second clock. One of the first and second clocks is designated as a master clock, and periodic clock adjustments of the other of the first and second clocks are performed to maintain synchronization with the designated master clock. The timing information of the measurement output generated by the interface station (20) takes into account any periodic clock adjustments applied between the occurrence of the measurement event and the generation of the measurement output. In this way jitter is reduced and measurement performance is improved.

Description

[Technical field]

[0001] The present invention relates to a frequency hopping wireless communication system comprising a measurement station and an interface station, in particular for passing metrology (measurement) data from a machine tool measurement probe to an associated probe interface. [Background technology]

[0002] Coordinate positioning apparatus (e.g. machine tools) are often equipped with measuring devices, for example touch-trigger probes, which measure the position of a point on the surface of a workpiece. A touch-trigger probe has a stylus which, when deflected by contact with the surface, generates a so-called trigger signal. This trigger signal is used to freeze the output of a measuring scale of the machine tool, thereby indicating the position of the probe and, with appropriate calibration, the position of a point on the surface of the workpiece.

[0003] In certain circumstances, for example where it is difficult to directly wire the probe to the machine controller, a radio transmission system is used to communicate the trigger signal via the probe interface to the machine controller. In particular, it is known to use a frequency hopping spread spectrum (FHSS) communication link between the probe and the probe interface. Such a communication protocol is described in US Pat. No. 5,399,323 and is also used in the "RMP" series of machine tool probes sold by Renishaw plc of Wotton-under-Edge, Gloucestershire, UK, and the associated "RMI" probe interface. A similar FHSS communication protocol is also used in the Primo series of measurement probes, also sold by Renishaw plc. The above FHSS system has been found to provide robust communication in often harsh (RF) operating environments.

[0004] A frequency hopping communication link relies on maintaining strict synchronization between the two stations at each end of the link, whether or not a transmission is taking place, to ensure that the two stations hop together (i.e., synchronously) between frequency channels according to a predefined hopping pattern. Maintaining such frequency channel synchronization is particularly important for systems that relay time-critical information, such as measurement probes, trigger event information transmitted between associated interfaces. For example, in the RMI / RMP system mentioned above, there is a periodic adjustment of the clock provided to the interface to maintain synchronization with the probe's master clock.

[0005] However, the inventors have discovered that the above-described real-time clock adjustment process can impact the measurement accuracy of an FHSS communication system, especially when communicating asynchronous measurement events such as trigger signals. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] International Publication No. 2004 / 057552 Brochure [Patent Document 2] International Publication No. 2018 / 134585 Brochure Summary of the Invention

[0007] According to a first aspect of the present invention there is provided a frequency hopping wireless communication system comprising a measurement station having a first clock and an interface station having a second clock, the measurement station is configured to transmit measurement information resulting from the measurement events, the measurement information including timing information relating the measurement events to a first clock; the interface station is configured to receive the measurement information from the measurement station and generate a measurement output including timing information defined relative to a second clock; one of the first clock and the second clock is designated as a master clock and periodic clock adjustments are performed on the other of the first clock and the second clock to maintain synchronization with the designated master clock; The timing information of the measurement output generated by the interface station is characterized to take into account any periodic clock adjustments applied between the occurrence of a measurement event and the generation of the measurement output.

[0008] The present invention thus relates to a frequency hopping spread spectrum (FHSS) radio communication system in which a wireless (FHSS) radio communication link is provided between a measurement station (e.g. provided as part of a battery-powered measurement device) and an interface station (e.g. provided as part of a measurement interface hard-wired to a machine tool). Both the measurement station and the interface station include their own clocks, referred to as a first clock and a second clock, respectively. As explained below, the first clock and the second clock may include oscillators and may be arranged to output a series of clock pulses at the same nominal intervals.

[0009] The measurement station is configured to transmit measurement information resulting from the measurement events. For example, the measurement station may transmit information about touch-trigger events generated by a touch-trigger sensor. The measurement information may also include timing information, which links the measurement events to timing derived from a first clock. The timing information may be provided as data (e.g., a time value) and / or the timing information may be encoded within the transmitted message (e.g., an edge of a transmitted frame may be synchronized to the first clock). In a preferred embodiment described below, the timing information includes a clock count that defines the time within a frame (e.g., relative to the starting edge of that frame) at which the measurement event occurred.

[0010] The interface station is configured to receive measurement information from the measurement station. In other words, the interface station is arranged to receive measurement information transmitted by the measurement station (i.e. via the FHSS link). The interface station is also configured to generate a measurement output including timing information defined relative to a second clock. In other words, a measurement output (e.g. a trigger signal) is output from the interface station to an associated device (e.g. a machine tool) at a specific time. The interface station times all events relative to its own clock (i.e. the second clock) and therefore the measurement output is made at a time defined relative to the second clock.

[0011] As mentioned above, the measurement station and the interface station necessarily use different clocks. While such clocks may be nominally identical, it is typically impractical or impossible to provide clocks that remain perfectly synchronized over long periods of time. Thus, in order to maintain frequency hopping synchronization, the first and second clocks need to be periodically aligned or synchronized. To achieve this, one of the first and second clocks is designated as a master clock (i.e., all system timing is tied to this master clock). Periodic clock adjustments are then performed for the other of the first and second clocks to maintain synchronization with the designated master clock. As will be explained below, this synchronization can be performed by periodically transmitting timing messages between the measurement station and the interface station. These timing messages may be transmitted at regular or irregular intervals. Thus, multiple such timing messages can be transmitted while an FHSS communication link is established between the measurement station and the interface station. Such timing messages allow real-time synchronization of the first and second clocks to be maintained. This ensures that the frequency hopping of the measurement station and the interface station remains synchronized, thereby preventing clock drift that could otherwise cause the frequency hopping synchronization to be gradually lost. This maintenance of clock synchronization is performed "on the fly", i.e., it can be performed for periodic synchronization while the communication link is being used to transmit measurement information from the measurement station to the interface station.

[0012] In prior art FHSS systems such as those described above, periodic synchronization of clocks is performed to maintain synchronized frequency hopping at both ends of the communication link. However, it has been found by the inventors that performing such a clock synchronization step simultaneously or nearly simultaneously with communicating measurement information from the measurement station to the interface station can result in measurement errors. In other words, the applied clock corrections are not taken into account (i.e., their effect is ignored) when generating the measurement output. For example, nudging a first clock count up or down to maintain clock synchronization can lead to an incorrect conversion of a time measured relative to the first clock to a time measured relative to the second clock (i.e., the nudge to the clock count introduces a timing difference that appears as a timing error).

[0013] Thus, the system of the present invention takes into account (i.e., uses or incorporates) any timing changes used to maintain clock synchronization when calculating the time (relative to the second clock) at which the measurement output should be generated. Thus, the timing information of the measurement output generated by the interface station takes into account any periodic clock adjustments applied between the occurrence of the measurement event and the generation of the measurement output. In other words, any periodic clock adjustments applied between the occurrence of the measurement event and the generation of the measurement output are incorporated into the timing information of the measurement output generated by the interface station. Thus, any timing changes applied are taken into account by the interface station when calculating the time (relative to the second clock) at which the measurement output should be generated. This can be thought of as correcting the time (relative to the second clock) at which the measurement output is generated to take into account the applied timing changes. In this way, clock synchronization is maintained in real time without reducing measurement accuracy.

[0014] Advantageously, the measurement information transmitted by the measurement station is time-stamped relative to a time defined by the first clock. For example, the measurement information may include a time value (defined by the first clock) defining when a particular measurement event occurred. Thus, the timestamp may include data defining the time. In a preferred embodiment, the measurement station transmits information to the interface station using a series of frames, and the interface station operates using a similar series of frames. Each of the frames may be assigned a frequency channel according to a periodically repeating hopping pattern or sequence known to both the measurement station and the interface station. Thus, the measurement information may be time-stamped by the measurement station to define the time relative to an edge of the frame. For example, a 250 μs frame may be subdivided into 250 counts from the clock. In such an example, the timestamp associated with the measurement information may define the count (i.e., as a number between 0 and 249) between which the measurement event occurred.

[0015] Advantageously, the first and second clocks each generate a clock count. For example, each of the clocks may include an oscillator that increments a clock count value. Such clock count value may be periodic. One of the first and second clocks may be designated as a master clock. Periodic clock adjustments may then include incrementing or decrementing the clock count of the other of the first and second clocks to maintain synchronization (i.e., of the clock count) with the designated master clock. In this way, the measurement station and the probe station may generate a series of clock counts that are coordinated (synchronized) with each other. A frame may be defined by a number of such clock counts (for example, a number of such clock counts may be 250 clock counts of 1 microsecond duration that define a 250 microsecond frame), so that the edges of such frames also remain synchronized. Thus, the system may enable the first clock and the second clock to be synchronized to within 1 millisecond, more preferably within 250 microseconds (μs), more preferably within 100 microseconds (μs), or more preferably within 10 microseconds (μs).

[0016] The periodic clock adjustment may be performed in any suitable manner. Conveniently, the periodic clock adjustment comprises an exchange of messages containing timing information between the measurement station and the interface station. For example, the measurement station may transmit the current clock count value of the measurement station to the interface station. Alternatively, the measurement station may transmit a message at a specific time (or clock count) defined by the (first) clock of the measurement station. The interface station may receive a transmission from the probe station and compare the timing information provided by the message received within the time defined by the (second) clock of the interface station. This would allow any count difference to be determined between the first and second clocks. After defining any timing (e.g. count) difference, the interface station may then instruct the probe station to modify the first clock of the probe station to remove the difference. For example, the probe station could receive an instruction to add a clock count to a frame or to skip a clock count in a frame to restore synchronization of the count of the first clock and the count of the second clock.

[0017] Periodic clock adjustments may be performed, for example, on demand or at irregular intervals when messages (e.g., conveying measurement information) are exchanged between the probe and the interface station. Periodic clock adjustments are preferably performed at regular (periodic) intervals. This ensures that synchronization is maintained even when no other messages are being exchanged by the probe and the interface station. In addition to any regular clock adjustments, ad hoc clock adjustments (e.g., when conveying measurement information) may also be performed. Thus, multiple such clock adjustments may be performed.

[0018] The first or second clock may be used as a master clock for system timing. Advantageously, the second clock is the master clock and the periodic clock adjustment comprises adjusting the first clock to keep it synchronized with the second clock. Thus, the master clock may be the (second) clock provided to the interface station. The second clock may also be used to define the timing of other components of the system, such as devices to which the interface station is connected.

[0019] Advantageously, the measurement station may be provided as part of the measurement device. The measurement device may also include a measurement sensor. The measurement sensor may generate metrology data. The measurement sensor may be of any type. For example, the measurement sensor may include at least one of a touch-trigger sensor, a scanning sensor, an ultrasonic sensor, or an imaging sensor. The measurement sensor may enable measurement of dimensions, position, temperature, etc. The measurement sensor may be a contact sensor (i.e., the measurement sensor may physically contact the object to be measured) or the measurement sensor may be a non-contact sensor (e.g., the non-contact sensor may use optical sensing, inductive sensing, capacitive sensing, etc.).

[0020] In a preferred embodiment, the measurement sensor of the measuring device includes a touch-trigger sensor. The touch-trigger sensor may sense a trigger event when a particular spatial relationship is reached with a point on the surface of the object. For example, the trigger event may be sensed when the stylus is deflected. The measuring device may be provided as a measuring probe of a machine tool. The measuring device may be battery powered. It is therefore advantageous for the measuring device to use as little power as possible to maximize battery life. The measuring probe may also include at least one of a stylus, a shank, and a battery.

[0021] The measurement information transmitted by the measurement station may relate to a touch trigger event sensed by the touch trigger sensor. Such a touch trigger measurement event may occur at any time. The measurement output generated by the interface station may include a trigger signal. For example, such a touch trigger signal may be output by latching a voltage on a line connected to a skip input of the controller, or by generating a pulse or a series of pulses passed to the skip input. Alternatively, the trigger signal may be passed to the controller via a digital data bus (as described, for example, in US Pat. No. 5,399,363). Thus, the measurement event may include an asynchronous measurement event. In other words, a measurement is not recorded at a specific time, but instead a measurement event may occur at an undefined point in time (e.g., when the stylus contacts the object). The time at which such a measurement event occurs can thus be used to measure the position of a point on the surface of the object. Therefore, the delay between an asynchronous measurement event being recorded by the probe station and the generation of a measurement output by the interface station is desirably not affected by any periodic clock adjustment applied between the measurement event and the generation of the measurement output. In other words, the latency of the system is desirably not affected by a periodic clock adjustment process.

[0022] The communication link between the probe and the interface station is preferably a bidirectional communication link. The probe station and the interface station both transmit and receive radio signals over multiple frequency channels that preferably spread across a particular frequency band (e.g., across the 2.4 GHz radio band). Data packets may then be transmitted and received during each of the frames on the designated frequency channel. The probe station and the interface station may be operable in multiple different modes. For example, multiple communication modes using different frame rates may be implemented. Multiple different hopping patterns may also be used. For example, the interface station may communicate with different measurement stations using different hopping patterns. However, it should be remembered that both the probe station and the interface station that communicate with each other need to use the same hopping pattern to maintain hopping synchronization. In other words, the measurement station and the probe station are preferably equipped with frequency hopping spread spectrum modems that operate using the same hopping pattern. An acquisition mode may also be implemented in which the measurement station passively receives information (hopping between channels) until an activation message is received from the measurement station.

[0023] The probe station and interface station may include a communication module implemented in various ways. For example, such a communication module may include a suitably configured circuit or processor. The communication module may also include an RF antenna or be connectable to an associated RF antenna. The communication module may also include a processor for calculating the hopping pattern (e.g., from the unique identifier). The communication module may be formed as a circuit board (e.g., an FPGA) for mounting within the measurement probe and / or within the measurement probe interface. The circuit board may implement other functions connected to frequency hopping communication or for other processes. Other components may be included within the probe station or interface station.

[0024] The multiple frequency channels used by the system may include frequency channels within the 2.4 GHz radio frequency band (spanning 2.4000-2.4835 GHz), also known as the Industrial, Scientific, and Medical (ISM) band, which is license-free in most countries. It should be noted that each of the frequency channels may be 2 MHz wide, but any transmission may only be over a portion of the channel's respective frequency range. In a preferred embodiment, 39 channels are provided (each 2 MHz wide) spanning from 2.404 GHz to 2.480 GHz. It should be noted that the present invention may be used in any radio frequency band and is not limited to ISM band operation. The system is preferably configured to operate in a standard license-free radio frequency band. It should be noted that such frequency bands may vary from country to country or over time depending on regulatory rules, but this does not change the advantages associated with the present invention. It should also be noted that frequency hopping wireless communication systems and their components may be classified as Digital Transmission Systems (DTS) or Hybrid Systems, rather than FHSS systems, according to the official definition of the Federal Communications Commission (FCC).

[0025] According to a second aspect of the present invention, there is provided a method of using a frequency hopping wireless communication system comprising a measurement station having a first clock and an interface station having a second clock, the method comprising the steps of transmitting measurement information from the measurement station to the interface station, the measurement information transmitted by the measurement station including timing information resulting from a measurement event and relating the measurement event to the first clock, receiving the measurement information at the interface station to generate a measurement output including timing information related to the second clock, and designating one of the first and second clocks as a master clock and periodically adjusting the other of the first and second clocks to maintain synchronization with the designated master clock, characterized in that when generating the timing information of the measurement output generated by the interface station, any periodic clock adjustments applied between the measurement event and the generation of the measurement output are taken into account. In other words, the method may include a step of incorporating into the timing information of the measurement output (i.e. generated by the interface station) any periodic clock adjustments applied between the occurrence of the measurement event and the generation of the measurement output. The method may also include any of the features described above in relation to the related apparatus.

[0026] Also described herein is a frequency hopping wireless communication system. The system may include a measurement station. The measurement station may have a first clock. The system may include an interface station. The interface station may have a second clock. The measurement station may be configured to transmit measurement information resulting from a measurement event (e.g., a touch-trigger event). The measurement information may include timing information relating the measurement event to the first clock. The interface station may be configured to receive the measurement information from the measurement station. The interface station may be configured to generate a measurement output including timing information relating the measurement event to the second clock. One of the first clock and the second clock may be designated as a master clock, and periodic clock adjustments of the other of the first clock and the second clock may be performed to maintain synchronization with the designated master clock. The timing information of the measurement output generated by the interface station may account for any periodic clock adjustments applied between the occurrence of the measurement event and the generation of the measurement output. In other words, a timing correction that accounts for the clock adjustments may be implemented. [Brief description of the drawings]

[0027] The invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0028] [Figure 1] FIG. 2 is a diagram illustrating a machine tool measurement probe and a probe interface. [Diagram 2] FIG. 2 is a diagram showing a clock count of a frame. [Diagram 3] A diagram illustrating how the probe clock and the interface clock can drift and how this is periodically corrected. [Figure 4] FIG. 13 illustrates how trigger events measured by a probe are used to generate trigger outputs from an interface. [Diagram 5] FIG. 13 illustrates how correction for clock drift can be taken into account when generating a trigger output. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0029] Referring initially to Figure 1, there is illustrated a touch-trigger measurement probe 10 mounted on a machine tool spindle 12. The measurement probe 10 has a deflectable stylus 14 with a tip 16 which contacts the workpiece. The measurement probe also includes a first frequency hopping wireless communication module or station 18 which would normally be integrated within the body of the measurement probe, but is shown separately in Figure 1 for clarity. A measurement probe interface 20 is mounted to a fixed part 22 of the machine tool structure and connected (typically via a cable) to a machine tool controller 24. The measurement probe interface 20 also includes a second frequency hopping wireless communication module or station 26.

[0030] Data is transmitted between the measurement probe 10 and the probe interface 20 via a frequency hopping wireless communication link established between the first frequency hopping wireless communication module 18 and the second frequency hopping wireless communication module 26. As mentioned above, the first frequency hopping wireless communication module and the second frequency hopping wireless communication module need to hop between synchronized frequency channels to allow a communication link to be established. The sequence in which the various different frequency channels are used is usually called a "hopping pattern". In prior art systems, a hopping pattern describes the rule by which different frequency channels are always used, or the sequence in which different frequency channels are always used, and is used to ensure that, after proper synchronization, both ends of the wireless link are transmitting and receiving on the same frequency channel at the same time (assuming both ends use the same frame rate). In other words, the hopping pattern describes the sequence of "hopping" between the different frequency channels. For further information, reference is made to US Pat. No. 6,399,433, the contents of which are incorporated herein by reference.

[0031] In a frequency hopping system, both ends of the link (i.e., the probe station and the interface station) must maintain strict synchronization. As one frame rolls over to the next, both ends of the link change to the next channel in the hopping sequence or pattern, regardless of whether a transmission is taking place. This allows both ends of the link to be ready to create and receive data packets on the correct channel at the correct time.

[0032] Referring to FIG. 2, a number of frames (F1 to F8) of the probe station are shown. The frequency channel used to transmit and receive information is changed every frame according to a so-called hopping pattern. In this example, each of the frames has a duration of 250 microseconds (μs). The duration of the frames is timed using a clock in the probe that counts from 0 to 249 in intervals of 1 microsecond (μs). As shown in the inset to FIG. 2, the 250 counts generated by the probe's clock define each of the frames. A clock count of 0 defines the start of a frame and a clock count of 249 is the final count of that frame. Thus, the clock count increments from 0 to 249 before being reset to zero for the start of the next frame.

[0033] In the above system, where the probe and associated interface communicate over a frequency hopping link, there are different clocks in the probe and the interface. Most oscillators used in electronic circuits to provide clock signals are characterized under certain defined conditions. At these defined operating parameters, the oscillators have different tolerances (i.e., permissible deviations from the ideal operating frequency) for each oscillator device. Any deviation from the defined conditions, such as deviations in operating temperature, leads to differences in oscillation frequency. If the probe and interface are used such that one of the units is at a different temperature than the other, the difference in oscillation frequency becomes even more pronounced. The quartz crystals used in the oscillators are also prone to change (drift) in oscillation frequency over time. This is known as aging. It is therefore necessary to design a system that is resistant to the drift of the quartz crystals, but can also compensate for it during operation, so that the highest degree of measurement accuracy is not affected even after many years of service.

[0034] With reference to FIG. 3, the effect of very small differences in the oscillation frequencies of the clocks introduced within the probe and the interface are illustrated. Probe frames F1-F9 are shown along with corresponding frames F1'-F9' of the interface. Although roughly aligned in time, the inset to FIG. 3 shows that drift between the probe and interface clocks has shifted the start of frames F2 and F2' slightly by the amount indicated by arrow 36. This slight frame shift is unlikely to affect communication of data, but it can increase (accumulate) over time until the probe and interface frames no longer align, at which point the probe and interface are no longer using the same frequency channel together (i.e., frequency hopping synchronization will be lost) and communication will fail.

[0035] This problem of clock drift has been previously addressed in the prior art frequency hopping systems mentioned above. In particular, it is known that there is a periodic transmission between the probe and the interface (labeled 38 in FIG. 3) to resynchronize the clocks. Briefly, the probe may send a message at a particular count within a frame. For example, the probe may send a status message at count 74 within a particular frame. The interface may read this message and compare the probe's clock count with the interface's own count. If the counts are different, the interface may send a message back to the probe telling it to adjust its clock count to synchronize the probe's clock count with that of the interface's clock. For example, the interface may tell the probe to skip or add a particular number of counts until the next frame to provide the necessary clock realignment.

[0036] Considering a more concrete example, a probe may send a so-called heartbeat message to its associated interface once every 32 milliseconds (ms) (regardless of frame rate). If a message is missed, the probe retries every frame. The interface waits until one of two things occurs: (a) a message sent by the probe is received by the interface, or (b) 63 milliseconds (ms) pass without a heartbeat message, in which case the interface outputs an error. After outputting the error, the interface starts searching for a probe to reconnect to (i.e., the interface attempts to establish a communication link from scratch).

[0037] For such an example, the maximum allowable drift between the two periodic oscillators would therefore occur in the latter case (b). To calculate this drift, some detailed information about the crystal oscillators used to drive the modems in both the probe and the interface is used. For example, the oscillation frequency is 32 MHz with a calibration tolerance of ±15 ppm. The frequency stability over the temperature range of -20°C to 85°C is ±15 ppm, with aging effects less than ±3 ppm / year. If the worst case scenario is considered, the following assumptions can be made: (i) the probe and the interface are at opposite ends of the calibration tolerance, (ii) the probe and the interface are at opposite ends of the usable temperature range, and (iii) this correlates to opposite ends of the frequency stability range. Furthermore, (iv) the probe and the interface are both 5 years old and age in opposite directions to each other according to the worst case scenario.

[0038] Taking all of the above into account, this example results in a drift of approximately 2.8 picoseconds (ps) between the two clocks on every single clock cycle. If we assume that the maximum number of clock cycles that can occur between calibrations is approximately 2 million cycles, then the amount of drift that could occur is approximately 5.67 microseconds (μs). This drift gets worse as the probe ages and can be up to 7.56 μs for a 10 year old probe.

[0039] The above-mentioned previous RMI / RMP and Primo measurement probe systems will adjust for such clock drift to maintain frame timing and robust communication. However, the small adjustments of the previous RMI / RMP and Primo measurement probe systems were not compensated for when measurement points occurred close together or simultaneously. The inventors have found that this manifests itself as jitter, i.e., small variations in switching times that manifest as small differences in measurement location. To achieve the most repeatable and accurate system, the present invention compensates for any timing adjustments when measurement points occur. Roughly, and ignoring other effects, this has been found to improve the timing accuracy of measurement events by a factor of about 20. This will now be explained in more detail.

[0040] Referring to Figure 4, a collection of touch trigger points using the above probe and interface is described. As outlined above, the probe and interface cycle through a series of frames together. In Figure 4, probe frames F1-F14 are shown aligned with interface frames F1'-F14'.

[0041] The probe may sense a trigger event at any time. In this example, trigger event 40 occurs during frame F2. The clock count in frame F2 (i.e., measured using the probe's clock) at the time the trigger event is detected is recorded. An accurate touch trigger measurement process relies on the interface always issuing a trigger signal at a specific (univariable) time after the trigger event occurs. In this example, trigger signal 44 is issued by the interface during frame F13' of the interface. The delay (or latency) between trigger event 40 and trigger signal 44 is denoted in FIG. 4 as "t latency " is marked.

[0042] Information about the trigger event is passed between the probe and the interface by an exchange of messages. For example, in frame F4, the probe may send a message 42 informing the interface that a measurement event has been detected and that it occurred at a particular clock count in frame F2. This message may be several frames after the event occurred, with the option to resend the message multiple times if communication on a particular frequency channel fails (e.g. due to interference). The interface can acknowledge receipt of this message (i.e. the probe knows that it was safely received). The interface can then act on the information in the message. In particular the interface then knows the frame and the clock count within that frame associated during which the trigger event occurred. This allows the interface to calculate the time required for the trigger signal 44 to be issued (i.e. the frame and the clock count within that frame) (i.e. the time required for the trigger signal 44 to be issued is calculated as t latency (This ensures that the delay is

[0043] The above relies on the probe's clock and the interface's clock being synchronized (i.e. the interface needs to map the timing of the probe's clock to the timing relative to its own (the interface's) clock). Therefore, since changes in the probe's state (i.e. touch points) are transmitted as timestamps relative to the start of a synchronized frame edge, it can be seen that the accuracy of the clocks used at the probe and interface has a direct impact on the ultimate measurement capabilities of the system.

[0044] While clock adjustments have been used in the past to maintain clock synchronization, such prior art techniques treated clock synchronization separately from the communication of trigger event information. In other words, the clock adjustment (i.e., synchronization) process could occur at any time relative to the communication of the trigger event (such trigger events being asynchronous events that can occur at any time). Thus, there could be clock timing changes at or around the time that a measurement point is recorded. Thus, the clock adjustment process was found to add error to latency, which appeared as jitter in the trigger signal. As detailed above, the present invention eliminates this source of error by taking any clock adjustments into account when generating the trigger signal.

[0045] Referring now to FIG. 5, a technique for taking clock adjustments and drift into account when communicating touch trigger events from the probe to the interface (also called RMI-QE) is described.

[0046] As explained above, the objective of the measurement system is to assert the RMI-QE skip or status output (i.e. the trigger signal coming out of the interface) at a fixed latency after the probe receives a trigger. The latency must be as short as possible and have very low jitter. The latency is shown in the figure as t LATENCYand in this example is 2.5 milliseconds (ms).

[0047] Both the probe and interface frame timers run at 250 microseconds (μs), so there are 10 frames with a latency of 2.5 milliseconds (ms). This allows 1 frame for the trigger, 1 "housekeeping" frame to allow the transmission to be scheduled, 1 frame to send the message, and 7 frames for retries.

[0048] In this example, the probe first sends a regular periodic message to the RMI-QE interface. The periodic message from the probe always occurs at a specific offset in the frame (t PROBE_SEND ) in this example, the offset is 74 microseconds (μs). The RMI-QE modem (interface) receives this message towards the end of the 75 μs, calculates that the probes have shifted by -1 μs, and adjusts the two frame timers to synchronize -1 μs (t ADJUST ) adjustment needs to be applied. The RMI-QE will send an adjustment message to the probe after a few microseconds and the probe will apply the adjustment, which will take effect at the end of the next frame.

[0049] As the next frame nears its end, a trigger is received (t TRIGGER ). This is shown at 247 microseconds (μs) in the frame. Note that no adjustments have been made to the frame length to match the RMI-QE's frame timer. It is clear from the figure that, to reduce jitter, the trigger occurred at 247 microseconds (μs) on the probe's frame timer, but the RMI-QE needed to calculate that the trigger occurred at 249 microseconds (μs) on its own frame timer.

[0050] In the second frame after the trigger occurs, the probeTRIGGER ) and the adjustment applied by the probe (t ADJUST ) and the number of retries to the modem for the RMI-QE. In this example, the probe frame timer drifts, causing the probe frame timer to again drift to 75 microseconds, resulting in the RMI-QE receiving a 1 microsecond (μs) drift in the probe (t DRIFT ) is calculated.

[0051] Note that it is desirable for frame timer adjustments not to be applied to frames after a trigger event and before the event acknowledgment is received from the RMI-QE interface. For example, if the trigger event occurs in the same frame as a periodic message containing the probe's adjustment value, no adjustment should be applied. Instead, the interface calculates the required drift compensation when the interface receives the event message, and this is used to correct for latency.

[0052] The acknowledgement to the event message contains the probe adjustments so that they can be used to correct the probe clock. During the frame, at Serial Peripheral Interface (SPI) intervals (225μs in our example), the modem sends all this information back to the FPGA on the RMI-QE interface. The information transmitted is as follows: t TRIGGER =247 t ADJUST =‐1 t DRIFT =‐1 RETRIES =0

[0053] After receiving this data, the FPGA outputs the skip (t TIMER ) needs to be asserted. To do this, the FPGA calculates the time remaining until the wait time (t LATENCY = 2.5 milliseconds (ms)) minus any delays in the transmission system. The delay is t OFFSETFrom the figure, t OFFSET is the offset in the frame at which the FPGA starts time (in the example, t TIMER_START , 225 microseconds (μs), the timestamp from the end of the frame in which the timestamp occurred (t TRIGGER ) to the distance (t TRIGGER_INV ), and any retries that occurred.

[0054] To implement the timer in the FPGA with minimal latency and to allow for maximum clock drift, it is also necessary that the timer be started at least 20 microseconds (μs) before the end of the frame in which the event data is received: TIMERSTART_MAX = 230 μs. If this is not possible, then a drift allowance must be built into the timer to prevent negative timer values ​​being requested.

[0055] To obtain the correct timestamp (249 μs), we use the timestamp from the probe (t TRIGGER Note also that the timestamp ( , 247 μs) needs to be fine-tuned. To do this, the adjustment value and the drift value need to be subtracted from the timestamp.

[0056]

number

[0057] Therefore, t TIMER To calculate:

[0058]

number

[0059] It becomes.

[0060] t LATENCY The value of varies depending on whether the computation is performed with the RMI-QE interface. For RMI-QE this means: t LATENCY =t OUTPUT_LATENCY -(n FRAMES_BEFORE_RX xt FRAME ) t LATENCY =2.5ms - (2x250μs) t LATENCY =2.0ms

[0061] t TRIGGER INV can be calculated by removing the trigger timestamp from the frame length.

[0062] Therefore (RMI-QE): t TIMER =t LATENCY -t TIMERSTART -(t FRAME -(t TIMESTAMP -t ADJUST -t DRIFT ))‐R xt FRAME t TIMER =2.0ms‐230μs‐(250μs‐(247μs‐‐1μs‐‐1μs))‐0 x 250μs t TIMER =2000μs‐230μs‐(250μs‐(249μs)) t TIMER =2000μs-(231μs) t TIMER =1769μs It becomes.

[0063] The above technique therefore allows clock adjustment to ensure that the probe clock and the interface clock remain synchronized, but this can be done without affecting measurement accuracy, in particular eliminating the jitter previously seen due to the clock adjustment process.

[0064] Although the above examples relate to touch-trigger measurement probes and associated probe interfaces, the same techniques can be used to transmit any measurement data. For example, ultrasound, scan point or other metrology data could be transmitted. Similarly, operation in the 2.4 GHz radio band is provided merely as an example and is not intended to limit the scope of the invention. Other features of the '611 patent may also be included in the system.

Claims

1. 1. A frequency hopping wireless communication system comprising a measurement station having a first clock and an interface station having a second clock, the measurement station is configured to transmit measurement information resulting from measurement events, the measurement information including timing information relating the measurement events to the first clock; the interface station is configured to receive the measurement information from the measurement station and generate a measurement output including timing information defined relative to the second clock; and one of the first clock and the second clock is designated as a master clock and periodic clock adjustments of the other of the first clock and the second clock are performed to maintain synchronization with the designated master clock; 11. A frequency hopping wireless communication system, comprising: a first interface station for receiving a first signal from a first sub-interface station and a second interface station for receiving a second signal from a first sub-interface station;

2. 2. The frequency hopping wireless communication system of claim 1, wherein said measurement information transmitted by said measurement station is time-stamped with respect to said time defined by said first clock.

3. 3. The frequency hopping wireless communication system of claim 2, wherein the measurement station transmits information to the interface station using a series of frames, and the measurement information is time-stamped to define a time relative to the edge of the transmitted frame.

4. 4. The frequency hopping wireless communication system of claim 1, wherein each of the first and second clocks generates a clock count, one of the first and second clocks is designated as a master clock, and the periodic clock adjustment comprises incrementing or decrementing the clock count of the other of the first and second clocks to maintain synchronization with the designated master clock.

5. 2. The frequency hopping wireless communication system of claim 1, wherein said periodic clock adjustment comprises said exchange of messages containing timing information between said measurement station and said interface station.

6. 2. The frequency hopping wireless communication system of claim 1, wherein said periodic clock adjustments are performed at regular intervals.

7. 2. The frequency hopping wireless communication system of claim 1, wherein the second clock is the master clock and the periodic clock adjustment comprises adjusting the first clock to stay synchronized with the second clock.

8. 2. The frequency hopping wireless communication system of claim 1, wherein the measuring station is provided as part of a measuring device, the measuring device also including a measuring sensor.

9. The frequency hopping wireless communication system of claim 8 , wherein the measurement sensor comprises a touch-trigger sensor.

10. The frequency hopping wireless communication system of claim 9 , wherein the measurement event comprises a touch-trigger event sensed by the touch-trigger sensor.

11. 11. The frequency hopping wireless communication system of claim 10, wherein the measurement output produced by the interface station comprises a trigger signal.

12. The frequency hopping wireless communication system of claim 1 , wherein the measurement events include asynchronous measurement events.

13. 13. The frequency hopping wireless communication system of claim 12, wherein the delay between the asynchronous measurement event recorded by the probe station and the generation of the measurement output by the interface station is not affected by any periodic clock adjustments applied between the measurement event and the generation of the measurement output.

14. 2. The frequency hopping wireless communication system of claim 1, wherein said measurement station and said probe station comprise frequency hopping spread spectrum modems that operate using said same hopping pattern.

15. 1. A method of using a frequency hopping wireless communication system comprising a measurement station having a first clock and an interface station having a second clock, the method comprising: transmitting measurement information from the measurement station to the interface station, the measurement information transmitted by the measurement station including timing information resulting from measurement events and relating the measurement events to the first clock; receiving the measurement information at the interface station to generate a measurement output including timing information relative to the second clock; designating one of the first and second clocks as a master clock and periodically adjusting the other of the first and second clocks to maintain synchronization with the designated master clock; Equipped with A method comprising: taking into account any periodic clock adjustments applied between the measurement event and the generation of the measurement output when generating the timing information for the measurement output generated by the interface station.

Images