Note: Descriptions are shown in the official language in which they were submitted.
CA 02248839 1998-08-19
wo 97/31437 PcT/Isg7/00l44
DIGITAL SONIC AND ULTRASONIC
COMMUNICATIONS NETWORKS
1. Field of the Invention
The present invention relates to digital communications networks, and to
5 communications using acoustic energy.
2. State of the Art
An industrial economy depends heavily on transportation infrastructure.
The United States enjoys one of the most advanced highway systems in the
world. Nevertheless, this system, designed principally in the 1950s, is
10 beginning to show signs of age. Furthermore, because of current budgetary
pressures, very few new highways are being planned or built. Instead, attention
has been focussed on ma~imi7ing the utilization of existing highways through
the application of computer and communications technologies. This effort is
referred to generally as the Intelligent Transit System (ITS).
The tacit underlying assumption concerning the application of
communications technology to transit has been that Radio-Frequency (RF)
communications will be used. The widespread use of RF commllnir~tions in
transit applications, however, suffers in concept from a number of
disadvantages. The ITS initiative appears to have gained critical momentum
20 just at a time when the scarcity of RF bandwidth is being felt most acutely.
The RF spectrum is, quite literally, "cluttered" with a wide variety of users all
competing for scarce bandwidth. Federal regulatory approval is therefore
required for most RF communications. Furthermore, a great deal of traffic is
interstate and even international (particularly in Europe). The result is a
25 patchwork of rules, regulations and practices, from jurisdiction to jurisdiction,
concerning RF communications.
What is needed, then is additional bandwidth that may be applied within
the context of the ITS and other similar transit applications. Preferably, such
bandwidth should be "clutter-free" and unregulated so as to allow for the
C~NFIRM~ION COPY
CA 02248839 l998-08-l9
W O 97/31437 PCTAB97/00144
consistent commercial use of such bandwidth from jurisdiction to jurisdiction.
The present invention addresses this need.
SUMMARY OF THE INVENTION
In accordance with the present invention, generally speaking, a digital
S co~ lullications network is provided using digital sonic and ultrasonic
communications-i.e., communications using acoustic energy instead of RF
energy. The "acoustic spectrum," as opposed to the RF spectrum, is
uncluttered and unregulated, allowing for unfettered cornmercial development of
equipment for ITS applications as well as a wide variety of other applications,
10 including applications that currently employ RF communications. Exemplary
applications include electronic toll boothing, controlled entry systems, border
crossing systems, etc. Coding and processing techniques are employed that
allow acoustic communications, including the communication of digital data, to
be reliably transmitted and received even in noisy acoustic environments.
More particularly, in accordance with one embodiment of the invention,
a digital acoustic communications apparatus includes one or more digital
acoustic commllnications devices comprising a data processor; memory coupled
to the data processor and storing digital data; and means for transmitting and/or
receiving digital data acoustically; wherein the acoustic digital communications20 apparatust during operation, transmits and/or receives digital data acoustically.
In accordance with further aspects of the this embodiment of invention, the
memory stores at least one of an identifying code word and a command, and
the means for transmitting and/or receiving transmits and/or receives at least
one of said identifying code word and said command acoustically. The means
25 for tr~n~mitting and/or receiving may be an acoustic digital communications
tr~n~mitter operating in the human audible range or may be an acoustic digital
communications transmitter operating in the ultrasonic range. Alternatively, themeans for transmitting and/or receiving may be an acoustic digital
communications receiver comprising an analog-to-digital converter, wherein the
30 data processor comprises a digital signal processor coupled to the
analog-to-digital converter for filtering a digital representation of a received
CA 02248839 1998-08-19
W O 97131437 PCT~B97/00144
acoustic signal and for recovering digital data symbols encoded therein. The
acoustic digital comml-nic~tions receiver may operate in the human audible
range or may further comprising a downconverter, whereby the acoustic digital
~ comrnunications receiver operates in the ultrasonic range. Still further, the
S means for tr~n.cmitting and/or receiving may be a digital acoustic transceivercomprising an input sound transducer, an analog-to-digital converter coupled to
the input sound tr~n~d~cer, an output sound tr~n~duc-er, and a digital-to-analogconverter coupled to the output sound tr~n~ cer; in which case the data
processor may be a digital signal processor coupled to the analog-to-digital
10 converter for filtering a digital representation of a received acoustic signal and
for recovering digital data symbols encoded therein, and coupled to the memory
and to the digital-to-analog converter for tr~ncmitting the identifying code word
or the command stored in memory acoustically. The acoustic digital
communications transceiver may operate in the human audible range or in the
15 ultrasonic range. A system in accordance with another aspect of the present
invention comprises a plurality of digital acoustic communications devices
including a plurality of acoustic digital tr~n~mi~ters and at least one acousticdigital receiver for, when one of said acoustic digital tr~n.~mitters is within
range and transmitting digital information, receiving said digital information.
20 The system preferably further comprises a computer and at least one wide areanetwork communications link established between the acoustic digital receiver
and the computer. More preferably, the system comprises multiple acoustic
digital receivers and multiple wide area network communications links, one
such link being established between each of a plurality of said acoustic digital2~ receivers and said computer.
~ n accordance with another aspect of the present invention, a method of
digital cornmunications comprising the steps of generating a carrier signal;
mod~ ting the carrier signal in accordance with digital information to produce
a modulated signal; and applying the modulated signal to an acoustic transducer
30 to produce a coded acoustic signal. The coded acoustic signal is propagated
across a distance many times a wavelength of the coded acoustic signal.
CA 02248839 1998-08-19
Wo 97/31437 PcTlIs97/ool44
Further steps include receiving the coded acoustic signal and transducing the
coded acoustic signal to produce a modulated signal; and demodulating the
modulated signal to produce the digital infollllation.
Uses of the communications method are many and varied. One such use
5 comprising the steps of providing an acoustic digital communications tr~nsmitter
to be carried with a moving object; providing an acoustic digital
comml-nirations receiver in proximity to a controlled area; tr~n.smitting from
the acoustic digital communications tr~nsmitter at least one of an identifying
code word that identifies the acoustic digital communications tr~n~mitter and a
10 comm~n~l; receiving at the acoustic digital communications receiver the
identifying code word or comm~n~; and in response to at least one of the
identifying code word or comm~n-l, allowing physical access of the moving
object to the controlled area. Another such use comprises the steps of
providing an acoustic digital communications tr~n~mitt~r to be carried with a
15 moving object; providing an acoustic digital communications receiver within an
area to be monitored; transmitting from the acoustic digital communications
tr~n.~mitter an identifying code word that identifies the acoustic digital
comm~lnic~tions transmitter; receiving at the acoustic digital communications
receiver the identifying code word; and when the code word is not received
20 within a predetermined interval of time, producing an alarm indication. Still a
further use comprises the steps of providing a first acoustic digital
c~ unications transceiver to be carried on an object; providing a second
acoustic digital communications transceiver at a fixed location; tr~n.cmitting
from one of the first and second acoustic digital communications transceivers a
25 query message; receiving the query message at another of the first and secondacoustic digital communications transceivers and transmitting a response
message; determining a one-way propagation time between the first and second
acoustic digital communications transceivers; and deterrnining a distance
between the first and second acoustic digital communications transceivers. The
30 location of the object may be intended to remain fixed for a time, in which case
the foregoing steps are repeated multiple times; a determination is made
CA 02248839 1998-08-19
Wo 97/31437 PcTlIsg7lool44
whether the location of the object has changed; and if the location of the object
has changed, an alarm indication is produced. Alternatively, the first acoustic
digital cornmunications transceiver may be a mobile acoustic digital
comml-nirAtions transceiver carried on a moving object, and the second acoustic
digital communications transceiver may be a base acoustic digital
communications transceiver, in which case the foregoing steps are repeated
multiple times; and a rate of change of location of the object is deterrnined.
The foregoing steps may be repeated at multiple base acoustic digital
transceivers and results communicated from the multiple base acoustic digital
transceivers to a common site. In this manner one or both of a location and a
heading of the object may be determined.
BRIEF DESCRIPTION OF THE DRAWING
The present invention may be further understood from the following
description in conjunction with the appended drawing. In the drawing:
Figure 1 is a timing diagram of a Coded Audio Signal transmitted at
fixed intervals;
Figure 2 is a block diagram of a Coded Audio Sound Generator;
Figure 3 is a plot of a correlation output for a codeword UW1 in no
noise;
Figure 4 is a plot of a correlation output for codeword UW1 in noise~
with the noise level set at twice the signal level;
Figure S is a block diagram of a Coded Detector Module System;
Figure 6 is an equivalent functional block diagram of a portion of the
Coded Detector Module of Figure S realized by the DSP;
Figure 7 is a timing diagram of a Coded Audio Signal tr~n~mittecl at
fixed intervals;
Figure 8 is a diagram of an Audio Command Packet;
Figure 9 is an equivalent functional block diagram of a portion of the
Coded Detector Module realized by the DSP;
Figure 10 is a block diagram of an ultrasonic Transit Control Module;
Figure 11 is a block diagram of a Coded Audio Transceiver Module;
CA 02248839 1998-08-19
wo 97/3t437 PcTnsg7/ool44
Figure 12 is a diagram illustrating a communir~tions sequence allowing
a t~ t~n~e ranging operation to be performed using acoustic energy;
Figure 13 is a timing diagram illustrating the timing of query and
response messages for purposes of performing a ~li.ct~nre ranging calculation;
Figure 14 is a block diagram of an acoustic digital communications
network, in particular a network for geolocation; and
Figure 15 is a diagram of a portion of a ~le~ic~tf d short-range
cornrnunications system.
DETAILED DESCRIPTION OF THE P~EFERRED EMBODIMENTS
Building blocks of the present digital sonic and ultrasonic
communications networks include sonic tr~n~mit~çrs (sound generators), sonic
receivers (detectors) and sonic transceivers (sound generators and detectors).
Different embodiments of these building blocks may possess varying degrees of
sophistication. Whereas the sound generators are relatively simple in their
construction, the sound detectors rely on Digital Signal Processing techniques to
achieve accurate detection over moderate ~li.ct~nr.es (on the order of one mile).
Three principal embodiments of a coded audio detector are described.
The first embodiment provides the ability to uniquely detect a particular coded
sound generator. The second embodiment adds to this unique detection
capability the further ability of a vehicle operator to have the detector take
specific actions. In a third embodiment, an ultrasonic downconverter module is
provided, allowing the coded sound detector to operate in the ultrasonic range.
The invention will be described primarily in terms of transit
applications. It should be understood, however, that the communications
techniques described, besides being applicable to vehicular communications, are
equally application to personal communications, the tagging of goods, etc.
In the first embodiment, a Coded Audio Detector Module is DSP-based.
A vehicle is equipped with a special Coded Sound Generator which issues
"codewords" at fixed intervals, or on command of the driver. Between the
transmission of these special codewords, the Coded Sound Generator need not
emit any sound. The DSP-based Coded Audio Detector Module receives the
CA 02248839 1998-08-19
WO 97131437 PCTAB97/00144
foregoing codewords, decodes the codeword to determine if it is one of a
pre-determined set of codewords recognized as being valid, and then issues a
signal to a controller if the codeword assigned to that vehicle is valid.
The Coded Audio Detector Module, or CADM, is used as part of a
5 two-part system. The first part is an audio-based tr~n.~mister system on each
vehicle that is to control or interact with the system. Referring to Figure 1,
when the vehicle operator enables the Coded Sound Generator, the tr~n~mitter
sends a codeword at specific time intervals, say 5 seconds, using binary FSK
modulation of the audio carrier for example.
Each vehicle is provided with a Coded Sound Generator. The Coded
Sound Generator may be a simple audio generator the output of which is input
to the microphone input of an amplifier. This Coded Sound Generator
generates the applo~,iate codeword.
Referring more particularly to Figure 2, a programmed microprocessor
201 is coupled to a Digital to Analog Converter (DAC) 203. An output of the
DAC 203 is coupled to an amplifier 209, which is coupled to a speaker 211. A
mode selection switch 213 and a manual signal switch 215 are also provided
and are coupled to the microprocessor 201.
The microprocessor 201 reads the mode selection switch 213 to
20 deterrnine if the operator wants the Coded Sound Generator to be activated
continuously at intervals or only upon user command. The microprocessor 201
generates synthetic digital waveforrns representing the desired codeword. These
signals are converted to an analog voltage by the DAC 203 and then input to
the amplifier 209. The manual signal switch 215 allows the operator to
25 generate codeword signals at will rather than at timed intervals.
The CADM will only issue a control signal when a coded signal which
meets specific conditions is detected. This feature allows for greater security
and reliability of operation.
In the first embodiment, the codewords used in the Coded Audio
30 Detector Module are binary patterns of a specific length. The patterns are
chosen such that they have desirable autocorrelation function characteristics-
CA 02248839 1998-08-19
WO 97/31437 PCT/IB97/00144
specifically low auto-correlation sidelobes. Furthermore, in choosing a family
of codewords, attention should also be paid to the cross-correlation properties
of the codewords. In particular, in addition to there being a low degree of
correlation between the codewords, their cross-correlation functions should have5 low sidelobes. Por example, the following seven codewords represent a family
of codewords which satisfy the foregoing requirements:
UW1 = 11011 1010100000
UW2 = 101 1 101 10100000
UW3 = 101 1 1 1001 100000
UW4 = 1 10101 101 100000
UW5 = 101 1 1 1010010000
UW6 = 1 1 1 100101010000
UW7 = 10101 1 101000100
The audio codeword may be tr~n~mitted using simple Frequency Shift
15 Keying (FSK) modulation at some carrier frequency, where fc is the center
frequency, f~ + ~f is the frequency for the tr~n~mi.~sion of a binary 1, and fc-~f
is the fre~uency for the transmission of a binary 0. The codeword consists of a
stream of binary digits sent using one of these two tones.
The CADM receives the codeword using a microphone system and then
20 demodulates the audio codeword. Demodulation is performed using an FSK
demodulator. The CADM first synchronizes to the incoming bit stream by
performing a symbol timing recovery operation on the codeword. Once
synchronized, the FSK CADM searches for the codeword. The search may be
done by binary correlation with threshold detection, using the stored reference
25 codewords (UW1 to UW7) as a reference. If the codeword is received with
more bits matching the stored reference pattern than the threshold value, it will
be processed and the desired comm~n~l~ will be in~e~leted and issued to the
controller. If the packet was received with an uncorrectable number of errors,
the command will be rejected and no signals will be sent to the controller.
3~ To test the perforrnance of the detector in searching for the codeword,the 15-bit codeword UW1 was used an example. The codeword was sent as a
CA 02248839 1998-08-l9
W O 97/31437 rCTAB97/00144
15-bit sequence using the binary PSK modulation scheme discussed earlier, and
was preceded and followed by silence. The signal was processed using a binary
correlation algorithm, and the correlation output was plotted as a function of
time as shown in Figure 3. In this case, the maximum in the correlation is seen
5 to occur at about sample 101, which marks the location of the codeword in
time. The maximum of the correlation value is 60, since the 15-bit codeword
was sampled at 4 samples per bit. If a threshold value of, say, 56 was taken as
a detection threshold, then only correlation outputs in excess of 56 would causethe microphone system to indicate the presence of the codeword.
The real performance advantage of binary correlation with threshold
detection using the foregoing codewords is obtained under noisy conditions.
Consider the same 15-bit codeword in a noisy environment where the noise
level is twice the level of the codeword sound received from the vehicle.
Referring to Figure 4, it may be seen that the correlation peak is still quite
15 prominent and distinct from any correlation peaks generated by the noise itself.
In this example, the correlation peak near bit 100 is still very prominent and the
peaks resulting from the correlation of the stored reference with the noise are
still very small.
It is possible to use longer length codewords to achieve even better
20 performance in a noisy environment. Longer codewords also reduce the
probability of false detection, i.e., the probability of the chance situation where
received noise just happens to look like the stored reference signal and falselycauses the detection threshold of the binary correlator to be exceeded. The
following table shows the probability of false detection and the probability of
25 missed detection (the probability that the codeword was in fact tr~n.cmit~ed, but
that noise corrupted a sufficiently large number of bits that the binary correlator
missed the codeword), for codeword lengths of 15, 20 and 32 bits, where the
correlator threshold is set to tolerate two bit errors (a 1% bit error rate
channel). It is readily seen that for the 32-bit codeword case tolerating two
30 errors, there will be very few cases of false detection. Assuming that eventshappen at the bit interval and that the bit rate is 20 bps, then there would be on
CA 02248839 l998-08-l9
W O g7t31437 PCT~B97tO0144
average one false detection approximately every 4.7 days. This probability can
be reduced even more by using the loudness of the received codeword to trigger
a signal pre-emption event (i.e., the correlation output must exceed the
threshold, and the sound level must exceed a sound level threshold, indicating
5 that the vehicle is in proximity to the microphone).
Table 1. Binary Correlation with Threshold Detection
For Various Codeword Lengths.
Codeword Bit
Length Errors Pfalse Pmissed
10(bits) Tolerated
1 4.88 x 10-4 9.63 x 10-4
1 2.00 x 10-5 1.69 x 10-3
2 2.01 x 10-4 1.00 X 10-~
32 2 1.23 x 10-7 3.99 x 10-3
15 32 3 1.28 x 10-6 2.87 x 10-4
In a preferred embodiment, the functionality of the CADM as described
is implemented using subs~nti~lly the same hardware platform as the
DSP-based siren detector of WO 95/24028 (McConnell et al), published
September 8, 1995, incorporated herein by reference. Only the DSP software
is changed. The detection algorithm may be based on the limiter/discriminator
approach of McConnell et al., but includes in addition a low-rate demodulator
to perform symbol timing recovery and codeword detection.
Referring more particularly to Figure 5, a Coded Sound Generator 501
is coupled to a loudspeaker 503. At the receiver, the coded signal produced by
the Coded Sound Generator and 501 is picked up by a transducer 505 and input
to a DSP-based logic board 507.
The DSP-based logic board 507 processes the coded signal and outputs
pre-empt signals to a controller 509 based on that processing.The DSP-based
CA 02248839 1998-08-19
wo 97/31437 PCT/IB97/00144
logic board 507 realizes a Coded Audio Detector Module that uses the same
limiter discriminator operations as described in McConnell et al. to perform
FSK demodulation of the FSK signal. The software is modified to incorporate
the following additional functions:
Symbol Timing Recovery - this may be based on a simple
early/late-gate symbol synchronizer.
Codeword Search - this may be based on a binary correlation with
threshold detection technique, using the pre-stored reference codewords as
templates for the binary correlation.
In addition to these functions, the software is modified to include a
comm~nll parser to determine which codeword was received and to then take
apploL~Iiate action based on the command and data in the cornmand packet.
An equivalent functional block diagram of the CADM is shown in
Figure 6. An output signal from a microphone 601 is filtered in a band-pass
filter 603. The filtered signal is then input to a combination of a discriminator
605, a decimator 607 and a median filter 609. An output of the median filter
609 is coupled to a symbol synchronization block 611, followed by a codeword
search block 613. An output of the codeword search block is input to a block
615 to control whether a signal is issued to the controller. Also input to the
block 615 is the output of the decimator 607, indicative of the received signal
level.
As compared to the DSP-based detector of McConnell et al., the
discriminator, decimator, and median filter operations are the same, to ensure
the highest sensitivity possible for the CADM based on the excellent signal
detection capability inherent in that technique. Signal detection is followed bythe operations of blocks 611, 613 and 615, required to decode the codeword
and then execute the command associated with that codeword.
The CADM may be provided with a multiplicity of channels. Different
channels are allocated to different approaches to an installation. In the vast
majority of cases, a four channel detector system will suffice. Cases with more
than four approaches may be dealt with by assigning additional channels.
CA 02248839 1998-08-19
W O 97131437 PCT~B97100144
In accordance with a second embodiment, the DSP-based Coded Audio
Detector Module offers increased functionality, above and beyond that of the
first embodiment. As in the first embodiment, the CADM is based on the DSP
siren detector module of McConnell et al., with the following exceptions.
5 First, the vehicle which is to activate the control function is equipped with a
special Coded Sound Generator which issues coded "frames" at fixed intervals
or on command of the driver. Between the tr~n~mis.~ion of these special coded
frames, the Coded Sound Generator need not emit any sounds. Second, the
DSP-based siren detector is modified to receive the coded frames, decode the
10 frame content, and then issue a control signal if the unique address ~s~ign~d to
that vehicle matches one of a list of addresses that the detector recognizes as
being valid.
As before, the Coded Audio Detector Module, or CADM, is part of a
two part system. The first part is an audio based tr~n~mitter system on each
15 vehicle that is to control or interact with the system. Referring to Figure 7,
when the vehicle operator enables the Coded Sound Generator, the transmitter
at specific time intervals, say S seconds, sends a packet frame, using binary
FSK modulation of the audio carrier for example.
Each vehicle is provided with a Coded Sound Generator. The Coded
20 Sound Generator may be a simple audio generator the output of which is input
to the microphone input of an amplifier system. This Coded Sound Generator
generates the appropriate packet frame.
The physical hardware used to realized the Coded Sound Generator may
be the same as previously described in relation to Figure 2. Referring again to
25 Figure 2, the microprocessor reads the mode selection switch to determine if
the operator wants the sound generated at intervals or only upon user actuation
of a switch, for example. The microprocessor generates synthetic digital
waveforms representing the packet frame (as opposed to a singular codeword as
in the previous embodiment). These signals are converted to an analog voltage
30 by the Digital to Analog Convertor (DAC) and then input to the amplifier. A
CA 02248839 1998-08-19
Wo 97/31437 pcT/Is~7lool44
manual signal switch is also available to allow the operator to generate packet
frame signals at will rather than at timed intervals.
The CADM will only issue a control signal when a packet frame which
~ meets specific conditions is detected. This feature allows for a greater security
5 and reliability of operation.
The audio comm~nd packet is structured in a fashion similar to a
standard X.25 packet frame, described for example in Kuo, Prolocols and
Techniques For Data Communications Networks, Prentice Hall, 1981,
incorporated herein by reference. Referring more particularly to Figure 8, the
10 audio command packet is transmitted using a simple Frequency Shift Keying
(FSK) modulation at some carrier frequency, where fc is the center frequency,
fc+~f is the frequency for the tr~nsmicsion of a binary 1, and fc-~f is the
frequency for the tr~n.~mi~sion of a binary 0. The packet consists of a stream
of binary digits sent using one of these two tones. The purpose of the various
15 segments of the packet are as follows:
Preamble - to allow the CADM to synchronize to the symbol centers of
the binary data signal. The preamble is typically an alternating binary
sequence, such as 1010101010.
Frame Synch - to provide word alignment to the control, data, and
20 parity portions of the command packet. The frame synch is typically a short
binary sequence such as a Barker code, Lindner Sequence, Maury-Styles
Sequence, etc. One suitable frame synch word consists of the binary sequence
0010 0000 0111 0101.
Header - this field of the frame contains binary address information that
25 is unique to each vehicle in the fleet, as well as a packet type identifier, and
control flags. The actual binary sequence depends on the values given to the
elements of the header field. In an exemplary embodiment, the field is 20 bits
in length.
Data - this field may consist of anywhere from 0 bits to say 256 bits of
30 information. In order to process the comm~n~.~ expeditiously, this field willtypically be kept small in practice and may only consist of 8 bits of data.
CA 02248839 1998-08-19
wo 97/31437 PcT/Isg7/ool44
Parity - parity inforrnation to be used for error correction and detection
in the packet. If a CRC-16 is used, then 16 parity bits are required.
Postamble - a known sequence of data used to allow clearing of buffers
in the receiver. The postamble would typically be a short sequence of
alternating binary 1's and 0's, such as 101010.
As an added security measure, the header, data, and parity bits may be
exclusive OR'ed with a known but secret pseudorandom Number (PRN)
sequence to provide scrambling and a limited degree of security. Other more
elaborate encryptiori schemes may also be applied, if desired.
The CADM module receives the packet using a microphone system and
demodulates the audio packet command. Demodulation is perforrned using an
FSK demodulator. The CADM first synchronizes to the incoming bit stream by
performing a symbol timing recovery operation on the packet, with the
preamble being used to assist this synchronization. Once synchronized, the
FSK CADM searches for the frame synch word to achieve frame
synchronization, after which it applies the PRN descrambling sequence, then
extracts the comunand, data, and parity fields from the packet. It then uses theparity bits to perform error correction and/or detection on the control and datafields of the packet. If the packet is received without errors or with a
correctable number of errors, it will be processed and the desired comm~n~s
will be interpreted and issued to the controller. Examples of such commAn(ls
are commAn~lc to open a gate, etc. If the packet was received with an
uncorrectable number of errors, the command will be rejected and no signals
will be sent to the controller.
As in the previous embodiment, the functionality of the CADM as
described is implemented using substantially the same hardware platform as the
DSP-based siren detector of McConnell et al. (See Figure 5.) Only the DSP
software is changed. The detection algorithm may be based on the
limiter/discriminator approach of McConnell et al., but includes in addition a
low-rate demodulator to perform symbol timing recovery and synch word
CA 02248839 1998-08-19
W O 97/31437 PCT~B97100144
detection. The header and data fields are processed by the CPU for activation
of transit control signals, for example.
The CADM uses the same limiter discriminator operations as described
in McConnell et al. to perform FSK demodulation of the FSK signal. The
5 software is modified to incorporate the following additional functions:
Symbol Timing Recovery - this may be based on a simple
early/late-gate symbol synchronizer.
Frame Synchronization - this may be based on a binary correlation
with threshold detection technique.
Desc~l.bling - this may be an Exclusive OR of the header, data, and
parity fields with the known PRN sequence.
Frame Extraction - as per conventional practice.
Error Correction/Detection Coding - this may be based on any of a
number of well established error correction coding schemes, such as ~mming
15 codes, Golay codes, BCH, etc.
In addition to these functions, the software is modified to include a
command parser to extract, interpret, and to then take appropriate action based
on the command and data in the command packet.
An equivalent functional block diagram of the CADM is shown in
20 Figure 9. An output signal from a microphone 901 is filtered in a band-pass
filter 903. The filtered signal is then input to a combination of a discriminator
905, a decimator 907 and a median filter 909. An output of the median filter
909 is coupled to a symbol synchronization block 911.
Thus far, the block diagram of the enhanced CADM is identical to that
25 of Figure 6. The symbol synchronization block 911, however, is followed by a
frame synch block 913 and an optional descrambler 915. An output of the
descrambler 915 is input to an error correction block 917. The error-corrected
packet frame is then input to a command parser 919. The command is input to
a block 920, where the validity of the command is checked. If the command is
30 valid, a block 921 is notified, which controls whether a signal is issued to the
CA 02248839 1998-08-19
W O97/31437 PCT~B97/00144
controller. Also input to the block 921 is the output of the decimator 907,
indicative of the received signal level.
Again, as compared to the DSP-based siren detector of McConnell et
al., the discriminator, decimator, and median filter operations are the same, to5 ensure the highest sensitivity possible for the CADM based on the excellent
signal detection capability inherent in that technique. Signal detection is
followed by the operations required (blocks 911 through 92i) to decode the
packet frame, and then execute the command and associated data for that
packet.
The foregoing description has assumed operation in the audible
frequency range. However, both the CADM of the first embodiment and the
C~DM of the second embodiment may be based on ultrasonic sound energy
operating above the threshold of human hearing (i.e. > 20 kHz). The module
when used to perforrn detection of ultrasonic control signals in transit
15 applications is referred to herein as a Transmit Control Module, or TCM. In apreferred embodiment, the TCM detects a digital packet which is binary FSK
coded onto an ultrasonic audio carrier at approximately 44 kHz, although
multilevel FSK (i.e. 4-level FSK) could also be used. The packet structure
may be the same as that the CADM of the second embodiment of the invention,
20 described previously in relation to Figure 8.
The reason for using ultrasonic audio frequencies is to reduce the
nl-i.c~nre value of the transit control signals issued by the transit vehicle. These
frequencies are beyond the upper limit of the human hearing range, typically in
excess of 20kHz. Although for purposes of the present description a frequency
25 in the vicinity of 44 kHz is assumed, other ultrasonic frequency ranges could be
used.
Referring to Figure 10, the TCM, like the CADM previously described,
is implemented using a DSP/CPU logic board 1001. The DSP/CPU logic
board issues control signals to a controller 1003. In the case of the TCM,
30 however, the DSP/CPU logic board is preceded by a conventional audio
down-conversion board 1005 which converts the 44 kHz ultrasonic carrier to a
16
CA 02248839 1998-08-19
Wo 97/31437 PcT/Isg7/ool44
nominal carrier frequency of 1 kHz. The ultrasonic carrier may be mod~ ted
by +500 Hz to represent binary 1 or 0.
The down-converter sirnply converts the carrier frequency, in this case
44 kHz, down to an interme~ te frequency which is suitable for processing by
the TCM (300 Hz to 2500 Hz). In this case, the intermediate frequency is
chosen to be 1 kHz. A low pass filter (LPF) 1009 follows a mixer 1007 to
ensure that the only the difference frequency is processed by the TCM.
As in the CADM previously described, preferably the sound level of the
ultrasonic sound is also used in determining if the received command is valid.
This feature ensures that only vehicles in close proximity to the TCM can
actually issue a cornmand at a specific in~t~ tion.
A Coded Audio Transceiver Module (CATM) results by apl~lopliately
combining elements of the Coded Sound Generator of Figure 2 and the Coded
Audio Detector Module of Figure 5. Referring more particularly to Figure 11,
a DSP-based logic board 1103 is coupled to an input sound tr~n~ducer 1101
through an A/D converter 1105. The DSP-based logic board is also coupled to
a Coded Sound Generator consisting of a D/A converter 1107 and an output
sound tr~n~dllcer 1109. If the CATM is to operate in the ultrasonic region, a
down-converter 1102 and an up-converter 1108 may also be provided. In many
instances, the CATM will be coupled to a controller 1111.
The foregoing description has focussed primarily on hardware and
software for vehicular and personal transponders using digital sonic and
ultrasonic communications. The following description will focus on exemplary
applications of such transponders in systems and digital sonic and ultrasonic
communications networks incorporating the same.
In transit applications, it is often desired to know the distance of a
vehicle from a particular location (distance ranging). It may further be desiredto know the actual location of the vehicle, its speed, heading, etc. In a bus
system, for example, passengers at a bus stop would be interested to know how
far away the next bus assigned to a particular route is, and its estimated time of
CA 02248839 1998-08-19
WO g7/3l437 PCT/IB97/00144
arrival at the bus stop. All of the foregoing information may be obtained
through the use of a digital sonic or ultrasonic communications network.
Referring to Figure 12, distance ranging is based on the two way
mess~ging concept used in the Transit Detector Module. It requires that an
5 intersection-mounted unit and a vehicle-mounted unit both support a transmit
and receive capability. In brief, what the system does is as follows. The
intersection-mounted unit (called the base) sends out periodic Query messages
which contain a message sequence number. These are short tr~n.~mi~sions
which are addressed to all vehicle-mounted units (called mobiles). Upon
10 receiving this mobile Query message, a mobile will send a Response message
back to the base which contains that vehicle's identity code and the message
sequence number to which it is responding. If the mobile always transmits the
Response message a fixed time after receiving the Quer~ message, the base unit
can calculate the ~li.ct~nre of that particular mobile from the intersection. If the
15 mobile responds to a number of sequential Q~ery messages, the distance of themobile over time can be established, hence enabling the vehicle velocity to be
derived.
The time for the Query message to cover the distance between the base
and the mobile is simply given by:
T= d/340
where d is the distance in meters, 340 is the speed of sound in meters/second,
and T is the time for the Query message to cover the distance between the base
and the mobile. The time for the Response message to cover the distance
between the base and the mobile is simply given by same equation. If TP is
25 used to represent the total processing time of the message by mobile and the
base, then the total time TTOTAL for a Query message to be sent by the base
and receive a Response message back from the mobile is:
TTOTAL = T + T + Tp
18
CA 02248839 1998-08-19
W O 97/31437 PCT~B97/00144
By rearranging the above equation, the one-way propagation time T can
be solved for as follows:
T= (TrorAL~Tp)/2
Since the nominal speed of sound in air is 340 meters/second, the
S one-way distance between the base and mobile simply becomes:
D = (TTor,~L-T~)/(2 340)
The Query message acts as an ALL CALL message to any vehicle in the
vicinity of the base that it wants any vehicle hearing the Query to respond withthe message sequence number and the vehicle ID. The reason for the message
sequence number in the Query and Response messages is to avoid a situation
where a distant vehicle hears the Query and sends a response message which
arrives after the base has issued a second Query message. In such a situation,
the base could falsely hlle~ et the mobile as being closer than it actually is.
In another application, the "mobile" units may be stationary, and it may
be desired to know if they actually move when they are not expected to. An
example of this is in monitoring cargo trailers in storage facilities. These
trailers are not expected to move unless a controller or yard manager has issueda release for that particular trailer. If the base detects a trailer as moving when
it shouldn't, it may raise an alarm to indicate possible theft of the trailer unit.
Figure 13 illustrates the timing of messages between the base and
mobile, and how the base unit measures speed and distance using this
technique. In this example, the time duration T represents the one-way time of
travel for a message between the mobile and the base, and TP represents the
processing time. This example shows the mobile receiving a Query message
with the Sequence Number 1. Upon receiving the Query, the mobile responds
T + TP seconds later with a Response message cont~ining the Sequence
Number of the Query it received plus the MOBILE ID number. The base
receives this Response T seconds after transmission by the mobile.
19
CA 02248839 l998-08-l9
W O 97/31437 PCT~B97/00144
If the base sends Query messages at fre~uent rate, say once every 5
seconds for example, then it can estimate the vehicle velocity by dividing the
distance of the vehicle at two sequential locations by the time between the
queries. This capability may be used to provide management information to
5 determine what the vehicle velocity was at fixed distances from the base. Thisinformation is useful in determining if the mobile approaching the location
where the base unit is located is: approaching faster than a recommended speed;
approaching slower than a recommended speed; accelerating as it approaches
the location; decelerating as it approaches the location, etc.
This technique can be applied using the two-way cornmunications device
at audible or inaudible (e.g., ultrasonic) frequencies, with the only change being
that the speed of sound used in determining the distance is chosen to correspondto the speed of sound at the frequency of operation used.
As indicated, the technique can be applied to vehicles in motion or those
that are stationary. However, its application is not limited to vehicles. In
general, it is applicable to any object the distance and or speed of which is tobe determined as a function of time. Other exemplary uses may include,
without limitation:
~ Location of marine objects or craft on the water-to detect the
presence of vehicles and/or objects in areas where they shou]d not be located,
or help in locating specific cargo.
~ Location of vehicles and/or objects on an airport loading ramp-to
detect the presence of vehicles and/or objects in areas where they should not belocated.
~ Personal locators to locate people.
The discussion to this point has focussed on a single base unit. If the
scope is broadened to include several base units, it is possible to obtain not only
speed and distance information, but also actual location and heading
information. Referring to Figure 14, in such a system, each base (1412a, b and
c) sends Query messages which contain the ID of the base. A mobile 1413
sends a Response to every Query it hears from every base unit, the Response
CA 02248839 l998-08-l9
W O 97/31437 PCT~B97/00144
including not only the time t but also the ID of the base unit it heard, along
with the Sequence Number and Mobile ID. Each base unit 1412 calculates the
distance and speed pertaining to the MOBILE ID, and sends this information to
a central computer 1416 along with the BASE ID of the base that sent the
S Query and the Time at which the measurement was made. The central
computer 1416 receives similar information from the other base units. The
central computer 1416 then has information pertaining to each MOBILE ID
which would consist of: the distance to a particular based identified by the
BASE ID; and the time the mobile was at that distance.
Since the central computer 1416 knows the coordinates of every base
1412 in the system, it is able to triangulate the actual coordinates of the mobile
1413 at various instants in time. By calc- l~tin~ the location at various points in
time, the actual he~din~ and route of the mobile 1413 may easily be calculated
and used for various purposes, such as audit records, notification of speed
15 violations, notification of movement when not authorized, tracking, etc.
More generally, the system of Figure 14 provides a two-way audio
communication system in which one or more vehicles may comrnunicate with a
network of fixed or mobile units (which could be one unit) to exchange
command, control, and information between the devices. The vehicles may be
20 mobile or fixed. The comrnunication network is based on an acoustic
communication medium and not an electromagnetic one.
Each network element typically comprises an acoustic transceiver, i.e.,
an acoustic tr~ncmitter and receiver. It may send comm~n~c or requests to
mobile/fixed vehicles, or even to other network elements. The network element
25 may do a number of things based on these comm~n-l~/requests, such as: relay
this information to some application program running on a computer somewhere
in the communications network, and relay a response back to the device over
the network element; or convey this comm~n-l/request from a mobile/fixed
vehicle to another mobile/fixed vehicle from the same network element or via
30 another network element elsewhere in the network
CA 02248839 1998-08-l9
W O 97/31437 PCT~B97/00144
Communication between the mobile end systems and the network occurs
using acoustic energy, which is typically in the range of about 100 Hz to over
100kHz. It is important to note that this is an acoustic link and not an
electromagnetic link as in radio.
S The system uses the fact that acoustic trAnsmicsion has a limited range of tr~n~mi~sion di~t~n~e, which is used to advantage by keeping all
communication local to some small area. Wide area coverage is provided by
using a conventional wide area communication network to link all the elements
of the comml-nic-~tion network. Hence in Figure 14, for example, lines 1414a,
b and c linking the base stations 1412a, b and c to the central computer 1416
may be part of the wide-area telephone network, such as dial-up lines or leased
lines.
One particularly advantageous method of connecting acoustic
transponders a wide-area network is through the use of wireless CDPD
(Cellular Digital Packet Data) modems, for example of a type sold by Sierra
Wireless of Vancouver, British Columbia. The CDPD network is IP (Internet
Protocol) -based, allowing for nearly seamless interface with the Internet as
well as certain transit network that are also based on a variant of IP.
Many or all of the foregoing principles may be applied in the area of
siren detection and preemption as described generally in the aforementioned
PCT application. Network connectivity and the ability to detect speed are
particularly advantageous in this application. Digital acoustic transmitters aremounted on emergency vehicles, e.g., within the grill or bumper area. Digital
acoustic receivers are mounted on semaphore overheads at intersections. As an
emergency vehicle approaches, the traffic signals may be preempted to give the
emergency vehicle a green light and other vehicles and pedestrians red lights.
Furthermore, taking advantage of network connectivity, an indication of the
location of the emergency vehicle may be tr:~n.~mitted to a computer at a central
traffic control center. The location of emergency vehicles may be displayed for
viewing by traffic control personnel. Furthermore, the central computer may
perform anticipatory control of other traffic lights in the vicinity of the
CA 02248839 1998-08-19
wo 97/31437 PcT/Is97lool44
emergency vehicle. As the emergency vehicles takes one of several possible
anticipated routes, the vehicle therefore finds traffic already cleared. As the
path of the vehicle is communicated back to the central computer, the
anticipatory preemption of traffic lights along paths not taken is then reversed.
S Another promising application of digital acoustic cornmunications is toll
collection and vehicle spacing. Electronic toll booths, widespread in Europe,
are just beginning to achieve commercial acceptance in the United States. Also
proposed are toll highways requiring vehicles spacing equipment to achieve
maximum safe utilization of the highway. Using directional acoustic
transponders mounted in the front and rear bumper areas of a vehicle, both
electronic toll boothing and vehicle spacing may be achieved. The electronic
toll booth transmits at regular intervals a query signal. When a vehicle
approaches the electronic toll booth, its front-facing acoustic transponder detects
the query and replies. A debit or billing transaction then ensues. As the
vehicle gets underway on the highway, an "autopilot"-like program is engaged.
The front-facing and rear-facing acoustic transponders engage in query/response
communications with vehicles in front of and in back of the subject vehicle
(assuming such vehicles are present). As a result, each vehicle has available toan on-board computer the distance to the vehicle in front and the distance to the
vehicle in back. Speed control is executed to m~int~in the appropriate distance
(not too great, not too small) from the vehicle in front.
The same type of arrangement may be used for collision avoidance,
whether on a toll highway or public highway. The distance, front and back, to
the next vehicle, and the rate of change of distance, is monitored. If an alarm
limit is reached, an alarm may be sounded to the driver. A further alarm limit
may be set to (assuming forward movement of the vehicle) cause the vehicle to
brake (in the case of immin~nt forward contact) or accelerate (in the case of
imminent rearward contact).
Digital acoustic communications may also be used to advantage to
provide information services to vehicle occupants. Increasingly, vehicles are
equipped with CD-ROMs for use in navigation. Typically, the user is required
CA 02248839 1998-08-l9
W O 97/31437 rCT~B97/00144
to ascertain and enter into a computer the vehicle's location, in addition to
desired d~stin~tion. The computer will then retrieve and display a map of the
locale, showing a selected route to the destination. In strange surrol~ntling.~,however, ascertaining one's location is not always an easy task. Referring to
Figure 15, a Dedicated Short-Range Communications System (DSRCS) is
shown. A }arge number of DSRCSs may be supported by a wide area
communications network, e.g., a typical land-line type of network which
supports ITS applications such as fleet management, emergency management,
etc. The DSRCSs provide a wireless communications link between the
land-side network and the vehicles using the services of the network. As a
vehicle moves along the road system, its physical location changes with time.
Many Road-Side Systems (RSSs) are placed along the road so that the
vehicle-based communication element can be in contact either continuously or
intermittently as it travels along the road. Likely locations for RSSs include
major intersections and key points along roads and highways, similar to a
cellular radio system where many cell sites are placed over a large geographic
region to provide wide area coverage for cellular radio users. Unlike cellular
radio, however, the illustrated network is based on digital acoustic
communications. Furthermore, the RSSs are only placed along a roadway, and
there may be small to large gaps in coverage along the roadway. Each RSS in
the network is located at a physically unique location, with this location having
a known geographic location (i.e.~ Iatitude/longitude). Each RSS broadc.~.ct~ a
specific packet (RSS Identifier Packet) which announces the presence of the
RSS to any nearby vehicles which can hear the RSS. The RSS Identifier Packet
contains a unique Cell Identifier associated with the RSS. A unique Cell
Identifier is assigned by the network management system to each RSS.In the
example of Figure 15, five RSSs provide coverage at key points along the road.
Between cells 1 and 99, there is an area with no coverage. Cells 3, 11 and 65
provide almost continuous coverage along a main artery.
A computer system in the vehicle, upon receiving the RSS Identifier
Packet, associates the unique Cell Identifier with a geographic location
24
CA 02248839 1998-08-l9
W 0 97/31437 PCT~B97/00144
(latitude/lon~ de) for that cell using some form of mass storage such as
CD-ROM~ The geographic location may then be used to retrieve information
of use to a driver in the vicinity of that RSS. For example, the geographic
location may be used to recall stored map information for the area in the
5 vicinity of the RSS location and display it to a driver. Or, in the case of a
tourist, display points of interest to a driver. Similarly, hotel, restaurant orother information may be recalled and displayed to the driver. In the case of a
utility or service vehicle, Graphical Information System (GIS) data could be
recalled and displayed to a driver. This could include such information as
10 location of gas lines, location of electrical lines, sewer information, etc. In the
case of emergency vehicles, information specific to the emergency service may
be recalled from the mass storage device. In the case of a fire department
vehicle, for example, such information could include the location of hazardous
or toxic materials, information on nearby hydrants, etc. CD-ROMs cont~inirlg
15 information of interest may be purchased by drivers or installed in vehicles by
fleet operators. The low cost and high storage capacity of CD-ROM makes it
very attractive for this purpose. Furthermore, the storage capacity of
CD-ROM-like media (DVD, etc.) may be expected to increase, allowing for the
storage and retrieval of media-rich information such as maps, photos, video,
20 and audio in addition to text.
Other transit-related applications of the digital sonic and ultrasonic
communications networks include controlling access to controlled areas. The
controlled area may be a structure such as a garage, a toll-bridge or toll-road,etc. Alternatively, the controlled area may be a geographic area, as in the case25 of border crossings between states or countries. Acoustic transponders may
also be carried on one's person. Exemplary applications of personal acoustic
transponders include access control and personnel monitoring. An acoustic
transponder may be used to provided access to a locked building, for example.
An acoustic transponder may also be used to monitor the whereabouts of
30 children or pets. Numerous other applications of sonic systems as described
herein will be apparent to one of ordinary skill in the art.
. .
CA 02248839 1998-08-19
W O 97/31437 PCT~B97/00144
It will be appa,Gllt to those of ordinary skill in the art that the invention
can be embodied in other specific forms without departing from the spirit or
essential character thereof. The foregoing description is therefore considered in
all respects to be illustrative and not restrictive. The scope of the invention is
5 in~licatecl by the appended claims rather than the foregoing description, and all
changes which come within the me~ning and range of equivalents thereof are
intended to be embraced therein.
26
