CA 022629~ 1999-02-03
WO 98/39882 PCT/US97/03182
IMPROVEMENTS IN ULTRAFAST TIME HOPPING CDMA-RF
RA~R~OUND AND BRIEF DESCRIPTION OF THE lNv~ lON
The present invention relates to a wireless RF time
hopping code-division-multiple-access (CDMA) and time-
division-multiple-access (TDMA) spread spectrum
communications systems, and specifically to ultrafast
systems, which use individual ultrashort pulses (monocycle)
or small number of cycles (packets) signals in the
picosecond (10-1~) through nanosecond (10-9) to microsecond
(10-6) range. Before transmission and after reception, the
system functions as a digital communications system. The
carrier for such wireless communications systems is neither
a frequency, amplitude, phase nor polarization carrier, but
is due to the precise timing arrangements in a sequence of
individual pulses provided by the digital coding schemes.
Whereas most wireless RF communications systems in the
art use frequency-domain receiver designs based on the
heterodyne, or super heterodyne principle, the receiver of
the present invention is a time-domain homodyne receiver.
Whereas prior art uses coding, e.g., in direct sequencing
or frequency hopping, to achieve spreading and despreading
.
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
of the signal with resultant processor gain, the present
invention uses an ultrashort pulse or packet as an
individual signal as well as coding which determines the
timing of such individual pulses within a sequence.
Information is carried in a transmission by the pulse
position modulation technique, i.e., by precise micro-
deviation from the pulse sequence timing set by the channel
code.
Due to the use of orthogonal coding schemes and the
use of ultrafast pulse sequence techniques, it is possible
to provide extremely high data rate wireless point-to-point
communications, as well as wide area multimedia
communications.
The invention in my above-identified patent
significantly increases the data rate of wireless RF
communications by using orthogonal coding schemes in
ultrafast time hopping CDMA communications both in point-
to-point and broadcast mode; it provides a communications
system which can coexist without interfering with, or
causing interference to conventional RF transmissions or
other ultrafast time hopping CDMA or TDMA users; it also
provides wireless communications system which can interface
with digital, e.g., optical fiber, communications systems;
and a communications system which is robust against
environmental notched filtering of frequency components in
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
the transmitted signal; and provides communications system
which has substantial range at modest power, is small in
size, weight and is not costly to manufacture.
Briefly, the above features are achieved in an RF
ultrafast time hopping CDMA and TDMA wireless
communications system, which uses individual pulses and
packets in a sequence of such pulses or packets, those
individual pulse/packets being so short in duration (e.g.,
in the picosecond and nanosecond range) that the individual
pulse signal energy is spread over very many frequencies
simultaneously or instantaneously (instead of sequentially)
with respect to a slow sampling system. A time hopping
sequential code is also used to position these
pulse/packets precisely in sequence providing optimum use
of time-frequency space and also providing noninterfering
transmission channels due to the orthogonality of the
coding schemes used. The ultrashort nature of the
individual pulses/packets used also permits the time
duration of a frame to be divided into very many
microintervals of time in which the signal could occur.
This division into very many microintervals in a frame
permits the availability of many possible coding schemes as
well as many noninterfering transmission channels. Thus
the ultrashort nature of the individual pulses, together
with orthogonal coding schemes, permits the highest
CA 022629~ 1999-02-03
WO 98/39882 PCI/US97/03182
multichannel or aggregate data rates of any wireless
communications system.
In one embodiment of the invention in my above patent,
a communications system uses: (i) orthogonal codes which
can be slaved to a single acquisition system/matched filter
and which captures and assigns each code to unique decoding
modems; (ii) correlators/acquisition systems/matched
filters which are able to detect the ultrafast signals and
retain memory of such capture over superframes of the order
of a millisecond; (iii) pulsed power sources, antennas,
encoding modems, oscillator-clocks, intelligence/data
encrypters; and (iv) EPROMs to provide coding information
to both encoding and decoding modems.
The present invention addresses optical time hopping
code-division-multiple-access (CDMA) and time-division-
multiple access (TDMA) spread spectrum communications
systems, and specifically ultrafast systems and high data
rate systems, which use individual ultrashort signals in
the femtosecond (10-15) range, but could apply to signals of
other temporal lengths. Before transmission and after
reception, the system functions as a digital communications
system. As in my above-identified patent, the carrier for
such wireless communications systems is neither a
frequency, amplitude, phrase nor polarization carrier, but
is due to the precise timing arrangements in a sequence of
., . _
CA 022629~ 1999-02-03
W098139882 PCT~S97103182
individual pulses or wave packets provided by the digital
coding schemes, i.e., a macrocode.
The present invention addresses a system and method in
which information is transmitted by a pulse or packet
position modulation technique, i.e., by precise micro-
deviation from the pulse/packet sequence timing set by the
channel code or microcode, e.g., by pulse/packet
modulation.
Due to the use of orthogonal coding schemes and the
use of ultrafast pulse sequence techniques, it is possible
to provide extremely high data rate optical fiber broadcast
as well as point-to-point communications.
OBJECTS OF THE PRESENT INV~;r. 110N
Accordingly, it is a further object of the present
invention to significantly increase the data rate of
optical fiber communications (Gagliardi & Karp, 1995;
Spirit & O'Mahony, 1995) by using orthogonal coding schemes
in ultrafast time hopping CDMA or TDMA communications both
in network broadcast and point-to-point mode.
It is a further object of the invention to provide an
optical fiber communications system which can interface
with digital, e.g., RF or wired line, communications
systems.
CA 022629~ l999-02-03
WO 98/39882 PCT/US97/03182
S~M~Y OF TEIE PRESENT INVENTION
Briefly, the above and other objects of the present
invention are achieved in an ultrafast time hopping CDMA
and TDMA wireless communications system, which u~es
individual wave packets in a sequence of such wave packets
or pulses. A time hopping sequential code is also used to
position these wave packets or pulses precisely in sequence
providing optimum use of time-frequency space and also
providing noninterfering transmission channels due to the
orthogonality of the coding schemes used. This time
hopping sequential code is referred to as the "macrocode"
and identifies each information subchannel within the total
channel. The data is encoded on each pulse or wave packet
in modulation schemes which can be either parallel or
serial -- the latter case, a pulse position modulation
scheme. The data encoding scheme is referred to as a
"microcode~.
As in the original invention, the ultrashort nature of
the individual pulses used also permits the time duration
of a frame to be divided into very many microintervals of
time in-which the signal could occur. This division into
very many microintervals in a frame permits the
availability of many possible coding schemes as well as
many noninterfering transmission channels. Thus, the
ultrashort nature of the individual pulses or wave packets,
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
together with orthogonal coding schemes, permits the
highest multichannel or aggregate data rates of any optical
communications system.
In one embodiment of the present invention, a
communications system uses: (i) orthogonal codes which can
be slaved to a single acquisition system/matched filter and
which captures and assigns each code to unique decoding
modems; (ii) correlators/acquisition systems/matched
filters which are able to detect the ultrafast signals and
retain memory of such capture over superframes of the order
of a millisecond; (iii) pulse train sources, encoders,
oscillator-clocks, intelligence/data encrypters; and (iv)
devices, e.g., acousto-optic modulators and holograms, to
provide coding information to both encode and decode. The
methods (o) - (iii) are also used by the original
invention. The methods (iv) are specific to the present
invention.
BRIEF DESCRIPTION OF THE DRAWINGS
The above and other objects, advantages and features
of the invention will become more apparent when considered
with the following specification and accompanying drawings
wherein:
CA 022629~ 1999-02-03
W098/39~2 PCT~S97/03182
FIG. la is a block diagram of a transmitter
incorporating the invention configuration for an ultrafast
time hopping CDMA wireless communications system, and FIG.
lb is a block diagram of a receiver configuration thereof,
5FIG. 2 illustrates frames and subframes in an
ultrafast time hopping CDMA wireless communication system
incorporated in the invention.
FIG. 3 is a diagrammatic exposition of the subframe,
frame and superframe,
lOFIG. 4 is a diagrammatic exposition of one method for
achieving correlation and subframe sampling,
FIGS. 5a, 5b and 5c illustrate two orthogonal codes
(Fig. 5a and 5b), and their auto- and cross-correlation
(Fig. 5c),
15FIG. 6 illustrates a hyperbolic congruence code, p =
ll, a = l, lO X lO matrix,
FIG. 7 illustrates a hyperbolic congruence code, p =
ll, a = l, 50 X 50 matrix,
FIG. 8 illustrates the autoambiguity function for the
20hyperbolic congruence codes, p = ll, a = l and p = ll, a =
3, lO X lO matrix,
FIG. 9 illustrates the crossambiguity function for the
hyperbolic congruence codes, p = ll, a = l and p = ll, a =
3, lO X lO matrix,
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
FIG. lOa illustrates a time-frequency representation
of an ultrashort pulse of l nanosec. and a synchronous
(e.g., heterodyne) receiver for a narrow-band sinusoid, and
FIG. lOb is a bird's-eye-view of Fig. lOa.
FIG. ll is a cut through the three-dimensional of Fig.
10,
FIG. 12 ill~strates an acquisition system according to
the invention,
FIG. 13 illustrates a detail of the acquisition system
and decode modems,
FIG. 14 illustrates an optical mode broadcast network
incorporating the invention,
FIG. 15 illustrates an optical mode point-to-point
system incorporating the invention,
FIG. 16 is a diagrammatic illustration of macrocode
and macrocode embedding incorporated in the invention,
FIG. 17 is a diagrammatic illustration of data
recovery incorporated in the invention, and
FIG. 18 illustrates the interfacing aspects of the
present invention.
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
DE~ATT~n DESCRIPTION OF THE PREFERRED EMBODIMENT
There are many possible embodiments of an ultrafast
time hopping CDMA and TDMA system incorporating the
invention. The following is an embodiment which permits
multichannel (high data rate) use.
1. Oscillator Clock 10, 10'. This circuit can use,
e.g., GaAs MMIC technology, or other semiconductor
technology, to convert DC power to a 2 GHz signal. The
output signal of the oscillator clock will have sufficient
power to drive the data gate circuitry and transmitter
amplifier (during transmission of a pulse or packet). The
oscillator clock is a crucial subcomponent requiring an
accuracy >20 picoseconds in 1 msec., or about 20 parts in
109.
The signal can be generated by a voltage-controlled
oscillator phase-locked to a frequency stable reference
signal.
2. Pulse Emitter and Antenna Module TA. During the
transmission of an on-going pulse a sample of the
oscillator signal is amplified and transmitted out of the
antenna. An RF switching circuit (in pulse generator PG)
driven by the comparator COMP trigger permits the
oscillator clock to drive the transmitter amplifier chain
for the duration of the pulse. The transmitter amplifier
CA 022629~ 1999-02-03
W098l39882 PCT~S97/03182
chain delivers the resulting RF pulse or packet to the
antenna at a power level required by the system.
The amplifier can be, e.g., a cascaded set of GaAs
MMIC chips, or other semiconductor technology. The
bandwidth and impedance matching of these amplifiers can be
achieved by, e.g., a distributed network of parallel
MESFETs, or by other semiconductor methods. The input and
output parasitic capacitances of the devices is absorbed by
series inductances which in effect form a lumped element 50
ohm transmission line.
The antennas per se for both transmit and receive used
can either be, e.g., of the nonresonant kind, e.g., or
nondispersive TEM horn designs. In many cases, printed
circuit methods can be used to fabricate the antennas on
lS the circuit boards, as well as other methods of
fabrication.
3. Acquisition Module AM. The acquisition module can
be based on designs, e.g., using associative string
processor modular technology or other means. This module
is described in detail below.
4.- Modems/Encoders and Modems/Decoders (Data Gate
Circuitry). The data gate circuitry is common to both the
transmitter and receiver. It can consist of, e.g., very
high precision GaAs digital circuitry, or other
semiconductor circuitry. The subframe counter is a free
. .
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
running counter driven by the clock oscillator. The output
of the counter is compared to the look-up code
corresponding to the frame counter.
The digital gate circuitry can be achieved using,
e.g., ECL compatible source coupled logic on GaAs, or other
semiconductor technology. Gate length and width can be
chosen to reduce the parasitic capacitances such that
loaded gate speed of less than 50 picoseconds can be met.
The receiver data gate counters are reset when a
transmission is received. A high speed data ~atch is
triggered to capture the output of the pulse detector
during the subframes triggered by the code. The output of
the data latch contains the transmitted data including
error correction which corresponds to the position of the
pulse within the subframe.
The transmitter data gate subframe and frame counters
are free running. Whenever the subframe counter and codes
match the pulse/packet generator is triggered causing a
high speed pulse to be transmitted. The pulse/packet
position in the subframe corresponds to the data and error
correction codes of the least significant bits at the
inputs to the Acquisition Module.
5. Code EPROMs (Code Look-up 14, 14'). The code
generation function can be performed by EPROMs in code
look-up 14, 14' in the transmitter and receiver. Once per
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
frame, a pulse/packet is generated by the frame counter
EPROM. The code specifies in which subframe the
pulse/packet will occur. With the use of more than one
code (data rate on demand) the EPROM will provide more than
one code to the transmitter and receiver. Alternatively,
phase-shift registers can be used to generate the codes.
6. Pulse Detector (Rise TLme Trigger 15). Background
interference can be rejected by a rise-time triggering
circuit, which is not merely a high pass filter. In order
to achieve rise-time triggering, the RF signal can be
passed through an envelope detector which is then fed
through a high-pass filter before reaching the trigger
threshold circuit. The high pass filter then
differentiates the envelope and passes transients while
rejecting 510w changes.
7. Receiver (Fig. lb). The receiver is a homodyne
receiver (not a heterodyne receiver). The receiver
preamplifier (not shown) needs a maximum of 40 db gain, no
AGC, and a noise figure of approximately 5 db. The
preamplifier of the receiver antenna RA feeds a
pulse/packet detector 15 which output and ECL pulse for
each detected pulse/pac~et. The pulse/packet detector 15
feeds the Acquisition Module AM, which includes correlator
CO' which output triggers to the frame counter FC' and
subframe counter SC'. The remainder of the receiver is
CA 022629~ 1999-02-03
WO 98/39882 PCT/US97/03182
similar or complementary to the transmitter. In a
preferred embodiment, a high-speed counter SC' gates a data
latch DL when the counter value matches the current main
code value. The high-speed counter SC' wraps around at
each frame interval. This wrap increments a frame counter
FC', which is used to look up the code commencement in the
EPROM 14'. The frame counter FC' wraps around at each
superframe interval. The data latch DL' feeds the FEC
decoder FEC' and optional decryptor DEC, which operates at
the frame rate (about 1 Megabit per second). The receiver
design is further described below.
With the use of multiple codes (data rate on demand)
the EPROM 14', or phase-shift register, or other means of
code generation, will provide more than one code to both
the receiver and transmitter.
THE CODES:
The wireless communications network can be used in
either network or duplex arrangements. Two levels of
coding are used in systems of the present invention. the
major code is used to time the pulse transmission and allow
multiple channels. Additionally, a forward error
correction (FEC) code CAN BE applied to the informational
data before transmission. There is a large choice of error
correcting codes (see Cipra, 1994).
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
The use of orthogonal codes permits the coexistence of
multiple channels slaved together in the same superframe of
a matched filter. Representative such codes are Quadratic
Congruence (QC) codes, Hyperbolic Codes (HC) codes and
optical codes (Titlebaum & Sibul, 1981; Titlebaum et al,
1991; Kostic et al, 1991). The discussion of coding
requirements will be based on these codes.
The method for generating the placement operators for
the QC code family provides a sequence of functions defined
over the finite field, JP, where:
JP = {0,1,2, . . . ,P - 1},
and p is any odd prime number. The functions are defined
as:
y(k;a,b,c)=[ak2 + bk + c]~odp,k~Jp,
where a is any element of Jp except O and b,c are any member
of JF. The parameter a is called the family index.
The difference function for the HC codes is the ratio
of two quadratic congruences. The denominator polynomial
of the ratio cannot have any zeros and the numerator is
quadratic and has at most two zeros. Therefore, the HC
codes have at most two hits for any subframe or frame
shift. A sequence, u~(i),i=0,1,2,...,n - 1, which is a
member of a time hopping code can be constructed according
to a method shown in Fig. 2. For example, pulses received
in the first interval of the macro-window signify ~ and
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
those received in the second interval signify '0". The
number of bits in the subframe (microwindow) is determined
by the precision of the data gate circuitry.
Alternatively, the subframe can be used to encode analog
information.
Fig. 5 shows the auto- and cross-correlation of two
orthogonal codes. The autocorrelation is excellent
indicating excellent transmission/reception capabilities.
The crosscorrelation is extremely flat, indicating
excellent cross-channel interference rejection.
Fig. 6 and Fig. 7 shows two HC codes, p = ll, a = 1,
10 X 10 matrix, Fig. 6, and, p = 11, a = 1, 50 X 50 matrix,
Fig. 7 and Fig. 8 shows the autoambiguity function, for the
HCC code, p = 11, a = 1, 10 X 10 matrix and Fig. 9 shows
the cross-ambiguity function, for the HC codes, - = 11, a
= 1 and p = 11, a = 3, 10 X 10 matrix.
The QC codes are defined as:
v(x) [ xfr+/)]
with l_a<p-l and O_x~p-l for p x p matrices.
The HC codes are defined as:
~(x)=-m~dp
where--is the multiplicative inverse in the field J~ and with I S ~ < p and I < x < p for
p - I x p - / matrices.
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
The QC and HC codes, due to Titlebaum and associates
(Albicki et al, 1992; Bellegarda & Titlebaum, 1988-1991;
Drumheller & Titlebaum, 1991; Kostic et al, 1991; Maric &
Titlebaum, 1992; Titlebaum, 1981; Titlebaum 7 Sibul, 1981;
S Titlebaum et al, 1991), are representative orthogonal codes
which can be used in the present invention, the ultrafast
time hopping CDMA and TDMA communication systems. Other
choices are available in the literature.
The major codes used in systems of the present
invention are equally applicable to optical orthogonal
coding procedures for fiber optical communications. The
use of orthogonal codes permits the coexistence of multiple
channels slaved together in the same superframe of a
matched filter. For ease of explanation, the following
terms are defined in Table 1.
l'able 1
Subframe The, e.g., - I nano~second interval during which a pulse is transmitted.
The pul.se is modulaled by adjusting its position within the interval to
one of two or morc po~s~sible times. For example, to send one bit per
subframe, the pul~se may be offset from the center of the subframe by
-250 picoseconds for a zero, or +250 picoseconds for a one.
Frame A, e.g, -I microsecond, interval, divided into approximately 1()00
subfralnes (or accor(iillg to the code length). A pulse is transmitted
during one subframe out of each frame. The pulse is sent during a
different subframe for each frame and according to the code.
CA 02262955 l999-02-03
W O 98/39882 PCTAJS97/03182
Superframe A, e.g, - I millisecond, interval, ICpl~S~ U;.~g one repetition of a code
pattern. ~n the present example, appro~imat~ly 1000 pulses are
transmitted during one supclrr.~ e, at pseud~landG~I- sp~(in~
Channel One unidirectional data path using a single orthogonal code. The raw
(uncorrected) capacity of one channel using a code of length 1020 is
approximately 0.5 mbs. Using all 1020 codes, the channel data rate is
approximately 500 mbs.
An example of the relations between subframe, frame, supe.rldllle and channel is given in
Table 2 for a code of length 1020, Table 3 for a code of length 508, and Table 4 for a code
of length 250.
'I'able 2
lar~est count of code modulation C' I
code period 1021 - I ~ 20 frames
hop slot duration (the subframe) ng~ picoseconds
frame time interval (the frame) 927 x 10-12 x 1020 50 nanoseconds
time of one complete code period (the 1020 x 950 x 10-9 0.97 milli~econds
superframe)
fraction of frame time for encoding 1020 x 695 X 1O-12 0 75
950 X lO-Y
forward error correcting redundancy 2
data interval 2 x 950 x 10-9 1.9 microseconds
data encoding interval subframe 2 x 695 picoseconds 1.39 nanoseconds
data rate I ~526 kbs
l.9x 10~
data frame subinterval 347.5 picoseconds
dataframe bandwidth 695 = 2 I bit
347.5
single channel data rate I x526x 103 -526kbs
maximum number of codes 1020
multichannel data rate using 1(~ 10x 526 x 103 5.26 mbs
codes
multichannel data rate using 1020x526x 103 537 mbs
the maximum number of codes
'rable 3
largest count of code modulation So9
code period 509- 1 508 frames
hop slot duration (the subframe) 695 picoseconds
frame time interval (the frame) 927 x 10-l2 x 508 472 nanoseconds
time of one complete code period (the 508 x 472 x 10-9 0.24 milliseconrlc
superframe)
r
CA 022629~ 1999-02-03
WO 98/39882 PCT/US97/03182
fraetion of frame time t'or encoding 508 x 695 x 10-" 0 75
472x lo-Y
forward error correcling re(lundancy 2
data interval 2 x 472 x 10-9 0.94 mieroseconds
data encoding interval ~subl'rame 2 x 695 picoseconds I 35~ nanoseeonds
data rate I ~ 11)59 kbs
0.~4x 1~)~
data frame subinterval 347.5 picoseconds
dataframe bandwidlh '55~5 =2 1 bit
347.5
single channel data rate I x lOS9x 103 -1059kbs
maximum number ol' codes 509
multichannel data rate using 1~ x 1059 x 103 10.59 mbs
codes
multichannel data rate usin~ 509 x 1059 x 103 539mbs
the maximum number Or codes
l'able 4
largest count of code modula~ion 251
code period 251 - I ' 50 frames
hop slot duration (the subtrame) ,95 picoseconds
frame time interval (the trame) 927 x 1o-12 X 250 233 nanoseconds
time of one complete code perio(l (the 250 x 233 x 10-9 0.06 milliseconds
superframe)
fraction of frame time tor encoding 250 x 695 x /o-'2 0.75
233 x / O~Y
torward error correcting r e~lun(lallcy 2
data interval 2 x 233 x 1o-9 0.47 microseconds
data encoding interval subframe 2 x 695 picoseconds 1 3~ nanoseconds
data rate I ~2130 kbs
0.47 x 10~
~lata trame subinterval 347.5 picoseconds
dataframe bandwidth 695 2 I bit
347.5
sin~le channel data rate I x2130x 103 ~2130kbs
maximum number of code~ 251
multichannel data rate using 11~ x 2130x 103 21.3 mbs
codes
multichannel data rate using 250x 2130 x 103 533 mbs
the maximum number of codes
The ~subframe. frame and ~superframe relations are shown in Pig. 3.
CA 022629~ 1999-02-03
WO 98/39882 PCT/US97/03182
THE RECEIVER:
In conventional frequency domain heterodyne receivers
the mixer is by far the preferred front-end component. In
general, mixers are used to convert a low-power signal from
one frequency to another by combining it with a higher-
power local oscillator (LO) signal in a nonlinear device.
Usually, the difference frequency between the RF and LO
signals is the desired output frequency at the intermediate
frequency (IF) at subsequent IF amplification. Mixing with
local oscillators downconverts to intermediate frequencies
and in the IF section narrowband filtering is most easily
and conveniently accomplished. Subsequent amplification
and detection is based on the intermediate frequency
signal.
The operation of a detector in the mixing code results
in a much lower conversion loss and is the reason for the
excellent sensitivity of the superheterodyne receiver. The
mixing action is due to a nonlinear transfer function:
I = f(V) = aO + a,V + a2V~ + a3V3,...anVn,
where I and V are the receiver current and voltage If VRF
sin~RFt is the RF signal and VLO sin~LOt is the LO signal,
then the mixing products are:
~1 I ( RF ~RFt Vr n sin Cl~OI ) + a2 ( VRF sinc~)rrFt+ VLO sincl~LOt) +
+al(vRFsin~)RFt+v~(7sincL~lnt) ~ n(vRFsino~Ft+vLosin~)Lot)n
CA 02262955 1999-02-03
WO 98139882 PCTIUS97/03182
The primary mixing products come from the second-order
term. However, many other mixing products - may be present
within the IF passband. Mixing produces not only a new
signal but also its image, i.e., ~LO + ~RF. However, in the
case of ultrafast time domain signals filtering could
severely limit the amplitude of the signal and hence its
range.
For example, the second-order term for a narrow-band
frequency domain signal is:
a2 (VRFsin~RFt+vLOsin~,ot)2
but for a broad-band time domain ultrafast signal it is:
a2(V~F 'Sin~RFt + V~F. ~in~r,t + V~F~ .sin~Flt+ V~F .sin~rt+.. V,Osin~t)
The output is then:
~V~F Sj n~)~rt~ r .S in~)r~,I.
1~ which possesses too many intermodulation products for use
as an IF input.
Therefore, due to the broadband nature of the
ultrafast, ultrawideband individual signal, the synchronous
(super) heterodyne receiver should not be the choice in
receivers of the present invention due to the number of
mixing products produced unless frequency selectivity for
CA 022629~ 1999-02-03
WO 98/39882 PCTlUS97/03182
packets is desired. The receiver of choice for the present
invention is a homodyne receiver.
A problem in definition arises in the case of the
homodyne receiver. We take our definitions from optical
physics (Born & Wolf, 1970; Cummins & Pike, 1974), not from
radar engineering. Essentially, the heterodyne method
requires a local oscillator to be mixed with the received
signal and is a ''self-beat~l or autocorrelation method. The
homodyne method is inherently a coherent method [cf. Born
& Wolf, 1970, page 256]. The heterodyne method can be used
with autocorrelation methods, e.g., after the mixing
operation. The heterodyne method can even use a ~Icoherent~
local oscillator, but only with a narrowband signal. The
distinguishing features between the two methods are that
the homodyne method is a coherent (correlative) signal
acquisition method with (a) no restrictions on the
bandwidth of the received signal and (b) restrictions on
the absolute timing of the signal bandwidth components.
Conversely, the heterodyne method is a signal acquisition
method with (a) no restrictions on the timing of the
received signal and (b) restrictions on the bandwidth of
the signal frequency components.
The various definitions of the heterodyne and homodyne
methods are not consistent. For example, the IEEE Standard
Dictionary of Electronic Terms (Jay, 1988) defines
CA 022629~ l999-02-03
W098~9882 PCT~S97/03182
'homodyne reception" as "zero-beat reception or a system of
reception by the aid of a locally generated voltage of
carrier frequency"; and the McGraw-Hill ~ictionary of
Science and Technology (Parker, 1989) defines "homodyne
reception" as "a system of radio reception of suppressed-
carrier systems of radio telephony, in which the receiver
generates a voltage having the original carrier frequency
and combines it with the incoming signal. Also known as
zero-beat reception.
Essentially, these definitions (Jay, 1988; Parker,
198g) refer to the generalized use of homodyning in a
receiver with more than one mixing stage. "Synchronous"
detection is achieved by a method called "homodyning",
which involves mixing with a signal of the same frequency
as that being detected either by external or internal
(i.e~ with a phase loop) methods. Thus, recently the term
homodyne~l has come to mean a method for the detection of
narrowband signals and for restoring a suppressed carrier
signal to a modulated signal.
Clearly the waters have been muddied concerning the
definitions of the heterodyne and homodyne methods.
However, the original optical physics definitions are
specific in equating heterodyning as a method of signal
acquisition using a local oscillator, and homodyning as a
method of signal acquisition using a coherent method such
CA 022629~ 1999-02-03
W098/3~82 PCT~S97/03182
24
as autocorrelation. Ultrafast, ultrashort pulse/packet
signal acquisition requires the homodyne method because it
is a coherent method and preserves timing information. It
would not be right to coin new terms, because the present
terms continue to survive relatively unambiguously in
optical physics from whence they came. Therefore, we shall
use the terms in the optical physics sense of those terms,
but cautiously recognizing the danger of triggering the
wrong associations.
The distinction between homodyne and heterodyne
reception is significant and bears on the claims that the
present invention is noninterfering to conventional, i.e.,
heterodyne, receivers. In Figs. lOa and lOb, is shown both
ultrafast, ultrashort pulse homodyne reception and narrow-
band synchronous signal heterodyne reception in time-
frequency space. Fig. lOa is a time-frequency
representation of an ultrashort pulse of l nanosec. and a
synchronous (e.g., heterodyne) receiver for a narrow-band
sinusoid. Looking only from the frequency axis, the
(exaggerated) spike of the ultrashort pulse would appear to
overlap with the rising ridge of the narrow-band heterodyne
receiver, i.e., the heterodyne receiver would appear to
receive any of the ultrafast, ultrashort signals. However,
looking at the total time-frequency plane representation,
it can be seen that the ridge representation of the narrow-
CA 022629~ 1999-02-03
W O 98/39882 PCTrUS97/03182
band synchronous heterodyne receiver does not extend down
to the representation of the ultrafast signal
representation. The synchronous, heterodyne receiver takes
some time to respond and requires a number of cycles of a
S signal to receive that signal.
The distinction between the homodyne reception of the
present invention and conventional heterodyne reception is
shown in Fig. 11, which is a cut through Fig. 10a and
viewed from the time side. The ultrafast, ultrashort pulse
signal is shown to be diminished in amplitude for
conventional heterodyne receivers of all attac~ (rise)
times, even if the ultrashort signal's average frequency is
at the center frequency of the heterodyne receivers. On
the other hand, homodyne reception preserves signal
amplitude and timing. While homodyne reception is
preferred, heterodyne reception can be used.
THE ACQUISITION SYSTEM:
The acquisition system/matched filter of the present
invention recognizes multiple codes (channels) over a
superframe time period tl msec. for a 1020 length code).
Fig. 12 shows an Acquisition System, ~, receiving F
wireless signals from four channels Sl-S4 in asynchronous
wraparound and triggers the receiving system decoding
modems, Sl-S4.
CA 022629~ 1999-02-03
WO 98/39882 PCT/US97103182
Fig. 13 shows an embodiment in which the superframe of
each transmission, Sl-Sn, e.g., with codes of length 1020,
is preceded by a preamble frame, e.g., of length 1~. This
preamble must be received in double wraparound for the case
in which the channels, S!-Sn, are unsynchronized in their
transmissions. In this embodiment, the preamble is the
same code for all channels, even if unsynchronized, and
acts as a synchronization alert to the Acquisition System,
which performs recognition of the channel code and assigns
a decoding modem. Unlike in the embodiment of Fig. 12, in
the embodiment of Fig. 13 the Acquisition System is not
functioning in wraparound mode, but is alerted to the
beginning of a transmission of a superframe by the
preamble, which is in double wraparound.
T~E NETWORgS:
The network applications of the present invention are
diverse and range from high data rate duplex systems, to
building-to-building systems, to the linking of optical
fiber networks between such buildings, to within building
communications, to LANs and WANs, to cellular telephones,
to the ~llast mile" of Global Grid communications, and to
smart highway" applications (Varaiya, 1993) (e.g.,
intelligent windshields, etc), etc.
CA 022629~ 1999-02-03
W098139882 PCT~S97/03182
APPLICATION AREAS:
Wireless WANs and LANs;
Personal Communications Networks;
Cellular Telephones;
Building Automation/Security Systems;
Voice communications;
Bridge & Router Networking;
Instrument Monitoring;
Factory Automation;
Remote Sensing of Bar Codes;
Vehicle Location;
Pollution Monitoring;
Extended-Range Cordless Phones;
Video TeleConferencing;
Traffic Signal Controls;
Medical Monitoring and Record Retrieval Applications;
Remote Sensing;
Factory Data Collection;
Vending Machine Monitoring;
"Last Mile" Global Grid Communications.
The invention includes the following features:
a) Apparatus and methods of ultrafast, ultrashort
pulse/packet transmission.
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
b) Apparatus and methods of transmitting sequences of
such ultrafast, ultrashort pulses/packets.
c) Apparatus and methods of pulse/packet interval
modulating such sequences according to a macro-coded
5scheme.
d) Methods of pulse/packet interval modulating such
sequences within a microwindow of the macrowindow set by
the code such that information can be encrypted in that
microwindow.
lOe) Codes stored in matrix form as, e.g., associative
memories and with superframes of received signals matched
against the stored codes.
f) Codes which are orthogonal codes and the temporal
coding of the sequence of ultrafast, ultrashort
15pulses/packets constitutes the carrier for the
transmission.
g) A homodyne, not a synchronous heterodyne, receiver.
h) An acquisition system/matched filter/correlators
which synchronizes to a superframe transmission and assigns
20such transmissions to an appropriate decoding modem on the
basis of code recognition.
i) Multichannel operation which provides high overall
or aggregate data rate (e.g., ~500 mbs for maximum
multichannel operation).
.
CA 022629~ 1999-02-03
W098t39882 PCT~S97/03182
29
The attached paper entitled "Comparison of
Communications... and disclosure statement~ and the paper
entitled "Reference" filed herewith are incorporated herein
by reference.
Summarizing, an ultrashort pulse/packet time hopping
code-division-multiple-access (CDMA) and time-division-
multiple access (TDMA) RF communications system in the time
frequency domain comprises a transmitter including:
a) a short duration pulse/packet generator for
generating a short duration pulse/packet in the picosecond
through nano-second to micro-second range and a controller
for controlling the generator,
b) code means connected to the controller for varying
the time position of each short pulse/packet in frames of
pulses/packets in orthogonal superframes of ultrafast time
hopping code division or time division multiple access
format,
c~ precise oscillator-clock for controlling such
timing,
d) encoding modems for transforming intelligence into
pulse/packet position modulation form,
e) antenna/amplifier system connected to said means
for generating for receiving and broadcasting said short
duration pulse/packets as a coded broadcast signal,
receiver means, said receiver means including:
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
a) antenna/amplifier system for receiving the
broadcast signal,
b) homodyne receiver for receiving and decoding the
coded broadcast signal, and
c) one or more utilization devices connected to the
homodyne receiver. The coding means generates sequences of
ultrafast, ultrashort pulses/packets, and an interval
modulator for interval modulating the sequences according
to a macrowindow encoded format. The macrowindow encoded
format set by the assigned code includes microwindows and
the pulse/packet interval modulator modulates the position
of an individual pulse/packet within a microwindow of each
macrowindow set by the code such that information can be
encrypted in each of the microwindow. The code means
includes codes stored in matrix form as, e.g., associative
memories and with superframes of received signal
representing the ful~ assigned code, orthogonal to other
assigned codes, matched against the stored codes.
Preferably, the codes are orthogonal codes with the
temporal coding of the sequence of ultrafast, ultrawideband
pulses constituting the carrier for transmission by the
antenna system.
The homodyne receiver includes a bank of
decoder/modems, an acquisition system/matched filter for
synchronizing to a superframe transmission, identifying
r
CA 022629~ 1999-02-03
WO 98/39882 rCT/US97/03182
coded sequencers in the superframe and assigning the
transmissions to a selected decoder/modem on the basis of
code recognition. The system is adapted for multichannel
operation and provides a high overall data rate in the 500
mbs range for maximum multichannel (aggregate channel)
operation.
TH~ PRESENT l~.v~ ON
The present optical mode of the invention is
illustrated in general physical terms in Figs. 14 - 17:
Fig. 14 illustrates a broadcast network mode in which
data sources (A) - either RF, IR, optical or electrical
line - are given an optical orthogonal macrocode, and the
data, encoded in microcodes (wave packets) within the
macrocode packet sequence, is transmitted asynchronously
lS along optical fibers to a hub system (C) which identifies
each orthogonal macrocode/channel and assigns a decoder (B)
to each macrocode or sequence of wave packets. The data is
then recovered at the decoders in RF, IR, optical or
electrical form. The broadcast network mode can also have
an embadiment in which no hub system is required and the
data sources (A) interact with the data recovery points (B)
without intermediary systems. In this embodiment, the data
stream is continuously sampled by the data recovery units
and wave packets are identified for subsequent decoding by
CA 022629~ 1999-02-03
W098/39~2 PCT~S97/03182
their occupancy in a macrocode "wave packet" stream or
train.
Fig. 15 illustrates a point-to-point mode, in which
data sources (A) - either RF, optical or electrical line -
are given optical orthogonal macrocodes which are slaved
together, and the data, encoded in microcodes (wave
packets) within the macrocode, is transmitted synchronously
along optical fibers to data recovery decoders (B) for each
macrocode/channel or sequence or train of packets. The
data is then recovered at the decoders in RF, optical or
electrical form.
Fig. 16 illustrates the data encoding in the microcode
embedded in the user's/channel's macrocode. The macrocode
defines the user (if only one code is allocated to a user)
or a one channel within a multichannel system. The
macrocode is an orthogonal optical code and is identified
by a first matched filter. The microcode can be an error
correcting code and is identified by a second matched
filter behind the first.
Fig. 17 illustrates the data recovery in both the
broadcast and point-to-point modes: A. the macrocode or
wave packet sequence identification either by the hub
system or by the individual data recovery unit. In this
embodiment, the hub system and the individual data recovery
units are represented by acousto-optical modulators tAOM).
CA 022629~ l999-02-03
WO 98/39882 PCT/~JS97/03182
Each black dot in the code matrices (see original patent)
represents a wave packet carrying the microcode. The
individual data recovery units only address the wave
packets arriving at the time appropriate to the
individually assigned orthogonal macrocodes. B. The data
is encoded at the microcode level. A representative 1011
~word" or wavepacket is shown corresponding to a position
in the macrocode indicated by a black dot. The microcode
word can be processed or decoded either in sequential form
or in parallel.
Fig. 18. The optical methods addressed in the present
invention can be interfaced with the methods addressed in
the original patent. In this figure, the data can arrive
at the transmit switch in optical, IR, electrical wire or
RF form and be decoded at the receive switch in optical,
IR, electrical wire and RF form, with the intervening link
in RF form as addressed in my Patent No. 5,610,907.
In general physical terms the present invention is a
system described in Figs. 14 - 17. The various component
parts are described in the System section and the specifics
of the coding schemes are described in the Codes section.
As in my Patent No. 5,610,907, the code is the carrier.
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
THE ~Y~
There are many possible embodiments of an ultrafast
optical time hopping CDMA and TDMA system, all of which use
laser light sources and an optical fiber network (cf.
Gagliardi & Karp, 1995; Spirit & O'Mahony, 1995). The
following is an embodiment of the invention which permits
multichannel or aggregate (high data rate) use.
1. An optical wave packet stream can be encoded in
a macrocode form or optical orthogonal code form by a
variety of methods, e.g., by acousto-optical modulators, or
holograms or clock devices, both real or emulated (c.f.
Weiner et al, 1992; Ford et al, 1994; Hillegas, 1994; Sun
et al, 1995).
2. The data can be encoded into the individual wave
packets in a microcode form by a variety of methods, e.g.,
holographically,k or by clocking or by acousto-optical or
other optical methods such as spatial light modulation.
(Fig. 16 illustrates one embodiment).
3. The macrocodes can be received at a hub system
for identification and assignment to individual data
recovery units, or directly to the individual data recovery
unit, which then sample and identify the unique macrocode
for the specific unit. This function can be performed in
a number of ways, e.g., by acousto-optical modulators,
CA 022629~ 1999-02-03
W098/39882 PCT~S97/03182
holograms or clock devices, both real or emulated. (Fig.
14 illustrates one embodiment.)
4. The macrocodes can be slaved together for
ultrahigh data rate transmission in point-to-point
operation. (Fig. 15 illustrates one embodiment.)
5. The data can be decoded from the individual wave
packet microcode into either a sequential bit stream or in
parallel form by a variety of methods, e.g.,
holographically, or by clocking or by acousto-optical or
other optical methods such as spatial light modulation.
(Fig. 17 illustrates one embodiment.)
6. The system can be used in an all-optical network,
an all-RF network, an IR-RF network, an all IR network,
optical-RF network, a wire-optical network or a wire-RF
network, etc.... (Fig. 18 illustrates one embodiment.)
THE CODES:
The optical orthogonal codes which define the
macrocode are similar to the RF orthogonal codes of the
original invention except that whereas an ~F signal can
have two polarities, +l and -l, as well as the value of
zero, 0, an optical orthogonal code can only take on the
two values of +l and 0, or -l and 0. Therefore, the
statistical representation of optical orthogonal codes are
related to, but differ in some respects, from RF orthogonal
CA 022629~ 1999-02-03
W098139~2 PCT~S97/03182
36
codes (cf. Nguyen et al, 1992; Chung & Kumar, 1990; Chung
et al, 1989).
Application areas include:
Extremely high data rate all-optical fiber
communication links;
Extremely high data rate optical fiber
c~m-~nication links interfacing with RF and electrical
wire communication links.
The present invention includes:
(a) methods of extremely high data rate optical fiber
transmission.
(b) methods of transmitting sequences of such
ultrashort wavepacket sequences or trains in a
macrocode.
(c) methods of encoding data within each wavepacket
of a wavepacket sequence or train as a microcode.
(d) codes which are optical orthogonal codes and the
temporal coding of the sequence or train of
ultrashort pulses or wavepackets constitutes the
- carrier for the transmission channel;
(e) macrocode recognition schemes and microcode data
recover units.
(f) multichannel operation which provides high
overall data rate (e.g., ~500 gigabits/sec -
CA 022629~ 1999-02-03
W098/39882 PCTNS97/03182
~terabits/sec for ~ximllm multichannel or
aggregate operation)~
References
Chung, H. & Kumar, P.V., Optical orthogonal codes - new
bounds and an optimal construction. IEEE ~rans. Information
Theory, 36, 866-873, 1990.
Chung, R.K., Salehi, J.A. & Wei, V.K., Optical orthogonal
codes: design, analysis, and application. IEEE Trans.
Information Theory, 35, 595-604, 1989.
Ford, J.E., Fainman, Y., & Lee, S.H., Reconfigurable array
interconnection by photorefractive correlation, Applied
Optics, 33, 5363-5377, 1994.
Gagliardi, R.M. & Karp, S., Optical Communications, 2nd
Edition, Wiley, New York, 1995.
Hillegas, C.W., Tull, J.X., Goswami, D., Strickland, D. &
Warren, W.S., Femtosecond laser pulse shaping by use of
microsecond radio-frequency pulses. Optics Letters, 19,
737-739, 1994.
Nguyen, Q.A., Gyorfi, L., Massey, J.L., Constructions of
binary constant-weight cyclic codes and cyclically
permutable codes. IEEE Trans. Information Theory, 38, 940-
949, 1992.
Spirit, D.M. h O'Mahony, M.J., High Capacity Optical
Transmission Explained, Wiley, New York, 1995.
Sun, P.C., Mazurenko, Y.T., Chang, W.S.C., Yu, P.K.L. &
Fainman, Y., All-optical parallel-to-serial conversion by
holographic spatial-to-temporal frequency encoding, Optics
Letters, 20, 1728-1730, 1995.
Weiner, A.M., Leaird, D.E., Reitze, D.H. & Paek, E.G.,
Femtosecond spectral holography. IEEE J. Quantum
Electronics, 28, 2251-2261, 1992.
While preferred embodiments of the invention have been
illustrated and described, it will be appreciated that
CA 02262955 1999-02-03
W098/39~2 PCT~S97/03182
other embodiments, adaptations and modifications of the
invention will be readily apparent to those skilled in the
art and embraced by the claims appended hereto.
T
