 |
Claims  |
|
|
I claim:
1. A method of transmitting information by generating a spread-spectrum
modulated signal and transmitting the signal over a communications link,
the method comprising:
determining selected chipping sequences from a plurality of possible
chipping sequences for message bundles in response to data to be
transmitted;
determining starting positions of said selected chipping sequences in
response to data to be transmitted;
modulating a carrier signal with the selected chipping sequences and
beginning the selected chipping sequences at the starting positions to
perform simultaneously code shift key modulation and pulse position
modulation for the message bundles; and
repeating at least a portion of said selected chipping sequences beginning
at the starting positions in every message bundle for at least as long as
multipath delays over the communications link.
2. A method as claimed in claim 1, further comprising transmitting another
carrier in phase quadrature with the first carrier, the second carrier
being modulated by a second plurality of possible chipping sequences, said
second plurality of possible chipping sequences being orthogonal to the
first plurality of possible chipping sequences.
3. A method as claimed in claim 1, further comprising: determining
amplitudes of the carrier signal in response to data to be transmitted;
and
modulating an amplitude of the carrier signal in response to the determined
amplitudes for the message bundles.
4. A method as claimed in claim 1, further comprising repeating the
selected chipping multiple times to allow detection of small frequency
offsets by means of spectral analysis of repeated correlation peaks at a
receiver.
5. A method as claimed in claim 1, further comprising inducing frequency
shifts in said carrier for message bundles in response to said data to be
transmitted, thereby further performing frequency shift key modulation.
6. A method for transmitting information across a communications link by
generating a spread-spectrum modulated signal temporally divided into
message bundles, said method comprising:
generating a first modulated subcarrier by linearly modulating a first
subcarrier with a first selected chip sequence being a pseudo-random
sequence of plural bits in which starting positions of the transmission of
the first selected chip sequence for each message bundle between a first
through last bit are determined in response to data to be transmitted,
wherein said first selected chip sequence is one of a first plurality of
mutually orthogonal chip sequences chosen in response to data to be
transmitted;
generating a second modulated subcarrier by linearly modulating a second
subcarrier with a second selected chip sequence being a pseudo-random
sequence of plural bits in which starting positions of the transmission of
the second selected chip sequence between a first through last bit are
determined in response to data to be transmitted, wherein said second
selected chip sequence is one of a second plurality of mutually orthogonal
chip sequences chosen in response to data to be transmitted;
generating a combined signal by combining said first modulated subcarrier
and said second modulated subcarrier in phase quadrature; and
inducing frequency shifts in said combined signal corresponding to one of a
plurality of frequencies in response to data to be transmitted.
7. A method as claimed in claim 6, further comprising repeating at least a
portion of said first and said second selected chip sequences for each
message bundle so that all multipath components will be simultaneously
present at a receiver for a time corresponding to at least one full
repetition of the first and second selected chip sequences.
8. A method as claimed in claim 6, wherein a relative phase shift of said
first modulated subcarrier and said second modulated subcarrier is
determined in response to data to be transmitted.
9. A method as claimed in claim 6, wherein the frequency shifts are on the
order of a few to tens of Hertz.
10. A method as claimed in claim 7, further comprising repeating the entire
first and second selected chip sequences multiple times to allow detection
of small frequency offsets by means of spectral analysis of repeated
correlation peaks at a receiver.
11. A method as claimed in claim 7, further comprising determining the
frequency shifts means by spectral analysis of repeated correlation peaks.
12. A method for demodulating a spread-spectrum modulated signal to receive
transmitted information, said method comprising:
cross-correlating a received chip sequence of the spread-spectrum modulated
signal with a plurality of possible chip sequences each beginning at a
predetermined bit;
determining to which one of said plurality of possible chip sequences said
received chip sequence corresponds;
determining a time delay until a correlation peak is generated;
determining a frequency shift in a frequency of the spread-spectrum
modulated signal; and
decoding said time delay, said received chip sequence, and the frequency
shift as transmitted information.
13. A method as claimed in claim 12, wherein the step of determining the
frequency shift comprises
decoding a frequency shift in message bundles of said transmission by
performing a full spectral integration and determining into which of a
plurality of Doppler bins a peak of in the received Doppler spectra falls
by cross correlation with a Doppler spectrum of previous message bundles.
14. A method as claimed in claim 12, further comprising detecting plural
repetitions of said received chip sequence for message bundles to ensure
that all multipath components of said spread-spectrum modulated signal are
available for coherent integration of the entire received chip sequence.
15. A method as claimed in claim 13, wherein the frequency shift is on the
order of a few to tens of Hertz.
16. An apparatus for transmitting information over a communications link,
the apparatus comprising:
a synthesizer for generating a carrier signal;
a modulator for spectrally spreading said carrier signal with a selected
chipping sequence, information being encoded into said carrier signal by
determining said selected chipping sequence from a plurality of possible
chipping sequences and determining a start bit among the bits comprising
said selected chipping sequence for each message bundle, the modulator
further repeating at least a portion of said selected chipping sequence
beginning at the start bit in every message bundle for at least as long as
multipath delays over the communications link; and
a transmitter for transmitting the modulated carrier signal over the
communications link.
17. An apparatus as claimed in claim 16, further comprising an amplifier
for modulating an amplitude of the modulated carrier signal transmitted
over the communications link for each message bundle.
18. An apparatus for transmitting information over a communications link,
the apparatus comprising:
a synthesizer for generating a carrier signal
a modulator for phase shift key modulating said carrier signal by a
selected chipping sequence, information being encoded into said carrier
signal by determining the selected chipping sequence from a plurality of
possible chipping sequences and by determining a start bit among the bits
comprising said selected chipping sequence for each message bundle; and
a modulator for modulating a frequency of said carrier signal encode
information.
19. An apparatus as claimed in claim 18, further comprised of said
modulator selecting said chipping sequence from a plurality of mutually
orthogonal sequences in response data to be modulated.
20. An apparatus as claimed in claim 18, wherein the frequency modulator
modulates the frequency of the carrier signal on the order of a few to
tens of Hertz.
21. An apparatus for transmitting information over a communications link,
the apparatus comprising:
a synthesizer for generating a first subcarrier and a second subcarrier in
phase quadrature;
a first modulator for phase shift key modulating said first subcarrier by
first chipping sequences selected from a first set of orthogonal
sequences, the first modulator including a first selector for selecting a
start bit among the bits comprising said first chipping sequences for each
message bundle;
a second modulator for phase shift key modulating said second subcarrier by
second chipping sequences selected from a second set of orthogonal
sequences, each one of said chipping sequences of said first set being
orthogonal to every chipping sequence of said second set, the second
modulator including a second selector for selecting a start bit from among
the bits comprising said second chipping sequences for each message
bundle;
an adder for combining the modulated first subcarrier and the modulated
second subcarrier to generate a composite signal; and
a frequency modulator for modulating a frequency of the composite signal.
22. An apparatus as claimed in claim 21, further comprising:
a first amplitude modulator for amplitude modulating said first subcarrier;
and
a second amplitude modulator for amplitude modulating said second
subcarrier.
23. An apparatus as claimed in claim 21, further comprising means for
modulating a phase relationship between said first subcarrier and said
second subcarrier.
24. An apparatus as claimed in claim 22, further comprising means for
modulating a phase relationship between said first subcarrier and said
second subcarrier.
25. A demodulator for receiving information over a communications link
contained in a spread-spectrum signal, the demodulator comprising:
a filter for cross-correlating a received chipping sequence with a
plurality of possible chipping sequences each beginning at a predetermined
bit;
a comparator for determining to which one of said plurality of possible
chipping sequences said received chipping sequence corresponds;
a delay detector for determining a time delay until a correlation peak is
generated by said filter relative to a received chipping sequence
beginning at the predetermined bit; and
a frequency detector for determining a frequency of the spread-spectrum
signal;
a decoder for decoding said time delay, said received chipping sequence,
and the frequency as transmitted data.
26. A demodulator as claimed in claim 25, wherein the frequency detector
performs a full spectral integration of repeated correlation peaks and
determines into which of a plurality of Doppler bins a peak of a received
Doppler spectra falls by cross correlation with a Doppler spectrum of
previous message bundles.
27. A demodulator as claimed in claim 25, further comprising an amplitude
detector for determining an amplitude of the spread-spectrum signal; the
decoder further decoding the detected amplitude as transmitted data.
28. A demodulator as claimed in claim 25, wherein the frequency detector
detects changes the frequency on the order of a few to tens of Hertz.
29. A method transmitting information by generating a spread-spectrum
modulated signal and transmitting the signal over a communications link,
the method comprising:
determining selected chipping sequences from a plurality of possible
chipping sequences in response to data to be transmitted;
determining starting positions of the selected chipping sequences in
response to data to be transmitted;
determining frequency shifts in response to data to be transmitted;
modulating a carrier signal with the selected chipping sequences, beginning
the selected chipping sequences at the starting positions, and modulating
a frequency of the carrier signal in response to the determined frequency
shifts to perform simultaneously code shift key modulation, pulse position
modulation, and frequency shift key modulation.
30. A method as claimed in claim 29, further comprising repeating at least
a portion of said selected chipping sequences beginning at the starting
positions in every message bundle for at least as long as multipath delays
over the communications link.
31. A method as claimed in claim 29, further comprising: determining an
amplitude of the carrier signal in response to data to the transmitted;
and
amplitude modulating the carrier signal in response to the determined
amplitude to further perform amplitude shift key modulation.
32. A method as claimed in claim 29, wherein the carrier signal is phase
shift key modulated in response to the selected chipping sequences. |
|
 |
|
 |
Claims |
|
|
|
Description |
|
|
BACKGROUND OF THE INVENTION
Spread-Spectrum refers to a broad class of modulation techniques in which a
bandwidth of a carrier is enlarged beyond that necessary to convey the
information to be transmitted. One technique for "spreading" the carrier
signal is to replace each bit of the original message, with a determinate
pseudo-random sequence of one's and zero's commonly known as a chipping
sequence and then bi-phase or phase-reversal modulate the carrier signal
in response to the sequence. As a result, a pattern of ten or possibly
hundreds of bits of the chipping sequence may represent one or only a few
bits of the original message. The sets of chipping sequences are generally
mathematically orthogonal code patterns known by both the transmitter and
receiver or generated by identical key generators seeded with the same
predetermined value. The use of matched filters enables the receiver to
enhance the signal upon reception thus allowing recovery of the
signal-to-noise ratio (SNR) lost because of the increased bandwidth of the
spread spectrum transmission.
The redundancy inherently present in spread spectrum modulation lowers
channel information capacity, which tends to restrict useful applications
due to its consumption of such large portions of the frequency spectrum
while not conveying any more information. The information capacity
limitation is further compounded in severely time dispersive channels.
SUMMARY OF THE INVENTION
The present invention concerns a composite spread-spectrum modulated signal
having a multi-dimensional signal space, comprising:
a pulse position modulation dimension generated by varying starting
positions of cyclically repeated chipping sequences;
a frequency shift key dimension generated by varying a carrier frequency
between a plurality of frequency offsets; and
a code shift keyed dimension generated by selecting between a plurality of
different chipping sequences.
This signal may further comprise a phase shift key dimension generated by
assigning a phase relationship of the subcarriers.
The present invention also concerns a modulator for generating a composite
signal, the modulator comprising:
a sinusoidal signal generator for generating a carrier signal being
temporally divided into message bundles;
a modulating section for linearly modulating the carrier signal with a
chipping sequence, information being encoded into the carrier signal by
selecting a start bit among the bits comprising the chipping sequence for
each message bundle; and
a section for modulating a frequency of the carrier signal.
Also the present invention concerns a demodulator for demodulating the
composite signal, the demodulator comprising:
a filter for cross correlating a received chip sequence with a plurality of
possible chip sequences each beginning at a predetermined bit;
a comparator for determining to which one of said plurality of possible
chip sequences said received chip sequence corresponds;
a means for determining a time lag until a correlation peak is generated by
said cross correlating means; and
a decoder for decoding said time lag and said received chip sequence as
transmitted data.
The above and other features of the invention including various novel
details of construction and combinations of parts will now be more
particularly described with reference to the accompanying drawing and
pointed out in the claims. It will be understood that the particular
composite signal, modulator, and demodulator embodying the invention is
shown by way of illustration only and not as a limitation of the
invention. The principles and features of this invention may be employed
in various and numerous embodiments without departing from the scope of
the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 illustrates a time domain representation of chipping sequences used
to modulate the subcarriers of the composite signal;
FIG. 2 illustrates the auto correlation function of the 63 bit Kasami
sequence;
FIG. 3 illustrates the signal space dimensions of subcarriers of the
composite signal;
FIG. 4 is a schematic diagram of a modulator for manufacturing the
composite signal;
FIG. 5 is a schematic diagram for a demodulator for decoding the composite
signal;
FIG. 6 illustrates the maximal length sequence auto-correlation function;
FIG. 7 illustrates the method for decoding the composite signal;
FIG. 8 illustrates a schematic diagram of an alternative embodiment for the
modulator; and
FIG. 9 illustrates a non-optimal correlation resulting from a receiver's
failure to synchronize on the beginning of a sequence repetition.
DETAILED DESCRIPTION OF THE INVENTION
Turning now to the figures, a composite signal constructed according to the
principles of the present invention is described.
The inventive composite signal utilizes a combination of four modulation
schemes and multiple subcarriers yielding a multi-dimensional signal
space. Briefly, the composite signal includes two independent subcarriers
in phase-quadrature. Each of these subcarriers are spread spectrum
modulated in two dimensions by selecting first the chipping sequence (code
shift key modulation) for each message bundle and then by selecting a
start bit in a cyclic and repeated transmission of the selected chip
sequence for each message bundle. Since shifting the starting point in the
cyclic code results in a corresponding shift in the position of the
received correlation peak this constitutes pulse position modulation. An
additional dimension is provided by making a slight shift in the frequency
of the sub-carriers (frequency shift key modulation), another dimension is
provided by selecting from a plurality of possible amplitudes, and a final
dimension is generated by modulating the phase relationship between the
quadrature subcarriers. Each of these dimensions will be discussed in
detail below.
Code Shift Key Modulation
Each subcarrier is assigned a unique set of orthogonal chipping sequences.
Specifically, the first subcarrier is assigned the first eight of the
sixteen orthogonal 255 bit Kasami sequences, identified for the purposes
of this description as the set-1 sequences. The second subcarrier is
assigned the remaining eight Kasami sequences, the set-2 sequences. Then,
for every message bundle, each subcarrier is bi-phase modulated by one of
the Kasami sequences in its assigned set which are selected in response to
the information to be transmitted. Since any one of eight different Kasami
sequences may be selected for each subcarrier for each message bundle,
three bits may be encoded into each subcarrier making six CSK bits in each
message bundle.
Although the present embodiment utilizes bi-phase modulation of the
subcarriers by the Kasami sequences, any form of modulation that provides
linear detection at the receiver could be used. For example, such linear
modulation techniques as: minimum shift key modulation, amplitude
modulation, frequency shift key modulation, and continuous phase
modulation represent a few alternative techniques that would satisfy the
criterion.
Each message bundle lasts long enough to accommodate N (e.g. eight)
repetitions of the same selected Kasami sequence. Normally, the number of
code repetitions will be a power of 2 to facilitate the fast Fourier
transform spectral analysis enabling frequency shift key demodulation.
FIG. 1 illustrates a time domain representation of exemplary chipping
sequences used to modulate each subcarrier for two full message bundles
plus the beginning of a third. The first subcarrier is consecutively
modulated by the third, fifth, and first Kasami sequences selected among
the first through eighth set-1 Kasami sequences assigned to that
subcarrier; each block labeled "3rd," for example, represents one
repetition of the third 255-bit Kasami sequence yielding a total of N or
eight, full repetitions per message bundle. The second subcarrier is
modulated by sequences selected from the ninth through sixteenth set-2
Kasami codes, i.e., the ninth, twelfth, and fifteenth.
Although other pseudo-random sequences could be used, the Kasami sequences
minimize Welch bounded leakage from one of the sequences to another.
Further, the Kasami sequences maintain their ideal autocorrelation
function regardless of a start bit when the sequences are cyclically
repeated.
Pulse Position Modulation
Each subcarrier is also pulse-position modulated by selecting the start bit
in the modulating Kasami sequence for each message bundle. Seven bits can
be encoded by selectively beginning the modulation of the subcarrier at
one of 128 bits in the 255 bits of the selected Kasami sequence. 255 start
positions could be defined but it is more reliable to have a one-shift
guard band since in general, signal delays will fall between the exact
centers of the time delay resolution cells. Since the demodulator can
identify the selected Kasami code independently of the bit on which the
transmission was started, the pulse position modulation does not affect
the code shift keyed modulation. A total of 14 bits are encoded using the
pulse position modulation, seven for each subcarrier.
FIG. 1 also illustrates the pulse position modulation by which, during the
first message bundle, the first subcarrier is repetitively modulated by
the third Kasami sequence beginning at the 70th bit. The first repetition
of the code progresses through code bit then rolls over to 1 and continues
to code bit 69 constituting a single repetition. Immediately after the
69th bit of the first repetition, the code bit 70 follows thereby starting
the second repetition. The message bundle ends with the 69th bit of the
third Kasami sequence, that is to say, exactly 8 repetitions are generated
to ensure that the message bundle has a fixed temporal length so that the
receiver can maintain synchronization. Then, to encode the second bundle,
the first subcarrier is modulated by the fifth Kasami sequence beginning
at the 98th bit. The second subcarrier is similarly pulse position
modulated by starting with the 34th bit of 9th Kasami sequence during the
first message bundle and the 190th bit during the second bundle.
The attractive property of the Kasami sequence is their auto correlation
functions illustrated in FIG. 2 for a 63 bit Kasami sequence. This ideal
(Welch Bounded, see Proakis, J., Digital Communications, Ch. 8, McGraw
Hill, 1983) correlation is obtainable at the receiver only if it can be
synchronized to begin sampling the instant the message bundle starts to
arrive. However, if time of arrival uncertainties result in sampling
before the arrival or after the termination of the code sequence, a
correlation more like the function illustrated in FIG. 9 will result. Time
of arrival uncertainties can be compensated for by sampling the end of a
preceding repetition of the same code or into the next repetition of a
subsequent identical code. This will maintain the ideal characteristics of
FIG. 2.
If time dispersion or multipaths, are present in the propagation medium, a
receiver can not synchronize with the beginning of the message bundle for
all multipaths, simultaneously. Therefore, the transmitter must cycle
through the selected Kasami sequences a second time for at least as long
as the maximum expected multipath delay. For instance, if one sequence
repetition takes 8.5 msec and the maximum anticipated time dispersion over
the channel is 5 msec, then one sequence plus 5 msec into the second
repetition must be transmitted to ensure that all multipath components are
present at the receiver for demodulation. Since the composite signal
cycles through the modulating Kasami sequence N times, for instance, eight
times, a single repetition being longer than the largest expected
multipath delay, the presence of all multipath components at the receiver
is ensured by the time the second repetition starts to arrive.
The composite signal is also adapted to perform in environments containing
multipath by the provision of a dead time during multipath overlap between
each message bundle. A dead time is selected to exceed the longest
multipath delay in the environment, for instance greater than 5 msec,
allowing the message bundle via all of multipaths to terminate before the
receiver begins to receive the next message bundle. Therefore, the
receiver can begin decoding the next message bundle without interference
from the last bundle received via the longest multipath. Since the
sequence length and the dead time both must be longer than the largest
multipath delay, they could both be made the same period (e.g. 8.5 ms).
Therefore, the dead time can be neatly incorporated into the first
repetition of each subsequent message bundle by the following means. Since
most time domain window functions (used to condition data for a Fourier
transform) such as a Hanning window are zero at one point we can make this
zero weight coincide with the first code repetition, thus eliminating all
samples from the first repetition and therefore multipath contamination.
This results in the continuous transmission illustrated in FIG. 1.
Frequency Shift Keyed Modulation
The first and second subcarriers are also collectively frequency shift
keyed modulated to one of four frequencies for each message bundle.
Therefore, an additional two bits can be encoded by selection of the
frequency shift in the composite signal.
The frequency shift key modulation of the two subcarriers may appear to
complicate the demodulation. On many channels of interest, however, a
Doppler shift of an unknown magnitude will continuously change the phase
of the carrier during transmission. The multiple samples of each
repetition of the Kasami sequence can be integrated only if the carrier is
coherent from one repetition to the next. Therefore, in a channel in which
a Doppler shift may be present, the signal integration must be performed
by a full spectral integration such as a Fourier transform. This is
generally called Doppler processing or Doppler integration.
Since the signal should be analyzed by the full spectral integration simply
to maintain coherence, information can be transmitted at "no extra cost"
by artificially inducing a "Doppler shift" by offsetting the transmitter
by a frequency corresponding to 1,2 . . . . N Doppler bins of the
receiver. This typically corresponds to a few Hz or 10's of Hz. Here, four
or N/2 frequency offsets rather than eight were selected to limit the
frequency offset and thereby avoid decorrelating the waveform. Therefore,
two bits of information can be transmitted by moving the position of the
peak in the received Doppler spectrum to any of the N/2 Doppler bins.
Since an actual Doppler shift may already be present on the channel, the
information encoded by the frequency shift key modulation is best encoded
differentially, using the previous message bundle as a reference for the
next. The Doppler spectrum is inherently circular, in that any signal that
exceeds the positive limit of the spectrum appears at the negative
extreme, i.e. wraps around. Therefore, regardless of the current position
of the Doppler peak, any shift of 0 through N can be made on the next
message symbol without running off the edge of the Doppler spectrum.
As was explained above for pulse position modulation, the Doppler spectrum
may not contain just a single peak, each multipath component may exhibit a
different Doppler shift, therefore, the differential shift carrying the
message information may only be detectable by cross-correlating it with
the Doppler spectrum from the previous message symbol. Note however, that
most multipath components are isolated by the range, that is, time delay,
resolution of the pulse compression technique, which occurs prior to the
computation of the Doppler spectrum. Therefore, there will be only one or
a few Doppler peaks per time delay bin in all but the most disturbed
propagation environments.
Amplitude Shift Key Modulation (ASK)
Selection of the transmitted signal amplitude from a plurality of possible
amplitudes can provide more information capacity. This can be provided by
placing a digitally controlled attenuator or variable gain amplifier in
each subcarrier channel prior to the summation of the two subcarriers.
Like the other modulation dimensions, the level is most reliably detected
differentially by comparing it to the level of the signal during the
preceding message bundle rather than against some absolute standard. Four
levels per subcarrier would allow two bits of information to be encoded on
each for a total of four additional bits per message bundle.
Implementation of the amplitude shift key dimensions should be avoided for
communications links exhibiting complicated fading or high noise.
Phase Shift Key Modulation
During quadrature modulation, the phase of one of the subcarriers is
advanced by ninety degrees relative to the other one. At demodulation, the
two subcarriers are distinguishable by their assigned Kasami codes.
Therefore, one final bit can be encoded by controlling whether the phase
of the first subcarrier leads or lags the phase of the second subcarrier
by ninety degrees for each message bundle.
FIG. 3 illustrates the multidimensional signal space for each subcarrier of
the composite signal. In our example, each subcarrier defines one of 4096
possible message bundles, eight possible Kasami sequences represented by
the eight divisions along the z-axis, 128 pulse positions indicated by the
128 divisions along the y-axis, and four frequency shifts into the
available Doppler bins designated by the four divisions along the x-axis,
totaling 12 bits in a three dimensional signal space. The other subcarrier
only adds 10 bits since the frequency shifts cannot be independently
selected for each subcarrier, or the signal envelope would then "beat"
causing an amplitude modulation. Finally, the modulation of the phase
relationship of the two subcarriers in the quadrature modulation adds
another bit, yielding a message bundle of 23 bits. The 4 bits provided by
the amplitude modulation are not shown because of the difficulty in
illustrating the additional dimensions.
In summary, by utilizing two orthogonal sets of 8 (16 total) 255 length
chipping sequences, repeated 8 times, there are 8 bit fields in the 27-bit
input data word,
1. FSK, 2-bits for the frequency offset selection;
2. CSK, 3-bits for selection of an orthogonal spreading code from set-1;
3. PPM, 7-bits for the rotation of the set-1 code's starting position;
4. ASK, 2-bits for selection of signal amplitude for the first subcarrier;
5. CSK, 3-bits for selection of a set-2 code;
6. PPM, 7-bits for the rotation of the set-2 code;
7. ASK, 2-bits for selection of the signal amplitude for the second
subcarrier; and
8. Differential PSK, 1-bit for the first subcarrier phase relative to the
second subcarrier (+90 or -90.degree.).
Modulator
FIG. 4 illustrates an inventive modulator for manufacturing the
above-described composite signal. More specifically, a direct digital
synthesizer 410 receives two bits of data on line 411 every message bundle
period and generates a sinusoidal signal on output line 413 having one of
four frequency offsets determined by the two bits; the remaining fourteen
frequency selection terminals labelled "carrier frequency select (fixed)"
are fixed to select the overall frequency of the carrier. The sinusoidal
signal generated by the synthesizer 410 is the second subcarrier. The
first subcarrier is generated by advancing the phase of the second
subcarrier by ninety degrees in the 90 degree phase shifter 490.
A first presettable counter 420 receives seven bits defining the bit on
which the chipping sequence for the first subcarrier will start, thus
comprising the pulse position modulation. These seven bits are loaded into
the counter's register in response to a message bundle start trigger
signal received at a preset/load terminal of the counter 420. Upon receipt
of this signal, the presettable counter 420 initiates counting up for
every cycle of the bit rate clock.
The count generated by the presettable counter 420 is used to address a
code storage Read-Only-Memory 430 storing the eight set-1 Kasami
sequences. The first code storage ROM 430 stores, at each address, the
corresponding bits of each set-1 Kasami sequence. For example, address 28
holds the twenty-eighth bits of each of the first through eighth Kasami
sequences in set-1. The eight data bits generated for each address
designated by the first presettable counter 420 are provided to a first
8-to-1 data selector 440 which selects the bit corresponding to the Kasami
sequence designated by the three code select bits it receives, code shift
key modulation. The selected bits are provided to a first mixer 450 to
phase-shift key modulate the first subcarrier after being level shifted by
level shifter 447 to make a bipolar code signal which results in an equal
signal envelope for 0 and 1 bits.
The second presettable counter 425, second code storage ROM 435, and the
second 8-to-1 data selector 445 operate substantially the same as the
first presettable counter 420, first code storage ROM 430, and the first
8-to-1 data select | | |
