Frequency synthesizer for improving a unique DDS characteristic

Miscellaneous active electrical nonlinear devices – circuits – and – Signal converting – shaping – or generating – Synthesizer

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C327S129000, C327S355000, C708S270000

Reexamination Certificate

active

06664819

ABSTRACT:

PRIORITY
This application claims priority to an application entitled “Frequency Synthesizer” filed in the Japanese Patent Office on Mar. 2, 2001 and assigned Serial No. 2001-58395, the contents of which are hereby incorporated by reference.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to a frequency synthesizer for generating a sine wave and a cosine wave by digital signal processing, and a frequency synthesizer capable of reducing spurious signals.
2. Description of the Related Art
When a direct digital synthesizer (DDS) is used in generating a local signal of a receiver, spurious signals generated by operation of the DDS deteriorate an adjacent channel interference characteristic and an outband interference characteristic. Likewise, when the DDS is used to generate a local signal of a transmitter, the spurious signals interfere with adjacent frequencies.
FIG. 6
illustrates how spurious signals are generated in a conventional DDS. The spurious signals caused by insufficient operation accuracy of the DDS are generated because of: (i) a phase requantization error e
p
due to a difference between an operation word length j of a phase operator comprised of an adder
71
and a phase register
72
, and an address length k of a ROM (Read Only Memory)
73
for converting phase data to amplitude data; and (ii) an amplitude quantization error e
a
of output bits of the ROM
73
. As illustrated in
FIG. 7
, the spurious signals are uniformly distributed around a frequency fc of a desired signal. If the operation word length j is set to be equal to the address length k (j=k) to improve the operation accuracy of the DDS, no spurious signal will be generated. Further, if an output data with a width m of the ROM
73
, as illustrates in
FIG. 6
, is set to a sufficiently large value, the spurious signals caused by the amplitude error will have a negligible level.
However, when the operation word length j of the phase operator is set to be equal to the address length k of the ROM
73
, the ROM size is doubled each time the address length k is increased by one bit. Therefore, it is difficult to realize the ROM
73
when the operation word length is relatively long. Accordingly, a method using the addition theorem of a trigonometric function has been proposed as a method for realizing an equivalent large ROM size with an actual small ROM size. For example, if j0-bit frequency setting data is F=A+B, where A represents data of j1 bits on the MSB (Most Significant Bit) side and B represents data of remaining j2 bits on the LSB (Least Significant Bit) side, then cos(F) and sin(F) are defined as
cos(
F
)=cos(
A+B
)=cos
A
·cos
B
−sin
A
·sin
B
sin(
F
)=sin(
A+B
)=sin
A
·cos
B
+cos
A
·sin
B
That is, the frequency setting data F can be generated by synthesizing the frequency setting data A and the frequency setting data B in accordance with the above formulas. For example, in the case where the frequency setting data F is comprised of 16 bits, even though the output data width m of the ROM
73
is 1 bit (m=1), the ROM
73
requires a capacity of 64 K words×2, as shown in the following formulas.
(1) For cos(F), 2
16
words=65,536 words
(2) For sin(F), 2
16
words=65,536 words
However, if the 16-bit frequency setting data F is divided into 8-bit frequency setting data A and 8-bit frequency setting data B, the required ROM capacity becomes 256 words, a square root of 65,536 words, as shown in the following formulas.
(2) For cos(A), 256 words
(2) For sin(A), 256 words
(3) For cos(B), 256 words
(4) For sin(B), 256 words
That is, since the required total capacity becomes 256 words×4, the required ROM size becomes {fraction (1/128)} times the ROM size of the conventional DDS.
FIG. 8
illustrates a structure of a DDS using the addition theorem of a trigonometric function according to the prior art. As illustrated, when j0-bit frequency setting data F represented by a phase variation width &Dgr;&phgr; is received, a phase operator comprised of an adder
51
and a phase register
52
accumulates the frequency setting data F into phase data Ff. The j0-bit phase data Ff is separated into j1-bit phase data Af and j2-bit phase data Bf starting from the MSB side, and k1 bits in the separated j1-bit phase data Af on the MSB side are applied as address signals to a coarse cos(A) ROM-A
53
and a coarse sin(A) ROM-B
54
, in each of which a table for converting phase data to amplitude data is stored. The ROM-A
53
and the ROM-B
54
sequentially output m-bit amplitude data, respectively. Here, the ROM-A
53
and the ROM-B
54
register quantized cosine and sine waves of a frequency corresponding to the j1 bits on the MSB side of the phase data Ef, respectively.
Meanwhile, k2 bits in the remaining j2-bit phase data Bf are applied as address signals to a fine cos(B) ROM-C
55
and a fine sin(B) ROM-D
56
, in each of which a table for converting phase data to amplitude data is stored. The ROM-C
55
and the ROM-D
56
sequentially output m-bit amplitude data, respectively. Here, the ROM-C
55
and the ROM-D
56
register quantized cosine and sine waves of a frequency corresponding to the remaining j2 bits of the phase data Ef, respectively. A complex mixer
57
synthesizes the m-bit amplitude data outputs from the ROM-A
53
, ROM-B
54
, ROM-C
55
and ROM-D
56
, and generates output signals cos(n) and sin(n) of the frequency synthesizer. In order to calculate a real-part output signal, the complex mixer
57
includes a multiplier
58
for multiplying a real-part input signal T
1
by a real-part input signal T
3
, a multiplier
59
for multiplying an imaginary-part input signal T
2
by an imaginary-part input signal T
4
, and a subtracter
60
for synthesizing an output of the multiplier
58
and an output of the multiplier
59
. Further, in order to calculate an imaginary-part output signal, the complex mixer
57
includes a multiplier
61
for multiplying the real-part input signal T
1
by the imaginary-part input signal T
4
, a multiplier
62
for multiplying the real-part input signal T
3
by the imaginary-part input signal T
2
, and an adder
63
for synthesizing an output of the multiplier
61
and an output of the multiplier
62
. The output of the ROM-A
53
is connected to a terminal T
1
of the complex mixer
57
, the output of the ROM-B
54
to a terminal T
2
of the complex mixer
57
, the output of the ROM-C
55
to a terminal T
3
of the complex mixer
57
, and the output of the ROM-D
56
to a terminal T
4
of the complex mixer
57
. As a result, the frequency synthesizer of
FIG. 8
outputs carrier signals cos(n) and sin(n) with a frequency corresponding to the frequency setting data F=A+B.
However, when the frequency setting data F needs 32 bits, 2
32
=4,294,967,296 words and a square root of 4,294,967,296 words is 65,536 words. Even though the frequency setting data F is divided into data A and data B, a 64K words×4=256K word-ROM is required. Thus, the DDS cannot implement the high-speed operation. When a desired operation word length of the frequency setting data F is increased, it is difficult to set an address length k1 of a ROM for converting frequency setting data A to amplitude data and an address length k2 of a ROM for converting frequency setting data B to amplitude data such that j1=k1 and j2=k2. As a result, j1>k1 and j2>k2. In this case, since the ROM size is smaller than when the frequency is not divided, it is possible to reduce error generated. However, since phase errors are generated in the frequency setting data A and the frequency setting data B, generation of the spurious signals is unavoidable.
In particular, the spurious signals of the DDS are uniformly distributed as illustrated in FIG.
7
. Thus, when used as a local signal generator of a radio communication apparatus, the DDS undergoes interference over a wide range in a receiver and causes interference over a wide range

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Frequency synthesizer for improving a unique DDS characteristic does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Frequency synthesizer for improving a unique DDS characteristic, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Frequency synthesizer for improving a unique DDS characteristic will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-3102636

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.