CTD programmable convolver

A programmable convolver comprising:

• a. first generator means for providing a pair of electronic output signals having magnitudes equal to the real and imaginary components of EXP[-j.pi.n.sup.2 /N],
• b. first multiplier means having inputs for receiving in parallel (1) successive samples of selected segments of electronic input signals and (2) said output signals of said first generator, for providing a first pair of product signals having magnitudes equal to the real and imaginary components of the complex product of the signals on said inputs of said first multiplier,
• c. first filter means, having a pair of inputs coupled to said first multiplier for receiving said first pair of product signals, having a complex impulse response of EXP[j.pi.k.sup.2 /N], and utilizing charge transfer devices to produce complex convolution signals having magnitudes equal to the real and imaginary components of the convolution of said first product signals and said impulse response,
• d. second multiplier means having inputs for receiving in parallel (1) said first pair of complex convolution signals, and (2) the real and imaginary components of an arbitrary complex signal H(kw.sub.p), for producing a second pair of product signals having magnitudes equal to the real and imaginary components of the complex product of the signals on said inputs of said second multiplier,
• e. second filter means, having inputs coupled to said second multiplier for receiving said second product signals, having a complex impulse response of EXP[-j.pi.k.sup.2 /N], and utilizing charge transfer devices to produce a second pair of complex convolution signals having magnitudes equal to the real and imaginary components of the convolution of said second product signals and said impulse response,
• f. second generator means for providing a pair of electronic output signals having magnitudes equal to the real and imaginary components of EXP[+j.pi.n.sup.2 /N], and
• g. third multiplier means, having inputs for receiving in parallel (1) said second pair of complex convolution signals, and (2) said output signals of said second generator, for producing output signals having magnitudes equal to the product of the signals on said inputs of said third multiplier.

A device for computing the discrete Fourier transform of complex input signals, said device including:

• a. first generator means comprised of charge transfer devices for providing a pair of output signals having magnitudes equal to the real and imaginary components of EXP[-j.pi.n.sup.2 /N],
• b. first multiplier means having two inputs for receiving successive samples of the real and imaginary components of said complex input signals, two other inputs connected to receive said output signals of said first generator;
• for providing a first pair of product signals having magnitudes equal to the real and imaginary components of the product of the signals on said inputs of said first multiplier,
• c. first filter means comprised of charge transfer devices, having a complex impulse response of EXP[j.pi.k.sup.2 /N], having inputs connected to receive said first pair of product signals for producing a first pair of complex convolution signals having magnitudes equal to the real and imaginary components of the convolution of said first product signals and said impulse response,
• d. second generator means comprised of charge transfer devices for providing a pair of electronic output signals having magnitudes equal to the real and imaginary components of EXP[-j.pi.n.sup.2 /N], and
• e. second multiplier means, having inputs connected to said first filter means to receive in parallel (1) said first pair of complex convolution signals and (2) said output signals of said second generator means, for producing a pair of final output signals having magnitudes equal to the real and imaginary components of the product of the signals on said inputs of said second multiplier.

A discrete Fourier transform unit comprising:

• a. first generator means comprised of charge transfer devices for providing a pair of output signals having magnitudes equal to the real and imaginary components of EXP[-j.pi.n.sup.2 /N],
• b. first multiplier means having inputs for receiving in parallel (1) successive samples of selected segments of complex electronic input signals, and (2) said output signals of said first generator;
• for providing a first pair of product signals having magnitudes equal to the real and imaginary components of the product of the signals on said inputs of said first multiplier,
• c. first filter means, having inputs connected to receive said first pair of product signals, having a complex impulse response of EXP[j.pi.k.sup.2 /N] obtained by utilizing charge transfer devices arranged as a dual impulse transversal filter;
• for producing a first pair of complex convolution signals having magnitudes equal to the real and imaginary components of the convolution of said first product signals and said impulse response,
• d. second generator means comprised of charge transfer devices for providing a pair of electronic output signals having magnitudes equal to the real and imaginary components of EXP[-j.pi.n.sup.2 /N], and
• e. second multiplier means, having inputs connected to said first filter means to receive in parallel (1) said first pair of complex convolution signals and (2) said output signals of said second generator means, for producing a pair of final output signals having magnitudes equal to the real and imaginary components of the product of the signals on said inputs of said second multiplier.

A device for computing the square of the magnitude of the discrete Fourier transform of complex input signals, said device including:

• a. first generator means comprised of charge transfer devices for providing a pair of output signals having magnitudes equal to the real and imaginary components of EXP[-j.pi.n.sup.2 /N],
• b. first multiplier means having two inputs for receiving successive samples of the real and imaginary components of said complex input signals, two other inputs connected to receive said output signals of said first generator;
• for providing a first pair of product signals having magnitudes equal to the real and imaginary components of the product of the signals on said inputs of said first multiplier,
• first filter means comprised of charge transfer devices having a complex impulse response of EXP[j.pi.k.sup.2 /N], having inputs connected to receive said first pair of product signals, for producing a first pair of complex convolution signals having magnitudes equal to the real and imaginary components of the convolution of said first product signals and said impulse response, and
• d. squaring circuit means, having inputs connected to receive said first pair of convolution signals, for producing a final output signal equal to the sum of the squares of the real and imaginary parts of said convolution signals.

A device for computing the square of the magnitude of N components of the discrete Fourier transform of an arbitrary input signal, said device comprising:

• a. first generator means comprised of charge transfer devices for providing a pair of output signals having magnitudes equal to the real and imaginary components of EXP[-j.pi.n.sup.2 /N],
• b. first multiplier means having inputs for receiving in parallel (1) successive samples of selected segments of complex electronic input signals, and (2) said output signals of said first generator, for providing a first pair of product signals having magnitudes equal to the real and imaginary components of the product of the signals on said inputs of said first multiplier,
• c. first filter means, having inputs connected to receive said first pair of product signals, having a complex impulse response of EXP[j.pi.k.sup.2 /N] obtained by utilizing charge transfer devices arranged as a dual impulse response transversal filter;
• for producing a first pair of complex convolution signals having magnitudes equal to the real and imaginary components of the convolution of said first product signals and said impulse response, and
• d. squaring circuit means, having inputs connected to receive said first pair of convolution signals, for producing a final output signal equal to the sum of the squares of the real and imaginary parts of said convolution signals.