Variable sized aperture window of an analog-to-digital converter

An improvement in sampling a high frequency input analog signal and converting it to a digital output signal is disclosed. This is accomplished by using a multitude of analog-to-digital converters in conjunction with a distributed sampling system. This combination of multiple converters and a distributed sampling system allows use of conventional device processing, such as that of 0.18 micron silicon, and also provides accurate sampling of very high frequency input signals. The distributed sampling system provides multiple samplings of the input signal by using multiple ADCs for multiple samplings, wherein each sampling is sequentially offset a fixed amount of time from the most recent preceding sampling. Each ADC has a designated central processing unit (CPU) to obtain sufficient data transfer capabilities. The samplings from the multitude of ADCs are a series of sequential digital output values. The digital output values could be the result of samplings all at the same frequency, or at different frequencies. Types of distributed sampling systems include a multitude of elongated trace patterns interconnected in series, a multitude of inverter pairs interconnected in series, a specific permittivity material device, and a sequencer or multiplier. A second enhanced sampling system includes a variable sized aperture window, wherein a width of a sample pulse is narrowed through a variable clock mechanism to produce a faster sampling rate. This variable sized aperture window system can be used by itself, or in combination with any of the presently described multiple ADC distributed sampling systems.

Skip to: Description  ·  Claims  · Patent History  ·  Patent History
Description

This application is a divisional of co-pending U.S. application Ser. No. 11/800,708, filed May 7, 2007 by the same inventors (issued Aug. 11, 2009 as U.S. Pat. No. 7,573,409), which claims priority to U.S. application Ser. No. 11/726,739, filed Mar. 22, 2007 by the same inventors (issued May 5, 2009 as U.S. Pat. No. 7,528,756), both of which are incorporated herein by reference in their entireties.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to the field of computers and computer processors, and more particularly to analog-to-digital converters (ADCs).

2. Description of the Background Art

An analog-to-digital converter (ADC) is an electronic circuit that converts continuous signals to discrete digital numbers. Typically, an ADC is an electronic device that converts an input analog voltage to a digital number.

The analog signal is continuous in time and it is necessary to convert this to a flow of digital values. It is therefore required to define the rate at which new digital values are sampled from the analog signal. The rate of new values is called the sampling rate or sampling frequency of the converter, and is typically reported as the number of samples per second (sps).

A continuously varying bandlimited signal can be sampled at intervals of time T, the sampling time, and measured and stored. The original signal can then be exactly reproduced from the discrete-time values by an interpolation formula. However, this reproduction is only possible if the sampling rate is higher than twice the highest frequency of the signal. This is sometimes referred to as the Shannon-Nyquist sampling theorem. Since a practical ADC cannot make an instantaneous conversion, the input value must necessarily be held constant during the time, called the conversion time, within which the converter performs a conversion.

It is often desirable to be able to sample analog signals in an integrated circuit (IC) at very high frequencies, for example in the range of several gigahertz (GHz). However, certain types of ICs are made with older semiconductor manufacturing and material technology that is capable of sampling signals only at lower frequencies, for example in the range of <1-2 GHz.

FIG. 1 is a diagrammatic representation of an example of an analog-to-digital (A-to-D) sampling system 100 as is currently known in the art. Embedded in chip 101 is the A-to-D block 102. A-to-D block 102 has a data output 105, typically, but not necessarily a parallel bus, and a sampling frequency control 104, which is used to sample input signal 103. The highest frequency component of the input component is f, and the sampling frequency fs must be at least twice the frequency of fi, preferably 2.2 times the frequency of fi for sampling that supports functions such as Fourier Transformations (FT) or Fast Fourier Transformations (FFT), etc. Therefore, if the desired input frequency fi is in the 10 GHz range, the chip must be able to clock the sampling frequency fs at approximately 20-22 GHz, based on the Nyquist Frequency. Building a chip with such a high sampling frequency is more costly, and the architecture of such chips does not permit embedding of large data functions such as CPUs, memory, etc. in such a chip.

Several analog-to-digital conversion methods are known. FIG. 1A is a schematic representation of a sample and hold circuit diagram for an ADC, which is also called a track and hold circuit. When the sample and hold switch 110 is open, the last instantaneous value of the input voltage is held on the sample and hold capacitor 111. When the sample and hold switch 110 is closed, the circuit is in track mode. Buffers 112 on the input and output isolate the sample and hold capacitor 111. A sample and hold ADC is simple and reliable, but is limited in its sampling frequency rate and it has a high error probability.

A second analog-to-digital conversion method is that which utilizes a phase detector ADC. A phase detector generates a voltage signal which represents the difference in phase between two signal inputs. When the two compared signals are completely in phase, the two equal inputs to an XOR gate will output a constant level of zero. With a one degree phase difference, the XOR gate will output a 1 for the duration of the signals being different ( 1/360th of the cycle). When the signals are 180 degrees apart, the XOR gate puts out a steady 1 signal. Integration of the output signal results in an analog voltage proportional to the phase difference. A phase detector contains a number of XOR gates that simultaneously measure a number of phase differences of the input signal. This has the advantage of being a fast acting device, but has the disadvantage of being a large power consumption device.

A third analog-to-digital conversion method is that which utilizes a flash ADC, which is also called a parallel ADC. FIG. 1B is a schematic representation of a flash ADC circuit diagram. A flash ADC is formed of a series of comparators 120, where each comparator 120 compares the input signal to a unique reference voltage. The comparator 120 outputs connect to the inputs of a priority encoder circuit 121, which then produces a binary output 122. As the analog input voltage exceeds the reference voltage at each comparator 120, the comparator 120 outputs will sequentially saturate to a high state. The priority encoder 121 generates a binary number based on the highest order active input, ignoring all other active inputs. The flash ADC is efficient in terms of speed, but contains a large number of components. For example, a three-bit flash ADC requires eight comparators, a four-bit version requires 16 comparators, and an eight-bit version requires 256 comparators.

A fourth analog-to-digital conversion method is a successive approximation ADC, schematically shown in FIG. 1C. The successive approximation ADC uses a successive approximation register (SAR) 130 as a sequence counter. This SAR 130 counts by trying all values of bits starting with the most significant bit (MSB) and finishing at the least significant bit (LSB). Throughout the count process, the SAR 130 monitors the comparator's output to see if the binary count is less than or greater than the analog signal input, and then adjusts the bit values accordingly. Different values of bits are tried from MSB to LSB to get a binary number that equals the original decimal number. The digital-to-analog converter (DAC) 131 output converges on the analog signal input much faster than with a regular sequence counter. The stoichastic renormalization group (SRG) 132 acts as a decimal to binary converter. The successive approximation ADC is a faster device, but has the disadvantages of high power consumption and a large number of components.

Various approaches have been taken to find an economical system that can sample high frequency input rates. In an article entitled, Design of a High-Performance Analog-to-Digital Converter, by Kevin Nary, published in CSD Magazine in October 1998, Nary discloses a folding and interpolating 8-bit 2-Gsps ADC. The number of comparators required for a 4-bit ADC is reduced from fifteen to six when switching from a flash to a folding architecture. This ADC increases the analog bandwidth and the maximum sample rate and consumes less power than a flash architecture ADC. One method of achieving a folding function uses cross-coupled, differential amplifiers, where a single fold is achieved with two cross-coupled, differential amplifiers. By adding more resistors and differential pairs, the number of folds may be increased. Nary reported results of a 2 GHz sampling frequency with 98 MHz input frequency.

Another approach has been disclosed in an article entitled, Capturing Data from Gigasample Analog-to-Digital Converters, by Ian King, published in I/O Magazine in January 2006, which discloses a method of de-multiplexing the digital output. For a 1.5 GHz sample rate, the conversion data will be output synchronous to a 750 MHz clock, where the data is presented to the outputs on both the rising and falling edges of the clock. Two latches are then used, wherein one latch is clocked on the rising edge of the phase-locked data clock and a second latch is clocked using a signal that is 180 degrees out of phase. This reduces the output to 375 MHz. After latching the incoming data, the clock domain is shifted using an intermediate set of latches so that all of the data can be clocked into a memory array on the same clock edge, which de-multiplexes the data rate to 187.5 MHz. A single-channel device can be put into a dual-edge sampling mode to increase the sampling speed from 1.5 Gsps to 3.0 Gsps, which increases the number of output data bits from 8 to 16. A system and method are clearly needed in which much higher sampling frequencies than 2-3 GHz can be converted.

SUMMARY OF THE INVENTION

It is an object of the present invention to adequately sample a very high frequency input analog signal using circuitry which, otherwise might not be able to sample at a sufficiently high rate.

An embodiment of the presently described invention includes a substrate with several ADCs and central processing units (CPUs), and a distributed sampling system. Each ADC works in conjunction with a designated CPU to form an ADC system. Each individual ADC system may contain conventional devices formed from 0.18 micron silicon, as an example. In this example, such individual systems are capable of sampling signals in the range of 1-2 GHz or less.

The description of the present invention illustrates how multiple conventional devices can be used to adequately sample a very high frequency input signal. A timing signal is passed through a distributed sampling system, also called a delay sampling system or a relay sampling system. When the timing signal reaches a first designated point along the distributed sampling system, a first ADC samples an input signal. When the timing signal reaches a second designated point along the distributed sampling system, a second ADC samples the input signal. The timing signal continues through the distributed sampling system until an established number of samplings have been taken by the same established number of ADC systems.

In the case when the devices are on a single chip, as in the present example, the timing signal is passed along the chip through the distributed sampling system. The occurrence of each subsequent sampling occurs at a clocked amount of time after the previous sampling. This is achieved by a plurality of sequential sampling prompts or taps originating from the distributed sampling system as the timing signal travels through the system. This results in cumulative samplings of the high frequency input signal by several ADCs, such that an adequate sampling necessary for optimum Nyquist-Shannon sampling is achieved. For example, if it is desired to adequately sample an input signal of 10 GHz using conventional systems capable of only 1 GHz sampling, then 20 ADC systems would be necessary in order to sequentially sample the input analog signal. In the present example, each ADC system obtains a sampling at a clocked 50 psec interval after the previous sampling. The sampling results of all 20 ADC systems are combined to obtain a result that produces essentially the same output as a single ADC system that is capable of sampling at 20 GHz.

Several distributed sampling systems are described. One such distributed sampling system includes several elongated trace patterns or additional lengths of wire, which are electrically interconnected in series. A timing signal travels through a first additional length of wire, after which a timing signal tap or prompt causes a sampling of the input signal to be taken by a first ADC system; this occurs at a specified period of time, given by Δt. The timing signal continues through a second additional length of wire, after which a timing signal tap or prompt causes a second sampling of the input signal to be taken by a second ADC system; this occurs after a second period of time, Δt. The timing signal continues through an established number of lengths of wire, which causes a cumulative sampling from the same established number of ADC systems. The results of the sequential samplings are a series of sequential digital output values from a plurality of ADCs. The digital output values could be the result of samplings all at the same frequency, or at different frequencies.

Another example of a distributed sampling system includes a specified permittivity material device, such as a SAW device. The material of the device determines the rate at which a timing signal travels through it. Samples of an input analog signal are taken by a plurality of ADC systems when a timing signal reaches a plurality of equidistant points along the device. The results of the sequential samplings are a series of sequential digital output values from a plurality of ADCs. The digital output values could be the result of samplings all at the same frequency, or at different frequencies.

Still another example of a distributed sampling system uses a sequencer or multiplier, such that a timing signal is multiplied a set number of times in order to produce an incremental period of time, Δt for each stage. The ADC systems sample an input analog signal after each period of time, Δt. The input signal sampling results from the multiplier sampling system are a series of sequential digital output values from a plurality of ADCs. The digital output values could be the result of samplings all at the same frequency, or at different frequencies.

An example of an ADC differential op amp circuit, which provides large common mode rejection is also described. By sampling the input signal out of phase, the input signal is completely differentiated and set apart from the background noise. This provides a cleaner signal, and therefore more accurate sampling results.

Yet another example of an ADC circuit discloses an A-to-D cell, which is based on a voltage controlled oscillator (VCO) circuit connected to an input. The VCO output goes into a counter, where the output is then compared to, or timed with a reference frequency through a gate, such as an XOR gate. The output then connects to a CPU, which also controls resets of the counter.

An example of a variable sized aperture window sampling system is also described. This example is achieved through the utilization of a variable aperture clock, such as a resistor capacitor differentiator comprised of a voltage controlled resistor and a capacitor. This variable aperture clock can modify the pulse width of a sample pulse to form a narrower pulse width, and therefore a faster sampling rate. This variable sized aperture window sampling system can be used by itself for ADC sampling, or it can be combined with any of the previously described multiple ADC distributed sampling systems.

These and other objects and advantages of the present invention will become clear to those skilled in the art in view of the description of modes of carrying out the invention, and the industrial applicability thereof, as described herein and as illustrated in the several figures of the drawings. The objects and advantages listed are not an exhaustive list of all possible advantages of the invention. Moreover, it will be possible to practice the invention even where one or more of the intended objects and/or advantages might be absent or not required in the application.

Furthermore, those skilled in the art will recognize that various embodiments of the present invention may achieve one or more, but not necessarily all, of the described objects and/or advantages. Accordingly, the objects and/or advantages described herein are not essential elements of the present invention, and should not be construed as limitations.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagrammatic view of a conventional ADC system;

FIG. 1A is a circuit diagram of a sample and hold ADC;

FIG. 1B is a circuit diagram of a flash ADC;

FIG. 1C is a circuit diagram of a successive approximation ADC;

FIG. 2 is a block diagrammatic view of a general ADC system according to the present invention;

FIGS. 3a-3b are representations of the timing relationship between samplings taken of an input analog signal and taps made in a timing signal distributed line according to a first embodiment of the presently described invention;

FIG. 4 is a representation of the timing relationship between samplings taken of an input analog signal and taps made in a timing signal distributed line according to a second embodiment of the presently described invention;

FIGS. 5-6 are block diagrammatic views of a third embodiment of the presently described invention;

FIGS. 7a-7b are circuit diagrams for an ADC that could be used with the presently described invention;

FIG. 8 is a diagrammatic view of a computer array, according to the present invention;

FIG. 9 is a detailed diagram showing a subset of the computers of FIG. 8 and a more detailed view of the interconnecting data buses of FIG. 8;

FIG. 10 is a block diagram depicting a general layout of a stack computer;

FIGS. 11a-11c are diagrammatic views of an ADC and computer system array according to the present invention;

FIG. 12a is a circuit diagram of an ADC sampling system according to the present invention;

FIG. 12b shows input voltage vs. output frequency characteristics of a CMOS silicon process; and

FIG. 13 is a circuit diagram of an enhanced ADC sampling system according to the present invention.

DETAILED DESCRIPTION

This invention is described with reference to the figures, in which like numbers represent the same or similar elements. While this invention is described in terms of modes for achieving this invention's objectives, it will be appreciated by those skilled in the art that variations may be accomplished in view of these teachings without deviating from the spirit or scope of the presently claimed invention.

The embodiments and variations of the invention described herein, and/or shown in the drawings, are presented by way of example only and are not limiting as to the scope of the invention. Unless otherwise specifically stated, individual aspects and components of the invention may be omitted or modified for a variety of applications while remaining within the spirit and scope of the claimed invention, since it is intended that the present invention is adaptable to many variations.

FIG. 2 shows an example of an ADC system 200 according to the present invention. An input signal 204 is passed into a number of analog-to-digital converter cells 202a through 202n of a chip 201. An external sampling clock 205 is shown in this example, but an internal clock could have been utilized as well. The sampling clock 205 is run at a substantially lower frequency, for example, 10 or 20 times lower frequency, than the intended sampling rate. By providing sequential time periods from time distribution apparatus 206a through 206n, it is possible to increase the net sampling rate by n number of times. In this example, the time periods are provided by an external source, although as discussed above, an internal timing source could also be utilized. If the input signal 204 with a frequency up to 10 GHz is to be accurately sampled, then a sampling clock 205 at 20 or 22 GHz would be necessary for optimum Nyquist-Shannon sampling. However, in the present inventive system, the sampling clock 205 can run at, for example, 1 GHz for n=20 or 22, respectively. The time periods, provided by the time distribution apparatus 206a through 206n would be in an increment of 1/20, 1/22, or similar increment of the sampling frequency, so that each ADC 202 would sample the input signal 204 at a slightly delayed point, resulting in a sampling that would be equivalent to using a single ADC, sampling at a rate of 20 or 22 GHz. The time periods, provided by the time distribution apparatus 206a through 206n occur as a result of tap line connections 207a through 207n between an individual distribution station (such as 206a) and a corresponding individual ADC (such as A/D 202a). When a timing signal (produced by sampling clock 205) travels through a plurality of serially connected distribution stations or distribution apparatus 206a through 206n a series of taps or sampling prompts are sent through tap line connections 207a through 207n to corresponding ADCs 202a through 202n, respectively.

Such an approach would require a multitude of ADCs or A-to-D channels 202, for example, in this case at least 20 or 22, but it would allow use of an older technology chip 201, for example 0.18 micron silicon, and it would permit sampling of signals running in the 10 GHz range or thereabouts. By increasing the number of A-to-D channels 202 even more, the sampled signal frequency (or its highest Fourier Transform component) could be even further increased.

Names of items 202a through 202n as ADCs, converter cells, or channels have been used in this example interchangeably. Typically, to be able to process the amount of data without losing sample data during processing, each A-to-D channel 202 must have sufficient data transfer capabilities, for example its own CPU 203a through 203n corresponding to A-to-D channels 202a through 202n.

The time period between each ADC sampling of the input signal could be achieved in various ways, as exemplified by the following embodiments. FIG. 3a discloses a time relationship between samplings taken of an input signal 301 and taps in a trace pattern 303 of a time distributed sampling system in a first embodiment of the invention. The trace pattern 303 contains a plurality of elongated wires connected in series. This time distributed sampling system has a plurality of ADC systems, wherein each ADC system comprises an ADC 202 and an associated central processing unit (CPU) 203, as discussed previously in relation to FIG. 2. When a timing signal 306 travels through a first length of wire 303a to a tap point of W1, a prompt to sample the input signal 301 after a measurable amount of time, given by Δt 304 is made. This timing is represented as ADC sampling point C1. As the timing signal 306 continues through a second length of wire 303b to a second tap point of W2, a prompt to sample the input signal 301 after a second period of time, Δt is made. This timing is represented by ADC sampling point C2. A separate ADC system samples the input signal 301 after the timing signal 306 reaches each trace pattern 303 at tap points, W1 through Wn for each tap in the distribution line. The results of the sequential samplings are a series of sequential digital output values from a plurality of ADCs. The digital output values could be the result of samplings all at the same frequency, or at different frequencies.

A more detailed explanation follows with reference to FIG. 3a. A timing signal 306 passes through a first length of wire, 303a to a tap point, W1. At that point in time, an input signal 301 is sampled by a first ADC system, represented in time by the ADC sampling point, C1. When the timing signal 306 travels through a second length of wire, 303b to a tap point represented by W2, the input signal 301 is sampled by a second ADC system, at the ADC sampling point, C2. The above described distributed sampling system continues to sample the input signal 301 at ADC sampling points 302, designated in time by C1, C2, etc. The input signal 301 is sampled after each sequential time period, Δt 304 as the timing signal 306 travels through the plurality of lengths of wire 303. A number of ADC systems are established on a chip in order to adequately sample an input signal 301 in order to meet the Nyquist-Shannon requirement.

Consider the following example, which is given to further clarify the present invention, wherein the given example is not to be construed as a limiting feature. If, for example, an input signal 301 of 10 GHz frequency was to be sampled, then the time difference 304 between ADC sampling points 302 would need to be at least 50 psec to meet the Nyquist-Shannon requirement for an adequate sampling rate of a 10 GHz input signal. Each successive ADC system would sample the input signal 301 at the sampling points C1, C2, etc., wherein each sampling would occur at 50 psec after the previous ADC sampling. The ADC sampling points 302 correspond in time to successive tap points of W1, W2, etc. along the trace pattern 303. If each ADC system was capable of capturing or taking a sample every 1 nsec, then a total of 20 ADC systems would be necessary to adequately sample an incoming 10 GHz signal. In this example, the distributed sampling system using multiple lengths of interconnected wires of the presently described invention is equivalent to using a single ADC, which is capable of sampling an input signal of 10 GHz at a sampling rate of 20 gsps.

FIG. 3b discloses a time relationship between samplings taken of an input signal 301 and taps made in a connected series of inverter pairs 305 of a time distributed sampling system in a second embodiment of the invention. Each clocked trace pattern 303 of FIG. 3a is replaced with a pair of inverters 305 in FIG. 3b. A timing signal 306 travels through a series of connected inverter pairs 305. When the timing signal 306 travels through a first inverter pair 305a, a prompt to sample the input signal 301 after a first time period, Δt 304 is made, which coincides with the ADC sampling point C1. As the timing signal 306 continues through a second inverter pair 305b, a prompt to sample the input signal 301 after a second time period, Δt 304 is made, which coincides with the ADC sampling point, C2. A separate ADC system samples the input signal 301 at each of the ADC sampling points, C1 through Cn, which occurs when the timing signal 306 travels through each inverter pair at points designated by W1 through Wn, respectively. The results of the sequential samplings are a series of sequential digital output values from a plurality of ADCs. The digital output values could be the result of samplings all at the same frequency, or at different frequencies.

A more detailed explanation follows with reference to FIG. 3b. When a timing signal 306 travels through a first inverter pair, 305a to a tap point represented by W1, an input signal 301 is sampled by a first ADC system at a first ADC sampling point, designated in time by C1. When the timing signal 306 travels through a second inverter pair, 305b to a tap point represented by W2, the input signal 301 is sampled by a second ADC system at a second ADC sampling point, C2. The above given distributed sampling system continues to sample the input signal 301 as the timing signal 306 travels through the plurality of inverter pairs 305. As the timing signal 306 travels through each of the inverter pairs 305, the input signal 301 is sampled after each sequential time period, Δt 304 at each ADC sampling point 302. A number of ADC systems are established on a chip in order to adequately sample an input signal 301 in order to meet the Nyquist-Shannon requirement.

FIG. 4 discloses a time relationship between samplings taken of an input analog signal 405 and line taps made in a specific permittivity material device 401 in a third embodiment of the invention. The time distribution sampling is achieved through the use of a specific permittivity material device 401, such as a surface acoustic wave (SAW) device. An input signal 405 is sampled after each measurable time period, Δt 403 as a timing signal 406 travels past each equi-distant point, given by S1 through Sn along the device 401.

The specific permittivity material device distributed sampling system represented by FIG. 4 works similar to the trace distributed sampling system of FIG. 3a. A separate ADC system contains an ADC and a corresponding CPU, as previously described with reference to FIG. 2. Each sequential ADC system samples the input signal 405, represented by ADC sampling points 402 when a timing signal 406 reaches each sequential equi-distant point along the device 401, corresponding to points S1 through Sn. As the timing signal 406 travels through the device 401, a prompt to sample the input signal 405 after each incremental time period, Δt 403 is made, wherein the value of Δt 403 is determined by the specific material of the device 401. When the timing signal 406 reaches a first sampling point 402 given by S1, a prompt to sample the input signal 405 by a first ADC system at a first ADC sampling point, C1 is made. When the timing signal 406 reaches a second sampling point, S2 within the device 401 after a second time period, Δt 403 a second ADC system is prompted to sample the input signal 405 at the corresponding second ADC sampling point, C2. The above described distributed sampling system continues to sample the input signal 405 at ADC sampling points 402, which correspond in time to points, S1 through Sn of the device 401. The results of the sequential samplings are a series of sequential digital output values from a plurality of ADCs. The digital output values could be the result of samplings all at the same frequency, or at different frequencies.

In an example of using an input signal 405 of 10 GHz, a timing signal 406 travels to a first point, given by S1 within the device 401. At this point, a first ADC system is prompted to sample the input signal 405 at the first ADC sampling point C1 after a first time period 403 of 50 psec. When the timing signal 406 travels to a second point S2 within the device 401, a second ADC system is prompted to sample the input signal 405 at a second sampling point, C2 which will occur after a second time period 403 of 50 psec. If each ADC system sampled the input signal 405 at a rate of 1 nsec, then 20 ADC systems would be required to adequately sample an input signal 405 of 10 GHz. In this example, the distributed sampling system using a specific permittivity material device of the presently described invention is equivalent to using a single ADC, which is capable of sampling an input signal of 10 GHz at a sampling rate of 20 gsps.

A fourth embodiment discloses a sequencer or multiplier distributed sampling system 601, and is described with reference to FIG. 5. An example of a sequencer distribution sampling system 601 could use an emitter coupled logic (ECL) as a sequencer 501. The sequencer 501 comprises a group of triggers 508 which are represented by w1 through wn. Each trigger 508 is connected to an ADC 502, each of which is then connected to an associated CPU 506. A timing signal 507 enters the sequencer 501, then each stage sequences or multiplies the timing signal 507 by the same incremental amount, given by Δt 503. Therefore, as a pulse 504 travels through a first ADC trigger, w1 the input signal 505 is sampled by ADC1. After a second time period Δt 503, a pulse 504 travels through a second ADC trigger, w2 wherein the input signal 505 is sampled by ADC2. The above described distributed sampling system continues to sample the input signal 505 by utilizing n number of triggers, w1 through wn and using ADC1 through ADCn converters, respectively. The sampling results are processed by n number of associated CPUs 506. The results of the sequential samplings are a series of sequential digital output values from a plurality of ADCs. The digital output values could be the result of samplings all at the same frequency, or at different frequencies. An important feature of the sequencer 501 is that time between each trigger 508 can be varied.

In an example of a 10 GHz input signal 505, the sequencer 501 comprises 20 triggers 508, as represented by w1 through w20. The input analog signal 505 will be sequentially sampled at intervals of 50 psec time periods, Δt 503. For example, ADC, will sample the input analog signal 505 when a pulse 504 travels through a first ADC trigger, w1 after a first time period, Δt 503 of 50 psec. Then ADC2 will sample the input analog signal 505 when a pulse 504 travels through a second ADC trigger, w2 after a second time period, Δt 503 of 50 psec. If each ADC 502 is capable of sampling an input analog signal 505 at a rate of 1 nsec, then 20 triggers 508 along with 20 associated ADCs 502 and 20 associated CPUs 506 would be necessary to adequately sample a 10 GHz input signal 505 at a sampling rate of 20 gsps. In this example, the distributed sampling system using multiple ADCs with a sequencer or multiplier of the presently described invention is equivalent to using a single ADC, which is capable of sampling an input signal of 10 GHz at a sampling rate of 20 gsps.

FIG. 6 is a block diagram of a sequencer or multiplier distributed sampling system 601 which was described with reference to FIG. 5, with the addition of a clock generating block 602. The clock generating block 602 could be internal or external, and could include, but is not limited to a phase locked loop (PLL), a delay locked loop (DLL), a voltage controlled oscillator (VCO), a ring oscillator, a crystal oscillator, or other type of oscillator. FIG. 6 also shows a timing signal 603.

FIG. 7a is a circuit diagram 707 of an ADC that could be used with the previously described inventions, which utilizes differential op amps. The differential op amp system that is shown in FIG. 7a has two input sources 701 which are utilized in conjunction with op amps 702a and 702b, wherein the op amp 702b is a voltage to current driver with a selectable gain multiplier. This configuration provides large common mode rejection for a very accurate reproduction of the input signal. The system of FIG. 7a further shows a counter 704, a CPU 705, and a digital output signal 706.

FIG. 7b is a circuit diagram 707 of an ADC which comprises a single-ended voltage controlled oscillator 703. The remaining elements are the same as for FIG. 7a. The inverter system of FIG. 7a has the advantage of separating out the desired input signal 701 to be sampled from the undesirable background noise; FIG. 7b has no noise immunity. However, the inverter system of FIG. 7b requires only one pin connection, whereas the inverter system of FIG. 7a requires two pin connections.

The ADC circuit diagrams of FIGS. 7a and 7b could be used with any of the previously described ADC/CPU distributed sampling systems for sampling an input analog signal.

The above described ADC/CPU distributed sampling systems could also be integrated with any of various architectures well known to the inventor. One mode for carrying out the invention is through utilizing an array of individual computers. An array is depicted in a diagrammatic view in FIG. 8 and is designated therein by the general reference character 10. The computer array 10 has a plurality (twenty four in the example shown) of computers 12 (sometimes also referred to as “cores” or “nodes” in the example of an array). In the example shown, all of the computers 12 are located on a single die 14. According to the present invention, each of the computers 12 is a generally independently functioning computer, as will be discussed in more detail hereinafter. The computers 12 are interconnected by a plurality (the quantities of which will be discussed in more detail hereinafter) of interconnecting data buses 16. In this example, the data buses 16 are bidirectional asynchronous high speed parallel data buses, although it is within the scope of the invention that other interconnecting means might be employed for the purpose. In the present embodiment of the array 10, not only can data communication between the computers 12 be asynchronous, but the individual computers 12 can also operate in an internally asynchronous mode. The individual computers 12 operate asynchronously, which saves a great deal of power since each computer 12 will use essentially no power when it is not executing instructions, and since there is no clock running therein.

One skilled in the art will recognize that there will be additional components on the die 14 that are omitted from the view of FIG. 8 for the sake of clarity. Such additional components include power buses, external connection pads, and other such common aspects of a microprocessor chip.

Computer 12e is an example of one of the computers 12 that is not on the periphery of the array 10. That is, computer 12e has four orthogonally adjacent computers 12a, 12b, 12c and 12d. This grouping of computers 12a through 12e will be used, by way of example, hereinafter in relation to a more detailed discussion of the communications between the computers 12 of the array 10. As can be seen in the view of FIG. 8, interior computers 12 such as computer 12e will have four other computers 12 with which they can directly communicate via the buses 16. In the following discussion, the principles discussed will apply to all of the computers 12, except that the computers 12 on the edge of the array 10 will be in direct communication with only three other computers 12, and the corner computers 12 will be in direct communication with only two other computers 12.

FIG. 9 is a more detailed view of a portion of FIG. 8 showing only some of the computers 12 and, in particular, computers 12a through 12e, inclusive. The view of FIG. 9 also reveals that the data buses 16 each have a read line 18, a write line 20 and a plurality (eighteen, in this example) of data lines 22. The data lines 22 are capable of transferring all the bits of one eighteen-bit instruction word simultaneously in parallel.

According to the present inventive method, a computer 12, such as the computer 12e can set high one, two, three or all four of its read lines 18 such that it is prepared to receive data from the respective one, two, three or all four adjacent computers 12. Similarly, it is also possible for a computer 12 to set one, two, three or all four of its write lines 20 high.

When one of the adjacent computers 12a, 12b, 12c or 12d sets a write line 20 between itself and the computer 12e high, if the computer 12e has already set the corresponding read line 18 high, then a word is transferred from that computer 12a, 12b, 12c or 12d to the computer 12e on the associated data lines 22. Then, the sending computer 12 will release the write line 20 and the receiving computer (12e in this example) pulls both the write line 20 and the read line 18 low. The latter action will acknowledge to the sending computer 12 that the data has been received. Note that the above description is not intended necessarily to denote the sequence of events in order. In actual practice, the receiving computer may try to set the write line 20 low slightly before the sending computer 12 releases (stops pulling high) its write line 20. In such an instance, as soon as the sending computer 12 releases its write line 20, the write line 20 will be pulled low by the receiving computer 12e.

Whenever a computer 12 such as the computer 12e has set one of its write lines 20 high in anticipation of writing it will simply wait, using essentially no power, until the data is “requested”, as described above, from the appropriate adjacent computer 12, unless the computer 12 to which the data is to be sent has already set its read line 18 high, in which case the data is transmitted immediately. Similarly, whenever a computer 12 has set one or more of its read lines 18 to high in anticipation of reading it will simply wait, using essentially no power, until the write line 20 connected to a selected computer 12 goes high to transfer an instruction word between the two computers 12.

As discussed above, there may be several potential means and/or methods to cause the computers 12 to function as described. However, in this present example, the computers 12 so behave simply because they are operating generally asynchronously internally (in addition to transferring data there-between in the asynchronous manner described). That is, instructions are generally completed sequentially. When either a write or read instruction occurs, there can be no further action until that instruction is completed (or, perhaps alternatively, until it is aborted, as by a “reset” or the like). There is no regular clock pulse, in the prior art sense. Rather, a pulse is generated to accomplish a next instruction only when the instruction being executed either is not a read or write type instruction (given that a read or write type instruction would require completion, often by another entity) or else when the read or write type operation is, in fact, completed.

FIG. 10 is a block diagram depicting the general layout of an example of one of the computers 12 of FIGS. 8 and 9. As can be seen in the view of FIG. 10, each of the computers 12 is a generally self contained computer having its own RAM 24 and ROM 26. As mentioned previously, the computers 12 are also sometimes referred to as individual “nodes”, given that they are, in the present example, combined on a single chip.

Other basic components of the computer 12 are a return stack 28 including an R register 29, an instruction area 30, an arithmetic logic unit (“ALU” or “processor”) 32, a data stack 34 and a decode logic section 36 for decoding instructions. One skilled in the art will be generally familiar with the operation of stack based computers such as the computers 12 of this present example. The computers 12 are dual stack computers having the data stack 34 and the separate return stack 28.

In this embodiment of the invention, the computer 12 has four communication ports 38 for communicating with adjacent computers 12. The communication ports 38 are further defined by the up port 38a, the right port 38b, the left port 38c, and the down port 38d. The communication ports 38 are tri-state drivers, having an off status, a receive status (for driving signals into the computer 12) and a send status (for driving signals out of the computer 12). If the particular computer 12 is not on the interior of the array (FIG. 8) such as the example of computer 12e, then one or more of the communication ports 38 will not be used in that particular computer, at least for the purposes described above. However, those communication ports 38 that do abut the edge of the die 14 can have additional circuitry, either designed into such computer 12 or else external to the computer 12 but associated therewith, to cause such communication port 38 to act as an external I/O port 39 (FIG. 8). Examples of such external I/O ports 39 include, but are not limited to, USB (universal serial bus) ports, RS232 serial bus ports, parallel communications ports, analog to digital and/or digital to analog conversion ports, and many other possible variations. No matter what type of additional or modified circuitry is employed for this purpose, according to the presently described embodiment of the invention, the method of operation of the “external” I/O ports 39 regarding the handling of instructions and/or data received there from will be alike to that described, herein, in relation to the “internal” communication ports 38. In FIG. 8 an “edge” computer 12f is depicted with associated interface circuitry 80 (shown in block diagrammatic form) for communicating through an external I/O port 39 with an external device 82.

In the presently described embodiment, the instruction area 30 includes a number of registers 40 including, in this example, an A register 40a, a B register 40b and a P register 40c. In this example, the A register 40a is a full eighteen-bit register, while the B register 40b and the P register 40c are nine-bit registers. Also depicted in block diagrammatic form in the view of FIG. 10 is a slot sequencer 42.

The data stack 34 and the return stack 28 are not arrays in memory accessed by a stack pointer, as in many prior art computers. Rather, the stacks 34 and 28 are an array of registers. The top two registers in the data stack 34 are a T register 44 and an S register 46. The remainder of the data stack 34 has a circular register array 34a having eight additional hardware registers therein numbered, in this example S2 through S9. One of the eight registers in the circular register array 34a will be selected as the register below the S register 46 at any time. The value in the shift register that selects the stack register to be below S cannot be read or written by software. Similarly, the top position in the return stack 28 is the dedicated R register 29, while the remainder of the return stack 28 has a circular register array 28a having eight additional hardware registers therein (not specifically shown in the drawing) that are numbered, in this example R1 through R8.

In addition to the registers previously discussed herein, the instruction area 30 also has an 18 bit instruction register 30a for storing an instruction word that is presently being used, and an additional 5 bit opcode register 30b for the particular instruction word presently being executed.

The previously described ADC/CPU distributed sampling systems could be integrated with the above described computer array, resulting in numerous system combinations of different type, size, and purpose. In addition, such systems could be processed as individual discrete components integrated together onto a substrate, or processed completely on a single chip, or a combination of the two processes.

The following description will give two examples of different ADC array possibilities, which are given to further clarify the present invention and are not to be construed as limiting features. FIG. 11a shows a chip or die 14 with several computers or nodes 12. The interior computers 12 are designated as general purpose computers (G) 94, which are interconnected and therefore, can share resources there between as previously described. The periphery of the die 14 contains several ADCs (A) 95. Each ADC (A) 95 has a dedicated computer, referred to as an ADC computer (C) 96. Each ADC computer (C) 96 has access to any or all of the general purpose computers (G) 94. The connections between the ADC computers (C) 96 may or may not be utilized.

FIG. 11b shows another embodiment of a die 14 with several computers 12. ADCs (A) 95 are formed at the periphery of the die 14, but there are no dedicated ADC computers (C) 96 as in FIG. 11a. Each ADC (A) 95 would have direct access to any or all of the interconnected general purpose computers (G) 94.

FIG. 11c shows a die 14 with a total of forty computers 12, wherein 20 computers 12 are ADCs (A) 95 and 20 computers 12 are general purpose computers (G) 94. FIG. 11c is an example of a die 14 which could be utilized in the previous examples of sampling a 10 GHz input analog signal. Each individual ADC was capable of sampling at a rate of 1 gsps; therefore, 20 such ADCs (A) 95 and 20 associated general purpose computers (G) 94 would be necessary to sample a 10 GHz input analog signal.

FIG. 12a is a circuit diagram of an ADC system 1200 according to another embodiment of the current invention. A-to-D cell 202 is based on a voltage controlled oscillator (VCO) circuit using VCO 1201 connected to input 204. The VCO output goes into a counter 1202, where the output is then compared to, or timed with a reference frequency 1203 through a gate, such as XOR gate 1204. The output then connects to a CPU 203, which also controls resets of the counter 1202 through line 1205.

The ADC circuit diagram of FIG. 12a has combined the advantages of the prior art ADC conversion methods that were previously discussed above, and decreased or eliminated their disadvantages. The ADC circuit diagram of FIG. 12a has the simplicity and reliability of the sample and hold circuit of FIG. 1A, and the speed and accuracy of the phase detector, flash (FIG. 1B), and successive approximation (FIG. 1C) circuits. The present inventive circuit of FIG. 12a has a small number of components, and uses very little power compared to the fast circuits. The input 204 of the present inventive ADC circuit is not limited to voltage sources, and it is not frequency dependent. There is no limitation on the range of the VCO 1201, and the counter 1202 is open to any speed or rate.

FIG. 12b shows as diagram 1211 characteristics of a VCO 1201 in a CMOS silicon process, such as 0.18 micron silicon. The input voltage range is from 0 to 1.8 volts, where the frequency moves from 1 GHz to 2 GHz. However, there is a narrow dynamic or useful range 1212 approximately 1 or 1.2 volts wide. Transfer curve 1213 shows the input voltage on the x-axis and the output frequency in GHz on the y-axis.

FIG. 13 shows an enhanced sampling system 1300 according to another embodiment of the present invention. In A-to-D converter cell 202, the input line 204 connects to an optional input buffer 1307. It then continues to an input sampling switch 1301, which connects to sample-and-hold capacitor 1302, whose voltage controls the VCO 1201. This approach allows the oscillator to run at a stable frequency between samplings. VCO 1201, in turn is connected to counter 1202, as described above which then has connections to CPU 203. CPU 203 also controls, in this example a sample pulse, which it sends to buffer 1306. A variable aperture clock system, such as a resistor capacitor differentiator made of a voltage-controlled resistor 1304 and a capacitor 1305 is utilized, where the resistor 1304 is voltage-adjustable. The CPU 203 causes a differentiated, shorter pulse which is buffered in buffer 1303, and the CPU 203 also controls the input sampling switch 1301. By controlling the resistor voltage, the CPU 203 can modify the pulse width of the sampling period and create a smaller aperture window, and thereby increase the sampling rate.

The voltage controlled resistor 1304 and capacitor 1305 create a resistor capacitor differentiator, which determines the aperture window size or the variable rate for the input sampling switch 1301. The CPU 203 modifies the pulse width of the sampling period by controlling the resistor 1304 voltage. The CPU 203 creates a differentiated, shorter pulse, and thereby controls the input sampling sample and hold switch 1301. A shorter sample aperture window, which provides a shorter VCO 1201 leads to the ability to sample higher frequency input signals. A variable sample aperture window also resynchronizes the sampling phases back together again, via a resynchronizing circuit.

Modifying the pulse width affects the settling time, etc. of the capacitor, and thus affects the accuracy of the sampling. There is a trade-off between speed and accuracy, where a higher speed leads to a less accurate measurement. Therefore, resistor 1304 allows the system to have a software control (not shown) for accuracy running as code in CPU 203.

The presently described invention of a variable width aperture window, which provides a variable sampling rate, can be used by itself or in combination with any of the previously described time distributed ADC sampling systems. Therefore, each ADC of a multiple ADC distributed sampling system could also comprise a variable aperture clock, such as a resistor capacitor differentiator to provide a shorter pulse and therefore, a shorter aperture window and a faster sampling rate. Likewise, the ADC variable rate aperture window sampling system could be used with any of the previously described multiple ADC distributed sampling system embodiments, including but not limited to the trace pattern embodiment described with reference to FIG. 3a, the inverter pair embodiment described with reference to FIG. 3b, the specific permittivity material device embodiment described with reference to FIG. 4, and the sequencer embodiment described with reference to FIGS. 5 and 6.

All of the above examples are only some of the examples of available embodiments of the present invention. Those skilled in the art will readily observe that numerous other modifications and alterations may be made without departing from the spirit and scope of the invention. Accordingly, the disclosure herein is not intended as limiting and the appended claims are to be interpreted as encompassing the entire scope of the invention.

Claims

1. An analog-to-digital converter (ADC) circuit, comprising:

a voltage controlled oscillator (VCO);
a counter;
a reference frequency source;
an input signal source; and
a logic gate.

2. The circuit of claim 1, further comprising a connection to a central processing unit (CPU).

3. The circuit of claim 2, wherein:

an output of said VCO goes into said counter.

4. The circuit of claim 3, wherein:

said output is compared to said reference frequency through said gate.

5. The circuit of claim 2, wherein:

said CPU controls resets of said counter.

6. The circuit of claim 1, wherein:

said logic gate is an exclusive-OR (XOR) gate.

7. A computer array, comprising:

a substrate;
an input signal source;
a first analog-to-digital converter circuit integrated on said substrate and coupled to said input signal source, said first analog-to-digital converter circuit including a VCO, a counter, a reference frequency source; and a logic gate;
a first computer integrated on said substrate, said first computer being coupled to said first analog-to-digital converter circuit; and
a second computer integrated on said substrate.

8. The computer array of claim 7, further comprising:

a second analog-to-digital converter circuit integrated on said substrate and coupled to said input signal source and to said second computer, said second analog-to-digital converter circuit including a VCO, a counter, a reference frequency source, a logic gate.

9. The computer array of claim 8, further comprising:

a sampling clock signal source operative to provide a clock signal; and
a first time distribution apparatus coupled to said sampling clock signal source; and
wherein said first analog-to-digital converter circuit is coupled to said first time distribution apparatus.

10. The computer array of claim 9, wherein said second analog-to-digital converter circuit is coupled to said first time distribution apparatus.

11. The computer array of claim 9, wherein said second analog-to-digital converter circuit is coupled to said time distribution apparatus, said time distribution apparatus being operative to increase the time it takes for said clock signal to reach said second analog-to-digital converter circuit with respect to the time it takes for said clock signal to reach said first analog to digital converter.

12. The computer array of claim 8, further comprising a data path connecting said first computer to said second computer.

13. The computer array of claim 8, further comprising:

a third computer integrated on said circuit substrate;
a fourth computer integrated on said circuit substrate; and
a data path connection said third computer to said fourth computer.

14. The computer array of claim 7, wherein said logic gate is an exclusive-OR (XOR) gate.

15. The computer array of claim 7, wherein said VCO is operative to provide an output signal to said counter

16. The computer array of claim 15, wherein said logic gate is operative to compare said output signal of said VCO with a reference frequency provided by said reference frequency source.

17. The computer array of claim 7, wherein said first computer is operative to control said counter.

18. The computer array of claim 7, wherein:

said VCO includes an input and an output;
said input of said VCO is coupled to said input signal source; and
said output of said VCO is coupled to said counter.

19. The computer array of claim 7, further comprising a second analog-to-digital converter circuit coupled to said input signal source and said second computer, said second analog-to-digital converter circuit including:

an input sampling switch;
a sample and hold capacitor;
a variable aperture clock;
a counter; and
a VCO.
Patent History
Publication number: 20100117880
Type: Application
Filed: Aug 10, 2009
Publication Date: May 13, 2010
Inventors: Charles H. Moore (Sierra City, CA), Leslie O. Snively (Castle Rock, CO), John Huie (Scottsdale, AZ)
Application Number: 12/462,828