WO2011069229A1 - Filtre à architecture de tisseur virtuel - Google Patents

Filtre à architecture de tisseur virtuel Download PDF

Info

Publication number
WO2011069229A1
WO2011069229A1 PCT/CA2009/001817 CA2009001817W WO2011069229A1 WO 2011069229 A1 WO2011069229 A1 WO 2011069229A1 CA 2009001817 W CA2009001817 W CA 2009001817W WO 2011069229 A1 WO2011069229 A1 WO 2011069229A1
Authority
WO
WIPO (PCT)
Prior art keywords
circuit
signal
sampling
filter
circuits
Prior art date
Application number
PCT/CA2009/001817
Other languages
English (en)
Inventor
Martin Mallinson
Dustin Forman
Original Assignee
Ess Technology, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ess Technology, Inc. filed Critical Ess Technology, Inc.
Priority to CN200980163431.7A priority Critical patent/CN102783015B/zh
Priority to PCT/CA2009/001817 priority patent/WO2011069229A1/fr
Priority to US13/515,259 priority patent/US8774749B2/en
Publication of WO2011069229A1 publication Critical patent/WO2011069229A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/06—Receivers
    • H04B1/16—Circuits
    • H04B1/30—Circuits for homodyne or synchrodyne receivers
    • H04B1/302—Circuits for homodyne or synchrodyne receivers for single sideband receivers
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
    • H03H15/00—Transversal filters

Definitions

  • This invention relates to signal processing. More particularly, this invention relates to the design of analog circuit effectively providing a quadrature-matched pair of mixers that are useful in implementing a virtual Weaver architecture filter.
  • Radio communications are based upon the observation that electro-magnetic energy is capable of radiating into empty space as a constant exchange of magnetic and electrical energy.
  • An oscillating electrical signal in a conducting material creates such radiation, to a greater or lesser degree dependent upon the so called “matching" of the conducting material to the frequency of oscillation, and when such radiation impinges upon another suitable material it induces current or voltage oscillations in that material, again to a greater or lesser extent depending on the matching.
  • This induced signal is very small, as perhaps can be appreciated by the fact that the solid angle subtended by the receiving material relative to the transmitting material can be very small indeed, and so the percentage of energy collectable in even an ideal system is tiny.
  • the physicist is inclined to consider the operation of a radio receiver as a thermodynamic device, inquiring into the total energy representing the signal at the receiver, and asking to what degree this signal exceeds the thermal noise present due to the finite temperature of the environment.
  • All energy storage means are coupled to the ambient temperature of the environment as is well understood from the 19 th century study of thermodynamics. Hence they all acquire a certain amount of energy within a given bandwidth and the wanted signal is additive to this thermal noise.
  • the extent to which the signal exceeds the noise is the signal-to-noise ratio.
  • Thermal noise exists in equal amounts across all possible frequencies of the oscillation of voltage or current. Indeed, one of the outstanding conclusions of thermodynamics is the 'equipartition of energy' theorem which states that there will be an equal quantity of energy in each possible bandwidth of radio receiver. Hence the total magnitude of thermal noise is assessed by summing over all frequencies of signal that are present at the receiver.
  • This last observation of the physicist is key to the engineer's design of a radio: the signal can be distinguished from the noise to a much higher degree if the bandwidth of the received signal is restricted. That is to say, if the engineer is able to make the radio receiver neglect all those frequencies that are not part of the desired signal, then the total noise will be less, since the noise must be added for every frequency present at the receiver output.
  • This aspect of the radio is called its "selectivity" and is critical in the separation of the signal from the noise. Higher selectivity means higher performance, lower noise, and better fidelity.
  • selectivity means higher performance, lower noise, and better fidelity.
  • the conceptually simplest viable radio receiver is of the so-called "tuned RF" type that arranges for a high Q circuit to amplify the desired signal frequency as much as possible and rapidly roll off its response to signals of any other frequency.
  • the tuned RF receiver suffers from a practical difficulty: in order to change the received frequency this tuned circuit must be changed, involving somehow adjusting the value of the inductor and/or capacitor, a task that is not impossible, but is inconvenient and somewhat costly. It would be far more convenient and less costly if the selectivity were always provided at a fixed frequency; a seemingly impossible task since the received frequency is required to change as different radio stations are selected.
  • the heterodyne principle invented in the latter years of the 19 th century, whereby the desired power of the received signal is shifted in frequency to a constant intermediate frequency and selectivity is achieved by circuit elements working at this fixed intermediate frequency.
  • the received signal at the desired frequency of reception is caused to interact with a signal generated locally in the receiver in such a way as to produce a third frequency, the intermediate frequency.
  • the heterodyne principle necessarily generates two output frequencies: one at 'frec+flo' and one at 'frec- flo' where 'free' is the received frequency and 'flo' is the local oscillator frequency.
  • the "interaction" of local oscillator and received signal is ideally multiplication and the electronic element that achieves this multiplication is called a "mixer".
  • the locally generated signal is adjustable in frequency of oscillation and so is able to transform any of a range of received frequencies to the same intermediate frequency.
  • Selectivity is provided at the fixed intermediate frequency by components that need not change; the choice of frequency to receive is set by the adjustable local oscillator.
  • This heterodyne principle or more precisely a variant of it that selects only one of the two heterodyne frequencies (typically the difference frequency 'frec-flo'), the so called "super-heterodyne" receiver, has come to dominate: it is ubiquitous in all applications of radio communication.
  • the "suitable phase shifting network” is itself not perfect: it is probably impossible to make a fixed phase shift over more than a narrow range of frequencies and the Hartley image rejection method is limited by the difficulty of making a wide frequency range phase shifting circuit.
  • Weaver In 1956 Donald Weaver described what has become known as the Weaver architecture for image rejection in his paper entitled "A third method of generation and detection of single-sideband signals", D.K. Weaver - Proc. IRE, 1956.
  • the first method implied in Weaver's title is the filter used prior to Hartley; the second method is Hartley's method.
  • Illustrated in Fig. 2, Weaver's method relies upon two sets of mixers; each set is called a quadrature modulator because it multiplies the signal by a local oscillator that has not one but two outputs. The two outputs are in quadrature to each other - they are 90 degrees apart in phase.
  • Weaver's architecture has been successful. Modern digital radios (cell phones, TV receivers, etc.) use his architecture and achieve image rejection to an adequate degree.
  • the Weaver architecture limited only by the quadrature nature of the signals, can reject an image by about 50db (about 1 part in 300).
  • Various digital enhancements can, with advanced signal processing in the digital domain, improve image rejection to perhaps 60db (1 part in 1000).
  • This disclosure first teaches how to make a Weaver architecture radio where no filter is required between the first and second set of mixers. Hence no phase error is incurred and image rejection is substantially higher than can be achieved by any analog implementation of the conventional Weaver architecture.
  • this disclosure documents a second innovation: the use of a time-division multiplexed second set of mixers, again aimed at the improvement of image rejection.
  • the filter sees a high speed signal as it must handle the two signals in sequence. Therefore, the signal seen by the filter is not the actual first mixer output - it is a chopped version of it. This cannot be filtered without maintaining separate state-variables for each part of the time division multiplexer.
  • no filter is necessary - and in this case time-division multiplexing in the analog domain can be used. This enables the design of a single circuit representing both of the second mixers, wherein no phase error is introduced and image rejection is enhanced.
  • the invention relies upon the use of a sampling mixer ("DSM") of the general type disclosed in Mallinson, US Patent No. 7,028,070 as the first set of mixers.
  • the sampling mixer in question is an analog circuit capable of the act of mixing and the generation of accurate quadrature signals.
  • the disclosure will show how a certain configuration of such a DSM can eliminate the need for a filter and yet be able to implement the Weaver trigonometric identities and remove the image.
  • the invention comprises a sampling mixer having a parallel array of processing elements, each of the elements comprising a sampling circuit for sampling an input signal.
  • the array is configured such that each processing element successively processes respective samples of the input signal in round- robin fashion.
  • Filter means are provided in the array.
  • Each multiplier of a set of multiplier circuits multiplies an output of respective ones of the processing elements, the set providing a summed output.
  • the filter means comprises a filter interposed between each of the sampling circuits and respective ones of the multiplier circuits.
  • the filter means includes a resistance shared by the inputs of each of the sampling circuits.
  • the invention comprises a processor for a radio frequency signal.
  • Each member of a first pair of quadrature mixers operates on the same samples of the signal, with a second pair of quadrature mixers operating on the outputs of respective ones of the first mixers.
  • the first pair of mixers comprises a parallel array of sampling circuits configured to operate in round-robin fashion to sample the signal.
  • a first and a second multipliers operate on the outputs of each of said sampling circuits and a plurality of such first multipliers have scaling factors selected to emulate the in-phase output of a local oscillator and a plurality of the second multipliers have scaling factors selected to emulate the quadrature output of a local oscillator.
  • the first pair of quadrature mixers comprises filter means and in a more specific aspect, the filter means comprises a series of filters between each of the sampling circuits and a common input of respective first and second multipliers. In yet a more specific aspect, the filter means includes a resistance shared by the inputs of each of the sampling circuits.
  • the foregoing processor scaling factors of one of the multipliers emulates a sinusoidal waveform and the scaling factors of the other of the multipliers emulates the same sinusoidal waveform out of phase by 90 degrees.
  • the invention comprises a circuit implementing an analog quadrature-matched pair of mixers.
  • Each element of a parallel array of processing elements comprises a sampling circuit for sampling an input signal.
  • the array is configured such that each processing element successively processes respective samples of the input signal in round-robin fashion.
  • the successive multiplier circuits of a first one of the sets has scaling factors that correspond to the coefficients in a Fourier series approximation of a sinusoidal waveform and the successive multiplier circuits of the second one of said sets has scaling factors that correspond to the coefficients in a Fourier series approximation of said sinusoidal waveform out of phase by 90 degrees.
  • the array comprises filter means, and in a more specific aspect, a filter is interposed between each of the sampling circuits and respective ones of the multiplier circuits.
  • the filter means includes a resistance shared by the inputs of each of the sampling circuits.
  • the invention further contemplates the use of the foregoing circuit in implementing the first pair of mixers in a Weaver architecture filter.
  • the invention comprises a method of filtering a radio frequency signal to reject an image frequency.
  • the method comprises procuring a successive series of samples of the signal, filtering each of the samples to produce filtered samples and successively multiplying each of the filtered samples by, on the one hand, a first plurality of respective scaling factors corresponding to the coefficients in a Fourier series approximation of a sinusoidal waveform, and on the other hand a second plurality of scaling factors corresponding to said coefficients representing a quadrature of said sinusoidal waveform.
  • the outputs of the multiplications are summed.
  • the invention comprises a circuit for processing a radio frequency input signal to achieve image rejection.
  • Switching means are provided for selecting between four sets of multiplier scaling factors.
  • the sets correspond to the coefficients of a Fourier series approximation of a sinusoidal signal, the inverse of said sinusoidal signal, a cosine of said sinusoidal signal and the inverse of said cosine respectively.
  • Each of the sets of scaling factors are applied to samples of the input signal to produce multiplied outputs and a summing circuit sequentially sums the outputs of each of the sets.
  • the invention comprises a circuit for processing a radio frequency input signal to achieve image rejection.
  • the circuit comprises a parallel array of processing elements with sampling circuits for securing a plurality of samples of the signal. Means are provided to sequentially process each of the processing elements and for filtering the outputs of the sampling circuits.
  • Each of the processing elements comprises an input switch, a capacitance associated with the input switch and two resistors each of which is selectively switchable to a first virtual ground amplifier, a second virtual ground amplifier or ground, the output of the processing element being defined as the difference between the outputs of the first and second virtual amplifiers.
  • a circuit sums the outputs of the processing elements.
  • the means for filtering of the preceding circuit includes a resistance shared by the inputs of the processing elements.
  • Fig. 1 is a circuit diagram of a prior art Hartley image rejection receiver
  • Fig. 2 is a circuit diagram of a prior art Weaver image rejection receiver
  • Fig. 3 is a circuit diagram of a prior art round-robin sampling mixer
  • Fig. 4 is an alternative diagrammatic representation of the circuit of Fig. 3;
  • Fig. 5 is a schematic diagram of an approximation of the Weaver architecture filter using a pair of round-robin sampling mixers
  • Fig. 6 is a circuit diagram of a round-robin sampling mixer including a set of filters between the sample and hold devices and the multipliers, according to an embodiment of the invention
  • Fig. 7 is a diagrammatic representation of an equivalent configuration of the circuit of Fig. 6;
  • Fig. 8 is a circuit diagram of a round-robin sampling mixer with filters interposed between the sample and hold devices and the multipliers according to an embodiment of the invention
  • Fig. 9 is a diagrammatic representation of how one might seek to implement time division multiplexing in a Weaver filter
  • Fig. 10 is a diagrammatic representation of a time quantized version of the second set of multipliers of Fig. 9 according to an embodiment of the invention.
  • Fig. 11 is a diagrammatic representation of the approach of Fig. 10, further resolved to apply a single quantization process to both of the second multipliers;
  • Fig. 2 is a diagrammatic representation of the approach of Fig. 11 , further resolved for the case where the coefficients of the round-robin sampling mixer are adjustable;
  • Fig. 13 is a circuit diagram of a filter implemented as a series resistance in the sampling capacitor according to the preferred embodiment of the invention.
  • Fig. 14 is a circuit diagram showing an exemplary notation for iterations of sample and hold devices according to the preferred embodiment
  • Fig. 15 is a circuit diagram of an alternative to the circuit of Fig. 14 implemented according to the preferred embodiment
  • Fig. 16 is a circuit diagram of a summing element to sum to a virtual ground according to the preferred embodiment
  • Fig. 17 is a circuit diagram of the round-robin sampling mixer according to an embodiment of the invention for one set of coefficients;
  • Fig. 18 is a modification of the circuit of Fig. 17 according to an embodiment of the invention;
  • Fig. 19 is a modification of the circuit of Fig. 18 according to an embodiment of the invention to select between two coefficients;
  • Fig. 20 is a circuit diagram showing a prototype design showing switches for selecting resistor values in repeated cells of the circuit of Fig. 19;
  • Fig. 21 is a circuit diagram showing a modification of the circuit of Fig. 20 according to the preferred embodiment to enable the reversal of coefficient values;
  • Fig. 22 is a table of values of coefficients and states of switches at the 4 quantized sample values of the preferred embodiment.
  • DSM sampling mixer
  • ⁇ ⁇
  • 'y' is the output
  • 'x' is a set of input parameters
  • 'w' a set of weighting factors.
  • the set of input parameters 'x' are the 'round-robin' samples of an input signal.
  • Fig. 3 which is Fig. 1 of US Patent No 7,028,070
  • a parallel array of sample and hold circuits 1 1 are activated in round-robin fashion (in this example by the delay line 14) to successively process samples of the input signal and present their outputs to respective multipliers 12 each having a scaling factor Cn.
  • a summing network 13 then adds all such products of coefficients (Cn) and samples to a common output (Aout).
  • the absence of rotation of the coefficients causes the circuit to operate as a correlator, multiplying the input signal by a virtual wave of successive coefficient values (corresponding to the coefficients of a Fourier series approximation of a sinusoidal waveform) and performing a running average over the results.
  • That running average may alternately be described as a "box car” filter or rectangular filter and results in the well known sin(x)/x response.
  • the DSM system can be represented as shown in Fig. 4.
  • Fig. 4 already resembles the first part of the Weaver circuit, and we can draw two DSM circuits and their associated filters used in the Weaver architecture as shown in Fig. 5.
  • the sin(x)/x filter is in the correct place and if it could function as the required low pass filter it would match each other very well, better than two separate analog filters since these sin(x)/x filters arise from the time taken to complete one round-robin sample and so are related to the clock.
  • the sin(x)/x response is not sufficient to adequately remove the upper product (i.e. the 'free' + 'flo' signal) because apart from the zeros of the transmission it is essentially a first order filter.
  • the first innovation addresses this problem of the poor performance of the sin(x)/x filter and its inability to perform the required Weaver filter action as shown in Fig. 6.
  • This filter is nominally identical in all channels, but need not be precisely so.
  • the effect of placing this filter in this position is to compound the existing sin(x)/x filter with this filter since multiplication and summation are linear operations, the filter is indistinguishable from a filter placed in series with the sin(x)/x filter as illustrated in Fig. 7. Therefore, by placing as many copies of a filter as there are coefficients after the sample and hold, we have created an apparently steeper (more effective) filter in series with the sin(x)/x filter.
  • Fig. 8 shows how two sets of coefficients, one encoding a sine wave and one encoding a 90 degree shift sine wave (i.e. a cosine) to emulate the quadrature outputs of a local oscillator can share the same set of sample and hold and filters.
  • Fig. 9 shows an obvious way that we may attempt this: the filter first gets used in the upper path, then in the lower path and so forth. Since it is the same filter it clearly matches. The problem is that this switching action requires the filter to handle first the upper path and the internal state variables begin to respond to that signal, then the lower path and the internal state variables begin to respond to that signal. Consequently the filter never settles and cannot perform the filtering action.
  • the one-of-four selector switch applies either the upper path directly or inverted, or the lower path directly or inverted. It is noted that in Fig. 11 , the filter has been moved to the input as discussed in the innovation earlier in this disclosure. However, it is important to realize that this drawing is over simplified in its use of that filter prior to the multiplication: this works when the invention as described earlier is used.
  • the output IF frequency is indistinguishable from that which would originate in a Weaver architecture radio with very well matched sine and cosine channels. No image is present in the output and the signal between the input and output multipliers of the Weaver do not exist in this implementation.
  • the Virtual Weaver circuit depends upon the presence of a filter in each of the DSM sampling sections.
  • the simplest form of such a filter can be created by a series resistance in the sampling capacitor as shown in Fig. 13. In this case when the switch closes to sample the input the bandwidth is limited ⁇ o ⁇ /2nRC .
  • V(out, outb) V(in) - Rf J 1_
  • this circuit implements the weighted sum of the DSM samples with either positive or negative coefficient values depending upon how R1 and R2 are defined within each repeated cell.
  • the equations to define R1 and R2 are: where Xi are the coefficient values and R1 i and R2i are the resistors in the cell connected as in the above prototype. In this equation and in all that follow, a practical limitation is apparent: as Xi more closely approaches zero but does not quite reach zero, the required R increases in value. High values can be impractical and consequently a reasonable maximum value of, for example, 2MOhms could be applied.
  • Fig. 17 The implementation shown in Fig. 17 then is the almost complete circuit of the DSM with one set of coefficients.
  • the values of R1 and R2 differ in the repeated block: if the desired coefficients are the Xi then the equations above show how to set the internal R1 and R2 values for each instance indexed by T.
  • This circuit despite its simplicity, is quite viable if we make one more addition. Note that as drawn above the load on the sampling capacitor will vary as the coefficient varies. If we can make that load constant, then we can make this simple all-passive circuit work. We will arrange that the input resistor that limits the bandwidth is for example, 1 k. And we will arrange that the load on the capacitor is always, for example, 10k. In this case we suffer approximately 10% signal attenuation but otherwise the circuit still operates.
  • R3 is constructed such that the load is always RT on the capacitor. Given these definitions this circuit is the complete band limited DSM with a single set of coefficients.
  • this circuit operates because the values of R1 , R2 and R3 in each of the 32 instances of 'SHA' are different; they differ as described by the equations on the top of the page. And furthermore, there are only ever actually two, not three, resistors in any given instance - one of the resistors R1 or R2 is always infinite, and hence non-existent in each cell. A single variable value resistor connected in one of two possible configurations is sufficient to make either R1 or R2 as required, as illustrated in Fig. 19.
  • RT R ⁇ R2; Note that the sign has been removed from the choice of resistors and is instead programmed into the switches. Specifically, suppose R1 in any given instance is derived from a positive value of X, then the switch S1 is connected in the lower position to drive A2 and 'Outb'. Conversely, if R1 is derived from a negative value S1 connects to the upper position and drives A1 and Out'.
  • the DSM is programmed for five cycles over the 32 elements, and that the initial phase has been set to a non-zero value to avoid the appearance of zeros in the coefficient values.
  • the table is interpreted as follows: for example, in instance 9 the value of R1 is 30.1 k, the value of R2 is 16.02k and the value of R3 is 229.405k.
  • the switch S1 in element 9 is connected to ground.
  • P1 it is connected to the lower position (to A2) to make a positive output.
  • P2 it is again connected to ground.
  • P3 it is connected to the upper position (to A1 ) to make a negative output.
  • the switch S2 is connected to the upper position in P0, to ground during P1 and P3, and to the lower position during P2.
  • This DC signal in the conventional implementation, would, due to the presence of active devices in the signal path, contain significant 1/F noise. It does not contain noise in this example. If, again as in the example, the virtual Weaver is clocked at 176 Mhz, then the output of 44 Mhz emerges for an input of 156.25 Mhz.
  • the equivalent intermediate signal between the first and second quadrature multipliers of the Weaver - which is avoided by this invention - is at zero frequency.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Superheterodyne Receivers (AREA)

Abstract

L'invention concerne un filtre à architecture de tisseur virtuel qui est réalisé en utilisant un mélangeur d'échantillonnage qui traite successivement les échantillons du signal d'entrée selon un modèle de répétition alternée et produit une somme des échantillons multipliés par des coefficients émulant des signaux sinusoïdaux en quadrature. Un oscillateur local, virtuel plutôt que réel, est réalisé de manière fiable sans désadaptation. Le filtrage entre les mélangeurs à tisseur est supprimé en faveur du filtrage au niveau de l'entrée d'échantillonnage et un multiplexage par répartition dans le temps efficace est obtenu en sélectionnant entre les combinaisons de résistance qui réalisent des coefficients de cadrage différents, ce qui résulte en une réalisation analogique efficace d'une architecture de tisseur virtuel.
PCT/CA2009/001817 2009-12-11 2009-12-11 Filtre à architecture de tisseur virtuel WO2011069229A1 (fr)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN200980163431.7A CN102783015B (zh) 2009-12-11 2009-12-11 虚拟韦弗结构滤波器
PCT/CA2009/001817 WO2011069229A1 (fr) 2009-12-11 2009-12-11 Filtre à architecture de tisseur virtuel
US13/515,259 US8774749B2 (en) 2009-12-11 2009-12-11 Virtual weaver architecture filter

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CA2009/001817 WO2011069229A1 (fr) 2009-12-11 2009-12-11 Filtre à architecture de tisseur virtuel

Publications (1)

Publication Number Publication Date
WO2011069229A1 true WO2011069229A1 (fr) 2011-06-16

Family

ID=44145054

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CA2009/001817 WO2011069229A1 (fr) 2009-12-11 2009-12-11 Filtre à architecture de tisseur virtuel

Country Status (3)

Country Link
US (1) US8774749B2 (fr)
CN (1) CN102783015B (fr)
WO (1) WO2011069229A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013067465A1 (fr) 2011-11-04 2013-05-10 Ess Technology, Inc. Conversion descendante de multiples canaux rf

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10680794B2 (en) * 2018-07-22 2020-06-09 SiliconIntervention Inc. Sample rate conversion by Gaussian blur
US11171682B2 (en) * 2019-01-30 2021-11-09 Swiftlink Technologies Inc. Dual polarization millimeter-wave frontend integrated circuit

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2192039A1 (fr) * 1995-12-15 1997-06-16 Herve Guegnaud Dispositif melangeur a rejet de la frequence image
US20050233725A1 (en) * 2004-04-20 2005-10-20 Khurram Muhammad Image reject filtering in a direct sampling mixer
US7028070B2 (en) * 2001-01-29 2006-04-11 Ess Technology, Inc. High speed filter
WO2008139390A1 (fr) * 2007-05-11 2008-11-20 Nxp B.V. Circuit mélangeur à deux modes et procédé

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6061551A (en) * 1998-10-21 2000-05-09 Parkervision, Inc. Method and system for down-converting electromagnetic signals
US7248625B2 (en) * 2002-09-05 2007-07-24 Silicon Storage Technology, Inc. Compensation of I-Q imbalance in digital transceivers
US7660374B2 (en) * 2004-05-21 2010-02-09 Honeywell International Inc. Method and apparatus for excision of narrowband interference signals in navigation or communication bands
WO2008026178A2 (fr) * 2006-08-31 2008-03-06 Nxp B.V. Estimation d'un déséquilibre i/q dépendant de la fréquence
US20090131006A1 (en) * 2007-11-20 2009-05-21 Mediatek Inc. Apparatus, integrated circuit, and method of compensating iq phase mismatch

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2192039A1 (fr) * 1995-12-15 1997-06-16 Herve Guegnaud Dispositif melangeur a rejet de la frequence image
US7028070B2 (en) * 2001-01-29 2006-04-11 Ess Technology, Inc. High speed filter
US20050233725A1 (en) * 2004-04-20 2005-10-20 Khurram Muhammad Image reject filtering in a direct sampling mixer
WO2008139390A1 (fr) * 2007-05-11 2008-11-20 Nxp B.V. Circuit mélangeur à deux modes et procédé

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013067465A1 (fr) 2011-11-04 2013-05-10 Ess Technology, Inc. Conversion descendante de multiples canaux rf
US8693972B2 (en) 2011-11-04 2014-04-08 Ess Technology, Inc. Down-conversion of multiple RF channels
CN103999369A (zh) * 2011-11-04 2014-08-20 Ess技术有限公司 多个rf通道的下转换
EP2774274A4 (fr) * 2011-11-04 2015-07-22 Ess Technology Inc Conversion descendante de multiples canaux rf

Also Published As

Publication number Publication date
CN102783015A (zh) 2012-11-14
CN102783015B (zh) 2015-01-14
US8774749B2 (en) 2014-07-08
US20120300888A1 (en) 2012-11-29

Similar Documents

Publication Publication Date Title
JP5269614B2 (ja) Ｖｈｆ／ｕｈｆ／ｇｓｍ／ｇｐｓ／ｂｌｕｅｔｏｏｔｈ／コードレス電話におけるトランシーバの開発
KR0130471B1 (ko) Ssb 신호 발생기
US6049573A (en) Efficient polyphase quadrature digital tuner
US4893316A (en) Digital radio frequency receiver
CA1089028A (fr) Filtre et dispositif de demodulation
AU591181B2 (en) Digital radio frequency receiver
CA2167702A1 (fr) Convertisseur de frequence dans un systeme de telecommunications
US7514993B2 (en) IQ demodulator
US8774749B2 (en) Virtual weaver architecture filter
US4015222A (en) Modulated passband signal generator
EP1693954B1 (fr) Circuit de démodulation pour un récepteur employant un schéma d'échantillonnage FI
EP0695028B1 (fr) Dispositif additionneur de signaux à échelle réduite et dispositif détecteur différentiel
WO2001071931A2 (fr) Syntoniseur numerique a frequence d'horloge optimisee, filtre cic parallele integre et oscillateur local
GB2513549A (en) Transmitter and receiver circuits
EP1089512A1 (fr) Dispositif de télécommunication à module analogique de transformation de Fourier
De Boer et al. Simplified harmonic rejection mixer analysis and design based on a filtered periodic impulse model
Agarwal et al. A fractional sample rate conversion filter for a software radio receiver on FPGA
Usman et al. Onthe complementary relationship between sampling and double sideband-suppressed carrier modulation
US4001510A (en) Digital modulator and demodulator system for fdm telephony
TWI411298B (zh) Frequency conversion device and conversion method and filter thereof
JP3230786B2 (ja) ディジタル化直交位相変調回路
Wang Implementation of a Filter and Multiplier for Next Generation Transceivers
Sun et al. Implementation aspects of generalized bandpass sampling
CS253508B1 (cs) Zapojení pro kvadraturní čtyřfázovou demodulaci signálu

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 200980163431.7

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09851950

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 13515259

Country of ref document: US

122 Ep: pct application non-entry in european phase

Ref document number: 09851950

Country of ref document: EP

Kind code of ref document: A1