Patent Grant
10382132
The present invention is directed towards radio-frequency (RF) transceivers, and more specifically towards transceivers that use direct digitization of RF signals, and digital processing thereof.
A simple RF signal consists of a narrowband baseband signal which is modulated onto a high-frequency RF carrier. A more complex, broader bandwidth RF signal may consist of multiple narrowband signals modulated on similar RF carriers. A conventional RF digital data receiver uses an analog mixer and local oscillator (LO) to translate the signal from the carrier to a lower frequency, the baseband or intermediate frequency (IF), where it is then typically digitized and further signal extraction is performed in the digital domain. Similarly a conventional RF digital data transmitter works in the reverse direction, converting a digital baseband or IF signal to analog, followed by mixing it with the LO signal to upconvert the signal. It is then amplified for transmission through the antenna.
In many wireless communications systems, the antennas are mounted on a tall tower, and the digital processing is carried out in basestations located remote from the antenna, e.g., on the ground. It is not practical or efficient to carry GHz-range (microwave) RF signals long distances over coaxial lines, since signal attenuation is too high. Therefore, the conventional partitioning of the system places amplifiers and analog mixers on the tower with the antenna, as shown in
Optic fiber systems capable of communicating microwave signals are known. These are typically applied in military radar applications.
A critical requirement in wireless communication systems is maintaining linearity on both transmission and reception. That is, the analog signal should not be subject to distortions that result in the alteration of signal composition within the band of interest. Spurious intermodulations from signal components within the band of interest can limit the useful dynamic range (called the spurious-free dynamic range or SFDR) of a receiver or transmitter system. Thus, in a narrow band system, non-linearities which result in spurious energy in frequencies outside of the band of interest are generally tolerable, since these are readily filtered or ignored. On the other hand, as the bandwidth of the signal of interest grows wider, the possibility of significant spurious signals in the band of interest due to non-linearities in the signal processing and transmission chain grows. These are not readily filtered or ignored, and therefore present limitations of the system.
In a radio system, non-linearities in the signal processing chain generally produce spurs or intermodulation products between components of the signal. The wider than the band of the signal processed, the greater the probability for intermodulation components (spurs) which lie within the band itself, and therefore cannot be eliminated with a band-pass filter. One strategy to avoid misinterpretation of such spurs is to employ a deconvolution process in the receiver to predict the effect of the non-linearity on the signal, and reverse its effects. However, this requires a receiver for the full bandwidth of the transmitter signal processing chain, which may be untenable. In general, it is preferred to avoid non-linearities, and where present, limit their effects as much as possible, so that the intermodulation products may be treated as noise.
A number of metrics specifying system performance are available. For example, the “Spur-free Dynamic Range” or SFDR, is a commonly used metric which presumes a relatively small number of high power signal components which produce intermodulation distortions, and therefore the parameter specifies the dynamic range of the system before spurs interfere with signal interpretation.
Spurious Free Dynamic Range is the usable dynamic range of a system before spurious noise interferes or distorts the fundamental signal. SFDR is the measure of the difference in amplitude between the fundamental signals and the largest harmonically or non-harmonically related spur within the band of operation. A spur is any frequency bin on a spectrum analyzer, or from a Fourier transform. Spur-free dynamic range (SFDR), as generally used, attempts to define receiver dynamic range in terms of two undesired interferers and the receiver noise floor. The spur-free dynamic range is the difference in dB between the receiver noise floor and the level of each of equal-amplitude signals that produce an in-band spurious product equal in power to the noise floor. Generally, the receiver third-order intercept point is used to predict the spurious product, but often the second-order distortion dominates.
The SFDR specification overlooks several important factors which influence dynamic range. First, it attempts to model interference by using just two (or perhaps up to 4) interfering signals. This overcomes some of the objections to single-tone testing, but the real signal environment is usually populated by a multitude of signals. Second, it does not reveal the effects of reciprocal mixing or compression like the desensitization dynamic-range test. Third, it does not effectively test the effects of receiver input filtering (preselection). Finally, SFDR, as it is ordinarily specified, considers only the third-order distortion. In fact, for many receivers, especially those with modest input filters, the second-order products may dominate.
Another metric employed is the “Noise Power Ratio” or NPR, which models the signal as white noise within a band, and then measures the noise floor at narrow ranges within the band resulting principally from intermodulation distortion.
These metrics are therefore extremes of a continuum which seeks to characterize a system to determine the impact of intermodulation distortion on in-band signals; the SFDR measures a specific effect of a small number, e.g., 2-4 signals, while the NPR measures the statistical effect of essentially an infinite number of signals which appear as white noise.
In a wideband system, the true SFDR measurement becomes difficult, since it may be difficult to determine a worst case effect without testing each different combination of signals, and the presumptions typically made for narrow band systems regarding the dominance of the third order intermodulation product. On the other hand, white noise itself is a statistical process, and the time of measurement is a relevant factor. For example, over short periods, intermodulation components from various spectral components within the white noise may cancel or reinforce each other, leading to a misleading measurement. Since each of these metric is presented as a simple ratio (often specified in decibels), care must be exercised in interpretation.
As used herein, the SFDR is intended to encompass the ratio of one or more respective signal components of equal amplitude within a band and their largest in-band intermodulation product, of any order, but excluding artifacts, such as a superposition of multiple intermodulation products. In cases where the SFDR is specified for a composite system with band-limiting filters, the input signals may take any value within the permissible input range, while the measurements are made after band-limiting.
The performance of high power amplifiers with many carriers (>10) is normally tested using a noise power ratio (NPR) measurement technique. In this test, white noise is used to simulate the presence of many carriers of random amplitude and phase. In a traditional test setup, the white noise is first passed through a bandpass filter (BPF) to produce an approximately square spectral pedestal of noise of about the same bandwidth as the signals being simulated. This signal is then passed through a narrow band-reject filter to produce a deep notch (typ. >50 dB) at the center of the noise pedestal. This noise signal is used to excite the test amplifier, which produces intermodulation distortion products that tend to fill in the notch. The depth of the notch at the output of the amplifier can be observed with a spectrum analyzer, and is the measure of the NPR.
NPR can be considered a measure of multi-carrier intermodulation ratio (C/I). NPR differs from multi-carrier C/I in that it is the ratio of carrier plus intermodulation to intermodulation (C+I/I). At higher ratios (C/I>20 dB), the two measures will approach the same value. The bandwidth of the noise source should be much wider the bandwidth of the BPF to insure the statistical distribution of the noise power resembles a random phase multicarrier source. The width of the noise pedestal is usually made equal to bandwidth of the channel under test. The width of the notch should be about 1 percent or less of the width of the noise pedestal.
As used herein, the NPR is intended to encompass the dynamic range of a white noise signal representing the full band range of a permissible input signal, with uniform amplitude, with an output signal from which the input signal is subtracted. Traditionally, such a measurement is obtained by providing a notch filter at the input for a narrow frequency, and then measuring the output within that notch band. On the other hand, digital processing techniques permit a more rigorous characterization. For example, a notch filter is not necessary; the output across the band can be compared with the input, and deviations characterized. In that case, the deviation can be expressed as a ratio to the input signal, and for example, a worst case deviation reported as the noise power ratio. In some cases, especially across a large band, the input amplitude power is not uniform. Since signal from one communication stream may interfere with signal from another, which has a different power spectrum, the NPR may be functionally specified based on real signal types, which may differ from equal amplitude noise across the band.
Thus, for example, in a wideband system which handles multiple smaller bands concurrently, a theoretical NPR may not yield fully functionally valuable information as compared to signal models. Typical signal types of interest are cellular communications, which include CDMA (and variants), GSM, LTE, WiMax, etc. Therefore, in addition to characterizing the performance with respect to a white noise across the wide band, it may be useful to determine a bit error rate (BER) of a specified model signal at a specified attenuation, or alternately, the limiting attenuation until a specified BER is achieved, in the presence of other types of signals in the remainder of the band. Since this type of characterization is communication protocol type-specific, it would not be generally useful for generic systems. Note that in some cases, the relative amplitude of signals within the broadband signal may be adjusted to optimize net performance; for example, if one signal interferes with that in an adjacent band, and both are not required to operate at maximum power, then one may be attenuated with respect to the other to achieve acceptable performance for both.
A primary source of nonlinearity in conventional transmission systems is the power amplifier (PA). In general, highly linear PAs (including semiconductor transistor PAs) are energy-inefficient, large and heavy. PAs that are compact and efficient tend to be highly nonlinear. A modern wireless communication system that requires many antennas and PAs on a given transmission tower may use, or even require the use of, relatively nonlinear PAs.
This nonlinearity limits the performance of the transceiver system, and must be corrected. Several linearization techniques are known in the prior art. Most of these techniques operate on the baseband signal before upconversion (see
It is well known in the prior art that there are important potential advantages to combining multiple RF signals in nearby bands to create single broadband signal. For example, this could substantially reduce the number of RF components required. However, the problems associated with nonlinear PAs and the inability to sufficiently correct their distortions severely limit the bandwidth of such a combined signal to less than about 100 MHz. Conventional technology does not have a solution to this problem, which makes broader band RF systems impractical.
Recently, a superconducting electronic technology (known as rapid-single-flux-quantum logic or RSFQ) has been demonstrated that can provide direct digitization of RF signals, as well as ultrafast digital processing at clock rates up to 40 GHz, with rates up to 100 GHz expected in the near future. This permits representation of RF signals in a directly digitized format, at broadcast frequencies, referred to as “Digital RF”™ (Hypres Inc.), wherein the sample rate is much larger than the carrier frequency, e.g., above the Nyquist rate. This provides a promising approach to very broadband multi-carrier RF communication signals (See, U.S. Pat. Nos. 7,280,623, 7,365,663, 7,362,125, 7,443,719, expressly incorporated herein by reference). However, RSFQ is an ultra-low-power technology (sub-mV level signals) that requires cooling to temperatures of 4 K (−269 C). Existing refrigeration technology makes it impractical to place these cryocooler systems on cellular towers, so they must be placed in ground stations. In that case, transmission losses of the RF signals between the ground and the tower largely eliminate the potential performance advantages. On the other hand, if the existing cryocoolers were sought to be placed on the tower, this would increase wind loading and power consumption on the tower, and potentially lead to increased maintenance costs due to potential servicing of the cryocooler in an inaccessible location. Since the tower electronics are standardized for various applications, even in cases where the tower components are accessible or increased maintenance and service permissible, the lack of economies of scale still make such installations infeasible.
Optical fibers would seem to be a promising alternative, since attenuation losses on fibers are quite small. However, the electro-optic (E/O) and opto-electronic (O/E) converters for prior-art multi-GHz analog signals had severe limitations in performance, again greatly reducing the performance advantages of the superconducting system. These converter limitations reflect both limits in bandwidth and limits in linearity. It is difficult to simultaneously achieve high bandwidth and high linearity, particularly in a reliable and inexpensive manufactured product. See, for example, U.S. Pat. No. 7,424,228, expressly incorporated herein by reference. There are several known methods for reducing nonlinearity (see, for example, U.S. Pat. No. 7,426,350, expressly incorporated herein by reference), but this remains a difficult problem.
The distinctions between digital and analog E/O (and O/E) modulators should be appreciated, which are directly analogous to the distinctions between digital and analog amplifiers. Digital amplifiers and modulators are nonlinear one-bit threshold devices, where only bit-errors are important. In contrast, analog amplifiers and modulators are generally required to be linear over a wide dynamic range, while nonlinearities generate intermodulations that reduce the useful spur-free dynamic range (SFDR). Digital optical modulators are long established for high-speed digital optical communications. In contrast, the technology of linear optical systems at microwave frequencies (microwave photonics or radio-over-fiber, or “RoF”) has been developing only over the past decade, with further improvements in performance continuing. For E/O converters, direct modulation of diode lasers is typically used for RF frequencies below a few GHz, while electro-optic interferometric modulation is used for higher frequencies. Photodiodes are used for O/E converters. The nonlinearity associated with such an optical link increases sharply with increasing bandwidth. A typical optical link specification may be expressed by SFDR=100 dB/B2/3, corresponding to 3rd-order intermodulation, where B is the bandwidth in dB-Hz. For a 10 GHz signal, corresponding to 100 dB-Hz, the link SFDR would be reduced to 33 dB. This would likely be quite sufficient for a digital signal, but would be too low for a high-dynamic range analog signal.
The low-power aspect of this superconducting technology requires the use of high-gain, high-bandwidth semiconductor amplifiers, which are near the limits of semiconductor technology, and commercial product is generally unavailable. Various non-ideal aspects of these amplifiers, especially nonlinear distortion, tend to degrade the performance of the overall system. Very recently, the same superconducting RSFQ technology has demonstrated the capability to provide very broadband digital predistortion directly on the digitized RF signal, for a digital-RF transmission system with a nonlinear PA. See, e.g., U.S. Pat. No. 7,313,199, expressly incorporated herein by reference. This can achieve much greater improvement in broadband system SFDR than would be possible using conventional baseband predistorters, and might be used to make these broadband systems practical. However, the link between the ground and tower units remains a problem.
One aspect of the present invention addresses these shortcomings and seeks to provide a hybrid system that combines these three technologies of superconducting RSFQ circuits, microwave-frequency optical links, and broadband semiconductor power amplifiers, in a way that builds on superior features of each and corrects the non-idealities that limit practical performance.
A new system architecture is presented for an RF transceiver in a wireless communication base station, wherein ultrafast multi-GHz digital data conversion and processing are carried out in a central ground station, and analog-RF amplification is carried out in tower-mounted units close to antennas. No analog mixing is necessary in the tower, since the signal communicated to the tower is a direct representation of the radio frequency signal to be transmitted. The ground station is linked to the tower units by means of broadband RF signals carried on optical fibers in either analog or digital format, in contrast to prior-art systems that carry only baseband (or IF) signals. High-speed digital processing of the full digital-RF signal is used to correct for nonlinearities in amplifiers and electro-optic converters in both transmitter and receiver, as shown in
The technology provides a new architecture for a radio system, leveraging the ultrafast digital-to-analog converter, analog-to-digital converter, and digital signal processing technologies presented herein.
In the case of low temperature superconductors, a cryocooler is required, which consumes energy to achieve cryogenic temperatures, i.e., <10K, and has a physical size. In addition, the cryocooler presents a possible service and maintenance issue, and a preference may therefore exist to have such a cryocooler located off an antenna, and, for example, in a base station located on the ground, or rooftop, or otherwise some distance from the antenna structure. On the other hand, the transmit channel of the radio system typically has at least a power amplifier located in close proximity to the antenna, and the receive channel has at least a receive amplifier, typically a low noise amplifier, located in close proximity to the antenna. Therefore, aspects of the present technology address efficient and effective communication of signals from the antenna electronics system to a remote location.
A particular issue addressed in accordance with the present technologies is an avoidance of frequency translation between an analog radio frequency interface and digital electronics, which leads to distortion and degradation of wideband radio frequency signals. For example, a wideband radio frequency signal may be greater than 19.5 MHz (20 MHz band spacing less guard-band), for example, 40 MHz (two adjacent 20 MHz bands) or 60 MHz (3 adjacent bands). Indeed, using high speed digital electronics, it is possible to process, at native frequency, signals having bandwidths of in excess of 100 MHz. A particular advantage is possible where the antenna system communicates on multiple discontiguous bands, for example, 3 bands within a 120 MHz range, since a system implemented in accordance herewith would be agnostic to the band selection, and therefore be more versatile and adaptive.
For high fidelity, large bandwidth radio frequency signals, the communication medium of choice is an optic fiber, though coaxial or other impedance controlled electrical cable may be employed. However, each of these suffers from distortion and attenuation of the signals, and requires amplification. It is noted that the issues for transmit and receive channels are not symmetric, and thus each issue is preferably considered separately. In particular, according to one aspect of the invention, the substantial high speed digital electronics are located distant from the antenna electronics system. Therefore, the receive channel may communicate an analog signal to the remote location. On the other hand, the transmit channel may communicate an analog or digital signal to the antenna electronics system, and advantages accrue if the communication is digital, since attenuation and distortion in the communications medium may be completely corrected; however, this scheme then requires a digital to analog converter operating at the data rate located within the antenna electronics package.
Three options are generally available for a transmit channel: a first option provides an analog signal communicated from the remote electronics station to the antenna electronics system, for power amplification and transmission. In this case, the remote electronics can predistort the signal to compensate for various types of distortion in the downstream electronics chain, though there is limited capability to compensate for intermodulation distortion. Thus, on this embodiment, the various elements of the communication link and transmission-side antenna electronics system should have high linearity.
A second option provides a baseband or intermediate frequency communication link from the remote electronics station, with frequency translation at the antenna electronics system. This option corresponds to a more typical cellular radio system. While employing lower frequency communications between the remote electronics system and antenna electronics system permits use of traditional coaxial cable, such systems have limited bandwidth.
A third option provides a digital communication between the remote electronics and the antenna electronics system. In this embodiment, we presume that the high speed digital electronics processing capabilities are not available in the antenna electronics system (though in some cases, they are, avoiding the need for significant communications between the modules), and therefore a fully processed, digital stream which is oversampled with respect to the radio frequency signal to be emitted from the antenna, and thus the antenna electronics system need only convert the digital stream into a corresponding signal, either before, during or after amplification. For example, the digital stream may be communicated through an optic fiber link, and converted from a digitally modulated optic signal to digital pulses. The digital pulses, if oversampled, may then be low pass filtered to produce an analog radio frequency waveform, which may then be amplified by an analog power amplifier.
According to another scheme, the digital stream may be communicated through an optic fiber link, and converted from a digitally modulated optic signal to digital pulses. The digital pulses are then used to drive a power digital-to-analog converter, which produces a high power, e.g., 10 Watt, analog radio frequency signal from the digital stream. In this case, a multibit digital signal may be provided, thus reducing the degree of oversampling required for high fidelity, and potentially facilitating the high power digital to analog converter operation. Thus, in contrast to an ideal digital to analog converter, the slew rate of the converter may be limited, thus permitting the higher order bit drivers, which correspond to higher power, to switch slower than the lower order bits, which correspond to lower power. Indeed, since the characteristics of the power digital to analog converter may be known by the remote electronics system, certain non-idealities may be precompensated. Based on the availability of precompensation, a matched power digital to analog converter may be provided such that the available compensation better matches the distortion. Thus, for example, a 4 bit dynamic range converter may have 5 or more digital input signals. In this way, cost and/or power consumption or other operating characteristics may be optimized.
As will be discussed further below, one or more upstream channels may be provided from the antenna electronics system to the remote electronics system. A principal application is for communication of a representation of a received radio frequency signal or signals. However, a feedback signal, such as a representation of the power amplified radio frequency signal transmitted or to be transmitted by the antenna may be returned to the remote electronics system, thus permitting a feedback-controlled pre-compensation of the signal communicated to the antenna electronics system, or more generally, an adaptive system capable of responding to changes in the system or environment to control the system operation. Thus, operating temperature may affect the output of the antenna electronics system. Internal temperature compensation may be provided, as a known option. On the other hand, temperature compensation (and other types of compensation) may be absent in the power amplification system, and these functions implemented digitally in the remote electronics system. This, in turn, permits, for example, a more efficient and/or less expensive power electronics design, and indeed potentially permits use of a power electronics module which requires under all or substantially all operating conditions, precompensation of the signal in order to produce a usable output signal.
In accordance with an aspect of the technology, the distortion, measured as a Digital Noise Power Ratio (DNPR) of the transmitted signal, defined as the maximum RMS distortion amplitude of the output within a narrow range, e.g., 1 kHz (yielding a DNPR1 kHz), over the entire bandwidth of a band, with respect to the RMS amplitude of a white noise signal with uniform RMS amplitude, extending over the entire band presented at the input, when measured over an arbitrarily long duration. The measurement may be made, for example, by digitizing the output and comparing it to the input with a digital signal processor, and therefore without requiring real filter implementations applied to the input or output signals. Typically, the effects of phase delay and group delay are excluded, and thus the distortion measurements are delay compensated, for example by use of an autocorrelator. Of course, the range of measurement can vary in accordance with the application. Typically, a transmitter system according to a preferred embodiment will provide a DNPR1 kHz of larger than 40 dB over a bandwidth of 120 MHz, e.g., from 2.4-2.52 GHz, more preferably larger than 50 dB, and perhaps extending up to 96 dB or greater.
The present technology also provides a receiver system. In this case, a radio frequency receiver will comprise an antenna or input from an antenna, a low noise amplifier, which may be cooled, for example <100K, and indeed, may be associated with high temperature superconductor filters operating at <78K, though will typically not be low temperature superconductor components. The signal from the low noise amplifier may drive an electrooptic converter, to thereby communicate an optical signal modulated with an analog radio frequency waveform, at its native radio frequency, to a remote location. As discussed above, this same optic link system, or a corresponding optic link, may also carry feedback signals, such as may be used to control over a transmitter system.
A receiver system in accordance with this embodiment does not require frequency translation components, and thus no mixer is used. Therefore, the level of spurious intermodulation components generated by the receiver may be minimized. Further, by ensuring that the pattern of intermodulation components with significant power within the band is relatively simple, digital deconvolution can be employed, if necessary. Digital deconvolution may also be employed in the receiver system to compensate for distortion in the corresponding signal transmitter. It is thus important to realize that while most radio systems treat intermodulation distortion as noise, and thus additive to the noise floor, these signal components are not stochastic, and to some extent that distortion can be reversed, in contrast to stochastic noise, which is truly random.
The optic link conveys the signal to the remote location, where an ultra-high speed analog-to-digital converter operates to digitize the radio frequency signal at its native frequency. In such a system, the SFDR can reach >90 dB (1 Hz bandwidth), even for wideband applications, from antenna to analog-to-digital converter, an amplified fiber optic link has an SFDR of, for example, >106 dB (in 1 Hz). The analog-to-digital converter and associated elements have, for example, an SFDR of −98.5 dBc over a 39 MHz bandwidth for a 20 GHz signal.
Other architectures may be used for the receiver as well, for example with the analog-to-digital converter located proximate to the antenna, frequency down-conversion, and the like.
A radio system is therefore provided, comprising an optoelectronic receiver adapted to receive an input from an optic communication medium comprising a digital domain representation of a radio frequency signal at a radio frequency data rate of at least 1.4 GHz; a converter adapted to receive a digital electronic signal from the optoelectronic receiver and convert the digital domain representation of the radio frequency signal into an analog domain radio frequency signal of at least 700 MHz; and an interface to an antenna adapted to transmit the radio frequency signal. It is noted that the 700 MHz band, which nominally begins at 698 MHz, is intended for new radio frequency mobile services, such as “Long Term Evolution” (LTE). The system, of course, may operate at higher frequencies, and for example, the converter, and other high speed digital electronic circuitry may operate at up to 40 gigasamples per second, or higher, permitting direct synthesis, without analog up-conversion, of signals of up to 20 GHz, or higher, and at lower frequencies, provides for greater oversampling opportunity. For example, the bands covered notably include the 700 MHz band, the 900 MHz band, 1.2-1.5 GHz bands, 1.8-1.9 GHz bands, 2.4-2.5 GHz bands, and the like. By avoiding analog mixers, a significant source of intermodulation distortion is avoided, thus permitting operation with a high spur free dynamic range. The interface may thus produce the analog domain radio frequency signal corresponding to the digital domain representation signal having a spur-free dynamic range of at least 40 dB over a bandwidth of at least 19 MHz. This 19 MHz represents, for example, a band having a band spacing of 20 MHz with a 0.5 MHz guard band on each side. Because the guard band does not generally include information signals, it may be acceptable for some intermodulation products or other spurious radiation to be present, though regulations typically control such emissions. Where a set of contiguous bands are combined, the entire block of frequencies may be treated as a single information signal.
The optoelectronic receiver and/or transmitter may comprise a multiplex transmitter and/or receiver, adapted to extract and/or inject information modulated on one of a plurality of information-carrying channels of the optic communication medium, substantially without interference with another of the information-carrying channels of the optic communication medium. For example the optic communication may be wavelength division multiplexed. In some cases, a digital signal is communicated on the optic communication medium, e.g., fiber optic, and various digital multiplexing schemes may be employed.
The converter, and other components may have an associated intermodulation distortion of the analog domain radio frequency signal and the digital domain representation of the radio frequency signal may be predistorted in a manner to compensate for at least a portion of the associated intermodulation distortion, to increase an effective spur-free dynamic range of the radio system. The predistortion may be adapted to improve the effective spur-free dynamic range by at least 3 dB, and more preferably by at least 6 dB. Larger degrees of reduction may be possible.
The system may further have an interface to an antenna adapted to receive a radio frequency signal of at least 700 MHz; a low noise amplifier and filter adapted to amplify the received radio frequency signal within a band; and an electrooptic transmitter adapted to transmit a signal corresponding to the received radio frequency signal within a band.
The converter may have an associated intermodulation distortion of the analog domain radio frequency signal and the digital domain representation of the radio frequency signal may predistorted in a manner compensate for at least a portion of the associated intermodulation distortion, to reduce an effective amplitude of at least one intermodulation product present in the analog domain radio frequency signal by at least 3 dB, and more preferably, by 6 dB. The system may further comprise an interface to an antenna adapted to receive, amplify and filter a received radio frequency signal; and an electrooptic converter adapted to generate an optic signal corresponding to the received radio frequency signal. A digital processor may be provided, adapted to increase an effective spur-free dynamic range of the received radio frequency signal by at least 3 dB, or at least 6 dB. Likewise, a digital processor may be provided, adapted to predistort the digital domain representation to increase an effective spur-free dynamic range of the analog domain radio frequency signal by at least 3 dB or at least 6 dB. The digital processor may comprise at least two Josephson junctions, and thus be part of a superconducting electronic circuit, such as a set of logic gates.
A radio system is also provided, comprising an optoelectronic receiver adapted to receive a signal from an optic communication medium comprising a radio frequency signal having at least one significant information-bearing component having a frequency of at least 698 MHz, and ranging, for example, from the 700 MHz band to the X band (7-12.5 GHz); an amplifier adapted to amplify the signal and generate a power radio frequency signal, substantially without frequency translation; and an interface to an antenna adapted to emit the power radio frequency signal, wherein the signal is predistorted to increase an effective spur-free dynamic range of the power radio frequency signal by at least 3 dB, and preferably at least 6 dB, by reducing a power of at least one intermodulation signal with respect to a corresponding signal absent predistortion. A transmitter may be provided, adapted to communicate a feedback signal corresponding to the power radio frequency signal, wherein the predistortion is based on at least the feedback signal. The power radio frequency signal is preferably generated with has a spur-free dynamic range of at least 40 dB over a bandwidth of at least 19 MHz, and preferably over 39 MHz, 59 MHz, or larger bandwidth. Likewise, the spur free dynamic range may be greater than 40 dB, and for example, may be 45 dB or more. The signal may have at least one significant information-bearing component having a frequency of at least 1.5 GHz, for example 1.8 GHz or 2.4 GHz.
The amplifier may comprise a power digital to analog converter, which directly generates an analog radio frequency signal from the at least one significant information-bearing component, having an amplitude of at least 27 dBm, and wherein the power radio frequency signal has an average power of at least 30 dBm.
The radio system may further comprise an interface to an antenna adapted to receive a radio frequency signal of at least 698 MHz; a low noise amplifier and filter adapted to amplify and filter the received radio frequency signal; and an electrooptic transmitter adapted to transmit a signal corresponding to the amplified and filtered received radio frequency signal.
A digital processor may be provided, adapted to predistort the signal to increase an effective spur-free dynamic range of the power radio frequency signal by reducing a power of at least one intermodulation product. Preferably, a largest amplitude intermodulation product (spur) is reduced in amplitude, but this is not always the case. Likewise, preferably the amplitude of a plurality of intermodulation products is reduced.
A radio transmitter is also provided, comprising an optoelectronic receiver adapted to receive an input from an optic communication medium comprising a digital domain representation of a radio frequency signal at a bit rate exceeding 2 Gigasamples per second and produce a digital electronic signal therefrom; a converter adapted to receive the digital electronic signal from the optoelectronic receiver and produce an information-bearing analog domain radio frequency signal of at least 1 GHz, substantially without frequency translation of the digital domain representation; and an interface to an antenna adapted to emit the radio frequency signal.
A radio system is further provided, comprising a receiver adapted to receive an input from a communication medium comprising a discrete time, quantized amplitude modulated representation of a signal, at an intersymbol modulation rate exceeding 2 giga-symbols per second; a converter adapted to receive the signal from the receiver and convert the signal into an analog domain radio frequency signal of at least 698 MHz, for example, greater than 1.5 GHz or 2.4 GHz and a bandwidth of at least 59 MHz; and an interface to an antenna adapted to emit the radio frequency signal, wherein the analog domain radio frequency signal has intermodulation products not expressly defined by the signal having an amplitude of at least −40 dB with respect to the signal; and a digital predistorter adapted to precompensate the input to reduce an amplitude of at least one intermodulation product by at least 3 dB, and preferably by at least 6 dB.
A wideband radio receiver is also provided, comprising at least one tower unit mounted on a tower, each tower unit comprising an antenna adapted to receive an incident wideband frequency distribution radio signal, an amplifier adapted to amplify the wideband frequency distribution radio signal, and an electro-optic converter adapted to convert the amplified wideband frequency distribution to an analog optical signal; an optical fiber communications link adapted to communicate the analog optical signal; and a base unit remote from the tower, adapted to receive the analog optical signal from the optical fiber communications link, comprising an opto-electric converter adapted to convert the optical signal received from the optical fiber link to an analog radio signal with a wideband frequency distribution corresponding to the wideband frequency distribution of the signal incident on the antenna, an analog-to-digital converter adapted to directly digitizes the entire wideband distribution, and a digital linearizer, adapted to digitally compensate the digitized wideband distribution for at least nonlinear distortion, to substantially reconstruct a digitally sampled equivalent of the frequency distribution of the signal incident on the antenna, having a spur-free dynamic range greater than a spur-free dynamic range of the optical signal.
A system is also provided for extracting information from a radio frequency signal, comprising: an input adapted to receive a radio frequency signal comprising information and significant radio frequency energy above 698 MHz; an amplifier adapted to amplify the radio frequency signal, wherein a bandwidth of the radio frequency signal comprising information is at least 19 MHz, and for example, 39 MHz, 59 MHz, 79 MHz, 100 MHz, 120 MHz, or larger bandwidth; an analog communication system having a communication length in excess of 1 meter, adapted to convey the amplified radio frequency signal, wherein the conveyed radio frequency signal has an associated distortion with respect to the radio frequency signal; a digitizer having a sampling rate in excess of 2 GHz, for example 10 GHz, 20 GHz or 40 GHz, adapted to digitize entire bandwidth of the conveyed radio frequency signal; and a digital signal processor adapted to correct at least a portion of the associated distortion, operating on the digitized conveyed radio frequency signal from the digitizer having a sampling rate in excess of 2 GHz.
A radio system is further provided, comprising: an optoelectronic receiver adapted to receive an input from an optic communication medium comprising a radio frequency signal having a bandwidth of at least 59 MHz at a band frequency higher that 698 MHz; an amplifier adapted to generate a power radio frequency signal corresponding to the radio frequency signal having an average power of at least 36 dBm (˜4 watts), and for example having an average power of at least 40 dBm (10 watts); and an interface to an antenna adapted to transmit the power radio frequency signal, wherein the power radio frequency signal has a distortion with an error vector magnitude of at least 1% with respect to the radio frequency signal communicated by the optical communication medium. A digital signal processor may be provided, for example, clocked at at least 5 GHz, and for example at 10 GHz, 20 GHz, 40 GHz, or faster, adapted to precompensate a digital data stream representing the radio frequency signal for at least a portion of the distortion. The error vector magnitude of the distortion may be, for example, at least 2% or at least 4%, and further comprising a digital signal processor adapted to precompensate a digital data stream representing the radio frequency signal to achieve a reduction in distortion of at least 3 dB, and preferably by at least 6 dB. For example, the error vector magnitude of the distortion is at least 4% and a digital signal processor is provided adapted to precompensate the radio frequency signal to achieve a reduction in error vector magnitude of at least 3%.
This compensation capability (predistortion) therefore allows the use of lower cost, lower power, higher power efficiency, or other advantageous types of devices, which may be mounted, for example, on an antenna tower.
A radio system is further provided, comprising a digital signal processor adapted to process a digital signal representing information to be communicated through a radio frequency transmission having a significant frequency component exceeding 698 MHz, having at least one input for receiving information, at least one logic unit adapted to define digital codes representing a modulated radio frequency signal oversampled with respect to the radio frequency transmission, and at least one digital logic unit adapted to produce a processed representation of the defined digital codes, by at least one of:
(a) predistorting the defined digital codes to increase a spur free dynamic range of the radio frequency transmission, based on a predicted non-linear distortion of a radio frequency transmitter;
(b) combining at least two sets of defined digital codes, each being oversampled with respect to a radio frequency transmission frequency in excess of 698 MHz;
(c) processing a signal representing at least one set of defined digital codes to reduce a peak to average power ratio of the radio frequency transmission after conversion to an analog radio frequency signal;
(d) processing the defined digital codes to provide source-encoding of a set of signals for driving a multiple antenna array, wherein preferably at least two antennas of the multiple antenna array receive respectively different signals; and
an output, adapted to present the processed representation.
The at least one digital logic unit may perform any two, or all three functions, in addition to other functions, in any order.
An optoelectronic modulator may be provided, adapted to communicate the output through an optic communication medium. A digital to analog converter may be provided, adapted to convert the defined digital codes into an analog radio frequency signal. Likewise, both a digital to analog converter adapted to convert the defined digital codes into an analog radio frequency signal, and an optoelectronic modulator adapted to communicate the analog radio frequency signal through an optic communication medium may be provided. The radio system may further comprise an optoelectronic demodulator, and a power amplifier to amplify the analog radio frequency signal to a power of at least 27 dBm, wherein the defined digital codes are predistorted to reduce an effect of a non-linear distortion of at least the optoelectronic modulator, optic communication medium, the optoelectronic demodulator, and the power amplifier. An optoelectronic modulator may be provided, adapted to communicate the output as a digital communication through an optic communication medium, an optoelectronic demodulator adapted to produce a communicated digital signal representing the output at a remote location from the optoelectronic modulator, and a digital to analog converter adapted to convert the communicated digital signal representing the output into an analog radio frequency signal. The analog radio frequency signal may be provided with a power of at least 27 dBm, and wherein the defined digital codes are predistorted to increase a spur free dynamic range of the radio frequency transmission, to reduce an effect of a non-linear distortion of the analog radio frequency signal. The defined digital codes may be processed to reduce a peak to average power ratio by at least 1 dB, and for example, 1.5 dB, 2 dB, 2.5 dB, or 3 dB. For example, at least two sets of defined digital codes are combined, and the combined sets of digital codes are processed to reduce a peak to average power ratio by at least 1 dB. The output may presents the processed representation an analog domain radio frequency signal having a spur-free dynamic range of at least 40 dB over a bandwidth of at least 19 MHz.
The digital logic unit may comprise at least one look-up table responding to look-up requests at a rate of at least 698 megasamples per second, and may provide a plurality of look-up tables operating in tandem, for example to implement a proportional-integral-differential processing algorithm in conjunction with a delay logic device, an integrator logic device, and a differentiator logic device.
The radio system may further comprise a digital crosspoint switch matrix. The radio system may further comprise a first optoelectronic modulator adapted to generate an optical communication of the output; an interface to an antenna adapted to receive a radio frequency signal of at least 698 MHz; and a second optoelectronic modulator adapted to optically communicate a signal corresponding to the received radio frequency signal. The defined digital codes may be processed by the digital logic unit to reduce an amplitude of at least one intermodulation product by at least 3 dB, preferably by at least 6 dB, based on a digital model of a non-linear process operating on the output. The digital signal processor may comprise an input adapted to receive a feedback signal representing a distorted of the output, and wherein the predistortion is responsive to the feedback signal. The digital signal processor comprises at least two Josephson junctions. Likewise, the digital signal processor may comprise an active superconducting logical device.
The output may represents at least one significant information-bearing component having a frequency of at least, for example, 1.0 GHz, or 1.5 GHz, or 2.0 GHz, or 3 GHz, or higher.
The radio system may further comprise an optoelectronic transmitter adapted to communicate the output as a signal in an optic communication medium; an optoelectronic receiver adapted to receive the signal from the optic communication medium comprising information defining a radio frequency signal having at least one significant information-bearing component having a frequency of at least 698 MHz; an amplifier adapted to amplify the signal and generate a power radio frequency signal, substantially without frequency translation; and an interface to an antenna adapted to emit the power radio frequency signal, wherein the signal is predistorted to increase an effective spur-free dynamic range of the power radio frequency signal by reducing a power of at least one intermodulation signal with respect to a corresponding signal absent the predistortion. A transmitter may be provided, adapted to communicate a feedback signal corresponding to the power radio frequency signal to the digital signal processor, wherein the predistortion is based on at least the feedback signal. The power radio frequency signal may have a spur-free dynamic range of at least 40 dB over a bandwidth of at least 19 MHz, and for example over a bandwidth of 39 MHz or 59 MHz. The amplifier may comprise a power digital to analog converter, which directly generates an analog radio frequency signal from the at least one significant information-bearing component, having an amplitude of at least 27 dBm, and wherein the power radio frequency signal has an average power of at least 30 dBm. The predistortion preferably improves the effective spur-free dynamic range by at least 3 dB, and more preferably by at least 6 dB.
A radio transmitter is provided, comprising a digital signal processor adapted to produce a digital domain representation of a radio frequency signal, sampled at a rate of at least 1.5 gigasamples per second; an optoelectronic transmitter adapted to communicate over an optic communication medium the digital domain representation of a radio frequency signal at a rate of at least 1.5 gigasamples per second; an optoelectronic receiver adapted to receive an input from the optic communication medium and produce a digital electronic signal therefrom; a converter adapted to receive the digital electronic signal from the optoelectronic receiver and produce an information-bearing analog domain radio frequency signal having significant frequency components exceeding 698 MHz; and an interface to an antenna adapted to emit the radio frequency signal.
The digital signal processor may be adapted to process a digital signal representing information to be communicated through a radio frequency transmission having a significant frequency component exceeding 698 MHz, having at least one input for receiving information, at least one logic unit adapted to define digital codes representing a modulated radio frequency signal oversampled with respect to the radio frequency transmission, and at least one digital logic unit adapted to produce a processed representation of the defined digital codes, by at least one of:
(a) predistorting the defined digital codes to increase a spur free dynamic range of the radio frequency transmission, based on a predicted non-linear distortion of a radio frequency transmitter;
(b) combining at least two sets of defined digital codes, each being oversampled with respect to a radio frequency transmission frequency in excess of 698 MHz;
(c) processing a signal representing at least one set of defined digital codes to reduce a peak to average power ratio of the radio frequency transmission after conversion to an analog radio frequency signal; and
(d) processing the defined digital codes to provide source-encoding of a set of signals for driving a multiple antenna array, wherein preferably at least two antennas of the multiple antenna array receive respectively different signals.
The at least one digital logic unit may perform any two, or all three, functions, in any order, in addition to other functions.
A power amplifier may be provided, adapted to amplify the analog radio frequency signal to a power of at least 27 dBm, wherein the digital signal processor predistorts the digital domain representation to reduce an effect of a non-linear distortion of at least the power amplifier.
The digital signal processor may predistort the digital domain representation to increase a spur free dynamic range of the radio frequency signal by at least 3 dB, and preferably at least 6 dB.
The defined digital codes may be processed to reduce a peak to average power ratio by at least 1.0 dB, 1.5, 2.0 dB, 2.5 dB, or 3 dB. At least two sets of defined digital codes may be combined, and the combined sets of digital codes are processed to reduce a peak to average power ratio by at least 1 dB, 1.5, 2.0 dB, 2.5 dB, or 3 dB.
The radio frequency signal may have, for example, a spur-free dynamic range of at least 40 dB over a bandwidth of at least 19 MHz, and preferably over 39 MHz or 59 MHz.
The defined digital codes may be processed by the digital logic unit to reduce an amplitude of at least one intermodulation product by at least 3 dB, preferably by at least 6 dB, based on a digital model of a non-linear process operating on the analog domain radio frequency signal.
The digital signal processor may have an input adapted to receive a feedback signal representing a distorted of the output, and wherein the predistortion is responsive to the feedback signal.
The digital signal processor may comprise an active superconducting logical device, such as a device implemented using Josephson junctions.
The radio frequency signal may be generated by a power digital to analog converter, which directly generates an analog radio frequency signal from the at least one significant information-bearing component, having an amplitude of at least 27 dBm, and wherein the power radio frequency signal has an average power of at least 30 dBm.
The radio frequency signal may have intermodulation products not expressly defined by the digital signal processor having an amplitude of at least −40 dB with respect to the radio frequency signal in the absence of predistortion, and the digital signal processor predistorts the digital domain representation of the radio frequency signal to reduce an amplitude of at least one intermodulation product by at least 3 dB.
The digital signal processor may comprise at least one look-up table, and may provide a plurality of look-up tables operating in tandem. The digital signal processor may, for example, comprise a delay element signal path, an integrator element signal path, and a differentiator element signal path, each signal path having at least one corresponding look-up table, and a combiner receiving an output of each respective signal path, to define a transformation of the digital domain representation of the radio frequency signal.
A radio system is further provided, comprising an optoelectronic receiver adapted to receive an input from an optic communication medium comprising an analog domain representation of a radio frequency signal having a frequency of at least 698 MHz; a converter adapted to digitize the analog domain representation at a sampling rate of at least 1.4 gigasamples per second; and a digital signal processor adapted to reduce an effective intermodulation distortion inferred to be present in the digitized analog domain representation based on strong signal components present in the digitized analog domain representation and their respective likely intermodulation products. The digital signal processor analyzes a frequency band greater than about 20 MHz, for example, 40 MHz, 60 MHz, 80 MHz, 100 MHz or 120 MHz. The digital signal processor preferably produces a representation of the analog domain radio frequency signal having a spur-free dynamic range of at least 40 dB over a bandwidth of at least 19 MHz, or 39 MHz, or 59 MHz. The digital signal processor preferably improves the effective spur-free dynamic range of the digitized analog domain representation by at least 3 dB, more preferably by at least 6 dB.
The radio system may further comprise an interface to an antenna adapted to receive a radio frequency signal of at least 700 MHz; and an electrooptic transmitter adapted to transmit a signal corresponding to the received radio frequency signal within the band on the optic communication medium.
The digital processor may be adapted to deconvolve a non-linear distortion of the digitized analog domain representation. Similarly, the digital processor may be adapted to reverse a non-linear distortion of the digitized analog domain representation, said digital signal processor comprising at least one look-up table, and for example may comprise a plurality of look-up tables operating in tandem. The digital processor may further comprise a delay element signal path, an integrator element signal path, and a differentiator element signal path, each signal path having at least one corresponding look-up table, and a combiner receiving an output of each respective signal path, to define a transformation of the digitized analog domain representation of the radio frequency signal.
The radio system preferably is implemented using superconducting digital electronic logic elements, and for example, the converter and/or digital signal processor may comprise at least one active superconductive electronic device. The converter may digitize the analog domain radio frequency signal at a rate of at least 5 gigasamples per second (GSPS), for example 10 GSPS, 20 GSPS or 40 GSPS.
A digital signal processor is provided, comprising an superconducting electronic circuit receiving as an input a digital signal having at least two radio frequency components having a center frequency above 698 MHz, representing an analog signal having information represented as signal energy within a band, and being subject to non-linear distortion, wherein the input further comprises at least one intermodulation product of the at least two radio frequency components present within the band, said superconducting circuit processing the input to reduce a signal energy within the band resulting from intermodulation distortion by at least 3 dB with respect to signal energy corresponding to information, and an output representing at least a portion of the information. The superconducting electronic circuit may comprises at least one look-up table, and for example may comprise a plurality of lookup tables. The superconducting electronic circuit may further comprise a digital correlator, which may be, for example, a multi-bit autocorrelator. The digital signal processor may further comprise a crosspoint switch matrix, and/or a digital channelizer.
A radio system is provided, comprising a receiver adapted to receive an input from a communication medium comprising radio frequency signal; a converter adapted to digitize the signal at a rate of at least 1.4 Gigasamples per second to digitally describe a radio frequency waveform having frequency components of at least 698 MHz and a bandwidth of at least 59 MHz; wherein the digitized signal has intermodulation products represented therein having an amplitude of at least −40 dB with respect to the signal; and a digital predistorter adapted to precompensate the input to reduce an amplitude of at least one intermodulation product by at least 3 dB.
These and other objects will become apparent through review of the detailed description and drawings.
A block diagram of a modern wireless basestation of the prior art is shown in
Consider first the transmitter, where the DBB Synthesizer might, for example, generate an OFDM (orthogonal frequency domain multiplexed) signal that is up to a few MHz in bandwidth, comprising many narrow band signals, properly timed and encoded. This digital signal is then used to modulate a diode laser or optical interferometer, and the optical signal is coupled into an optical fiber that is sent up the tower. Optical fibers generally exhibit relatively weak attenuation; the choice of fiber would depend on the distance to be covered, which could be quite far in some cases. The digital optical signal is demodulated at the tower, typically with a photodiode, and the resulting digital electrical signal recovers the initial DBB signal. This is then sent to a standard digital-to-analog converter (DAC) to obtain the analog baseband signal, which is then combined with an RF local oscillator in a mixer (with both I and Q signals in the standard way) to upconvert the signal and generate a low-power version of the signal to be transmitted. This passes through an appropriate bandpass filter (BPF), is amplified in a power amplifier (PA), and broadcast through an antenna. In typical wireless systems, the same antenna is used alternately for both transmission and reception, with a duplexer to isolate the two systems.
The prior-art receiver system in
The Prior Art IQ Transceiver shown in
The system of
One aspect of the present invention takes a substantially different approach from the prior art system in
Consider the transmitter in
For the receiver in
This technology employs integrated circuits comprised of many superconducting Josephson junctions, and at present is implemented preferably using niobium (Nb) as the superconductor, operating preferably at temperatures below about 5 K (−268 C). These circuits can be tested in the laboratory in a Dewar filled with liquid helium at a temperature of 4.2 K, but in the field would be mounted on a cryocooler inside an evacuated chamber. This cryocooler is a closed-cycle refrigerator, typically with two or more stages, which may be based on a pulse-tube, Gifford-McMahon, or Stirling cycle, and several models are sold commercially that can operate reliably for extended periods as long as electric power remains available. However, the power and weight budgets for present-day 4K cryocoolers may be too large for placement on a tower, hence a ground-based location in accordance with the present invention is preferable.
It is important to point out that RSFQ data converters (ADCs and DACs) are extremely linear, since they are based on a fundamental physical constant, the single flux quantum Φ0=h/2e=2.07 mV−ps, where h is Planck's constant and e is the charge on the electron. In fact, the international defining standard for the volt is now based on arrays of Josephson junctions, such that a data rate of 100 GHz corresponds to a voltage of 0.207 mV. This high linearity of data conversion is quite attractive for maintaining linearity in a digital receiver or transmitter. However, this linearity will be degraded by nonlinear distortion in the amplifiers and optical converters. Fortunately, the linearity can be restored through the use of appropriate “inverse distortions”, also possible using RSFQ circuits.
It should be noted that various technological implementations for the various components are possible, however, if a cryocooler is required for any RSFQ component, the use of additional RSFQ components is generally efficient.
Also shown in
The pre-amplifiers and optical converters can operate at standard temperatures, but in some cases the device noise may be reduced by cooling, typically to temperatures above the 4K operating temperature of the RSFQ circuits. Since most 4K cryocoolers are designed with thermal stages at an intermediate temperature of, say, 70 K, integration of these devices with the RSFQ circuits may be directly obtained within a single system having multiple temperature modules. A further advantage may be that optical fibers may be provided that carry less heat into the cryogenic environment than a metallic coaxial line with a similar bandwidth, and thus reduce the heat load on the low temperature portions cryocooler.
Similarly, although the low noise amplifier (LNA) and band pass filter (BPF) of the receiver in
The Receiver 80 in
The antennas for the various bands might be located on the same tower or on different towers. This may also include systems with multi-input, multi-output signals in the same frequency range, known as MIMO. Within the spirit of “software-defined radio” or “cognitive radio”, the switch matrix may be dynamically reprogrammed to redistribute signals according to the current traffic and availability of the various bands.
Likewise, more sophisticated antenna systems may be provided, such as electronically steerable antenna arrays, conformal antenna arrays, synthetic aperture antennas, and the like. In particular, the digital and analog processing is not limited to traditional cellular radio implementations.
MIMO systems exploit the spatial separation of antennas operating in the same band and having overlapping native coverage areas to extract or convey information, which can lead to a reduction in interference, increased channel information carrying capacity, or other advantages. Spatial multiplexing is a transmission technique in MIMO wireless communication to transmit independent and separately encoded data signals, called streams, from each of the multiple transmit antennas. Therefore, the space dimension is reused, or multiplexed, more than one time. If the transmitter is equipped with Nt antennas and the receiver has Nr antennas, the maximum spatial multiplexing order (the number of streams) is Ns=min(Nt, Nr), if a linear receiver is used. This means that Ns streams can be transmitted in parallel, leading to an Ns increase of the spectral efficiency (the number of bits per second and per Hz that can be transmitted over the wireless channel). Therefore, in accordance with the present technology, multiple antennas and mast-mounted electronics modules, with spatial separation and overlapping beam patterns and frequency bands of operation, which in accordance with aspect of the invention may be simplified with respect to generally used electronics modules, are employed. This may be particular advantageous with WiMax (IEEE 802.16), and 3G or later cellular technologies. Typically, the multiple radio frequency signals or their digital representations, are communicated as wavelength division multiplexed (WDM) signals on a single optic fiber, though multiple optic fibers may also be employed. Preferably, the signals are not frequency translated in an analog domain for modulation in a single optic channel, since this frequency translation generally degrades the SFDR and may produce distortion and bandwidth limitations on the signal.
In an open-loop MIMO system with Nt transmitter antennas and Nr receiver antennas, the input-output relationship can be described as y=Hx+n, where x=[x1, x2, . . . , xN
A precoding matrix W is used to precode the symbols in the vector to enhance the performance. The column dimension Ns of W can be selected smaller than Nt which is useful if the system requires Ns(≠Nt) streams because of several reasons, for example, if either the rank of the MIMO channel or the number of receiver antennas is smaller than the number of transmit antennas.
Preferably, these transformations may be performed by the digital signal processor processing the information at radio frequency sample rates, thus avoiding introduction of distortion due to analog downconversion or the like. In the transmit signal processing, the digital signal processor may perform a beamforming operation to control a radiation pattern of a set of common information transmitted by a plurality of antennas, this increasing an effective signal power in spatial regions of interest, while reducing an effective signal power where it is not needed. Likewise, the transmitted signals may be precoded with different information patterns on different antennas, especially in a multi-user environment (e.g., multiple remote or mobile receivers receiving information from a central antenna array). The receiver's task typically involves, in part, a beamforming task, as well as compensation for multipath effects and possible other distortions, through a plurality of antennas having overlapping native beam patterns and common operating frequency band. The MIMO processing therefore involves compensation for delay, matrix transformation, signal correlation and autocorrelation (e.g., for multipath echo detection and processing) and the like. These functions are each available in high speed digital processors, such as RSFQ circuits. See, e.g., U.S. Pat. No. 7,440,490, expressly incorporated herein by reference. See also, Wikipedia.
To achieve MIMO from a conventional system, several technologies are available:
Multi-user MIMO can leverage multiple users as spatially distributed transmission resources, requiring significant signal processing. Multi-user MIMO can generalized into two categories: MIMO broadcast channels (MIMO BC) and MIMO multiple access channels (MIMO MAC) for downlink and uplink situations, respectively. Single-user MIMO can be represented as point-to-point, pairwise MIMO.
Many Antennas is a smart antenna technique, which overcomes the performance limitation of single user MIMO techniques. In cellular communications, the number of the maximum considered antennas for downlink is 2 and 4 to support LTE and IMT-A requirements, respectively, though arrays of 8-64 or more antennas have been proposed. With large antenna arrays, such techniques as New SDMA: MU-MIMO, Network MIMO (COMP), Remote radio equipments; New beamforming: linear beamforming such as MF, ZF and MMSE and nonlinear beamforming such as THP, VP, and DPC; New antenna array: direct, remote and wireless antenna array; Direct antenna array: linear and 3D phased array, new structure array, and dynamic antenna array; and Remote and wireless antenna array: distributed antenna array and cooperative beamforming may be employed, alone or in combination or subcombination.
Note that the technology of digital communication on optical fibers lends itself to sending multiple independent digital signals on the same fiber between the base unit and the tower unit. This is accomplished by multiplexing two or more signals at slightly different wavelengths, and is known as wavelength division multiplexing, or WDM. These digital signals at different wavelengths could represent multiple bits of the same digital signal, or alternatively a deserialized bit stream, or even completely independent signals. The multiple signals would then be demultiplexed at the receiving end of the fiber, without significant crosstalk or interference among the signals. The decision on how to partition multiple digital signals among optical fibers may be determined in specific cases by considerations of performance and cost.
This system operates not on the amplitude and phase of the signal, as in conventional linearizers, but on the sampled RF signal itself. For this reason, it is not limited to narrowband signals, but rather accounts for harmonics of the RF signal. Since strong nonlinearities generate harmonics, this digital domain radio frequency rate approach is useful in broadband systems where the intermodulation products extend much more widely than the signal itself.
One consideration in addressing linearity issues is the tradeoff between linearity on the one hand and efficiency (as well as cost, weight, and system complexity) on the other hand. It is well known that one may improve linearity in PAs by selecting a PA with excess capacity, and operating well below saturation capacity in the linear regime, a procedure known as “backoff”. A similar constraint may be present in high-linearity optical converters. The advantage of using digital domain radio frequency rate linearizers is that a system designer has the flexibility to select devices and operate them in a regime that may be relatively nonlinear, while correcting these nonlinearities with digital processing. Indeed, a component with well characterized non-linearities may be used in preference to one which is more linear, but for which the residual non-linearities are more complex, or less characterized, or less predictable. This will permit an increase in system efficiency with decreased cost and weight of hardware on the tower.
A related issue is associated with combined RF signals having a large peak-to-average-power ratio (PAPR), which is inevitable in wideband signals based on orthogonal frequency-domain multiplexing (OFDM) and similar approaches. A large PAPR is undesirable in that it may lead to use of PAs (and optical converters) with excess dynamic range and capacity, which can be inefficient and expensive. The prior art has identified algorithms, operating on baseband signals, that can decrease PAPR somewhat, typically in exchange for some nonlinear spectral broadening. However, in accordance with an embodiment of the present invention, the digital domain radio frequency rate linearizers may be selectively programmed to reduce the PAPR by operating directly on the digital signal. This may permit an improved optimization and tradeoff for very broadband RF signals.
The prior art has also identified several multi-amplifier approaches for increasing the efficiency of PAs. These include Doherty amplifiers, polar modulation (envelope elimination and tracking), and outphasing amplifiers (linear amplification with nonlinear components, or LINC). While the examples provided herein demonstrate single-channel PAs, this in no way excludes the use of one or more of these more sophisticated amplifier techniques, which would provide correspondingly improved performance. Similar digital domain radio frequency rate techniques including digital domain radio frequency rate linearization and broadband optical links may be applied by one skilled in the art to systems comprising these alternative amplifier designs.
The present application is a Division of U.S. patent application Ser. No. 14/519,197, filed Oct. 21, 2014, now U.S. Pat. No. 9,548,878, issued Jan. 17, 2017, which is a Continuation of U.S. patent application Ser. No. 13/602,474, filed Sep. 4, 2012, now U.S. Pat. No. 8,867,931, issued Oct. 21, 2014, which is a Continuation of U.S. patent application Ser. No. 12/403,332, filed Mar. 12, 2009, now U.S. Pat. No. 8,260,145, issued Sep. 4, 2012, and is a Continuation of U.S. patent application Ser. No. 12/403,329, filed Mar. 12, 2009, now U.S. Pat. No. 8,260,144, issued Sep. 4, 2012, and is a Continuation of U.S. patent application Ser. No. 12/403,326, filed Mar. 12, 2009, now U.S. Pat. No. 8,260,143, issued Sep. 4, 2012, each of which claims the benefit of priority of Provisional Patent Application 61/035,932, filed, Mar. 12, 2008, the entirety of which are expressly incorporated herein by reference.
| Number | Name | Date | Kind |
|---|---|---|---|
| 3324295 | Harris et al. | Jun 1967 | A |
| 3999105 | Archey et al. | Dec 1976 | A |
| 4038600 | Thomas | Jul 1977 | A |
| 4306313 | Baldwin | Dec 1981 | A |
| 4561119 | Epworth | Dec 1985 | A |
| 4805235 | Henmi | Feb 1989 | A |
| 4841519 | Nishio | Jun 1989 | A |
| 4922452 | Larsen | May 1990 | A |
| 4953157 | Franklin et al. | Aug 1990 | A |
| 4975926 | Knapp | Dec 1990 | A |
| 5010346 | Hamilton | Apr 1991 | A |
| 5067173 | Gordon et al. | Nov 1991 | A |
| 5101501 | Gilhousen et al. | Mar 1992 | A |
| 5247309 | Reich | Sep 1993 | A |
| 5253309 | Nazarathy et al. | Oct 1993 | A |
| 5257283 | Gilhousen et al. | Oct 1993 | A |
| 5274381 | Riza | Dec 1993 | A |
| 5293545 | Huber | Mar 1994 | A |
| 5321849 | Lemson | Jun 1994 | A |
| 5325378 | Zorabedian | Jun 1994 | A |
| 5339184 | Tang | Aug 1994 | A |
| 5359412 | Schulz | Oct 1994 | A |
| 5371625 | Wedding et al. | Dec 1994 | A |
| 5373383 | LaGasse | Dec 1994 | A |
| 5400391 | Emura et al. | Mar 1995 | A |
| 5406588 | Birchler | Apr 1995 | A |
| 5412474 | Reasenberg et al. | May 1995 | A |
| 5416629 | Huber | May 1995 | A |
| 5450223 | Wagner et al. | Sep 1995 | A |
| 5457734 | Eryaman et al. | Oct 1995 | A |
| 5465264 | Buhler et al. | Nov 1995 | A |
| 5477368 | Eskildsen et al. | Dec 1995 | A |
| 5485009 | Meyzonnetie et al. | Jan 1996 | A |
| 5495484 | Self et al. | Feb 1996 | A |
| 5550667 | Krimmel et al. | Aug 1996 | A |
| 5566173 | Steinbrecher | Oct 1996 | A |
| 5574982 | Almgren et al. | Nov 1996 | A |
| 5579341 | Smith et al. | Nov 1996 | A |
| 5592327 | Gabl et al. | Jan 1997 | A |
| 5615034 | Hori | Mar 1997 | A |
| 5640414 | Blakeney, II et al. | Jun 1997 | A |
| 5644622 | Russell et al. | Jul 1997 | A |
| 5710651 | Logan, Jr. | Jan 1998 | A |
| 5737104 | Lee et al. | Apr 1998 | A |
| 5751242 | Goutzoulis | May 1998 | A |
| 5777773 | Epworth et al. | Jul 1998 | A |
| 5805235 | Bedard | Sep 1998 | A |
| 5812600 | Hess | Sep 1998 | A |
| 5852651 | Fischer et al. | Dec 1998 | A |
| 5856980 | Doyle | Jan 1999 | A |
| 5859611 | Lam et al. | Jan 1999 | A |
| 5917636 | Wake et al. | Jun 1999 | A |
| 5920416 | Beylat et al. | Jul 1999 | A |
| 5920821 | Seazholtz et al. | Jul 1999 | A |
| 5940196 | Piehler et al. | Aug 1999 | A |
| 5953139 | Nemecek et al. | Sep 1999 | A |
| 5963567 | Veselka et al. | Oct 1999 | A |
| 5974209 | Cho et al. | Oct 1999 | A |
| 6016426 | Bodell | Jan 2000 | A |
| 6081361 | Adams et al. | Jun 2000 | A |
| 6088349 | Zumkeller | Jul 2000 | A |
| 6096496 | Frankel | Aug 2000 | A |
| 6104851 | Mahgerefteh | Aug 2000 | A |
| 6108548 | Furukawa et al. | Aug 2000 | A |
| 6115162 | Graves et al. | Sep 2000 | A |
| 6115403 | Brenner et al. | Sep 2000 | A |
| 6125276 | Lupien | Sep 2000 | A |
| 6137612 | Startup | Oct 2000 | A |
| 6167269 | Kondo | Dec 2000 | A |
| 6181921 | Konisi et al. | Jan 2001 | B1 |
| 6188808 | Zhou et al. | Feb 2001 | B1 |
| 6201638 | Hall et al. | Mar 2001 | B1 |
| 6208861 | Suzuki | Mar 2001 | B1 |
| 6222861 | Kuo et al. | Apr 2001 | B1 |
| 6233454 | Sato | May 2001 | B1 |
| 6246500 | Ackerman | Jun 2001 | B1 |
| 6256130 | Bulow | Jul 2001 | B1 |
| 6256137 | Hironishi | Jul 2001 | B1 |
| 6271959 | Kim et al. | Aug 2001 | B1 |
| 6298186 | He | Oct 2001 | B1 |
| 6324391 | Bodell | Nov 2001 | B1 |
| 6330609 | Garofalakis et al. | Dec 2001 | B1 |
| 6331991 | Mahgerefteh | Dec 2001 | B1 |
| 6351651 | Hamabe et al. | Feb 2002 | B1 |
| 6359716 | Taylor | Mar 2002 | B1 |
| 6366309 | Siegle | Apr 2002 | B1 |
| 6366778 | Bender et al. | Apr 2002 | B1 |
| 6381234 | Sakoda et al. | Apr 2002 | B2 |
| 6459743 | Lipka | Oct 2002 | B1 |
| 6473214 | Roberts et al. | Oct 2002 | B1 |
| 6506342 | Frankel | Jan 2003 | B1 |
| 6525682 | Yap | Feb 2003 | B2 |
| 6535165 | Stephens | Mar 2003 | B2 |
| 6546249 | Imai et al. | Apr 2003 | B1 |
| 6563622 | Mueller et al. | May 2003 | B2 |
| 6563623 | Penninckx et al. | May 2003 | B1 |
| 6577013 | Glenn et al. | Jun 2003 | B1 |
| 6583906 | Schemmann | Jun 2003 | B1 |
| 6594070 | Sugata et al. | Jul 2003 | B2 |
| 6597836 | Johnson et al. | Jul 2003 | B2 |
| 6600794 | Agarossi | Jul 2003 | B1 |
| 6618513 | Evankow, Jr. | Sep 2003 | B2 |
| 6634811 | Gertel et al. | Oct 2003 | B1 |
| 6643417 | Strutz et al. | Nov 2003 | B2 |
| 6654564 | Colbourne et al. | Nov 2003 | B1 |
| 6654595 | Dexter | Nov 2003 | B1 |
| 6665351 | Hedberg | Dec 2003 | B2 |
| 6677877 | Johnson | Jan 2004 | B2 |
| 6687278 | Mason et al. | Feb 2004 | B1 |
| 6704353 | McCarty, Jr. | Mar 2004 | B1 |
| 6724523 | Yap | Apr 2004 | B2 |
| 6748133 | Liu et al. | Jun 2004 | B2 |
| 6760342 | Skones et al. | Jul 2004 | B1 |
| 6778307 | Clark | Aug 2004 | B2 |
| 6781435 | Gupta et al. | Aug 2004 | B1 |
| 6804471 | Ionov et al. | Oct 2004 | B1 |
| 6810047 | Oh et al. | Oct 2004 | B2 |
| 6826163 | Mani et al. | Nov 2004 | B2 |
| 6826164 | Mani et al. | Nov 2004 | B2 |
| 6834134 | Brennan, III et al. | Dec 2004 | B2 |
| 6836487 | Farmer et al. | Dec 2004 | B1 |
| 6847758 | Watanabe | Jan 2005 | B1 |
| 6898379 | Lenosky | May 2005 | B2 |
| 6901224 | Davis | May 2005 | B1 |
| 6947206 | Tsadka et al. | Sep 2005 | B2 |
| 6963685 | Mahgerefteh et al. | Nov 2005 | B2 |
| 6996345 | Chen | Feb 2006 | B1 |
| 7006726 | Hayes | Feb 2006 | B2 |
| 7013090 | Adachi | Mar 2006 | B2 |
| 7054538 | Mahgerefteh et al. | May 2006 | B2 |
| 7057716 | Yap | Jun 2006 | B2 |
| 7076170 | Choa | Jul 2006 | B2 |
| 7079780 | Rollins | Jul 2006 | B1 |
| 7123846 | Tateyama et al. | Oct 2006 | B2 |
| 7127172 | Chen | Oct 2006 | B1 |
| 7127175 | Mani et al. | Oct 2006 | B2 |
| 7146169 | Kahana | Dec 2006 | B2 |
| 7245833 | Volkening | Jul 2007 | B1 |
| 7251297 | Agazzi | Jul 2007 | B2 |
| 7263291 | Mahgerefteh et al. | Aug 2007 | B2 |
| 7266269 | Koste et al. | Sep 2007 | B2 |
| 7272327 | Epworth | Sep 2007 | B2 |
| 7280623 | Gupta et al. | Oct 2007 | B2 |
| 7280721 | McCallion et al. | Oct 2007 | B2 |
| 7302185 | Wood et al. | Nov 2007 | B2 |
| 7312199 | Burdick | Dec 2007 | B2 |
| 7313199 | Gupta et al. | Dec 2007 | B2 |
| 7324036 | Petre | Jan 2008 | B2 |
| 7362125 | Gupta et al. | Apr 2008 | B2 |
| 7365663 | Rylov et al. | Apr 2008 | B2 |
| 7366470 | Green | Apr 2008 | B1 |
| 7369715 | Darcie et al. | May 2008 | B2 |
| 7369722 | Yilmaz et al. | May 2008 | B2 |
| 7369779 | Croussore et al. | May 2008 | B1 |
| 7421168 | Goutzoulis | Sep 2008 | B1 |
| 7423252 | Zhao | Sep 2008 | B2 |
| 7424228 | Williams et al. | Sep 2008 | B1 |
| 7426350 | Sun et al. | Sep 2008 | B1 |
| 7440490 | Kidiyarova-Shevchenko et al. | Oct 2008 | B2 |
| 7443719 | Kirichenko et al. | Oct 2008 | B2 |
| 7446696 | Kondo | Nov 2008 | B2 |
| 7447436 | Yee et al. | Nov 2008 | B2 |
| 7477852 | Agarwal et al. | Jan 2009 | B2 |
| 7486247 | Ridgway et al. | Feb 2009 | B2 |
| 7495560 | Easton et al. | Feb 2009 | B2 |
| 7555217 | Hillis et al. | Jun 2009 | B2 |
| 7590354 | Sauer et al. | Sep 2009 | B2 |
| 7627250 | George et al. | Dec 2009 | B2 |
| 7649495 | Stephens | Jan 2010 | B2 |
| 7657132 | Yap et al. | Feb 2010 | B1 |
| 7660491 | Thaniyavarn | Feb 2010 | B2 |
| 7697847 | Matsui et al. | Apr 2010 | B2 |
| 7738602 | Langenbach | Jun 2010 | B2 |
| 7787773 | Kondo | Aug 2010 | B1 |
| 7787823 | George et al. | Aug 2010 | B2 |
| 7822075 | Liu | Oct 2010 | B2 |
| 7825364 | Hillis et al. | Nov 2010 | B1 |
| 7848654 | Sauer et al. | Dec 2010 | B2 |
| 7898464 | Anderson et al. | Mar 2011 | B1 |
| 8027588 | Kim et al. | Sep 2011 | B2 |
| 8059969 | Yap | Nov 2011 | B1 |
| 8103175 | Kowalczyk et al. | Jan 2012 | B1 |
| 8103178 | McKinney | Jan 2012 | B2 |
| 8111998 | George et al. | Feb 2012 | B2 |
| 8175459 | Thelen et al. | May 2012 | B2 |
| 8260143 | Gupta | Sep 2012 | B2 |
| 8260144 | Gupta | Sep 2012 | B2 |
| 8260145 | Gupta | Sep 2012 | B2 |
| 8275265 | Kobyakov et al. | Sep 2012 | B2 |
| 8515285 | DeSalvo et al. | Aug 2013 | B2 |
| 8538272 | Robinson | Sep 2013 | B1 |
| 8548330 | Berlin et al. | Oct 2013 | B2 |
| 8644844 | Shapira et al. | Feb 2014 | B2 |
| 8693875 | Banwell et al. | Apr 2014 | B2 |
| 8718478 | Thelen et al. | May 2014 | B2 |
| 8787776 | Carusone | Jul 2014 | B2 |
| 8831428 | Kobyakov et al. | Sep 2014 | B2 |
| 8867919 | Logan | Oct 2014 | B2 |
| 8867931 | Gupta | Oct 2014 | B2 |
| 8873585 | Oren et al. | Oct 2014 | B2 |
| 8913892 | Berlin et al. | Dec 2014 | B2 |
| 9020362 | Gupta | Apr 2015 | B2 |
| 9172471 | Hutchinson | Oct 2015 | B2 |
| 9335406 | Lohr | May 2016 | B2 |
| 20010005437 | Masuda | Jun 2001 | A1 |
| 20020060827 | Agazzi | May 2002 | A1 |
| 20020080898 | Agazzi | Jun 2002 | A1 |
| 20020111150 | Pleasant | Aug 2002 | A1 |
| 20020131662 | Strutz et al. | Sep 2002 | A1 |
| 20020154372 | Chung et al. | Oct 2002 | A1 |
| 20020159490 | Karwacki | Oct 2002 | A1 |
| 20020163454 | Yap | Nov 2002 | A1 |
| 20020175776 | Pleasant | Nov 2002 | A1 |
| 20020176659 | Lei et al. | Nov 2002 | A1 |
| 20020186094 | Pleasant et al. | Dec 2002 | A1 |
| 20030002120 | Choa | Jan 2003 | A1 |
| 20030030868 | Sasai et al. | Feb 2003 | A1 |
| 20030067952 | Tsukiji et al. | Apr 2003 | A1 |
| 20030099018 | Singh et al. | May 2003 | A1 |
| 20030147114 | Kang et al. | Aug 2003 | A1 |
| 20030175033 | Taga et al. | Sep 2003 | A1 |
| 20030193974 | Frankel et al. | Oct 2003 | A1 |
| 20030198477 | Kuri et al. | Oct 2003 | A1 |
| 20040008933 | Mahgerefteh et al. | Jan 2004 | A1 |
| 20040008937 | Mahgerefteh et al. | Jan 2004 | A1 |
| 20040022332 | Gupta | Feb 2004 | A1 |
| 20040036943 | Freund et al. | Feb 2004 | A1 |
| 20040070515 | Burkley et al. | Apr 2004 | A1 |
| 20040076199 | Wipiejewski et al. | Apr 2004 | A1 |
| 20040086275 | Lenosky | May 2004 | A1 |
| 20040096221 | Mahgerefteh et al. | May 2004 | A1 |
| 20040208590 | Nabors et al. | Oct 2004 | A1 |
| 20040218890 | Mahgerefteh et al. | Nov 2004 | A1 |
| 20050047780 | Hoshida | Mar 2005 | A1 |
| 20050052255 | Chiang | Mar 2005 | A1 |
| 20050100345 | Welch et al. | May 2005 | A1 |
| 20050111852 | Mahgerefteh et al. | May 2005 | A1 |
| 20050123301 | Kallstenius | Jun 2005 | A1 |
| 20050175356 | McCallion et al. | Aug 2005 | A1 |
| 20050206989 | Marsh | Sep 2005 | A1 |
| 20050260983 | DiPiazza | Nov 2005 | A1 |
| 20050271394 | Whiteaway et al. | Dec 2005 | A1 |
| 20050286829 | Mahgerefteh et al. | Dec 2005 | A1 |
| 20060002718 | Matsui et al. | Jan 2006 | A1 |
| 20060018666 | Matsui et al. | Jan 2006 | A1 |
| 20060029358 | Mahgerefteh et al. | Feb 2006 | A1 |
| 20060029396 | Mahgerefteh et al. | Feb 2006 | A1 |
| 20060029397 | Mahgerefteh et al. | Feb 2006 | A1 |
| 20060228120 | McCallion et al. | Oct 2006 | A9 |
| 20060232760 | Asbrock | Oct 2006 | A1 |
| 20060233556 | Mahgerefteh et al. | Oct 2006 | A1 |
| 20060239630 | Hase et al. | Oct 2006 | A1 |
| 20060274861 | Langenbach | Dec 2006 | A1 |
| 20060274993 | Mahgerefteh et al. | Dec 2006 | A1 |
| 20070008530 | Gibbs | Jan 2007 | A1 |
| 20070098417 | Peral et al. | May 2007 | A1 |
| 20070140703 | Fells | Jun 2007 | A1 |
| 20070206956 | Kikushima et al. | Sep 2007 | A1 |
| 20070297803 | Peral | Dec 2007 | A1 |
| 20080101444 | Gupta | May 2008 | A1 |
| 20080101501 | Gupta | May 2008 | A1 |
| 20080101503 | Gupta | May 2008 | A1 |
| 20080101798 | Healey | May 2008 | A1 |
| 20080107213 | Gupta | May 2008 | A1 |
| 20080212968 | Lindop | Sep 2008 | A1 |
| 20080212974 | Davies et al. | Sep 2008 | A1 |
| 20090232507 | Gupta | Sep 2009 | A1 |
| 20090232510 | Gupta | Sep 2009 | A1 |
| 20090238305 | Agazzi | Sep 2009 | A1 |
| 20090263137 | Hossein-Zadeh et al. | Oct 2009 | A1 |
| 20090263144 | McKinney | Oct 2009 | A1 |
| 20090297155 | Weiner et al. | Dec 2009 | A1 |
| 20100111540 | Caplan et al. | May 2010 | A1 |
| 20100166423 | Chang | Jul 2010 | A1 |
| 20100247104 | Yu et al. | Sep 2010 | A1 |
| 20100290551 | Agazzi | Nov 2010 | A1 |
| 20110064415 | Williams et al. | Mar 2011 | A1 |
| 20120141122 | Carusone | Jun 2012 | A1 |
| 20120315049 | Banwell et al. | Dec 2012 | A1 |
| 20120328301 | Gupta | Dec 2012 | A1 |
| 20130004180 | Gupta | Jan 2013 | A1 |
| 20130230328 | Fan et al. | Sep 2013 | A1 |
| 20140112374 | Braz et al. | Apr 2014 | A1 |
| 20140270807 | Hutchinson | Sep 2014 | A1 |
| 20150125155 | Gupta | May 2015 | A1 |
| 20170134091 | Gupta | May 2017 | A1 |
| Number | Date | Country |
|---|---|---|
| 2107147 | Apr 1983 | GB |
| Number | Date | Country | |
|---|---|---|---|
| 20170134091 A1 | May 2017 | US |
| Number | Date | Country | |
|---|---|---|---|
| 61035932 | Mar 2008 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 14519197 | Oct 2014 | US |
| Child | 15404950 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 13602474 | Sep 2012 | US |
| Child | 14519197 | US | |
| Parent | 12403332 | Mar 2009 | US |
| Child | 13602474 | US | |
| Parent | 12403329 | Mar 2009 | US |
| Child | 12403332 | US | |
| Parent | 12403326 | Mar 2009 | US |
| Child | 12403329 | US |
