|Publication number||USRE41130 E1|
|Application number||US 11/123,255|
|Publication date||Feb 16, 2010|
|Filing date||May 6, 2005|
|Priority date||Oct 22, 1999|
|Also published as||US6560445, WO2001031872A1|
|Publication number||11123255, 123255, US RE41130 E1, US RE41130E1, US-E1-RE41130, USRE41130 E1, USRE41130E1|
|Inventors||Bruce Fette, Peter J. Leahy|
|Original Assignee||Bruce Fette, Leahy Peter J|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (36), Non-Patent Citations (3), Referenced by (4), Classifications (6), Legal Events (3)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This invention pertains to radio frequency communication, in general, and to method and apparatus for modulating and demodulating information onto radio frequency carrier, in particular.
Spectrum is a precious resource. With increasing usage of spectrum, interference is of increasing concern. All radio frequency communication developed to date is subject to interference. The worst type of interference occurs with intentional jamming or overloaded spectrum usage. In addition, varying propagation conditions including propagation effects such as fading and multipath can produce substantial impairment of existing systems' ability to maintain exchange of information.
It is desirable to provide a type of modulation for radio frequency transmission that is relatively more robust at delivering information from transmitter to receiver in the presence of a large variety of interference. It is also desirable that such modulation type be robust to impairments introduced by the propagation effects of the channel. It is highly desirable to provide a communication system which will support many simultaneous users of a band of communication spectrum by distributing information within a modulation in ways that are different from all other modulations, thereby making other forms of modulations unlikely to interfere with communications utilizing this modulation.
The invention will be better understood from a reading of the following detailed description in conjunction with the drawing figures in which like reference designators are used to identify like elements, and in which:
The present invention is directed to an entirely new method of radio frequency modulation. The modulation process utilized in the radio frequency (RF) system and method of the invention is based on performing modulation of information in a manner similar to the way all animals generate acoustic sounds for some form of communication. Animals create sounds using nonlinear processes that generate a rich spectrum of harmonics. This nonlinear process in human speech is called pitch frequency, and in the English language, it is modulated up or down to indicate emphasis. Secondly, the harmonic spectrum is modulated with multiple resonant frequencies (formants), and the motion of these frequencies toward important targets, known as pronunciation gestures, conveys the linguistic information in the English language. It is recognized that a room full of people all talking at the same time are all able to communicate in spite of overlap in spectrum and communication properties; we call this the “cocktail party effect.” Humans use many techniques to achieve this ability. The present invention mimics many of the same functions performed acoustically in RF modulation technology to achieve the same ability to communicate utilizing RF in spite of all impairments of noise, propagation anomalies and other transmitters occupying the same bandwidth. In essence, the present invention is directed to a modulation in which RF signals are modulated with amplitude and frequency or phase versus time to create unique symbols in high dimensional space which can be recognized by any of several measurements able to be made upon the receivable signal components.
Animals use pulses of energy rather than constant energy output. The peak factor allows for communication over greater range at some average power level, although intervening silence intervals interrupt communications. This is a desirable tradeoff when the average power of transmission would not close the link over that extended range reliably.
In the description that follows, it is assumed that both transmitter and receiver have a clock that has been set to a common time as well as a common nonlinear process, such as cryptography, which can convert clock time into specifications of desired signal time slots allowing the receiver to focus on only parts of the received signal at the same time the transmitter might choose to transmit. In accordance with the invention the duty factor of such random burst transmissions (usually of data packets) is adjusted to match the number of members expected to participate in a net, thereby leaving each net member a time slot. When this is the case, all net members use the same time clock (for example Global Positioning System (GPS) time) and the same nonlinear algorithm (cryptographic algorithm) and will add an offset delay to that time which is based on their net number assignment which guarantees that no net members transmit at the same time; however, overlapped transmissions may occur at a receiver due to propagation delay. These minor overlaps may be removed by automated orthogonalization techniques. This burst time slot selection technique is not a requirement but may be used in conjunction with this invention and does emulate the nonlinear excitation properties of animal communications.
In signal processing literature a “cepstrum” is defined as the inverse Fourier transform of the log magnitude of the Fourier transform of a signal. In accordance with the principles of the invention, an analog input signal is converted by an analog to digital (A/D) converter 101 as seen in FIG. 1. The digital output of A/D converter 101 is converted into a time domain signal at window 103 and applied to a Fast Fourier Transform processor 105 to produce a Fourier spectrum output. The log or root magnitude is formed over the spectrum output at a log magnitude functional block 107. The Inverse Fast Fourier Transform (IFFT) is formed over the output of the log magnitude functional block 105 by IFFT 109 to produce outputs that exhibit cepstrum. Similarly the phase or phase change may be extended from the FFT and then an IFFT may produce the WASTRUM.
Carrier frequencies for this communication modulation are selected by a common time, a cryptographic process and a net member number. For example, if the communications band of interest consists of 16384 channels each capable of modulating 8000 bits per second, then receiver and transmitter pair choose a subset of those channels, say 128, which they will use at a given instance to modulate information. Since it is desirable for many other pairs of communicators to also be able to communicate over the same band at the same time and to minimize their intentional interference, it is desirable that all cooperating members of a network communication band use the common time and common cryptographic algorithm to select the carrier frequencies. Each pair of communicators will choose a net number and the net number will be combined with the cryptographic carrier selection in a way that guarantees no frequency overlap amongst the cooperating members. So, for example, if there are 128 pairs of communicators within a band, and each uses 128 carrier frequencies, this consumes the entire 16384 carrier frequencies without overlap in frequency, but provides a random set of frequencies for each pair of communicators. These carrier frequencies need not and preferably are not contiguous in the frequency spectrum.
An example of modulation in accordance with the invention is illustrated in the diagram of
Each receiver 1003R, 1001R selects the desired carrier frequencies f1 through f8 and brings together measurements of the modulation properties on all the desired carriers and discards all other carriers from the remainder of the band, thereby creating the spectrum of the channels intentionally received.
Each transmitter 1001T, 1003T modulates frequency or phase versus time in such a way that the modulation information is impressed over all parts of the spectrum transmitted and therefore recoverable from whatever parts of the spectrum are received without interference.
Modulation is imposed upon the frequencies in a cepstral fashion.
It has been previously recognized that various forms of interference in time domain and/or frequency domain can be recognized and suppressed. In accordance with the present invention, interference in the cepstral domain is recognized and appropriately minimized when it is recognized as interference. Interference is usually recognized as a large impulse that significantly exceeds the normally expected signal power.
In the time domain, RF communication interference often occurs due to lightning wherever it may occur in the world. It may also occur due to engine ignition noise, radar, or due to nearby burst transmissions not coordinated by automated orthogonalization. Impulses in the time domain are easily recognized as high energy events where the rms energy in a short window exceeds the average of rms energy in that window averaged over a longer time. When such impulses occur in the time domain, they may be recognized, and the time interval may be marked as interfered or may simply be zeroed. With a well chosen false alarm rate for detecting such events, suppression of time domain impulse interference may be achieved.
Similarly, narrowband transmissions from other communications systems not part of an automated orthogonalized network may be recognized as narrow frequencies in the frequency domain, which have much higher energy than the average energy over the desired spectrum of interest. These may also be marked as interference or may be zeroed. Similarly, 60 to 80 dB of suppression of such narrowband frequency interference is possible in the frequency domain. Conversion to the frequency domain is usually performed by windowing the time domain information with a suitable window (for example, a Hamming Window) and then performing a Fast Fourier Transform with a reasonably large number of points (1024, for example) and then taking the magnitude of the energy at each frequency. The noise floor of the spectrum and a threshold for detecting and removing narrowband interference are best accomplished with a histogram, which will clearly identify a large number of hits at the energy of the noise floor and relatively few high energy bits and the energy of those hits which then quantifies a threshold for detecting narrowband interference that can be removed.
As described with reference to
Cepstrum also discovers the properties of the communications channel due to multi-path and other propagation considerations. If a transmission has uniform energy over its entire bandwidth (as in white noise), the reflections of that signal off of buildings, cars, lamp posts, airplanes, etc. introduce a delayed replica of the signal. At some frequencies, this replica may cause destructive interference to the direct path signal and at other frequencies constructive enhancement. In the cepstral domain, these propagation effects show up as along term offset in cepstral coefficients at low frequencies (C1 through C10). If signals are modulated in the frequency domain, these channel propagation features may be removed by subtracting the average value of each low cepstrum coefficient (C1 through C10).
The spectral shape of received signals may be measured by examining the low order cepstral coefficients, i.e. cepstral coefficients C1 through C10. C1 represents the slope of the spectrum. Higher energy at lower frequencies and lower energy at higher frequencies (commonly occurring due to propagation) causes a positive C1. C2 represents the curvature of the spectrum from low to mid to high. Relatively more energy in mid spectrum will cause a positive C2 while relatively more energy at low and high spectrum will cause a negative C2. Similarly, cepstral coefficients C3 through C10 describe the spectral shape as relatively faster variations as the spectrum is scanned from low to high frequency.
Even if a few carriers of transmission are obscured, the cepstrum can recover the relative shape properties of the other received cepstrally modulated and transmitted signals. To maximize the separation of interference from other transmission types, a constellation of modulation properties is defined in the cepstrum and then converted back into a variety of magnitudes to impress onto the carrier frequencies that have been selected for transmission. This process can be used to define cepstral constellations that support from one bit to as many as 10 bits. The allocation of bits to cepstral modulation patterns need not be one bit to one cepstrum, but may preferably be done with a vector quantization strategy such that all cepstrum values convey all the bits, each showing up as a unique spectral shape (or equivalently a unique sound if the signals were audible rather than RF). This amounts to amplitude modulation of all channels in some way that the amplitude contour impressed over many carriers is recoverable from any reasonable subset of carriers actually received. Any carriers recognizably interfered are simply marked as interfered and not used in the cepstral analysis. If there are significantly more carriers than cepstral coefficients calculated and more cepstral coefficients calculated than bits delivered, there is high processing gain and high ability to communicate in high interference.
In a similar fashion, by introducing phase changes onto all carrier frequencies in a way that the information bits delivered cause effects in phase on all carrier frequencies received, it becomes possible to distribute the information bits so that they can be recovered over any reasonable subset of non-interfered received signal. While this is not strictly a cepstrum, the signal processing is performed in a similar manner, but rather than modulating the magnitude of the spectral shape we modulate all carriers with some type of carrier phase change. For example, suppose two bits are to be sent and that four phase change constellations are defined, one for each possible dibit. Dibit 00 may produce all positive angle phase changes and dibit 11 may produce all negative phase changes and the other two combinations may produce positive phase changes at low frequency and negative phase changes at high frequency or vice versa.
In accordance with one aspect of the invention, a simplified receiver design 1003R, 1001R as shown in
Under the worst possible interference, such as in the case of intentional jamming or overloaded spectrum the ability to separate the desired signal from the interference requires even greater focus of attention. In these cases a sequence of modulations is created that causes a sequence of changes of cepstral values. The ability to demodulate the desired signal and ignore the interference then arises from the ability to observe those parts of received symbols that correctly track paths defined for the cepstral coefficients and to ignore trajectories which show evidence of interfering phenomenon.
The cepstral concept can be used with amplitude modulation of each of the spectral lines. However, in accordance with the principles of the invention, delta phase modulation onto the plurality of carrier frequencies may be used as indicated by the phase modulation diagram of FIG. 10.
The trajectories of the cepstral coefficients may be used to define “1” and “0”.
Assume for purposes of explanation that an information symbol consists of a sequence of N partial symbols (as a large number of sounds in sequence make a word). Further assume that one bit of information consists of a sequence of 16 cepstrally defined partial symbols. Over this sequence of N partials, perhaps some cepstral coefficients have a continuous slope and perhaps others exhibit a curvature.
In addition to the interference suppression techniques described above for time, frequency and noise suppression, when there is intense babble similar to that of the intended modulation, interference must be suppressed in the cepstral demodulation process. Cepstral sequences may be compared against the symbol constellations in all defined dimensions. The sequence for C1 is compared against the defined sequence for C1, etc. Where sequences match within prescribed tolerances for each partial transition that can be taken as evidence for the reported symbol. When cepstral sequences fall outside of tolerance for a trajectory sequence, that feature can be ignored as having been interfered during the interval when it is out of tolerance. The net result of this trajectory modulation is that demodulation consists of all the standard time, frequency, cepstrum interference suppressions and in addition is an M+1 dimensional weighted pattern match, where M is the number of cepstral coefficients modulated with time over the symbol and the remaining dimension is time.
Trajectory modulation shown in
The modulation technique defined here enables communications over all known types of intentional jamming and poor propagation conditions and most importantly lends itself to a network of communications equipment all operating in automatic orthogonalization. This modulation is orthogonal to narrowband and to spread spectrum modulations and thereby provides an important third choice of modulation that will not interfere with the other two modulations.
This invention defines optimal modulation symbols to deliver information at the highest delivery rate practical, given the current link bandwidth and signal to interference ratio. Since the optimal modulation symbol set for a cepstral modulation system may change as a function of signal to interference ratio, this design process will locate global optimum symbol sets and allow these symbol sets to be selected for use in modulation during intervals experiencing a specific level of interference.
Each transmitter 1001T, 1003T of system 1000 in
After excising these interference signals, the cepstral domain allows identification of the channel link properties, which largely show up as direct current (DC) offsets to low order cepstrum C1, C2, etc.
Assuming that cepstral modulation is accomplished by either amplitude modulation of or by phase modulation of the plurality of carrier signals, or both, it is possible to create a simulation of link conditions resembling the current link conditions. Within this simulation current types of interference from other signals, current multi-path properties of the link and gaussian noise can all be modeled. Then the ability to detect modulation on each cepstral coefficient can be measured and converted to the information carrying capacity of that cepstral coefficient.
In the simulation, all cepstral coefficients are modulated with random noise, ranging over the values −1 to +1. This may be performed one coefficient at a time or in multiples of coefficients or even all coefficients, each with independent noise. The simulation of many baud of communication is performed (assume at least 10 times the number of bits to be delivered under these conditions). The receive simulation decodes and recovers the cepstral coefficients, which can now be correlated with those coefficients transmitted. This correlation now provides evidence of the signal to noise ratio (SNR) of each cepstral coefficient and its ability to convey information. This process will also produce a model of how the various forms of interference degrade or offset the cepstral coefficients. Also, certain types of interference may be detected and eliminated by recognizing that a certain cepstral coefficient was always modulated to be zero and that, by receiving as non-zero, the location of interference may be recovered in the spectrum and its artifact removed.
Modulation constellations for cepstrum can be one-dimensional (scalar) as shown in
Receiver 1003R includes a processor that initially, at step 901 selects a first SNR in the range of SNR for the expected communication environment of communication link 1005
The technique defined here can result in both transceiver 1001 (
Turning now to
The invention has been described in terms of various embodiments. It will be appreciated by those skilled in the art that various changes and modifications may be made to the various embodiments without departing from the spirit or scope of the invention. It is further intended that the invention not be limited in scope by the particular embodiments shown and described herein. It is intended that the invention be limited in scope only by the claims appended hereto.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3662108 *||Jun 8, 1970||May 9, 1972||Bell Telephone Labor Inc||Apparatus for reducing multipath distortion of signals utilizing cepstrum technique|
|US3810019||Sep 25, 1972||May 7, 1974||Sperry Rand Corp||Multifrequency communication system for fading channels|
|US4189779 *||Apr 28, 1978||Feb 19, 1980||Texas Instruments Incorporated||Parameter interpolator for speech synthesis circuit|
|US4209836 *||Apr 28, 1978||Jun 24, 1980||Texas Instruments Incorporated||Speech synthesis integrated circuit device|
|US4209844 *||May 12, 1978||Jun 24, 1980||Texas Instruments Incorporated||Lattice filter for waveform or speech synthesis circuits using digital logic|
|US4234761 *||Jun 19, 1978||Nov 18, 1980||Texas Instruments Incorporated||Method of communicating digital speech data and a memory for storing such data|
|US4304964 *||Apr 28, 1978||Dec 8, 1981||Texas Instruments Incorporated||Variable frame length data converter for a speech synthesis circuit|
|US4310831 *||Feb 4, 1980||Jan 12, 1982||Texas Instruments Incorporated||Pulse width modulated, push/pull digital to analog converter|
|US4319083 *||Feb 4, 1980||Mar 9, 1982||Texas Instruments Incorporated||Integrated speech synthesis circuit with internal and external excitation capabilities|
|US4323732 *||Feb 4, 1980||Apr 6, 1982||Texas Instruments Incorporated||Speech synthesis system with alternative coded or uncoded parameter formats|
|US4328395 *||Feb 4, 1980||May 4, 1982||Texas Instruments Incorporated||Speech synthesis system with variable interpolation capability|
|US4331836 *||Nov 16, 1979||May 25, 1982||Texas Instruments Incorporated||Speech synthesis integrated circuit device|
|US4335275 *||Feb 4, 1980||Jun 15, 1982||Texas Instruments Incorporated||Synchronous method and apparatus for speech synthesis circuit|
|US4344148 *||Feb 25, 1980||Aug 10, 1982||Texas Instruments Incorporated||System using digital filter for waveform or speech synthesis|
|US4354056 *||Feb 4, 1980||Oct 12, 1982||Texas Instruments Incorporated||Method and apparatus for speech synthesis filter excitation|
|US4357489 *||Feb 4, 1980||Nov 2, 1982||Texas Instruments Incorporated||Low voltage speech synthesis system with pulse width digital-to-analog converter|
|US4419540 *||Oct 5, 1981||Dec 6, 1983||Texas Instruments Incorporated||Speech synthesis system with variable interpolation capability|
|US4441201 *||Jan 25, 1982||Apr 3, 1984||Texas Instruments Incorporated||Speech synthesis system utilizing variable frame rate|
|US4443859 *||Jul 6, 1981||Apr 17, 1984||Texas Instruments Incorporated||Speech analysis circuits using an inverse lattice network|
|US4449233 *||Mar 5, 1982||May 15, 1984||Texas Instruments Incorporated||Speech synthesis system with parameter look up table|
|US4468805 *||Oct 13, 1981||Aug 28, 1984||Texas Instruments Incorporated||Variable frame length data converter for a speech synthesis circuit|
|US5537398 *||May 12, 1995||Jul 16, 1996||Motorola, Inc.||Apparatus for multi-rate simulcast communications|
|US5867528 *||Jan 16, 1997||Feb 2, 1999||Alcatel Alsthom Compagnie Generale D'electricite||Method and modem for adaptive allocation of the pilot carrier in a multi-carrier system|
|US5991277 *||Apr 9, 1998||Nov 23, 1999||Vtel Corporation||Primary transmission site switching in a multipoint videoconference environment based on human voice|
|US6219334 *||Mar 12, 1998||Apr 17, 2001||Kabushiki Kaisha Toshiba||Receiving apparatus for receiving orthogonal frequency division multiplexing signal and receiving method thereof|
|US6246672 *||Apr 28, 1998||Jun 12, 2001||International Business Machines Corp.||Singlecast interactive radio system|
|US6304611 *||Jun 18, 1998||Oct 16, 2001||Hitachi Denshi Kabushiki Kaisha||OFDM modulator and OFDM modulation method for digital modulated wave having guard interval|
|US6560445 *||Oct 22, 1999||May 6, 2003||General Dynamics Decision Systems, Inc.||Radio communication system and method of operation|
|US6826284 *||Feb 4, 2000||Nov 30, 2004||Agere Systems Inc.||Method and apparatus for passive acoustic source localization for video camera steering applications|
|US7024353 *||Aug 9, 2002||Apr 4, 2006||Motorola, Inc.||Distributed speech recognition with back-end voice activity detection apparatus and method|
|US7212815 *||Mar 3, 1999||May 1, 2007||Ascom (Schweiz) Ag||Quality evaluation method|
|US20030182113 *||Mar 24, 2003||Sep 25, 2003||Xuedong Huang||Distributed speech recognition for mobile communication devices|
|US20040030544 *||Aug 9, 2002||Feb 12, 2004||Motorola, Inc.||Distributed speech recognition with back-end voice activity detection apparatus and method|
|US20080031369 *||Jun 7, 2006||Feb 7, 2008||Li Ye Geoffrey||Apparatus and methods for multi-carrier wireless access with energy spreading|
|GB1605319A||Title not available|
|WO2001031872A1||Oct 6, 2000||May 3, 2001||Motorola Inc||Radio communication system based on cepstral modulation|
|1||*||Bingham, John A. C. Multicarrier Modulation for Data Transmission: An Idea Whose Time Has Come. IEEE Communications Magazine (May 1990), pp. 5-14.|
|2||Copy of International Search Report for Appl. No. PCT/US00/27869, mailed Feb. 13, 2001, 2 pages.|
|3||*||Ormsby, Charles D. Cepstral Processing for GPS Multipath Detection and Mitigation. Thesis. (Air Force Institute of Technology Dec. 1997).|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8823539 *||Jan 31, 2012||Sep 2, 2014||Fluke Corporation||Method of displaying multi-fiber test results|
|US8976906 *||Mar 29, 2012||Mar 10, 2015||QRC, Inc.||Method for spectrum sensing of multi-carrier signals with equidistant sub-carriers|
|US8982971 *||Mar 29, 2012||Mar 17, 2015||QRC, Inc.||System for spectrum sensing of multi-carrier signals with equidistant sub-carriers|
|US20130194103 *||Jan 31, 2012||Aug 1, 2013||Fluke Corporation||Method of displaying multi-fiber test results|
|U.S. Classification||455/91, 455/102|
|International Classification||H04B1/02, H04L27/30|
|Oct 4, 2010||AS||Assignment|
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:GENERAL DYNAMICS DECISION SYSTEMS, INC.;REEL/FRAME:025084/0449
Effective date: 20041118
Owner name: LEWIS SALES LLC, NEVADA
|Oct 25, 2010||FPAY||Fee payment|
Year of fee payment: 8
|Oct 28, 2014||FPAY||Fee payment|
Year of fee payment: 12