RU2837777C2 - Noise reduction using segmented, frequency-dependent phase suppression - Google Patents

Abstract

FIELD: noise reduction technologies.

SUBSTANCE: noise reduction in a signal stream containing an interference signal, called noise, is performed by obtaining a digitized noise signal and using a digital processor circuit to divide the resulting noise signal into different frequency band segments and thus generate multiple segmented noise signals. Then, individually for each segmented noise signal, the processor shifts the segmented noise signal in time by an amount depending on the selected frequency of the segmented noise signal to form a plurality of shifted segmented noise signals. The exact time shift applied to for each noise segment, it takes into account the frequency content of the segment and the processing time of the system. Amplitude scaling is applied individually for each segmented noise signal. The shifted and amplitude-scaled segmented noise signals are then combined to form a composite anti-noise signal, which is output to the signal stream to attenuate noise through destructive interference.

EFFECT: noise reduction through destructive interference.

20 cl, 22 dwg, 1 tbl

Description

Field of technology to which the invention relates

The present invention relates generally to electronic and automatic noise suppression technologies. More particularly, the invention relates to a noise suppression technology that generates frequency-dependent anti-noise components in a plurality of spectrum segments precisely calculated for the system and implementation.

State of the art

For decades, scientists and engineers have been working on a technical solution to the problem of electronic automatic noise cancellation (ANC). The basic physics of wave propagation suggest that it is possible to form an “anti-noise” wave that is 180 degrees out of phase with the noise signal and completely cancels the noise through destructive interference. This works well enough for simple, repetitive, low-frequency audio signals. However, this solution is not suitable for dynamically rapidly changing audio signals or sounds containing higher frequencies.

The best modern systems (using a hybrid structure that combines feedforward and feedback) can reduce repetitive noise (such as a motor or fan) at frequencies as low as 2 kHz using LMS (least mean square) adaptive filtering variants to shape the anti-noise signal by re-estimating the transfer function that produces the least practical noise at the output. Although companies continue to invest in improving ANC results, their efforts seem to be focused on improving these existing technologies. Moreover, despite the availability of significant computing power, it is clear that ANC using various adaptive filters has an upper frequency limit somewhere below 4 kHz, and the ability to attenuate signals from 10 to 30 dB.

Disclosure of the essence of the invention

In contrast to the traditional approach, the disclosed system is capable of suppressing virtually any frequency range in an autonomous mode and at least the entire audio spectrum in real time at commercially available processing speeds, more efficiently than currently used methods.

Processing speed and computing power continue to increase rapidly (e.g., Moore's Law has been in effect since 1965). Some commercial and military industries are less cost sensitive (than most consumer products) and can accommodate the higher cost of current maximum speeds/power. Additionally, the extraordinary power of quantum computing will be utilized in the near future. In view of the above, the inventors of the present invention have provided embodiments of the disclosed system and method that are expected to become increasingly commercially attractive over time. Thus, in the present invention, for simplicity of presentation, the number of embodiments has been reduced to five basic embodiments defined by the minimum number of different hardware system structures necessary to implement the plurality of implementations for the invention. The hardware system structures are part of the invention because they are integrated in a special way with variations of what is referred to as a "core processor" signal processing methodology. The basic elements of these five embodiments are depicted in Figs. 1-5. Broadly speaking, the five embodiments can be described as: airborne systems; Telecommunication and personal systems; autonomous signal processing systems; encryption/decryption systems; and signal signature recognition, detection and reception systems.

According to the disclosed technology, a desired accurate anti-noise signal is generated, used and applied across the entire signal spectrum in real time. The system/algorithm is flexible, allowing for higher or lower resolution and control in accordance with the implementation requirements (or, as far as is practically possible, taking into account the limitations of the cost of computing power or other limiting factors imposed on the development engineer implementing the invention). The integration of this versatile and efficient technology into specific hardware and software structures facilitates use in a wide range of applications. The inventors conditionally classify them into five areas based on the structures of the systems that were presented: airborne systems; telecommunication and personal systems; autonomous signal processing systems; encryption/decryption systems; and signal signature recognition, detection and reception systems. Obviously, this list should be considered as potentially possible for systems, since this list is not intended to limit the scope of the present invention.

In addition to the beneficial effect across the entire audio spectrum, the disclosed technologies can also be used for electromagnetic signals. Thus, the disclosed technologies are capable of suppressing noise in virtually any frequency range in an autonomous mode and, at least, across the entire audio spectrum in real time using currently commercially available processors. It is expected that, as processor speeds increase or by combining the power of several processors, the process of real-time processing of any electromagnetic signal will be practically realized according to the present invention.

The algorithm processes individual frequency segments individually, calculating the ideal anti-noise signal for the system or application, significantly improving noise cancellation performance across the entire audio spectrum. In fact, the algorithm can successfully cancel noise across the entire audio spectrum in stand-alone and signal processing applications. It is also more effective across the entire audio spectrum in headphones and radio systems, and can handle higher frequencies than any other deployed system. Processing discrete frequency segments (and allowing grouping of frequency ranges or bands, as described below) allows the algorithm to be tuned to work highly effectively for any particular implementation, inside or outside the audio spectrum.

Discrete frequency segment processing allows generation of an anti-noise signal for dynamic noise sources that change rapidly over time. (Current commercially available methods are limited to cyclic stationary type audio signals, such as engine noise.) Discrete frequency segment processing also reduces the need for multiple input microphones in headphones/earphones.

In the case of audio, the algorithm also reduces the number of microphones required for microphone noise cancellation and the need for complex "beamforming" algorithms to identify the desired speech signal from ambient noise. This aspect is particularly relevant to telecommunications headsets, because the anti-noise signal generated for headphones must also effectively suppress unwanted signals when added to the microphone input signal with a slight delay adjustment (possibly using passive components to provide the required delay). If desired, feedback can be adapted and stored in the presets.

Calibration mode reduces the need for expensive system tuning for different physical systems, both during product development and mass production.

The use of frequency bands or ranges in algorithm versions provides many advantages, including:

I. Reduce the amount of computing power and memory required;

II. Ensuring maximum system performance for specific applications with minimal latency and reduced complexity;

III. The ability to create and deploy presets for different types of noise, environmental sounds, etc.;

IV. The ability to use the algorithm to enhance the clarity of specific signals in ambient noise. This allows the algorithm to be used in hearing aids (e.g., better distinguishing speech in a noisy restaurant), in audio surveillance applications (analyzing speech from ambient noise), for recognizing device signatures on a network or in a noise field, or in encryption/decryption applications.

Additional areas of applicability will become apparent from the description provided herein. The description and specific examples herein are intended for illustrative purposes only and are not intended to limit the scope of the present invention.

Brief description of the drawings

The drawings described herein are intended only for illustrative purposes of selected embodiments and do not represent all possible implementations and are not intended to limit the scope of the present invention.

Fig. 1 is a block diagram of a first embodiment of a muffler device useful in providing noise attenuation or noise suppression in a system in an air environment.

Fig. 2 is a block diagram of a second embodiment of a jammer device useful for providing noise attenuation or noise suppression in a telecommunications microphone, telecommunications headset, or earphone/headphone system.

Fig. 3 is a block diagram of a third embodiment of a muffler device useful in providing noise attenuation or noise suppression in a signal processing system;

Fig. 4 is a block diagram of a fourth embodiment of a jammer device useful for encrypting and decrypting confidential information.

Fig. 5 is a block diagram of a fifth embodiment of a jammer device useful for suppressing noise in electromagnetic transmission and for separating signatures of specific equipment or messages from background noise of power lines (for use in power grid communication technology and smart grid applications, for example).

Fig.6 is a block diagram illustrating a method for programming a digital processor circuit to execute an algorithm of a core processor of a system used in a jammer device.

Fig. 7 is a flow chart further illustrating a method for programming a digital processor circuit to execute an algorithm of a core processor used in a jammer device.

Fig. 8 is a signal processing diagram illustrating the processing technologies implemented by the core processor algorithm of the system of Fig. 6.

Fig.9 is a detailed signal processing diagram illustrating the calibration mode used in connection with the system core processor algorithm.

Fig. 10 is a diagram of the system core processor process.

Fig. 11 is an exemplary single-unit low-power airborne silencer system configured as a desktop personal quiet zone system.

Fig. 12 is an exemplary single-unit low-power airborne silencer system configured as a window unit.

Fig. 13 is an exemplary single-unit low-power air-filled silencer system configured as a unit mounted in an air chamber.

Fig. 14 is an exemplary high-power multi-block air-to-air muffler system configured to reduce highway noise.

Fig. 15 is an exemplary high-power, multi-block air-to-air muffler system configured to reduce noise in a vehicle.

Fig. 16 is an exemplary high power multi-unit airborne jammer system capable of forming a cone of silence to protect a private conversation from being overheard by others.

Fig. 17 is an exemplary embodiment of the integration of a smartphone.

Fig. 18 is an exemplary embodiment of noise-canceling headphones.

Fig. 19 is another exemplary embodiment of noise-canceling headphones.

Fig.20 illustrates an exemplary implementation of the processor.

Fig.21 illustrates an exemplary embodiment of encryption-decryption.

Fig. 22 illustrates an exemplary signature detection concept.

The corresponding reference numbers designate the corresponding parts in the several drawing views.

Implementation of the invention

A more complete description of exemplary embodiments will now be given with reference to the accompanying drawings.

The disclosed muffler device can be used in various applications. For purposes of illustration, five exemplary embodiments will now be described in detail. It should be understood that these examples provide an understanding of some of the various applications for which the muffler device can be used. Other applications and other implementations are also possible within the scope of the appended claims.

Referring to Fig. 1, a first exemplary embodiment of a muffler device was illustrated. This embodiment is intended to provide noise suppression for a system in an air environment in which incoming environmental noise is detected and a noise suppression signal is generated and broadcast to the surrounding area. As illustrated, this embodiment includes a digital signal processor circuit 10 having an associated memory 12 in which configuration data referred to in this document as presets for applications are stored. The digital signal processor circuit can be implemented using a commercially available integrated circuit of a multimedia processor, such as a quad-core ARM Cortex A53 processor Broadcom BCM2837 or the like. A description of a method for programming the digital signal processor circuit is presented below. In a preferred embodiment, the digital signal processor circuit can be implemented using a Raspberry Pi computer, such as a Raspberry Pi 3 model B or better. This device includes a signal processor circuit 10, as well as a VideoCore IV graphics processor, embedded SDRAM, WiFi and Bluetooth transceiver circuitry, 802.11n wireless LAN circuitry, and Bluetooth 4.1 communication support. Twenty GPIO six ports are provided, as well as USB 2 four ports, an Ethernet 100Base-T port, DSI ports and CSI ports, a 4-pole composite video/audio port, and an HDMI 1.4 port. These ports can be used to provide connectivity between inputs and outputs in the signal processor circuit 10, as shown in the block diagrams of Figs. 1-4.

The airborne noise reduction system of Fig. 1 includes one or more input microphones 14 that are deployed in a physical location where they can detect a noise source that needs to be suppressed. Each of the microphones 14 is connected to a digital-to-audio converter or DAC 16 that converts the shape of an analog signal from the connected microphone into digital data, as by sampling. While various sampling rates can be used depending on the task, in the illustrated embodiment, a frequency of 48 kHz is used as a sampling frequency. The sampling frequency is selected taking into account the frequency range occupied by most of the sound energy of the noise, the distance between the input microphone and the feedback microphone, and other factors related to specific implementations and purposes.

Between the DAC 16 and the digital signal processor circuit 10, an optional gate circuit 18 is connected that passes noise energy above a predetermined threshold and blocks energy below this threshold. The gate circuit 18 may be implemented using software running on the processor or using a stand-alone threshold-type noise suppression integrated circuit. The gate circuit is configured to distinguish between a level of ambient background noise that is not considered unwanted and a higher noise level associated with unwanted noise. For example, if an airborne noise suppression system is deployed to eliminate intermittent road noise from a nearby highway, the gate threshold will be set to open when the sound energy of moving vehicles is detected and to close when the rustling of leaves of trees in a nearby grove is detected. In this way, the gate helps to reduce the load on the digital signal processor 10 and prevent unwanted pumping.

The valve may be user configurable, allowing the user to set a noise threshold so that only sounds above the set threshold are considered noise. For example, in a quiet office, the ambient noise level may be around 50 dB SPL. In such an environment, the user may set the noise threshold so that it only operates for signals with an SPL level greater than 60 dB.

A digital-to-analog converter or ADC 20 is coupled to the output of the digital signal processor circuit 10. The ADC serves as a complement to the DAC 16 by converting the output data of the digital signal processor circuit 10 into an analog signal. The analog signal is a specially designed noise suppression signal designed to suppress noise detected by the input microphone(s) 14. A suitable amplifier 22 and a speaker or transducer system 24 project or broadcast this noise suppression signal into the air where it will mix with and suppress a noise source that is heard from a vantage point within the effective transmission region of the speaker or transducer system 24. Essentially, the speaker system 24 is positioned between the noise source and the listener or receiving point such that the listener/receiver can receive a signal reaching his or her location except that signals from the noise source are suppressed by the noise suppression signal from the speaker or transducer system 24.

If desired, the circuit may also include a white or pink noise source that is fed to the amplifier 22, essentially mixing a predetermined amount of white noise or pink noise with the analog signal from the digital signal processing circuit 10 (via the ADC 20). This noise source helps to soften the effect of the noise-cancelling signal by masking otherwise audible transients that may occur when the noise-cancelling signal is combined with the noise source signal downstream of the loudspeaker.

A feedback microphone 26 is located in front of (downstream of) the loudspeaker system 24. The feedback microphone is used to sample the signal stream after the anti-noise signal has been introduced into the stream. This microphone provides a feedback signal for the digital signal processor circuit 10, which is used to adapt the algorithm that controls the process of generating an appropriate noise reduction signal by the digital signal processor circuit. Although not shown in Fig. 1, the output signal of the feedback microphone can be processed by suitable amplification circuits and/or an analog-to-digital converter to provide a feedback signal used by the noise reduction algorithm. In some implementations, where only the amplitude of the noise against the noise reduction signal is used, the output signal of the feedback microphone can be processed in the analog domain to obtain the amplitude of the signal voltage, which can be processed by averaging or other means, if necessary. In other implementations, where a more accurate estimate of the noise against the noise reduction signal is required, a phase comparison with the input signal of the microphone can also be performed. In most implementations of this embodiment of the system, both the phase and amplitude of discrete segments of the feedback signal will be analyzed in comparison with the microphone input signal or the desired result for the application (discrete segment generation and processing is discussed later in this document). The output signal of the feedback microphone may be sampled and converted to the digital domain using an analog-to-digital converter. The feedback microphone may be connected to either a microphone input or a line input connected to a video/audio port of the digital signal processor circuit.

Additional examples of the noise reduction system shown in Fig. 1 are given in the section “implementation variants for various use cases”.

A second embodiment of the jammer device is shown in Fig. 2. As will be explained, this embodiment has two signal paths: a received audio signal path that reduces noise in the user's headphones, earphones or loudspeakers 24a; and a transmitted audio signal path, where sounds captured by the microphone of the telephone 34 are processed to reduce environmental noise, also captured by the microphone of the telephone 34. Thus, the received audio signal path will be used to improve the quality of the sound in the user's headphones or through the loudspeakers, reducing or eliminating environmental sounds. This makes it easier to listen to music or talk on the phone. The transmitted audio signal path will be used to partially or completely suppress environmental sounds, such as wind noise, entering the microphone of the user's telephone 34. Of course, the same noise reduction methods can be used with other systems, not just with telephones, including recording sound through a microphone during concerts, broadcasting sound using microphones, and the like.

With reference to Fig. 2, the illustrated exemplary embodiment is adapted for use with a headset system and uses some of the same components as the embodiment of Fig. 1. In this embodiment, the input microphone is implemented using one or more noise-sensitive microphones 14a located outside the headset or headphones/earphones. An analog-to-digital circuit associated with or built into each noise-sensitive microphone converts ambient noise into digital signals that are fed to the digital signal processing circuit 10. The feedback microphone 26 is located inside the headphones/earphones, in an audio connection with the loudspeaker(s) 24a of the headset, or the feedback microphone can be completely eliminated, due to a more precisely controlled physical embodiment. In systems that include a talkback microphone, the talkback microphone data includes components of a composite signal that may include useful entertainment content (e.g., a music or audio/video soundtrack) and/or a speech signal, plus noise and anti-noise, and that may be compared to a noise and anti-noise signal, or a composite input and useful signal.

It should be noted that this system is significantly different from conventional systems, as the ability to effectively suppress sounds with frequency components above 2000 Hz reduces the need for acoustic isolation methods used in conventional noise-canceling headphones. This allows for lighter and less expensive products and facilitates efficient deployment in the form of an "earbud".

In this embodiment, the noise processing of the muffler device can be implemented independently for each earphone. For cheaper earphone/mini earphone/headset products, the processing can be implemented jointly for both earphones in a stereo system, or the processing can be performed by an external processor located in a smartphone or other device.

Another important difference in this embodiment is that a calibration mode will also be used to calculate the appropriate amplitude adjustments required for each frequency range band to compensate for the effect that the physical properties of the earphone or earbud design have on ambient/unwanted noise before it reaches the ear (the calibration mode and frequency ranges are discussed later in this document).

Similarly, the anti-noise signal generated by the system in this embodiment is mixed with the desired music or voice signal via the mixer 30 (the desired music or voice signal is supplied by the telephone 30, a music player, a communication device, etc.), and both will often be output together via a common speaker. In this case, the feedback microphone will only operate during the calibration mode and can be omitted in production devices under certain circumstances. Alternatively, the feedback microphone can function continuously in certain applications with multiple speakers of this embodiment (for example, in virtual reality headsets and gaming headsets).

In headset systems that include a speech microphone, the output signal of the digital signal processor circuit 10 can be fed to the speech microphone circuit as well as to the headphone loudspeaker circuit as described above. This is illustrated in Fig. 2, where a first output signal is fed from the digital signal processor circuit 10 to a first mixer 30 that supplies audio to the sound reproduction system (amplifier 22 and headphone loudspeaker 24a). A second output signal is fed from the digital signal processor circuit 10 to a second mixer 32 that supplies the signal to a telephone 34 or other voice processing circuit. Since the background noise signal is sampled at a point (points) remote from the position of the communication/voice microphone, the useful voice signal will not be suppressed. This alternative anti-noise signal may or may not have frequency band amplitudes adjusted in a calibration mode as described above, depending on the application.

For mission-critical communications applications such as multimedia broadcasting or military fighter communications, a separate signal processing circuit may be needed for microphone noise suppression. This would allow for precise suppression of known noise signatures, provide the ability to never suppress certain critical information frequency bands, and facilitate other customization for these mission-critical applications via user configuration or presets.

Further examples of the noise reduction system of Fig. 2 are given in the section “various use cases” below.

A third, more general embodiment is shown in Fig. 3. In this embodiment, the input signal may be obtained from any source and the output signal may be fed to a circuit or device (not shown) that typically processes the input signal. Thus, the embodiment shown in Fig. 3 is intended to be placed within or on-line with a signal processing or transmission device. In this embodiment, feedback will typically not be used. Known noise characteristics may be compensated for by system parameters (specified by settings labeled "presets" 12). If these characteristics are unknown, the system may be calibrated to suppress noise by processing a portion of the material that has noise but does not contain a signal (for example, the system may be calibrated for a noise signature using a "preview" portion of a video segment). Applications of this embodiment include removing noise from event recordings or surveillance audio, or removing them in a "live" situation with an appropriate amount of "broadcast delay", such as the "7-second" delay currently used to censor profanity during live broadcasts.

Although the above examples have illustrated a stand-alone digital signal processor circuit 10, it should be understood that a processor in a mobile device, such as a smartphone, may be used to execute signal processing algorithms; or the system core processor may be implemented as an "include" for a software package or integrated into another signal processing system. Thus, the description in this document should also use the term "system core processor" to refer to signal processing algorithms, as described more fully below, that are executed on a processor, such as a stand-alone digital signal processor circuit, or a processor integrated into a smartphone or other device. As shown in the embodiment of Fig. 3, the system core processor may be used to process noise in a stand-alone mode or in signal processing applications where an input microphone, output amplifiers and feedback microphones may not be required.

Additional examples of the noise reduction system of Fig. 3 are described below in the section “implementation variants for various use cases”.

A fourth embodiment is shown in Fig. 4. In this embodiment, there are two parties wishing to exchange information with each other via files, broadcasts, signal transmissions, or other means; and to restrict access to this information. This embodiment requires that both the "encoding side" and the "decoding side" have access to equipment containing this invention.

The encryption/decryption "key" is a means of setting the frequency range used to encode the information. The encryption and decryption "key" settings will be made taking into account the characteristics of the noise or other signal into which the information is to be embedded, and these frequency range settings include both frequency and amplitude information. This allows encoded signals to be embedded into very narrow segment(s) of the transmission, which may appear, for example, to be a harmless broadband transmission of some other material or a "white noise" signal. In an example involving intelligent transmission of what appeared to be white noise, encryption would require recording the appropriate length of white noise to carry the full message. This white noise recording will be processed by this invention only for a very narrow frequency range or set of ranges (i.e. a very narrow segment(s) of noise will be "cut out" by the core processor of the system), with frequencies not contained in that specific set of ranges being transmitted through the system unchanged (the default amplitude will be set to 1 for missing frequencies in the bandwidth definitions on the encryption side); and the amplitudes for the frequency range(s) that will contain the data that can be adjusted to facilitate masking in the noise. The information to be exchanged can be encoded in a "signal carrier" in the narrow frequency range(s) using frequency modulation or other methods (providing another level of encryption) that will be mixed with the output signal of the system (the noise with the segments "cut out"). This will effectively embed the information in what may appear to be a random white noise signal in this example (or another type of wideband transmission or signal transmission, if desired).

On the "decryption" side, it is necessary to know the frequency and amplitude settings for the frequency band range(s) describing where the information was recorded, to serve as the "decryption key". On the decryption side, the default amplitude for frequencies missing from the frequency band definition should be set to 0. This means that the system will not output any output for frequencies missing from the frequency band definition, so that only the useful signal to be decoded is output.

The ability to select a default amplitude for frequencies not present in certain frequency band ranges is one of the defining characteristics of this embodiment.

The security of the transmission of information is greatly enhanced if the encryption/decryption "key" is shared between the parties via an alternative means, but it may be contained in a "calibration header" in the transmission or file, calculated from a timestamp or other setting, etc.

Additional examples of the noise reduction system of Fig. 4 are given below in the section “implementation variants for various use cases”.

A fifth embodiment is shown in Fig. 5. In this embodiment, the invention is used to assist in recognizing, detecting or receiving transmissions or signatures of a device in a noise area, such as noise fields present in saturated electromagnetic fields in large cities and other areas that are inherent in power lines, etc.

The invention may be used to assist in detecting, recognizing or receiving signals in such noise by forming presets for bandwidth settings designed to pass only the target signal. These bandwidth settings include frequency and amplitude information necessary to identify "fingerprints" or "signatures" of target signals from background noise characteristics as determined by a preliminary analysis. This setting will be accomplished by excluding frequency components of the target signal from the bandwidth settings and using a default amplitude of 0 for frequencies not present in the bandwidth settings, effectively passing only the target signals through the system; and appropriately adjusting the amplitude and frequency of adjacent frequencies or harmonics to further enhance the target signal. This would help detect weak signals that would otherwise be undetected in the noise field.

For example, when an air conditioning compressor is turned on, it transmits a unique pulse to the grid. A power company substation can use the present invention in a system to help them anticipate peak loads by counting pulses from various products. In power line communication applications, the characteristics of "normal" noise and fluctuations can be minimized and useful communication signals can be enhanced by using presets designed for this task. Presets can also be designed to detect or improve distant or weak electromagnetic coupling. Likewise, presets can be designed to detect interference in a noise field identified with certain types of objects or other potential threats.

In this embodiment, multiple instances of a system core processor (or other multi-core or multiplexed device) may be deployed on a server to facilitate recognition, detection, or reception of multiple types of signals or signatures on a single node.

Additional examples of the noise reduction system shown in Fig. 5 are described below in the section “implementation variants for various use cases”.

Overview of the system core processor noise suppression algorithm

The essence of the noise reduction algorithm of the core processor of the system is to form an ideal anti-noise for a set of small discrete segments containing a noise signal. The digital signal processing circuit 10 (whether implemented as a stand-alone circuit or using the processor of another device, such as a smartphone) is programmed to execute a signal processing algorithm that accurately generates a specialized set of noise-reducing signals individually for each of the set of discrete frequency segments containing the target noise signal, or a part thereof.

Fig. 6 shows the basic architecture of the noise reduction algorithm of the core processor of the system. As shown, the received noise signal 40 is divided into different frequency band segments. In the presently preferred embodiment, the width of these segments (and in some embodiments, the amplitude scaling factor applied to the anti-noise) in the different frequency ranges can be set differently for each of the different frequency bands 42. These parameters for the frequency band can be set via the user interface, presets, or dynamically based on criteria in various embodiments. Each frequency band range is then further divided into frequency band segments of a selected width. Then, for each frequency band segment, the digital signal processing circuit shifts the phase of this segment by an amount depending on the selected frequency of the segment noise signal. For example, the selected frequency can be the center frequency of the band segment. Thus, if a particular band segment extends from 100 Hz to 200 Hz, the selected center frequency can be 150 Hz.

By segmenting the incoming noise signal into a plurality of different frequency segments, the digital signal processing circuit is able to adapt the noise reduction algorithm to the specific requirements of a given application. This is done by selectively controlling the size of each segment in accordance with the specific application. As an example, each segment in the entire frequency range of the incoming noise signal may be quite small (e.g., 1 Hz). Alternatively, different parts of the frequency range may be subdivided into larger or smaller segments, using smaller segments (with a higher resolution) where the most important information content is located, or if short wavelengths require it; and using larger (with a lower resolution) segments at frequencies that carry less information or have longer wavelengths. In some embodiments, the processor not only subdivides the entire frequency range into segments, but may also separately control the amplitude in a given segment differently based on settings in the frequency bands.

When extremely high noise reduction accuracy is required, the noise signal is divided into small segments (e.g., 1 Hz or other sized segments) across the entire spectrum or across the entire spectrum of the noise signal, as appropriate. Such fine-grained segmentation requires significant computing power. Thus, in applications where lower power and a cheaper processor are required, the noise reduction algorithm of the core processor of the system is configured to divide the signal into frequency bands or ranges. The number of frequency bands can be adjusted in the program code of the core processor of the system according to the needs of the application. If desired, the digital processor can be programmed to divide the received noise signal by applying wavelet decomposition to divide the received noise signal into different frequency band segments and thus generate a plurality of segmented noise signals.

For each specific application, the segment size and how/if the size will vary across the spectrum is an initial condition of the system, determined by defining parameters for the different frequency ranges. These parameters can be set via the user interface and then stored in memory 12 as presets for each application.

After the noise signal has been segmented according to a segmentation plan established by the digital signal processing circuit (automatically and/or based on a user configuration), phase correction is selectively applied to each segment to form a segment waveform that will substantially suppress the noise signal within the frequency band of that segment through destructive interference. In particular, the processing circuit calculates and applies a frequency-dependent delay time 46, taking into account the frequency of the segment and taking into account any system spread or delay time. Since this frequency-dependent delay time is calculated and applied individually to each segment, the processing circuit 10 calculates and applies these phase correction values in parallel or very quickly in series. The segmented phase-corrected noise signals (shifted in phase) are then combined in step 48 to form a composite anti-noise signal 50, which is then output to the signal stream to attenuate the noise through destructive interference. As shown in Fig. 6, the anti-noise signal may be introduced into the signal stream via a loudspeaker system with an amplifier or other converter 24. Alternatively, in certain applications, the anti-noise signal may be introduced into the signal stream using a suitable digital or analog mixer.

In some embodiments, the noise suppression may be further improved by using a feedback signal. Thus, as shown in Fig. 6, the feedback microphone 26 may be located in the signal stream, downstream from where the noise signal was introduced. In this way, the feedback microphone detects the results of destructive interference between the noise signal and the anti-noise signal. The feedback signal obtained from the feedback microphone is then fed to the processing circuit 10 for use in adjusting the amplitude and/or phase of the anti-noise signal. This feedback processing is illustrated as a whole in step 52 in Fig. 6. The feedback processing 52 comprises converting the feedback microphone signal into a suitable digital signal by means of an analog-to-digital conversion, and then adjusting the amplitude and/or phase of the anti-noise signal for maximum noise suppression, using the feedback microphone signal as a reference. When the anti-noise signal and the noise signal interfere destructively in an optimal manner, the feedback microphone signal will detect zero because the noise energy and the anti-noise energy optimally cancel each other out.

In one embodiment, the amplitude of the combined anti-noise signal 50 may be adjusted based on the feedback microphone signal. Alternatively, the amplitudes and phases of each frequency band segment may be adjusted individually. This may be done by comparing the amplitude and phase of the signal flow at the feedback point with the input signal and adjusting the anti-noise parameters. Alternatively, the frequency content and amplitude of the feedback signal itself may be examined to indicate adjustments needed for the noise reduction parameters to improve the results by fine-tuning the frequency-dependent delay time 46 and amplitude for each segment.

Definition of frequency-dependent time delay

The signal processing circuit 10 calculates a frequency-dependent time delay for each segment taking into account a number of factors. One of these factors is a calculated 180-degree phase shift time that is associated with a predetermined frequency (e.g., a segment center frequency) for each individual signal segment.

This calculation can be performed in calibration mode and stored in a table in memory 12 or continuously recalculated in real time, depending on the application and the available computing power. The exact time delay required to form the appropriate anti-noise, for each frequency "f", is calculated by the formula: (1/f)/2. That is:

where f is a predetermined frequency (e.g. center frequency) for that segment.

Another factor used by the signal processing scheme is the system offset time, which in turn depends on two factors: the air propagation time and the system propagation time.

In order to generate accurate noise cancellation signals, the processing circuit uses a priori knowledge of the speed of sound propagation in air, measured as the time it takes for a signal to travel from an input microphone to a feedback microphone. In this description, this travel time is referred to as the air travel time. The processing circuit is also based on a priori knowledge of the time it takes for the processor 10 and the associated input and output components (e.g., 14, 16, 18, 20, 22, 24) to generate a noise cancellation signal, referred to herein as the system travel time. This data is necessary to ensure accurate phase matching of the noise cancellation signal with the noise signal in order to ensure ideal cancellation results. The speed with which the noise signal travels through air depends on many physical factors, such as air temperature, pressure, density, and humidity. The processor computation time and the circuit throughput time depend on the processor speed, the speed of the memory access bus 12, and the signal delays in the input/output circuits associated with the processor 10 in various embodiments.

In a preferred embodiment, these airborne and system propagation times are measured and stored in memory 12 during a calibration mode. The calibration mode may be manually requested by the user via the user interface, or the processor 10 may be programmed to automatically perform calibration periodically or in response to measured air temperature, pressure, density, and humidity.

Thus, the preferred embodiment measures the propagation time in air from the moment the noise signal is detected at the input microphone 14 until it is subsequently detected at the feedback microphone 26. Depending on the application, the two microphones may be located at a constant fixed separation distance (for example, in the embodiment of the headset of Fig. 2), or they may be located at a separation distance that depends on where the two microphones happen to be in the field. The time delay, related to the time required to process the input signal, output to the loudspeaker system 24 (24a) and received at the feedback microphone 26, corresponds to the propagation time of the system.

After the air and system propagation times are measured and stored in the calibration mode, the signal processing circuit 10 calculates the system offset time as an arithmetic difference between the air propagation time and the system propagation time. This difference calculation can also be either calculated in real time or stored in the memory 12. In some fixed applications, such as headphones, an integrated calibration mode may not be required, since the calibration can be performed on the production line or established based on known fixed geometric dimensions of the headphones. The system offset time can be stored as a constant (or dynamically calculated in some applications) for use in the noise reduction calculations described in this document.

For each discrete frequency segment to be processed, an anti-noise signal is generated by delaying the processed signal by a time equal to the absolute value of the 180-degree phase shift time for that discrete frequency segment minus the system offset time. This value is referred to here as the applied time delay. The applied time delay for each frequency segment can either be stored in a table or continuously calculated in different implementations of the algorithm.

Fig. 7 illustrates in more detail the manner in which the signal processing circuit may be programmed to implement the noise reduction algorithm of the core processor of the system. The programmed process begins with a series of steps that populate a set of data structures 59 in the memory 12, where the parameters used by the algorithm are stored for access as needed. Additional details of the core processor process of the system are also discussed below with reference to Fig. 10.

Referring to Fig. 7, first a record containing a selected fragment size is stored in the data structure 59. The fragment size is the length of the time segment to be processed in each iteration performed by the system core processor, as represented by the number of samples to be processed as a group or "chunk" of data. The fragment size depends mainly on the application (the frequency range to be processed and the required resolution), the propagation time of the system, and the transit time between input and output for noise signals that are transmitted over the air or otherwise (the processing must be completed and the anti-noise injected into the signal stream before the original signal passes the anti-noise exit point).

For example, for a system in an air environment processing the full audio spectrum, the distance between the input and output microphones is 5.0 inches, the sampling rate is 48 kHz, and the system propagation time is 0.2 ms; a fragment size of 16 would be appropriate (at a sampling rate of 48 kHz, 16 samples are equivalent to ~0.3333 ms in time; and at standard temperature and pressure, sound travels ~4.5” in air in that time). The processor operation can be optimized to efficiently handle the desired fragment size by limiting system calls and state changes to one per fragment.

This fragment size record is typically saved at the very beginning when the noise reduction device is configured for a given application. In most cases, it is not necessary or desirable to change the fragment size record while the kernel processor noise reduction algorithm is running.

The frequency band ranges, the segment sizes in each frequency band range, and the output signal scale factor for each frequency band are set as initial conditions depending on the application and stored in the data structure 59. These parameters can be set in the user interface, used as presets in the system, or dynamically calculated.

The processing circuit then measures and stores in the data structure 59, at step 62, the system propagation time corresponding to the time spent by the processing circuit and its associated input and output circuits to perform the noise reduction processes. This is accomplished by operating the processing circuit in the calibration mode described below, where a noise signal is supplied to the processing circuit, which the processing circuit acts on to generate a noise signal and an output signal. The time elapsed between the input of the noise signal until the output of the anti-noise signal represents the system propagation time. This value is stored in the data structure 59.

Additionally, the processing circuit at step 64 measures and stores in the data structure 59 the air propagation time. This operation is also performed by the processing circuit in the calibration mode described below. In this case, the processing circuit switches to a mode in which it does not generate any output signal. The elapsed time between the reception of the signal at the input microphone and the reception of the signal at the feedback microphone is measured and stored as the air propagation time.

The processing circuit then calculates and stores in the data structure 59, at step 66, the system offset time, defined as the air propagation time minus the system propagation time. This value is needed later when the processing circuit calculates the applied time delay.

With the above mentioned calibration parameters calculated and stored in this way, the system core processor noise suppression algorithm can now perform pre-calculations for a specific segment (alternatively, these calculations can be performed in real time, provided that there is sufficient computing power available).

As shown, step 68 and the subsequent steps 70 and 72 are performed in parallel (or in rapid succession) for each segment in accordance with the bandwidth settings. If there are 1000 segments for a given application, then steps 68-70 are performed 1000 times, preferably in parallel, and the data is stored in the data structure 59.

At step 70, the phase shift time is adjusted by 180 degrees by subtracting the previously stored system offset time for each segment. The processor circuit calculates and stores the absolute value of this value as the applied time delay - thus, the applied delay time is a positive number representing the amount of phase shift that should be applied to the corresponding segment.

The system core processor uses this stored data to process the frequency segments more quickly (by applying pre-calculated time shifts for all frequency segments in advance). The processor circuit at step 72 performs a phase shift of the noise segment signal by shifting the noise signal of the segment in time by an amount stored as the applied time delay for this segment. In addition, if an amplitude correction (or phase fine-tuning) is required in accordance with the frequency range settings or the feedback processing 52 (Fig. 6), this correction is also applied here (in some embodiments, the phase shift and amplitude correction may be applied simultaneously, storing the information as a vector). All segments are processed in parallel or in fast serial mode, depending on the system architecture.

Once all segments for a particular fragment have been properly adjusted, the processing circuit at step 74 then recombines all processed segments to generate an anti-noise waveform for output to the signal stream.

To further understand the process performed by the processing circuit, we now turn to Fig. 8, which provides a more physical representation of how the noise signal is processed. Beginning at step 80, a noise signal 82 is obtained. In Fig. 8, the noise signal is depicted as a time-varying signal containing many different frequency components or harmonics.

In step 84, the portion of the noise signal spectrum is divided into segments 86 in accordance with the parameters 59 discussed in connection with Fig. 7. For the sake of illustration, Fig. 7 assumes that the time-varying noise signal 82 was expressed in a frequency domain where the lowest frequency components are allocated to the leftmost side of the spectral graph 86, while the highest frequency components or harmonics are allocated to the rightmost side of the spectral graph. For example, the spectral graph 86 may have a range of 20 Hz - 20,000 Hz, covering the generally accepted full range of human hearing. Of course, the spectrum may be distributed differently depending on the application.

It should be recognized that although the noise signal has been represented in the frequency domain in the spectrum 86, the noise signal is by its nature a time-varying signal. Thus, the amount of energy in each frequency domain segment will vary over time. To illustrate this variation, a waterfall diagram 88 is shown showing how the energy within each frequency segment may vary over time along the vertical axis.

For each segment individually, as in step 90, a frequency-dependent phase shift (i.e., an applied time delay) is applied. To illustrate this, waveform 92 represents the noise frequency in the segment before the shift. Waveform 94 represents the same noise frequency after the system offset time has been applied. Finally, waveform 96 represents the resulting noise frequency after the 180 degree phase shift time has been applied (note that this is for illustrative purposes only - in actual processing, only the applied time delay is applied, which is the absolute value of the 180 degree phase shift minus the system offset time). For this illustration, it is also assumed that no amplitude scaling is required for the segments being processed.

By combining the time-shifted components from each segment, an anti-noise signal 100 is formed at step 98. When this anti-noise signal is output to the signal stream, as at step 102, the anti-noise signal 100 is mixed with the original noise signal 104, causing destructive interference, effectively suppressing or weakening the noise signal. The remaining signal is any information-bearing signal 108 that can be extracted at step 106.

Calibration mode

Fig. 9 illustrates how the processing circuit 10, the input microphone 14, the loudspeaker 24 with amplifier and the feedback microphone 26 can be used to perform calibration by selectively using the algorithm of the system core processor when performing the measurement.

In a currently preferred embodiment, the air travel time is calculated with the noise reduction system of the core processor of the system and the output power turned off. In this state, the air travel time is calculated as the time difference between the moment when the input noise is detected at the input of the microphone 14 and the time when the noise is detected at the feedback microphone 26. The system travel time is measured using the noise reduction system of the core processor of the system. The same input is again input into the input microphone 14. This time it is processed by the core processor of the system and output through a loudspeaker (for example, the loudspeaker 24 or another suitable calibration loudspeaker or transducer) placed in front of the feedback microphone 26. During the processing of the input signal in the core processor of the system, its frequency can be changed such that the output pulse differs from the input impulse noise (or the time/phase of the two signals can be used to distinguish the output signal of the system from the original noise). Both the signals propagating in the air and the signals generated by the system will be fed to the feedback microphone. The propagation time of the system can then be calculated from the time for the input pulse signal to reach the feedback microphone and the time for the output signal to reach the feedback microphone.

It is worth noting that this calibration mode can effectively eliminate a significant amount of engineering time required to “tune” the system to account for minor variations among the microphones used or the physical geometry of the headphones or noise-canceling headphones. This can result in significant cost savings in product development. The calibration mode also eliminates the physical inconvenience of setting up individual sets on the production line due to tolerances and manufacturing variations in individual components (especially in microphones), providing an automated way to initially tune. This aspect represents another significant indication of production cost savings.

Using the system offset time, the processor calculates the segment-specific time delay that will be applied to each segment to generate the exact antinoise required for the segment. To calculate the exact segment time delay, the processor determines the time required to obtain a 180-degree phase shift at the center frequency of the specific frequency segment and adjusts it for the system offset time. Specifically, the segment time delay is calculated as the absolute value of [the 180-degree phase shift time minus the system offset time].

Once all the anti-noise segment time delays have been calculated, the digital signals for each segment are then time delayed by the calculated amount for that segment, and all the anti-noise segments thus generated are then combined into a single anti-noise signal, which is then output (e.g. to a speaker system).

In embodiments that use a feedback microphone or other feedback signal source, the processor 10 compares the input signal of the feedback microphone with the input signal of the input microphone in both phase and amplitude. The processor uses the phase comparison to adjust the applied time delay and uses the amplitude of the anti-noise signal to adjust the amplitude of the generated noise suppression signal. When adjusting the amplitude, the processor can control the amplitude individually for each segment in a range of frequency ranges (and thus effectively control the amplitude for each segment). Alternatively, the frequency composition and amplitude of the feedback signal itself can be used to determine the necessary adjustments to the amplitude and phase of each segment.

A detailed description of the operating process of the system core processor

Referring now to Fig. 10, a detailed explanation is given of how the signal processing circuit 10 implements the process of operating the core processor of the system. In particular, Fig. 10 shows in detail the software architecture implemented by the signal processing circuit in the preferred embodiment. The user can interact with the processor executing the process of operating the core processor of the system in several ways. If desired, the user can start the configuration mode of the jammer device in step 120. In this way, the user can also possibly configure the frequency bands and the corresponding widths and amplitudes of the segment, as well as the parameter of the gate noise threshold (part of the data structure 59 (see also Fig. 7). The fragment size can also be set as a parameter of the user interface.

Alternatively, the user may simply initiate the jammer device at step 132. In this case, the user may give a command to initiate the process of the system core processor to calibrate the device at step 134. The calibration process causes the software 124 of the system core processor to perform the calibration process. By calling the calibration engine 126, the portion of the software 124 of the system core processor operating in the signal processing circuit performs the calibration process described in detail above, thereby filling the data structure 59 with the propagation time of sound in air, the propagation time of the system and other calculated parameters. These stored parameters are then used by the anti-noise generation engine 128, which is also part of the software 124 of the system core processor. As shown, the anti-noise generation engine 128 supplies signals to the loudspeaker, which then injects an anti-noise signal into the signal stream at step 130.

Whether required as part of a calibration process or as part of a noise reduction process during use, the core processor of the system inputs signals from the input microphone and the feedback microphone, as shown in 136. To reduce noise during use, the user commands the device to suppress noise via the user interface as in step 138. As shown, this causes an anti-noise signal to be input into the signal stream (e.g., into the air 140). If used, white noise or pink noise may also be input into the signal stream in step 142.

To further illustrate the way the signal processing circuit operates according to the system core processor algorithm on sample input data, a table is provided below. The table below shows sample user-defined frequency ranges. As can be seen, the applied time delay can be represented as a floating point number corresponding to the delay time in seconds. As the data shows, typical applied time delays can be quite small, but each applied time delay is precisely calculated for each given frequency segment.

Unlike traditional noise reduction technologies, the core processor's noise reduction algorithm provides exceptional results for frequencies above 2000 Hz and helps achieve good results across the entire audio spectrum (up to 20,000 Hz) and beyond (provided the processing speed is sufficient). Conventional technologies currently in use are only effective up to approximately 2000 Hz and are essentially ineffective above 3000 Hz.

Implementation options for different use cases

The basic technology described above can be used in a wide range of applications. Some of these use cases are described below.

Variants of implementation of the system (audio) muffler in the air environment

The airborne jammer system can be implemented as shown in Fig. 1 and described above. Several different embodiments are possible. These include a single-unit system with low power consumption optimized to provide an individual quiet zone. As shown in Fig. 11, the components shown in Fig. 1 are mounted on a table or in a housing containing a loudspeaker with amplification that is directed at the user. Note that the feedback microphone 26 is located in the sound field of the loudspeaker. It can have a fixed configuration or can use a removable bracket that also functions as a microphone stand so that the user can place the feedback microphone closer to the place where the user is located. Another possible embodiment is to place the jammer system in a smartphone or add it as an application for a smartphone. In this embodiment, the built-in microphones and speakers of the phone, as well as the microphone from the headset, can be used to build the system.

Fig. 12 shows an alternative embodiment where the components of Fig. 1 are placed on a mounting frame adapted to be placed in a window of a room, where the window frame is designated by the letter W. In this embodiment, an input microphone 14 captures external sound and a loudspeaker 24 amplifiedly introduces an anti-noise audio signal into the room. The feedback microphone may be located on a bracket or conveniently located near the user. If desired, the feedback microphone may communicate wirelessly with the processing circuit 10 using Bluetooth or another wireless communication protocol. In this embodiment, the amplitude adjustments for the various ranges of the frequency band determined in the calibration model are likely to be significant (as in the embodiment for headphones) due to how the walls and windows affect the original noise.

Fig. 13 shows another embodiment where the components of Fig. 1 are placed in a camera mounting package adapted for installation in an HVAC system duct or in a fan system duct (e.g., a bathroom ceiling fan, a kitchen fan, an industrial ventilation fan, etc.). In this embodiment, sounds generated by the HVAC system or fan system are sampled by an input microphone 14, and an anti-noise signal is input into the duct system. In this embodiment, if desired, multiple speaker systems can be deployed on ventilation registers throughout the house or building so that the HVAC or fan system noise is further reduced at each location. Such an embodiment with multiple speakers can use separate feedback microphones 26 in each room, for example, next to each register. The processing circuit 10 can supply an amplifier volume control signal individually to each speaker with gain to adjust the sound pressure level of the anti-noise signal for each room. Rooms located closer to the noise source may require higher anti-noise signal gain than rooms located further away. Alternatively, a separate device can be installed in individual registers to provide specific control of conditions in a particular location.

The second class of airborne silencer devices includes high-power multi-unit systems designed to reduce noise from high-energy sources. These include systems to reduce noise generated at construction sites, noise generated along busy streets or highways, and noise generated by nearby airports. These same high-power, multi-unit systems can also be adapted to reduce noise around the perimeter of a school or stadium. High-power, multi-unit systems can also be used to reduce road noise in vehicle cabins (e.g., cars, trucks, military tanks, boats, airplanes, etc.).

Fig. 14 shows an exemplary high-power multi-unit system deployed to reduce noise on a highway. Individual suppressor devices, each implemented as shown in Fig. 1, are arranged such that they intercept highway noise using their respective input microphones 14 and inject sound energy of the anti-noise into the environment so that highway noise is suppressed by destructive interference located at a distance. Alternatively, the suppressor device may be located directly in the yard of a house to provide more direct coverage.

In an embodiment of the noise reduction on the highway, the individual suppressor devices are preferably mounted on vertical posts or other suitable structure so that the speakers are significantly above the head of any people standing nearby. The feedback microphones 26 can be placed at a significant distance from the suppressor device using WiFi or other wireless communication protocol to send feedback information to the processing circuit.

In addition, if desired, the individual muffler units may be wirelessly networked, such as a cellular network or a local area network, allowing the muffler units to exchange information about the local audio input signals and feedback signals received by each muffler unit. In the case of road noise, loud noise sources may be monitored by the input microphones of the respective muffler units. Thus, such as the braking sound of a semi-trailer truck or a motorcycle with ineffective engine mufflers along a section of noise-protected highway, a system of collective muffler units may communicate with each other and adapt their respective noise signals to enhance what would otherwise be possible using individual input and feedback microphones. This implements a form of diversity noise suppression made possible by using two different, mathematically orthogonal information sources: (a) the feedback microphone sources and (b) the collective input microphones shared via a cellular network or a local area network.

Fig. 15 illustrates how a powerful multi-unit system can be deployed in a vehicle, such as an automobile vehicle. Microphones with multiple inputs are located at noise input locations in the cabin. These inputs are processed individually, either using multiple cores of a multi-core processor 10, or using multiple processors 10 (shown here as an icon on the vehicle's infotainment screen), representing the case where the system is factory installed and integrated into the vehicle's electronic system. Since each input signal is processed individually, there is no need for each signal to be segmented in the same way. Indeed, each individual type of noise signal will typically have its own noise signature (tire noise is very different from muffler noise, for example). Thus, each input signal is segmented to best match the frequency spectrum and sound pressure levels at each individual noise location to provide the desired results at each occupant location.

Although special anti-noise speakers can be used in the vehicle interior, it is also possible to use the sound system already present in the vehicle. Thus, in the illustrated embodiment, the processing circuit 10 supplies a stereophonic or alternative surround sound audio signal, which is mixed with the sound coming from the in-car entertainment system. The mixing can be performed both in the digital and in the audio domain. However, in any case, the processing circuit 10 receives a data signal that supplies the processing circuit with information about what volume level the user has selected for the entertainment system. The processor uses this information to adjust the volume of the anti-noise signal so that it properly neutralizes noise sources regardless of what volume level the user has selected for the entertainment system. Thus, when the user increases the volume of the music, the processing circuit 10 reduces the noise signal supplied to the mixer for compensation. The processor is programmed to generate the correct sound pressure levels in the cabin without affecting noise, regardless of how the user sets the sound level for listening to entertainment content.

Another class of airborne jammer system provides an inverted air function. In this type of system, the jammer system is configured inverted to form a "cone of silence" that allows a private conversation to occur in public without others being able to clearly hear what is being said. As shown in Fig. 16, the jammer device is deployed with one or more outward-facing loudspeakers. Input microphones are located in the center of the loudspeaker arrangement, thus placing people in a private conversation on the "noise input" side of the audio stream. In this embodiment, feedback microphones 26 are deployed in locations where third-party listeners (unwanted listeners) cannot easily block these microphones and thus alter the generated noise signal to cancel out this conversation.

Telecommunications microphone, telecommunications headset and personal headphones/earbuds (audio)

The telecommunications/headphone system may be implemented as shown in Fig. 2 and described above. Several different embodiments are possible.

One such embodiment, shown in Fig. 17, is a smartphone application where the input microphone is on the back of the smartphone, the speaker is a smartphone receiver speaker, the system core processor is implemented using the smartphone processor, and the feedback microphone is not used (due to fixed geometry). In this embodiment, the same anti-noise can be added to the microphone transmission. An alternative to this embodiment includes a passive headset connected to a microphone/headphone jack, a flashlight jack, etc. For this alternative to be effective, the microphone on the phone must be exposed to ambient noise, and not located in a pocket, purse, or backpack.

Another consumer-grade embodiment may be noise-canceling headphones, a headset, or earbuds, wherein the core processor processing is performed using a processor and input microphone(s) included in the headset, headphones, or earbuds, as shown in Fig. 18. In this embodiment (as described previously), a common processor may be used for both ears of the stereo system for lower cost/performance products, and separate processors may be used for each ear for higher cost/performance products.

Commercial and military products will likely use faster processors, separate processing for each earpiece, and a separate processor for microphone noise reduction. For the most critical applications, an additional input microphone(s) will be used to capture high intensity ambient noise (due to crowds in a stadium, wind, vehicles, ammunition, etc.), the core processor of the system will be in close proximity to the actual microphone transmission, as shown in Figure 19. For example, Navy Seals on the F470 Combat Rubber Raiding Craft may be able to eliminate the throat microphones they currently use and may benefit from improved communications using this type of system. Likewise, sports commentators will enjoy a smaller, lighter, less intrusive, more "camera friendly" headset design.

Autonomous signal processing (audio)

The stand-alone signal processing system may be implemented as shown in Fig. 3 and described above. In this embodiment, known noise characteristics may be reduced or eliminated from recordings or in a real-time situation with an appropriate delay between the action and the actual broadcast. In this embodiment, the core processor of the system may be a "plug-in" of another editing or processing software system, embedded in another signal processor or as a stand-alone device, as shown in Fig. 20. If noise characteristics are to be removed (or, alternatively, a signal that will be transmitted in an unknown noise environment: excluding these frequencies from the bandwidth settings and setting the amplitude scaling factor to 1 for frequencies not in the bandwidth definitions will allow only these frequencies to pass), the system parameters may be set manually (or via presets) to effectively remove the noise and transmit the target signal. Alternatively, a calibration mode may be used to analyze the noise of the "pre-roll" portion to determine appropriate anti-noise settings. Use cases for this embodiment include removing noise from old recordings, reducing noise in a "live" situation without adversely affecting the quality of the speaker's speech, removing noise from a recording in a surveillance system, enhancing the sound in a surveillance recording, etc.

Encryption/decryption (audio range or out of range)

The encryption/decryption system may be implemented as shown in Fig. 4 and described above. The primary use case here is to transmit private information by secretly encoding it into narrow segments of a broadband transmission or noise signal in a manner that does not significantly affect the broadband signal. Another use case for this embodiment is to include additional data or information about the broadband transmission. As shown in Fig. 21, the encryption "key" will include frequency and amplitude information to "isolate" discrete "channels" in the broadband signal that will not significantly affect the broadband content (e.g., white noise). The encoded signal will be placed on "carriers" of appropriate frequencies by modulation, so that when the "carriers" are added to the broadband content, it will appear "normal" to observers. The "decryption key" will dictate the amplitude and frequency settings for the frequency band ranges that will result in the cancellation of all information except those "carriers" which can then be demodulated and decoded. It is expected that this will be achieved most often by excluding the "carrier" frequencies from the frequency band definitions and setting the default amplitude of the excluded frequencies to 0. The preservation of the "carrier" signals can be further improved by appropriately scaling the amplitude of the anti-noise generated in the immediate vicinity of the "carrier" frequencies as part of the "decryption key" definition.

In an alternative embodiment, the encryption/decryption system may not perform the step of "selecting" discrete channels in the spread spectrum signal. Rather, these discrete channels are simply identified by the processor, based on private a priori information, which portions of the frequency spectra should be selected. Such a priori information about the selected channels is stored in a memory accessed by the processor and is also communicated to the intended recipient of the message in a secret or private manner. The message to be sent is then modulated onto a suitable carrier that contains the message in these discrete channels but is mixed with the noise signal that would otherwise be present. The entire spread spectrum signal (including the discrete channels masked by the noise) is then transmitted. After reception, the spread spectrum signal is processed at the decoding end using a processor that is programmed to divide the spread spectrum signal into segments, identify the discrete channels carrying the message (based on the a priori information about the channel frequencies), and perform a noise reduction operation on the message channels.

Improved signal signature detection, reception or recognition process (audio range and above)

Improving the recognition, detection or reception of a signature can be implemented as shown in Fig. 5 and described above. This embodiment of the core processor of the system facilitates the recognition, detection or improvement of specific types of transmissions or signatures of a device in a noise field. Fig. 22 shows the possibility of recognizing or detecting multiple signatures in a single field of noise or data transmission by deploying multiple instances of the core processor of the system to examine this field. Unlike other embodiments, where a microphone is used to capture an incoming noise signal, in this embodiment, noise produced by a specific device is captured for a predetermined period of time, and then the captured data is processed by calculating a moving average or other statistical smoothing manipulation to develop a noise signature for this device. Depending on the nature of the device, this noise signature can be an audio frequency signature (for example, representing the sound of a fan motor), or it can be an electromagnetic frequency signature (for example, representing radio frequency interference generated by a commutated motor or an electronically controlled motor). The noise signature for that device is then used to generate an anti-noise signal. The noise signature developed in this way can be stored in memory (or in a database for access by other systems) and can be accessed by the signal processing circuitry as needed to reduce the noise of specific devices or classes of devices.

In addition to the ability to reduce noise from specific devices, the database of stored noise signatures can also be useful for identifying devices by the signature of the noise they generate, by setting the system core processor to process only these signatures. One use case includes the ability for energy companies to detect the activation of non-intelligent network products to help anticipate network loads caused by legacy products (HVAC systems, refrigerators, etc.). Another use case would be to detect or enhance distant or weak electromagnetic communications with known characteristics. Alternatively, frequency and amplitude parameters of frequency ranges can be set to detect interference in the transmission field or noise associated with specific incidents, such as: electromagnetic interference that may be caused by an unmanned aerial vehicle or other object crossing a transmission line or noise field, activation of surveillance or counter-surveillance equipment, interference with transmission sources or fields, atmospheric or terrestrial events, etc.

The above description of the embodiments is provided for purposes of illustration and description. This document is not exhaustive or limiting of the present invention. Individual elements or features of a particular embodiment are generally not limited to that particular embodiment, but, where applicable, are interchangeable and can be used in the selected embodiment, even if not specifically shown or described. The same can be changed in many respects. Such changes should not be considered as a departure from the present invention, and all such modifications should be within the scope of the invention.

Claims (53)

1. A method for reducing noise in a signal stream containing a false signal related to noise, comprising the steps of: receive a digitized noise signal from the environment in which there is a signal flow; using a digital processor circuit, the digitized noise signal is divided into a set of different frequency ranges; segmenting at least one frequency range from the set of frequency ranges into a plurality of noise segments; calculating, using a digital processor circuit, a frequency domain representation of the digitized noise signal, wherein the step of calculating the frequency domain representation is performed before or after the steps of dividing and segmenting; performing, using the digital processor circuit, individual shifting of frequency domain representations for one or more noise segments comprising the digitized noise signal in time, wherein the amount of time shift depends on the selected frequency in the frequency domain representation taking into account both the air propagation time of the noise signal and the system propagation time associated with the speed of the digital processor circuit and associated equipment, and the air propagation time is the transmission time for the signal to travel from the input microphone to the feedback microphone; converting the frequency domain representation for each of the noise segments of the digitized noise signal into the time domain to generate an anti-noise signal; and output an anti-noise signal into the signal stream to reduce noise due to destructive interference. 2. The method of claim 1, further comprising the step of segmenting each frequency range in the set of frequency ranges such that the number of noise segments of a given frequency range can vary for frequency ranges in the set of frequency ranges. 3. The method of claim 1, further comprising the step of selectively applying phase correction to one or more noise segments comprising the digitized noise signal. 4. The method according to claim 1, further comprising the step of shifting the frequency domain representation of the noise segments in the frequency range without shifting the frequency domain representation of the noise segments in another frequency range in the set of frequency ranges. 5. The method of claim 1, further comprising the step of selectively applying amplitude scaling to one or more noise segments comprising the digitized noise signal. 6. The method according to claim 1, further comprising the step of performing amplitude scaling in one frequency range without scaling the amplitude in another frequency range of the set of frequency ranges. 7. The method according to claim 1, further comprising the step of converting the anti-noise signal into an analog signal and outputting the anti-noise signal into a signal stream by mixing the anti-noise signal with noise in the signal stream. 8. The method according to claim 7, further comprising the step of mixing the anti-noise signal with noise in the signal stream by using loudspeakers with an amplifier or other converter located in the signal stream. 9. A system for reducing noise in a signal stream containing a noise signal, comprising: a microphone for receiving a noise signal from an environment in which there is an audio signal stream; an analog-to-digital converter for receiving a noise signal from a microphone and operating to convert the noise signal into a digitized noise signal; a digital signal processor for receiving a digitized noise signal from an analog-to-digital converter and processing the digitized noise signal by dividing, using a digital processor circuit, the digitized noise signal into a set of different frequency ranges; segmenting at least one frequency range from the set of frequency ranges into a plurality of noise segments; calculating a frequency domain representation of the digitized noise signal in the frequency domain, wherein the calculation of the frequency domain representation is performed before or after the splitting and segmenting steps; shifting the frequency domain representation of one or more noise segments comprising the digitized noise signal in time, wherein the amount of time shift depends on a selected frequency in the frequency domain representation taking into account both the air propagation time of the noise signal and the system propagation time associated with the speed of the digital processor circuitry and associated equipment, the air propagation time being the transmission time for the signal to travel from the input microphone to the feedback microphone; and transforming the frequency domain representation for each of the noise segments of the digitized noise signal into the time domain to generate an anti-noise signal; and A digital-to-analog converter for outputting an anti-noise signal from a digital signal processor and functioning to convert the anti-noise signal into an analog anti-noise signal. 10. The system of claim 9, wherein the digital signal processor is further configured to segment each frequency range in the set of frequency ranges such that the number of noise segments of a given frequency range can vary for frequency ranges in the set of frequency ranges. 11. The system of claim 9, wherein the digital signal processor is further configured to selectively apply phase correction to one or more noise segments comprising the digitized noise signal. 12. The system of claim 9, wherein the digital signal processor is further configured to selectively apply amplitude scaling to one or more noise segments comprising the digitized noise signal. 13. The system of claim 9, further comprising a mixer configured to receive an audio signal stream and an analog anti-noise signal and to operate to combine the audio signal stream with the analog anti-noise signal. 14. A method for calibrating a noise reduction system implemented in an audio device, comprising the steps of: turn off the noise reduction system; and obtain, using an audio device, a known signal from the environment surrounding the audio device; determine the air propagation time of a known signal when the noise reduction system is turned off, and the air propagation time is the transmission time for the signal to travel from the input microphone of the audio device to the speakers of the audio device; include a noise reduction system; and determine the propagation time of a signal in the system for a known signal in the noise reduction system when the noise reduction system is turned on, wherein the propagation time of a signal in the system is the speed of passage of the digital processor circuit in the noise reduction system. 15. The method according to claim 14, further comprising the steps of: calculate the system displacement time by subtracting the signal propagation time in the system from the signal propagation time in air; and storing the system offset time in a machine-readable memory of the audio device, wherein said machine-readable memory is accessible to the noise reduction system. 16. The method according to claim 14, further comprising the step of outputting, using the noise reduction system, an anti-noise signal to the output of the signal stream via the audio device, wherein the anti-noise signal is created using the offset time of the system. 17. The method according to claim 14, wherein the step of determining the propagation time of a signal in air for a known signal comprises the sub-steps of: determine the time of capture of a known signal by means of an input microphone; determining the signal capture time via a feedback microphone located near the loudspeaker of the audio device; and The propagation time of a signal in air is calculated by subtracting the time of capture of a known signal by the input microphone from the time of capture of the signal by the feedback microphone. 18. The method according to claim 14, wherein the step of determining the signal propagation time in the system comprises sub-steps in which: determine the time of capture of a known signal by means of an input microphone; determining the signal capture time via a feedback microphone located near the loudspeaker of the audio device; and calculate the propagation time of a signal in the system by subtracting the time of capture of a known signal by the input microphone from the time of capture of the signal by the feedback microphone. 19. The method of claim 18, further comprising the step of modifying the known signal after the known signal is received by the audio device but before the known signal is input to the noise reduction system. 20. The method according to claim 16, further comprising the steps of: segmenting a known signal into a plurality of signal segments, wherein each signal segment of the plurality of signal segments is correlated with a different frequency range; calculating, for each signal segment, a segment delay time, wherein the segment delay time is caused by a 180-degree phase shift of the center frequency of the frequency range associated with the given signal segment; calculate, for each signal segment, the applied delay time as the absolute value of the segment delay time minus the system offset time; and store, for each signal segment, the time of the applied delay in the machine-readable memory of the audio device.