JP2006502671A - Input signal electromagnetic processing device - Google Patents

Abstract

The present invention relates to an apparatus, method, and product for correcting electromagnetic waves by reducing noise in the electromagnetic waves. A transfer function for multiple segments is derived by identifying the desired output parameter. The transfer function is applied to the wave via a plurality of filters and other devices.

Description

  The present invention relates generally to electromagnetic signal processing, specifically to noise reduction in electromagnetic signal processing, and more specifically to noise reduction in segmented amplifiers used in RF transmitters (high frequency transmitters). About.

  Until recently, electromagnetic waves have been modified using analog technology. In other words, conventionally, no attempt has been made to individually correct characteristics such as current and voltage, and to convert these characteristics to correct the electromagnetic wave itself. However, recent digitization of wave correction techniques has made it possible to separate and directly correct the characteristics of electromagnetic waves in order to obtain desired results. Digital wave correction technology is a preferred technology because it is faster and more accurate than conventional methods and consumes less power.

  For example, filtering techniques have been improved by digitizing wave characteristics. This is because digital technology has made it possible to quickly and precisely generate and / or modify (process, enhance, separate, filter, etc.) frequencies and other wave characteristics.

  Therefore, if an apparatus, a method, and a product using the characteristics of digitized electromagnetic waves are provided to generate and / or correct electromagnetic waves, it will be useful for electromagnetic wave correction techniques.

  Embodiments of the present invention include devices, methods, and products for correcting electromagnetic waves by reducing electromagnetic noise. A preferred method embodiment determines a predetermined output wave parameter, separates the amplitude and phase characteristics of the input wave, derives a transfer function from the predetermined output wave parameter, and the transfer function is transformed into the amplitude characteristic of the wave. A transfer function application method characterized by being applied. After applying the transfer function, the amplitude characteristic and the phase characteristic can be combined to generate an output wave.

  FIG. 1 shows a preferred embodiment. An input wave a is supplied to the digital signal processing device 10. The digital signal processing apparatus 10 includes an A / D converter 11 that digitizes a wave transmitted using, for example, rectangular coordinates or I / Q data. The rectangular / polar converter 12 receives the I and Q data and converts them into polar coordinates. Here, in other embodiments, the digitized wave may be sent to a quadrature / polar converter as desired, and any known method may be used for digitizing the wave. Also, although this embodiment is described in relation to digitized waves, I, Q data, polar coordinates, other embodiments are not limited to this, and any digital or analog waveform, Alternatively, a combination thereof may be used.

  In the embodiment of FIG. 1, the rectangular / polar converter 12 outputs the digitized wave to polar coordinates. The polar coordinates are represented by, for example, R, P (sin), and P (cos), the R coordinates represent the wave amplitude characteristics, and P (sin) and P (cos) represent the wave phase characteristics. “Characteristics” described herein means electromagnetic wave characteristics such as frequency, voltage, amplitude (including magnitude and envelope (envelope)), phase, current, waveform, or pulse. In other embodiments, one or more characteristics may be derived from the input wave as desired.

Here, FIG. 2 will be briefly described. FIG. 2 shows a schematic diagram of a wave converted according to the embodiment of FIG. The input wave a is converted into an amplitude element m having the amplitude characteristic of the input wave during time t ₁ and a phase element p having the phase characteristic of the carrier wave at the same time. In the figure, a preferred amplified output wave b is shown. In this and other embodiments, this time may be set as desired. For example, to maximize the wave resolution and operating speed, the wave amplitude and phase characteristics may be derived from various sampling rates, and these sampling rates can be changed during operation. May be determined automatically. In the preferred embodiment, the input waves are synchronized in order to increase the accuracy of the output and minimize any distortion.

Returning to the description of FIG. The amplitude and phase characteristics are sent on separate paths. Amplitude characteristics of the input wave is converted to a digital pulse by through converter 13 routes a ^m. The digital pulse is composed of a digital word that is quantized from B _n-1 of the most significant bit (MSB) to B _{0 of the} least significant bit (LSB). In various embodiments, the length of the digital word may be different. In general, the longer the digital word is, the higher the reproduction accuracy of the input wave. As will be described in detail later, the digital word provides attenuation and / or amplification control. Of course, in other embodiments, digital words with different configurations may be used, and similarly, amplitude or other wave characteristics may be derived and / or provided in other ways.

Converter 13 divides the bits, each of which is a time domain square waveform, into separate paths 0-N-1. Each bit enters a low-pass filter bank 30 composed of filters F _{0 to} F _n−1 . As described below, the impulse response of the filter F ₀ ~F _n-1 are each _{h (t) 0 ~h (t} ) N-1.

In the embodiment of FIG. 1, seven control configuration lines 21a-g derived from the converter 13 are shown. In the preferred embodiment, the number of these control components depends on the resolution of the word. Incidentally, FIG. 1, for clarity, the control configuration line has led to integrated into the route a ^m of single control configuration line 22a-g, in the present embodiment, as described later, the control arrangement line Are not actually integrated and are sent individually to the control configuration line.

The phase characteristic passes through the path a ^p . The phase characteristic is first modulated into a wave via a D / A converter 18 and a synthesizer 20 (a voltage controlled oscillator in a particularly preferred embodiment). The synthesizer 20 supplies an output wave composed of phase information. This output wave has a constant envelope (envelopment line) and does not vary in amplitude, but has a phase characteristic of the input wave before conversion. It is done. The waves divided between these driver lines are sent to the current sources 25a to 25g and serve to control the potentials of the current sources 25a to 25g, as will be described in detail later. In other embodiments, a characteristic source other than the phase characteristic may be used.

  In the present embodiment, transistors are used as the current sources 25a to 25g. In other embodiments, one or more appropriately segmented transistors may be used as the current sources 25a-25g. Current sources 25a-25g must not be driven to saturation. This is because, when driven to saturation, the function as a current source is lost and the voltage source is used, so that a desired current combining the current sources cannot be obtained.

(Composed of the above-described control configuration line 21A～g) path a ^m terminates at the control configuration line 22a-g. In a particularly preferred form, the control component line is a switching transistor, preferably a current source, but as will be discussed later, other embodiments may use other wave characteristic sources, An adjustment system may be used. The control configuration lines 22a-g are adjusted by being switched by the bits of the digital word output from the amplitude element. When the bit is “1”, that is, “high”, the corresponding control element is switched on, so that the current flows from the control element via the bias control configuration line 23a-g to the corresponding current source 25a- flows to g. As already mentioned, in various embodiments, the length of the digital word may vary, resulting in different numbers of bits, control components, control components, driver lines, bias control lines, current sources, etc. Also good. Furthermore, the correspondence between the resolution, configuration, line, and current source of the digital word need not be one to one.

  The current sources 25a-g receive current from the control component when the control component is turned on, and each current source is adjusted according to that component. In a particularly preferred embodiment, the control component may be regarded as a bias control circuit and many control components are regarded as a bias network, as will be described later, since the corresponding control component supplies a bias current to the current source. May be. In some embodiments, it may be desirable to distribute one or more bias control circuits statically or dynamically to one or more current sources using a switching network.

  Returning to the description of FIG. Each current source has a role of a potential current source, can generate a current, and is output to each of the current source lines 26a to 26g. Each current source may or may not act as a current source, and therefore may or may not generate current, which is the value of the appropriate digital word that adjusts the control component. Because it is done by. Whether the current source is activated and current is generated from the current source is determined by the bit value obtained by digitizing the amplitude element for adjusting the corresponding control component.

  Here, in a preferred embodiment, the current source is not a single or multiple amplifiers, but is a function of multiple current sources as amplifiers as described herein. In preferred embodiments, amplification and / or attenuation may be viewed as a function of these embodiments, and thus amplification and / or attenuation may be viewed as an electrical component or system that amplifies and / or attenuates.

  The combined current is the sum of the currents output from the current sources 25a-g. Therefore, this embodiment can act as an attenuator and / or an amplifier. Useful output current can be provided because no additional circuitry or elements need to be added between the current sources to combine the current from each current source. Therefore, the combined current b output on the line 27 may be used as an amplifier or an attenuator for applying a load, for example.

  In a preferred embodiment, the current sources differ in current output and magnitude. For this reason, various weightings (weightings) are performed on the current supplied in potential by these current sources. For example, in a preferred embodiment, the size of the first current source is twice that of the next current source, the next current source is twice that of the next current source, and this relationship is the last current source. It continues until. Since the number of current sources matches the number of bits in the digital control word, the largest current source is controlled by the MSB (most significant bit) of the amplitude word, and the next bit in the word has the second largest current source. This control relationship continues until the least significant bit is sent to the current source with the smallest. Of course, as already mentioned, in other embodiments, the bits may be matched to the current source in other patterns, such as by using a switching network. Furthermore, in a particularly preferred embodiment, double current sources of the same size are provided, as are current sources of different sizes. In other embodiments, other wave characteristics may be supplied to other current sources to adjust these current sources.

  In order to determine the transfer function h (t) for the preferred embodiment, the following analysis is used. This analysis, and similar analyses, may be used to derive other transfer functions in various embodiments of the invention.

The output signal S _out from the transistor 25 can be expressed by the following equation.

Here, s (t) represents the input signal, and A represents the gain of the amplifier. Since S _out is composed of a combination of input signals S _i (t), when i = 1. The output signal can be expressed as:

At this time, since W _i represents a weight applied to each signal, the sum of the weights of all signals is equal to s (t).
When the gain is removed from equation (b), s (t) can be expressed by the following equation.

  As already mentioned, in the above preferred embodiment, if each segment is twice the adjacent segment, the weight can be defined as follows:

  Of course, in other embodiments, if the segment weights are different, the value of Wi will change accordingly. Since the signal Si (t) is a time-domain square waveform corresponding to N-bit obtained by quantizing the signal s (t) as described above, if the value of Wi in the equation (d) is put into the equation (c), the input The signal is expressed by the following equation.

Here, Si (t) = LSB (least significant bit), S _N (t) = MSB (most significant bit)
The output signal can also be expressed in polar coordinates as:

  Combining equations (c) and (f) yields:

  The noise of the output signal can be interpreted as a function of the amplitude characteristic and the phase characteristic, and the noise floor (noise level) in the output signal is the noise floor of the amplitude characteristic s (t) and the noise of the phase characteristic θ (t). It can be obtained by folding on the floor. Therefore, in order to reduce the noise floor in the synthesized RF signal, it is possible to reduce both the s (t) and θ (t) noise floors. As mentioned previously, in the preferred embodiment, only the signal s (t) is filtered, but in other embodiments, either or both of the amplitude characteristic s (t) and the phase characteristic θ (t). May be filtered or may be changed in other ways.

The filter value of s (t) is obtained by first _obtaining a desired noise floor in Sout. The synthesized filter signal can be expressed by the following equation.

  Therefore, when equation (g) is inserted into (h), the following equation is obtained:

  From the above it can be seen that each segment can be filtered to achieve filtering of the output signal.

  In other embodiments, once the value of h (t) is determined, it may be performed by methods other than filtering. For example, in other embodiments, h (t) may be executed by summing the partial values of h (t). In the preferred embodiment, for some segments of transistors, such as transistor 25 of FIG. 1, by changing the response speed of each segment to the input signal s (t) 1-s (t) n depending on each segment. , A partial value of h (t) may be incorporated into each segment. By slowing the response to excitation, a given segment acts like a filter with a partial value of h (t). Therefore, the combination of the segment outputs is the partial value of h (t) or the sum of h (t).

  As another embodiment, if you want to filter a digitized wave using the h (t) value, use the appropriate partial value of the given h (t) value and follow the digitized wave. You may filter by several positions. That is, there may be a series of filters based on the wavy lines, or the waves may be separated into a number of filtered spectra.

  Applying h (t) using a partial value of h (t) means applying h (t) through a plurality of signals that add up to an output signal. Therefore, a partial value of h (t) may be applied along a plurality of weighted signals before being summed into an output signal. For example, if the wave is split into one or more characteristics and / or otherwise, the partial value of the desired h (t) value that is relevant to determine the total value equal to the desired h (t) In use, h (t) may be applied to these properties and / or splits.

  It is possible to derive and apply all the partial values of h (t). That is, the partial values obtained through various weighted signals may be applied even if they are not equal.

  FIG. 2 illustrates the far-end noise reduction that can be achieved with the preferred embodiment of the present invention. In FIG. 2, the 7-bit digital segment amplifier of the embodiment of the present invention is used. The noise reduction filter having the h (t) value derived in this embodiment is a fourth-order Butterworth response with a bandwidth of 10 MHz. In this example, a CDMA2000 signal is used as the electromagnetic input signal. As can be seen from FIG. 2, the noise floor level of the filtered signal is reduced.

3 shows a series of screen shots showing the time domain waveform of the bits of the digital word representing the amplitude portion of the electromagnetic signal a ^m is depicted. As can be seen from FIG. 3, when the h (t) filter of this embodiment is used, the end of the rectangular waveform of the bits is rounded, and the transition speed from one state to the next state is slow. As a result, the quantization noise effect generated in the bit on the farthest noise floor of the restored IQ signal is reduced.

  In the preferred embodiment, the line amplitude and / or attenuation can be achieved over a relatively large frequency range, allowing a wideband amplitude change at the associated transmitter. Therefore, you may use this embodiment for transmitters, such as a cell-type communication apparatus. Furthermore, since this embodiment has a relatively low input capacitance (low capacitance) along the phase path, matching requirements are minimized.

  Furthermore, according to an embodiment of the present invention, the linearity of transmission does not depend on the linearity of the amplifier, but depends on how the current is linearly loaded. Is also efficient. Therefore, in order to obtain the maximum effect, each current source can be biased as a non-linear current source such as class B or C. Furthermore, since there is almost no static current drawing to the current source that can no longer be used, the efficiency can be further improved.

  As shown in the explanatory diagram, since the output current mainly depends on the signal drive level, the power control is simple. For example, if the signal drive level is increased or decreased using a variable gain amplifier or attenuator, the output current will increase or decrease accordingly. Further, when the bias for the drive controller is increased or decreased, the output current is also increased or decreased. Of course, the upper limit is determined by the segmentation of the transistor without being increased indefinitely.

  Also, as desired, suitable current sources, such as other transistor segments and / or formats, and other devices and methods may be used in any embodiment of the present invention.

  Various system architectures may be used to construct embodiments of the present invention. Accordingly, an embodiment of the invention or its various components and / or features may be construed as comprising the entire hardware, comprising the entire software, or a combination thereof. If desired, the semiconductor device may be provided with various elements such as an IC and special purpose IC components such as silicon (Si), silicon germanium (SiGe), and gallium arsenide (GaAs).

  The required accuracy level differs depending on the embodiment. For example, in the embodiment, the accuracy of wave digitization varies depending on the length of the digital word. Also, the number of control components, the number of transistor segments, etc. may all be changed as required. Further, in various embodiments, non-linear elements may be used as desired, but in this embodiment it is desirable to use non-linear components in the amplitude path after the signal is digitized.

  Various embodiments may take the form of an entirely hardware embodiment or an embodiment combining software and hardware aspects. Thus, each block or combination thereof in the figure supports a combination of means for performing a particular function and a combination of steps. As is well known to those skilled in the art, the individual blocks or combinations of blocks in the figure may be expressed in a wide variety.

  Although the present invention has been described using embodiments, it is envisioned that additional advantages and modifications will be made to the invention. Accordingly, the invention, which can be more widely applied, is not limited to the details described herein. For example, changes may be made without departing from the technical idea and scope of the present invention. Furthermore, preferred embodiments may include devices and / or methods and / or products specialized for specific input signals, carrier waves, and output signals. For example, RF, microprocessor, microcontroller, and / or computer device, CDMA, CDMA2000, W-CDMA, GSM, TDMA and other mobile phones, Bluetooth, 802.11a, -b, -g, radar, 1xRTT, bidirectional The embodiments may be applied to wireless / wireless devices such as GPRS, computers, computer communication devices, and portable devices such as PDAs. Accordingly, the present invention is not limited to the specific embodiments, but is interpreted based on the appended claims and includes equivalents.

It is a figure which shows preferable embodiment. It is a figure which shows preferable embodiment. FIG. 6 shows various results of a preferred embodiment. FIG. 6 shows various results of a preferred embodiment.

Claims (19)

Obtaining one or more weighted signals;
Identifying a desired transfer function for summing the weighted signals;
A transfer function application method comprising applying a weight value of the transfer function to each of the weight signals. Obtaining one or more weighted signals;
Converting the one or more weighted signals into first and second polar coordinates;
Identifying a desired transfer function for summing the first and second polar coordinates;
A transfer function application method comprising applying a weight value of the transfer function to each of the first and second polar coordinates. Determining the desired output wave parameters;
Separation of amplitude and phase characteristics of input wave parameters,
Deriving a transfer function from the desired output wave parameter;
Applying the transfer function to the amplitude characteristics of the input wave.   4. The method of claim 3, wherein the amplitude characteristic and the phase characteristic are combined to generate an output wave.   The method of claim 3, wherein determining the desired output wave parameter further includes determining an output signal noise parameter.   6. The method of claim 5, wherein deriving a transfer function from the desired output wave parameter further includes deriving an impulse response for the filter.   The method of claim 6, wherein applying the transfer function to the amplitude characteristic of the input wave includes applying the impulse response to the amplitude characteristic through a filter.   The method of claim 7, wherein the filter is a low pass filter.   8. The method of claim 7, wherein the filter is a fourth order batterworth filter. One or more weighted signal acquisition means;
Means for identifying a desired transfer function for summing the weighted signals;
Means for applying a weighted value of the transfer function to each of the weighted signals. One or more weighted signal acquisition means;
Means for converting the one or more weighted signals into first and second polar coordinates;
Means for identifying a desired transfer function for summing the first and second polar coordinates;
Means for applying a weighted value of the transfer function to each of the first and second polar coordinates. Means for determining desired output wave parameters;
Means for separating the amplitude characteristics and phase characteristics of the input wave parameters;
Means for deriving the transfer function from the desired output wave parameters;
Means for applying the transfer function to the amplitude characteristic of the input wave.   The apparatus of claim 12, wherein the amplitude characteristic and the phase characteristic are combined to produce an output wave.   The apparatus of claim 12, wherein the means for determining the desired output wave parameter further includes means for determining an output signal noise parameter.   The method of claim 14, wherein the means for deriving a transfer function from the desired output wave parameter further comprises means for deriving an impulse response for the filter.   16. The apparatus of claim 15, wherein applying the transfer function to the amplitude characteristic of the input wave includes applying the impulse response to the amplitude characteristic via a filter.   The apparatus of claim 16, wherein the filter is a low pass filter.   The apparatus of claim 16, wherein the filter is a fourth order Batterworth filter.   A noise reduction product in electromagnetic signal processing having a transfer function performed by a plurality of filters, the filter receiving a series of weighted signals and executing the transfer function on the series of signals. A noise reduction product in electromagnetic signal processing, characterized by being:

Images