Apparatus for quantized linear amplification with nonlinear amplifiers

Abstract

An apparatus for quantized linear amplification with nonlinear amplifiers that performs a linear amplification of variable-envelope single carrier (SC) or multi-carrier (MC) bandpass signals, based on sampled and quantized versions of its complex envelope, where the quantizer generates N.sub.b bits that are mapped into N.sub.mN.sub.b polar components, in which the quantized symbol can be decomposed, that are modulated as N.sub.m constant or quasi constant envelope signals and where each one is amplified by a nonlinear amplifier.

Claims

1. An apparatus for quantized linear amplification with nonlinear amplifiers comprising: A sampler that generates the time domain samples obtained from the complex envelope of a single carrier (SC) or multicarrier (MC) input signal; at least one quantizer generating the quantization bits that correspond to the quantized values of the time domain samples of the input signal; at least one mapper that generates a set of periodic signals with same frequency of the carrier frequency and different constant amplitudes in which the quantized amplitude value can be decomposed and that converts the quantization bits received from the quantizer into polar sequences that control the phases of the component signals in which the quantized symbol is decomposed; at least a set of amplification branches in parallel with the number equal at least to the half of the quantization bits, each one with a modulator that uses a polar sequence obtained from a quantization bit, a phase shifter and an amplifier that amplifies each one of the signals in which the quantized signal is decomposed; one combiner that combines the amplified signals connected to one transducer or a set of N.sub.m transducers directly connected to each amplification branch that generates an amplified version of the quantized signal.

2. An apparatus for quantized linear amplification with nonlinear amplifiers of claim 1, wherein each sample s(n) of the complex envelope is quantized in N.sub.b quantization bits that correspond to a quantization value s.sub.n,QT taken from a finite alphabet of 2.sup.N.sup.b possible quantization symbols, that is decomposed into the N.sub.mN.sub.b polar components with different constant amplitudes that are each one modulated as a BPSK signal specially designed to have good tradeoffs between reduced envelope fluctuations and compact spectrum (e.g., a GMSK (Gaussian Minimum Shift Keying) signal) and amplified by a set of N.sub.m nonlinear amplifiers.

3. An apparatus for quantized linear amplification with nonlinear amplifiers of claim 1, wherein each sample s(n) of the complex envelope is decomposed Into the in-phase and quadrature components that are quantized as s.sub.n,QT.sup.I and s.sub.n,QT.sup.Q, respectively, by two independent quantizers, each one with N.sub.b.sup.I and N.sub.b.sup.Q quantization bits that correspond to two finite alphabets of 2.sup.N.sup.b.sup.I and 2.sup.N.sup.b.sup.Q of possible quantization symbols, that are decomposed into N.sub.m.sup.IN.sub.b.sup.I and N.sub.m.sup.QN.sub.b.sup.Q polar components with different constant amplitudes by two mappers and amplified by N.sub.m.sup.I and N.sub.m.sup.Q amplifiers, respectively.

4. An apparatus for quantized linear amplification with nonlinear amplifiers of claim 1, wherein the mapping of the quantized bits can be done by a mapper that makes directly the conversion of the quantization bits into a set of BPSK signals with different constant amplitudes or other constant envelope signals and uses the quantization bits to control the phases of the constant envelope signals components.

5. An apparatus for quantized linear amplification with nonlinear amplifiers of claim 1, wherein the mapping of the quantized bits can be done by mappers that make directly the conversion of dibits into a set of QPSK (Quadrature Phase Shift Keying) or Offset-QPSK components with different and constant amplitudes or other constant envelope components, that are separately amplified.

6. An apparatus for quantized linear amplification with nonlinear amplifiers of claim 1, wherein each sample s(n) of the complex envelope is quantized in N.sub.b quantization bits that correspond to a quantization value s.sub.n,QT taken from a finite alphabet of 2.sup.N.sup.b possible quantization symbols, where each dibit of the set of N.sub.b/2 dibits is modulated by a OOSPK signal or a MSK signal with constant amplitude or other signal specially designed to have good tradeoffs between reduced envelope fluctuations and compact spectrum (e.g., a GMSK signal) and amplified by a set of N.sub.m/2 nonlinear amplifiers.

Description

BRIEF DESCRIPTION OF DRAWINGS

(1) The various aspects of embodiments disclosed here, including features and advantages of the present invention outlined above are described more fully below in the detailed description in conjunction with the drawings, where like reference numerals refer to like elements throughout, in which:

(2) FIG. 1 shows a diagram of the linear power amplification apparatus' structure where the input signals are sampled in 100 and the resulting quantization bits of 101 from the quantization of the signal's time domain samples are subsequently converted into N.sub.m antipodal format by the polar decomposition converter 102. These antipodal are mapped into N.sub.m polar sequences in 103 that are the inputs of BPSK modulators 104, whose outputs are continuous time BPSK signals that are the inputs of the phase shifter 105, and amplified by N.sub.m nonlinear amplifiers 106. The N.sub.m amplifiers' outputs are the inputs of 108. 108 can be implemented as a combiner followed by a transducer that can be an antenna or a laser diode or a hydrophone or as a set of N.sub.m transducers that can be a set of N.sub.m antennas, N.sub.m hydrophones or N.sub.m laser diodes, where each transducer is directly connected to each amplification branch. Depending on 108 structure the N.sub.m amplifiers' outputs can be summed by a combiner before being transmitted to the channel or can be directly transmitted to the channel by a set of N.sub.m transducers or other type of transducers, depending on the transmitting channel that is being considered.

DETAILED DESCRIPTION OF THE INVENTION

(3) The present application describes a linear amplification apparatus using a quantizer combined with a decomposition of the quantized symbols into N.sub.m polar components that are amplified individually by a nonlinear amplifier. Referring to the figures, it will now be described technology using different embodiments of the same technology, which is not intended to limit the scope of protection of this application. The embodiments are composed by a method of sequential steps as described below.

(4) The apparatus for linear power amplification with nonlinear amplifiers uses as input for the quantizer the time domain samples of the complex envelope that can be obtained from a MC or SC signal. The values of the samples of variable envelope signals are quantized by N.sub.b quantization bits that are converted into polar components in which the quantization symbol is decomposed as the sum of several polar components [37]. Each one is modulated as a serial OQPSK signal [38] with reduced envelope fluctuations or as constant envelope signal before being amplified by a separate amplifier.

(5) The basic structure of the apparatus for linear power amplification with nonlinear amplifiers considered in this application is depicted in FIG. 1. The input signals of the power amplification apparatus are sampled in 100 to generate the time domain samples s(nT.sub.s)=s(n).

(6) In 101, by performing the amplitude quantization of the time domain samples, the amplitude of each sample s(n) is transformed into a quantized symbol s.sub.n,QT taken from a finite alphabet of 2.sup.N.sup.b possible quantization symbols or levels. The power amplification apparatus disclosed in the present application employs a sampler, represented by 100, and a quantizer, that can be uniform or non-uniform, represented by 101, with N.sub.b quantization bits and M=2.sup.N.sup.b quantization levels. The N.sub.b quantization bits are employed in the definition of N.sub.mN.sub.b polar components in which each quantization symbol can be decomposed. In the polar converter of 102 the quantization bits (.sub.n.sup.(N.sup.b.sup.1), .sub.n.sup.(N.sup.b.sup.2), . . . , .sub.n.sup.(1), .sub.n.sup.(0)) are converted into polar form (b.sub.n.sup.(N.sup.b.sup.1), b.sub.n.sup.(N.sup.b.sup.2), . . . , b.sub.n.sup.(1), b.sub.n.sup.(0)) by b.sub.n.sup.(m)=(1).sup..sup.n.sup.(m). The finite set of quantized symbols ={s.sub.0, s.sub.1, . . . , s.sub.M1}, where M=2.sup.N.sup.b is the number of constellation symbols, follows the rule
(.sub.n.sup.(N.sup.b.sup.1),.sub.n.sup.(N.sup.b.sup.2), . . . ,.sub.n.sup.(1),.sub.n.sup.(0))s.sub.n,QT,
with (.sub.n.sup.(N.sup.b.sup.1), .sub.n.sup.(N.sup.b.sup.2), . . . , .sub.n.sup.(1), .sub.n.sup.(0)) denoting the binary representation of n with N.sub.b=log.sub.2(M) bits.

(7) Next, in 103 the polar version of the quantization bits (b.sub.n.sup.(N.sup.b.sup.1), b.sub.n.sup.(N.sup.b.sup.2), . . . , b.sub.n.sup.(1), b.sub.n.sup.(0)) are mapped in N.sub.mN.sub.b polar components, that are the result of the decomposition of quantization value s.sub.n,QT into polar components given by

(8) $\begin{array}{c}{s}_{n,QT}=g_0+g_1{b}_n^{\left(0\right)}+g_2{b}_n^{\left(1\right)}+g_3{b}_n^{\left(0\right)}{b}_n^{\left(1\right)}+g_4{b}_n^{\left(2\right)}+\mathrm{.Math.}\\ =\underset{i=0}{\overset{N_m-1}{.Math.}}{g}_i\underset{m=0}{\overset{N_m-1}{.Math.}}{\left(b_n^{\left(m\right)}\right)}^{{}_{m,i}}=\underset{i=0}{\overset{N_m-1}{.Math.}}{g}_i{b}_n^{\mathrm{eq}\left(i\right)},\end{array}$
with (.sub.N.sub.m.sub.1,i .sub.N.sub.m.sub.2,i . . . .sub.1,i .sub.0,i) denoting the binary representation of i, b.sub.n.sup.(m)=(1).sup..sup.n.sup.(m) denoting the polar representation of the bit .sub.n.sup.(m), b.sub.n.sup.eq(i)=.sub.m=0.sup.N.sup.m.sup.1(b.sub.n.sup.(m)).sup..sup.m,i denoting the i'th polar component of s.sub.n and N.sub.m is the number of non-zero g.sub.i coefficients of the referred decomposition equation.

(9) Each one of the N.sub.m polar components is modulated as a BPSK signal in 104, being each of these N.sub.m BPSK specially designed to have good tradeoffs between reduced envelope fluctuations and compact spectrum (e.g., a GMSK signal). Note that the pulse shape employed in 104 can be selected to achieve high spectral efficiency and constant envelope. For the i-th branch the peak amplitude of the corresponding BPSK signal is given by the corresponding |g.sub.i|.

(10) Next, in each branch the resulting signals are submitted to a phase shifter 105 where the signal obtained at the output of BPSK modulator 104 suffers a phase rotation before being amplified by the nonlinear amplifier 106. Thus, in each branch, the signal at the BPSK modulator's output is multiplied by a complex coefficient in the phase shifter 105 and then amplified by a nonlinear amplifier 106, which can operate in saturated mode or near to it. The amplification stage 107 is composed by N.sub.m amplifiers 106 in parallel whose outputs are the inputs of 108. 108 can be implemented as a combiner followed by a transducer connected to an antenna or a laser diode or a hydrophone or as a set of N.sub.m transducers that can be N.sub.m antennas, N.sub.m hydrophones or N.sub.m laser diodes, where each transducer is directly connected to each amplification branch. Depending on 108 structure the N.sub.m amplifiers' outputs can be summed by a combiner before being transmitted to the channel or can be directly transmitted to the channel by a set of N.sub.m transducers that can be N.sub.m antennas, or N.sub.m hydrophones or N.sub.m laser diodes or other type of transducers depending on the transmitting channel that is being considered.

(11) In another embodiment, the in-phase and quadrature components of the complex envelope samples of the input signals are separated and quantized by two independent quantizers each one with N.sub.b.sup.I and N.sub.b.sup.Q quantization bits that correspond to two finite alphabets of 2.sup.N.sup.b.sup.I and 2.sup.N.sup.b.sup.Q of possible quantization symbols for the in-phase quantized value s.sub.n,QT.sup.I and the quadrature quantized value s.sub.n,QT.sup.Q respectively. The in-phase quantization value s.sub.n,QT.sup.IN and the quadrature quantization value s.sub.n,QT.sup.Q are decomposed as a sum of N.sub.m.sup.IN.sub.b.sup.I and N.sub.m.sup.QN.sub.b.sup.Q polar components, respectively, and amplified by N.sub.m.sup.I and N.sub.m.sup.Q amplifiers before being combined in a combiner with N.sub.m.sup.I+N.sub.m.sup.Q inputs.

(12) In another embodiment, the N.sub.m quantization bits can be transformed into N.sub.m/2 sets of two polar components that modulated by a OQSPK or a MSK (Minimum Shift Keying) signal or other constant envelope signal, which is subsequently amplified by the non-linear amplifier 106, which can operate in saturated mode or near to it. In this case the amplification stage 107 is composed by N.sub.m/2 amplifiers 106 in parallel whose outputs are the inputs of 108.

(13) While preferred embodiments of the present disclosure have been described above, the present disclosure is not limited to the specific configurations described above. Various variations and modifications may be made without departing from the scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and alterations may occur depending on design requirements and other factors insofar as they are within the scope of the appended claims or the equivalents thereof.