阅读说明：本技术 混合多层信号分解系统与方法 (Hybrid multi-layer signal decomposition system and method ) 是由 胡兰 于 2019-01-18 设计创作，主要内容包括：一种混合多层方法,用于将源信号分解为多个分解信号,所述多个分解信号可用于共同表示所述源信号或恢复所述源信号。示例性实施例是包括多层(或多级)信号分解以生成恒定包络信号而不影响所述原始信号的方法。在示例性实施例中,所述方法包括信号分解以保持恒定包络特性和根据所述信号分解限制带宽扩展。所述方法包括将源信号分解为两个一级分解信号,每个一级分解信号具有恒定的包络幅值。所述方法还包括：在每次迭代时根据阈值幅值将所述恒定包络信号中的每个恒定包络信号迭代分解为后级分解信号。所述后级分解信号在每次迭代时具有与所述阈值幅值相关的包络幅值的恒定包络。(A hybrid multi-layer method for decomposing a source signal into a plurality of decomposed signals that can be used to collectively represent the source signal or recover the source signal. An exemplary embodiment is a method that includes multi-layer (or multi-level) signal decomposition to generate a constant envelope signal without affecting the original signal. In an exemplary embodiment, the method includes signal decomposition to maintain constant envelope characteristics and limiting bandwidth extension according to the signal decomposition. The method includes decomposing a source signal into two first-order decomposed signals, each first-order decomposed signal having a constant envelope magnitude. The method further comprises the following steps: iteratively decomposing each of the constant envelope signals into a post-level decomposed signal according to a threshold magnitude at each iteration. The post-level decomposed signal has a constant envelope of envelope amplitudes related to the threshold amplitude at each iteration.)

1. A method for decomposing a source signal, the method comprising:

decomposing the source signal into two primary decomposed signals, wherein the constant envelope amplitude of each primary decomposed signal is half of the maximum amplitude of the source signal;

decomposing each of the first-order decomposed signals into a first second-order decomposed signal and a second-order decomposed signal each having a constant envelope, wherein a constant envelope magnitude of the first second-order decomposed signal is equal to a threshold magnitude, and a constant envelope magnitude of the second-order decomposed signal is equal to the constant envelope magnitude of one of the first-order decomposed signals minus the threshold magnitude.

2. The method of claim 1, wherein the threshold magnitude is a predefined value.

3. A method according to any one of claims 1 and 2, wherein the threshold amplitude is half the maximum amplitude of either of the two first order decomposition signals.

4. A method according to any one of claims 1 to 3, wherein for each sample value k, the source signal is denoted x (k), wherein the two first order decomposition signals are:

and

wherein A is_mFor the maximum amplitude of the source signal, Φ (k) and (k) are functions of k, and j is a unit imaginary number.

5. The method according to any one of claims 1 to 4, wherein the two first order decomposition signals are:

and

wherein for each sample value k, the source signal is denoted as x (k), and e (k) is an error function.

6. The method of any one of claims 1 to 5, wherein the first secondary decomposed signal and the second secondary decomposed signal are:

and

wherein for each sample value k, the source signal is denoted as x (k) and d is the threshold amplitude.

7. The method of claim 6, wherein d is half of the maximum amplitude of one of the two constant first-order decomposition signals, wherein the first second-order decomposition signal and the second-order decomposition signal into which one of the two first-order decomposition signals is decomposed are:

and

8. the method according to any of claims 1 to 6, wherein for each sample value k, the source signal is denoted x (k), wherein the first and second secondary decomposed signals into which one of the two primary decomposed signals is decomposed are:

and

wherein d is the threshold amplitude, A_mThe sum of phi, which is the maximum amplitude of the source signal, is a function of k, and j is a unit imaginary number.

9. The method of any one of claims 1 to 8, further comprising:

iteratively performing at least one post-level decomposition on each of the two-level decomposition signals as input signals, as follows:

the threshold amplitude of the present level decomposition is determined,

decomposing the input signal into a first present-level decomposed signal and a second present-level decomposed signal each having a constant envelope, the constant envelope magnitude of the first output signal being equal to the threshold magnitude of the present-level decomposition, the constant envelope magnitude of the second output signal being equal to the constant envelope magnitude of the input signal minus the threshold magnitude of the present-level decomposition.

10. The method of claim 9, further comprising: determining that the constant envelope amplitude of any of the output signals has reached a predetermined value, and in response to said determining, ending the iterative execution.

11. The method of claim 10, wherein the iterative execution ends when the constant envelope magnitudes of all of the output signals reach the predetermined value.

12. The method of any of claims 9 to 11, further comprising: determining that the iterative execution has performed a decomposition of a predetermined order of magnitude, and in response to the determination, ending the iterative execution.

13. The method of any of claims 9 to 12, further comprising: determining that a predetermined period of time has elapsed, and in response to the determination, ending the iterative execution.

14. The method according to any of claims 9 to 13, wherein the threshold amplitude value of each stage of decomposition is half of the constant envelope amplitude envelope value of one of the input signals.

15. The method according to any of claims 9 to 14, wherein the threshold amplitude value of each stage of decomposition is smaller than a constant envelope amplitude envelope value of one of the input signals.

16. The method of any one of claims 1 to 15, further comprising: filtering each of the two-level decomposition signals using at least one filter.

17. The method of claim 16, wherein the at least one filter comprises at least one low pass filter or at least one band pass filter.

18. The method of any one of claims 1 to 17, further comprising: transmitting each of the secondary decomposed signals To at least one subsystem including a power amplifier, a Digital-To-Analog Converter (DAC), a transmitter, or a transmission line.

19. The method of any one of claims 1 to 18, further comprising: the other constant envelope signal is stored to a memory.

20. The method of any one of claims 1 to 19, wherein the threshold amplitude is less than the maximum amplitude of the two first order decomposition signals.

21. A method for decomposing a source signal, the method comprising:

decomposing the source signal into two primary decomposed signals, wherein the constant envelope amplitude of each primary decomposed signal is half of the maximum amplitude of the source signal;

decomposing each of the two primary decomposed signals into a secondary decomposed signal having a constant envelope with a constant envelope amplitude equal to half of the maximum amplitude of one of the primary decomposed signals from which the source signal can be recovered.

22. The method of claim 21, wherein for each sample value k, the source signal is represented as x (k), and wherein the secondary decomposed signal from the primary decomposed signal is:

or

Wherein A is_mThe sum of phi, which is the maximum amplitude of the source signal, is a function of k, and j is a unit imaginary number.

23. The method according to any of claims 21 and 22, wherein for each sample value k, the source signal is denoted x (k), wherein the two first order decomposition signals are:

and

wherein A is_mThe sum of phi, which is the maximum amplitude of the source signal, is a function of k, and j is a unit imaginary number.

24. The method of any one of claims 21 to 23, further comprising: each two-level decomposed signal is stored to a memory.

25. The method according to any of claims 21 to 24, wherein each of the two primary decomposed signals is decomposed into only one corresponding secondary decomposed signal.

26. An apparatus for decomposing a source signal, the apparatus comprising:

at least one controller for:

decomposing the source signal into two primary decomposed signals, wherein the constant envelope amplitude of each primary decomposed signal is half of the maximum amplitude of the source signal;

decomposing each of the first-order decomposed signals into a first second-order decomposed signal and a second-order decomposed signal each having a constant envelope, wherein a constant envelope magnitude of the first second-order decomposed signal is equal to a threshold magnitude, and a constant envelope magnitude of the second-order decomposed signal is equal to the constant envelope magnitude of one of the first-order decomposed signals minus the threshold magnitude.

27. The apparatus of claim 26, further comprising a receiver for receiving the source signal.

28. A non-transitory computer-readable medium containing instructions to decompose a source signal, the non-transitory computer-readable medium comprising instructions executable by a processor of a communication device, the instructions comprising:

instructions to decompose the source signal into two primary decomposed signals, each primary decomposed signal having a constant envelope amplitude that is half of a maximum amplitude of the source signal;

instructions to decompose each of the first-level decomposed signals into a first second-level decomposed signal and a second-level decomposed signal each having a constant envelope, wherein a constant envelope magnitude of the first second-level decomposed signal is equal to a threshold magnitude, and a constant envelope magnitude of the second-level decomposed signal is equal to the constant envelope magnitude of one of the first-level decomposed signals minus the threshold magnitude.

Technical Field

The exemplary embodiments relate generally to the field of communications and signal processing technology and, more particularly, to a method and system for signal decomposition.

Background

Signal decomposition generally refers to converting one source signal into a plurality of decomposed signals, which collectively represent the original source signal. For example, the multiple decomposed signals may contain useful individual information of the source signal, may be processed separately, or may be recombined into the original source signal.

The latest wireless standards such as the currently proposed fifth Generation (5G) communication standard aim to improve communication, for example, to achieve greater capacity. Implementing certain aspects of a 5G system may result in high dynamic signal amplitudes with high Peak-to-Average Power Ratio (PAPR) waveforms. Signal conditions for higher PAPR typically require more bits to represent the signal, which may also be referred to as bit resolution or bit quantization. Signal conditions of higher PAPR often reduce power efficiency, which is generally undesirable. The signal conditions for higher PAPR may exceed the effective dynamic range of some subsystems.

Some existing systems decompose the signal, but these systems may have a wide signal bandwidth and a high PAPR. For example, a difficulty with some existing signal decomposition systems is that they may have bandwidth extension issues.

A difficulty with some subsystems that can be used in a signal decomposition system is that their dynamic operating range is limited.

It is desirable to provide a system and method for decomposing a source signal into a decomposed signal having a lower PAPR, higher power efficiency, fewer quantization bits, and limited bandwidth.

This background information provides information that applicants believe is likely to be relevant. Any of the above information is not admitted to be or should not be construed as constituting prior art.

Disclosure of Invention

A hybrid multi-layer (multi-stage) decomposition method may be used to decompose a source signal into a plurality of decomposed signals that may be used to collectively represent the source signal or recover the source signal. The Peak-to-Average Power Ratio (PAPR) of the decomposed signals is lower than the source signal, and generally requires fewer quantization bits to represent each decomposed signal. Exemplary embodiments of the method include multi-layer signal decomposition to generate a constant envelope signal with no or minimal impact on the original signal.

In an exemplary embodiment, the method includes decomposing the source signal into two first-order decomposed signals, each first-order decomposed signal having a constant envelope amplitude that is half of a maximum amplitude of the source signal. The method further comprises the following steps: iteratively decomposing each of the first-order decomposition signals into a post-order decomposition signal according to a threshold magnitude at each iteration. Each post-stage decomposed signal has one or more constant envelopes whose constant envelope magnitudes depend on a threshold magnitude. The decomposed signals output by each stage have constant envelope characteristics.

It is an object of at least some example embodiments to provide a method and system for multi-layer signal decomposition that maintains constant envelope characteristics for each layer and limits the bandwidth extension of the signal decomposition.

It is an object of at least some example embodiments to reduce PAPR using signal decomposition with lower Error Vector Magnitude (EVM) compared to other existing decomposition methods.

It is an object of at least some example embodiments to maintain a constant amplitude envelope at each layer of a multi-layer signal decomposition without affecting or with minimal impact on the recoverability of the original source signal.

It is an object of at least some example embodiments to limit the input amplitude range of a subsystem that can better handle the limited amplitude range of a received input signal.

One exemplary embodiment is a method for decomposing a source signal, the method comprising: decomposing the source signal into two primary decomposed signals, wherein the constant envelope amplitude of each primary decomposed signal is half of the maximum amplitude of the source signal; decomposing each of the first-order decomposed signals into a first second-order decomposed signal and a second-order decomposed signal each having a constant envelope, wherein a constant envelope magnitude of the first second-order decomposed signal is equal to a threshold magnitude, and a constant envelope magnitude of the second-order decomposed signal is equal to the constant envelope magnitude of one of the first-order decomposed signals minus the threshold magnitude.

In an exemplary embodiment of the method, the threshold amplitude is a predefined value.

In an exemplary embodiment of any of the above methods, the threshold amplitude is half of a maximum amplitude of any of the two first-order decomposition signals.

In an example embodiment of any of the above methods, for each sample value k, the source signal is denoted as x (k), where the two first order decomposition signals are:

and

wherein A is_mFor the maximum amplitude of the source signal, Φ (k) and (k) are functions of k, and j is a unit imaginary number.

In an exemplary embodiment of any of the above methods, the two first order decomposition signals are:

and

wherein for each sample value k, the source signal is denoted as x (k), and e (k) is an error function.

In an exemplary embodiment of any of the above methods, the first secondary decomposed signal and the second secondary decomposed signal are:

and

wherein for each sample value k, the source signal is denoted as x (k) and d is the threshold amplitude.

In an exemplary embodiment of any of the above methods, d is half of a maximum amplitude of one of the two constant first-order decomposition signals, wherein the first second-order decomposition signal and the second-order decomposition signal into which one of the two first-order decomposition signals is decomposed are:

and

in an exemplary embodiment of any of the above methods, for each sample value k, the source signal is denoted as x (k), wherein the first and second secondary decomposed signals into which one of the two primary decomposed signals is decomposed are:

and

wherein d is the threshold amplitude, A_mFor the maximum amplitude of the source signal, Φ and is a function of k, j is the unit imaginary number.

In an exemplary embodiment of any of the methods above, the method further comprises: iteratively performing at least one post-level decomposition on each of the two-level decomposition signals as input signals, as follows: determining a threshold amplitude of a present-level decomposition, decomposing the input signal into a first present-level decomposition signal and a second present-level decomposition signal each having a constant envelope, the constant envelope amplitude of the first output signal being equal to the threshold amplitude of the present-level decomposition, the constant envelope amplitude of the second output signal being equal to the constant envelope amplitude of the input signal minus the threshold amplitude of the present-level decomposition.

In an exemplary embodiment of any of the methods above, the method further comprises: determining that the constant envelope amplitude of any of the output signals has reached a predetermined value, and in response to said determining, ending the iterative execution.

In an exemplary embodiment of any one of the above methods, said iterative execution ends when said constant envelope amplitudes of all said output signals reach said predetermined value.

In an exemplary embodiment of any of the methods above, the method further comprises: determining that the iterative execution has performed a decomposition of a predetermined order of magnitude, and in response to the determination, ending the iterative execution.

In an exemplary embodiment of any of the methods above, the method further comprises: determining that a predetermined period of time has elapsed, and in response to the determination, ending the iterative execution.

In an exemplary embodiment of any of the above methods, the threshold amplitude value of each stage of decomposition is half of the constant envelope amplitude envelope value of one of the input signals.

In an exemplary embodiment of any of the above methods, the threshold amplitude value of each stage of decomposition is less than a constant envelope amplitude envelope value of one of the input signals.

In an exemplary embodiment of any of the methods above, the method further comprises: filtering each of the two-level decomposition signals using at least one filter.

In an exemplary embodiment of any of the above methods, the at least one filter comprises at least one low pass filter or at least one band pass filter.

In an exemplary embodiment of any of the methods above, the method further comprises: transmitting each of the secondary decomposed signals To at least one subsystem including a power amplifier, a Digital-To-Analog Converter (DAC), a transmitter, or a transmission line.

In an exemplary embodiment of any of the methods above, the method further comprises: the other constant envelope signal is stored to a memory.

In an exemplary embodiment of any of the above methods, the threshold amplitude is less than a maximum amplitude of the two first-order decomposition signals.

Another exemplary embodiment is a method for decomposing a source signal, the method comprising: decomposing the source signal into two primary decomposed signals, wherein the constant envelope amplitude of each primary decomposed signal is half of the maximum amplitude of the source signal; decomposing each of the two primary decomposed signals into a secondary decomposed signal having a constant envelope, wherein the constant envelope amplitude of the secondary decomposed signal is equal to half of the maximum amplitude of one of the primary decomposed signals from which the source signal is recoverable.

In an example embodiment of any of the above methods, for each sample value k, the source signal is denoted as x (k), wherein the secondary decomposed signal from the primary decomposed signal is:

or

Wherein A is_mFor the maximum amplitude of the source signal, Φ and is a function of k, j is the unit imaginary number.

In an example embodiment of any of the above methods, for each sample value k, the source signal is denoted as x (k), where the two first order decomposition signals are:

and

wherein A is_mFor the maximum amplitude of the source signal, Φ and is a function of k, j is the unit imaginary number.

In an exemplary embodiment of any of the methods above, the method further comprises: each two-level decomposed signal is stored to a memory.

In an exemplary embodiment of any of the above methods, each of the two primary decomposed signals is decomposed into only one corresponding secondary decomposed signal.

Another exemplary embodiment is an apparatus for decomposing a source signal, the apparatus comprising: at least one controller for: decomposing the source signal into two primary decomposed signals, wherein the constant envelope amplitude of each primary decomposed signal is half of the maximum amplitude of the source signal; decomposing each of the first-order decomposed signals into a first second-order decomposed signal and a second-order decomposed signal each having a constant envelope, wherein a constant envelope magnitude of the first second-order decomposed signal is equal to a threshold magnitude, and a constant envelope magnitude of the second-order decomposed signal is equal to the constant envelope magnitude of one of the first-order decomposed signals minus the threshold magnitude.

In an exemplary embodiment of the apparatus, the apparatus further comprises: a receiver for receiving the source signal.

Another exemplary embodiment is a non-transitory computer-readable medium for decomposing a source signal, the non-transitory computer-readable medium comprising: instructions to decompose the source signal into two primary decomposed signals, each primary decomposed signal having a constant envelope amplitude that is half of a maximum amplitude of the source signal; instructions to decompose each of the first-level decomposed signals into a first second-level decomposed signal and a second-level decomposed signal each having a constant envelope, wherein a constant envelope magnitude of the first second-level decomposed signal is equal to a threshold magnitude, and a constant envelope magnitude of the second-level decomposed signal is equal to the constant envelope magnitude of one of the first-level decomposed signals minus the threshold magnitude.

Drawings

Embodiments will now be described, by way of example, with reference to the accompanying drawings, in which like reference numerals may be used to refer to like features, and in which:

FIG. 1 illustrates, in block diagram form, a signal processing system including a multi-layer signal decomposition module provided by an exemplary embodiment;

FIG. 2 shows a diagram of an original input signal and a first layer decomposition;

FIG. 3 shows a diagram of an original input signal, a first layer decomposition, a second layer decomposition and a third layer decomposition;

FIG. 4 illustrates a detailed block diagram of the signal processing system of FIG. 1 provided by an exemplary embodiment;

FIG. 5 illustrates a logic diagram for layer control of the signal processing system of FIG. 1 provided by an exemplary embodiment;

FIG. 6 illustrates a detailed block diagram of the signal processing system of FIG. 1 implementing the layer control of FIG. 5 provided by an exemplary embodiment;

FIG. 7 illustrates a detailed block diagram of the signal processing system of FIG. 1 with a symmetric decomposition provided in an exemplary embodiment;

FIG. 8 illustrates a graph of simulation results for a first module of a second layer of the signal processing system of FIG. 1 provided by an exemplary embodiment;

FIG. 9A shows a graph of performance spectrum for a prior art system with threshold decomposition only;

FIG. 9B illustrates a spectral plot of the original signal and the first layer output of the signal processing system of FIG. 1 provided by an exemplary embodiment;

FIG. 10A shows a Peak-to-Average Power Ratio (PAPR) plot of the performance of a prior art system with threshold decomposition only;

FIG. 10B illustrates a Peak-to-Average Power Ratio (PAPR) graph of the performance of the signal processing system of FIG. 1 provided by an exemplary embodiment;

FIG. 10C illustrates an error versus spectrum plot of the performance of the signal processing system of FIG. 1 provided by an exemplary embodiment;

FIG. 11 illustrates a block diagram of an example reconstruction module of the signal processing system of FIG. 1 provided by an example embodiment;

fig. 12 shows a block diagram of another example reconstruction module of the signal processing system of fig. 1 in case of symmetric decomposition according to another example embodiment.

Detailed Description

The exemplary embodiments describe a method and system for signal decomposition. The system includes a disaggregated architecture that provides a flexible system architecture design.

In an example embodiment, the method includes multi-layer signal decomposition to generate a constant envelope signal without affecting or with minimal impact on the original source signal recovery.

In an exemplary embodiment, the method includes multi-layer decomposition to reduce PAPR and the number of quantization bits. For the first layer decomposition or first level decomposition, the method includes decomposing the source signal into two signals, each signal having a constant envelope amplitude that is half the maximum amplitude of the source signal. The high-level decomposition uses a threshold decomposition algorithm, where a target amplitude threshold can be used to further reduce the dynamic range of the constant envelope signal.

Since the decomposed signal has a limited bandwidth characteristic, the method may include filtering the decomposed signal to remove higher frequencies and retain most of the original signal.

Referring to fig. 1, fig. 1 illustrates a signal processing system 100 provided by an exemplary embodiment. The system 100 includes a multi-layered signal decomposition module 102, one or more subsystems 104, and a reconstruction module 106. The decomposition module 102 is used to receive and process an original input signal (source signal 'Sin') 108. The decomposition module 102 includes a plurality of decomposition layers 110a, 110b … … 110n (individually or collectively denoted as '110'), also referred to herein as stages, for collectively performing multi-layer signal decomposition as described in more detail herein, resulting in a plurality of decomposed signals decomposed from the original input signal 108. Each of the plurality of decomposed signals may be filtered using one or more filters 112. One or more subsystems 104 may perform other functions on the filtered, decomposed signal. In an exemplary embodiment, the reconstruction module 106 is configured to reconstruct or recombine the received signals from one or more of the subsystems 104 by, for example, summing the received signals or reversing the function of the decomposition module 102. The reconstruction module 106 thus generates a restored signal 114.

Referring to the multi-layer decomposition module 102, in the exemplary embodiment, a first layer 110a (also referred to as layer 1 or level 1) is based on a constant envelope algorithm and a higher layer is based on a threshold algorithm that further decomposes the signal using a threshold amplitude. For example, as will be described in greater detail herein, the second layer 110b (also referred to as layer 2 or stage 2) through nth layer 110n (also referred to as layer n or stage n) are based on a threshold algorithm that maintains a constant envelope characteristic.

In an exemplary embodiment, one or more filters 112 may also be used to filter a specified bandwidth due to the bandwidth limitation created by the constant envelope of each layer 110. In an exemplary embodiment, the one or more filters 112 include a low pass filter or a band pass filter, or both. In the exemplary embodiment, a respective filter 112 is used for each of the final decomposed signals. In another exemplary embodiment, the same filter 112 is used for each of the final decomposed signals.

In an exemplary embodiment, the one or more subsystems 104 are power amplifiers, Digital-to-Analog converters (DACs), transmitters, transmission lines, and the like. At least one of the subsystems 104 may have a limited dynamic operating range and better performance within that operating range. Not all of the decomposed signals need be sent to one or more subsystems 104.

In an exemplary embodiment, the system 100 resides on one device or apparatus. In another example, the system 100 is implemented by two or more devices or apparatuses. For example, the reconstruction module 106 may be located on a different device than the decomposition module 102.

In an exemplary embodiment, the system 100 receives the raw input signal 108, for example, via direct transmission or via a wired or wireless network using a receiver of the communication subsystem. In another exemplary embodiment, the original input signal 108 is generated by the system 100 itself. Similarly, in an exemplary embodiment, the recovered signal 114 may be used by the system 100 itself for further processing or storage to memory, or may be transmitted to another device or network.

In an exemplary embodiment, as shown in FIG. 1, each module of each layer 110 produces at most two outputs or branches. In another exemplary embodiment described in more detail herein, for the higher layers 110b to 110n, only the first of the two branches needs to be determined for each module, and since the second branch has the same attributes or recoverability as the corresponding first branch, the second branch need not be determined or stored. For example, only the determination signals of the first branch of the layers are stored in a memory. Thus, the second branch is optionally not determined or stored in memory.

Referring to fig. 1, the first layer 110a will be described in more detail according to an exemplary embodiment. The first layer 110a is a constant envelope algorithm. The first layer 110a output is a decomposition of the input signal 108 into two constant envelope signals.

In general, the input signal 108 may be defined as:

x(k)＝|x(k)|e^j(k)(1.1)

in equation (1.1), | x (k) | is the amplitude of the original signal, k is the sample value, and j is the unit imaginary number.

The intermediate signal is introduced as follows:

in the intermediate signal, A_mFor the maximum amplitude of signal 108, Φ (k) is a function of k.

The original signal 108 may be rewritten as:

x(k)＝A_mcos(k)e^j(k)

in this equation, (k) is a function of k. The original signal 108 may be rearranged as:

thus, the original signal can be decomposed into two constant envelope signals:

x(k)＝x₁₁(k)+x₁₂(k) (1.3)

for equation (1.3), the two constant envelope signals are:

and

as will be apparent to those skilled in the art, the respective signals defined by (1.4) and (1.5) each have a constant envelope with a constant envelope magnitude equal to A_m/2。

In another exemplary implementation of the signals of (1.4) and (1.5), it is possible to rearrange the signals as follows:

and

in equations (1.6) and (1.7), e (k) is an error function defined as follows:

in equation (1.8), cos Φ (k) has been defined above.

Subsequent signal synthesis of the decomposed signal involves signal addition of equations (1.6) and (1.7). This results in the positive and negative error function values cancelling out each other when added. Since the error function will be cancelled out in the signal synthesis stage, the presence of the error function only affects the dynamic range of the signal envelope, and not the final combined signal result.

Note that equations (1.6) and (1.7) can be simplified because e (k) does not need to be calculated to obtain an accurate value, but rather an estimated value. For example, Coordinate Rotation digital computer (CORDIC) algorithms may be used to estimate the error function, as is understood in the art.

Replacing equation (1.8) with equations (1.4) and (1.5) yields:

and

the following procedure proves that the above equations are equivalent to equations (1.4) and (1.5). The above equation can be simplified as:

and

the exponential function can be expressed in terms of cosine and sine functions as follows:

and

equations (1.6) and (1.7) are equivalent to equations (1.4) and (1.5) as evidenced by the substitution of equations (1.11) and (1.12) for equations (1.9) and (1.10).

Referring also to fig. 1, the higher layers 110b … … 110n are each used to perform further signal decomposition. As described in more detail herein, since each of the two signals received from the first layer 110a has a constant envelope, the second layer 110b through the nth layer 110n may also be used to output decomposed signals having constant envelopes.

Referring to fig. 2, fig. 2 shows a graph 200 of the amplitude envelope values of the original input signal 108 and the first layer 110 a. Graph 200 is a plot of amplitude versus sample. The first layer 110a decomposition algorithm may be completely decomposed into two constant envelope signals, each having a signal amplitude dynamic range of 'dy'. It is recognized herein that merely iterating this algorithm cannot further reduce the signal amplitude dynamic range 'dy' shown in the graph 200. In an exemplary embodiment, a threshold algorithm is used to further decompose the two constant envelope signals to further reduce the dynamic range of each signal without affecting or with minimal impact on the envelope characteristics of each signal (e.g., keeping PAPR small, keeping constant envelope).

The threshold algorithm of the second layer 110b through the nth layer 110n is used to further decompose the signal to reduce the dynamic range and signal amplitude. The threshold algorithm is determined from the threshold'd', as follows:

and

in an exemplary embodiment, the value of'd' is in the range 0 < d < A_mA predefined value of. A. the_mIs the maximum value of the amplitude of the original input signal 108.

In another exemplary embodiment, the value of'd' may be selected or programmed to have different magnitudes of the decomposed signal in each layer as the target magnitude. As shown in fig. 3, the signal amplitude may decrease after several layers.

Fig. 3 shows a graph 300 of the original input signal 108 and the amplitude envelope values of the first layer 110a, the second layer 110b and the third layer 110 c. The graph 300 is similar to the graph 200 of fig. 2 and further illustrates the amplitude envelope values of the second layer 110b and the third layer 110 c. As shown in fig. 3, in the exemplary embodiment of symmetric decomposition, each'd' value is calculated to be half the amplitude of the input signal received from the upper layer. In another exemplary embodiment, not shown, there is an asymmetric decomposition, where the value of'd' is not half of the input signal received from the previous layer.

In an exemplary embodiment, the same threshold algorithm is used iteratively in the second tier 110b and the higher tier. The number of layers can be determined according to the design requirements of the system. The value of'd' of each layer 110 may be selected as a value greater than zero and less than the constant envelope amplitude of the input signal of the current layer 110.

Each layer 110 outputs a decomposed signal having a reduced dynamic range compared to the input signal received from the previous layer 110. Further, the decomposed signal of each layer 110 may be represented using a smaller number of quantization bits than the input signal received from the previous layer 110.

Referring to fig. 4, fig. 4 shows the system 100 in more detail. The second layer 110b processes two input signals received from the first layer 110a, respectively. The result of the second layer 110b performing the decomposition is output as two output signals per input signal, four output signals in total. In an exemplary embodiment, the second layer 110b decomposition may be defined using two identical functional modules.

Referring again to fig. 4, the output signals of the second layer 110b and the module 1 are:

and

the output signals of the second layer 110b and the module 2 are as follows:

and

more than two layers, the principle is the same. Each input signal is decomposed into two output signals.

For the subscript of x in fig. 4, the first number represents a layer, the second number represents a module number or module ID of the layer, and the third number represents a branch of the module number. For example, for 'x' shown in FIG. 4_{31_2}(k) ', ' 3 ' denotes signal outputs of the third layer 110c, ' 1 ' denotes a first module (' module 1 ') of the third layer 110c, and ' 2 ' denotes a second branch of the module 1 of the third layer 110 c.

The maintenance of a constant amplitude envelope at each layer 110 is described below. For the second layer 110b, equation (1.4) is the signal input and may be inserted into equations (2.1) and (2.2) as follows:

and

the above equation can be simplified as:

and

the signal maintains a constant envelope characteristic but reduces the dynamic range. The same results are possible for other branches and layers. The mathematical equation map is shown in fig. 4.

Referring now to fig. 5 and 6, fig. 1 illustrates layer control of a system 100 provided by an exemplary embodiment. Fig. 5 illustrates a logic diagram 500 of a layer control, and fig. 6 illustrates a block diagram of the system 100 of fig. 1 implementing the layer control. The source signal is iteratively decomposed into further decomposed signals until a specified condition is determined.

The number of layers can pass through the target dynamic range A_tgAnd (5) controlling. Block 502 receives the raw input signal 108. Block 504 determines a constant envelope amplitude. Block 506 checks whether the constant amplitude is greater than a specified time period T_tgTarget dynamic range A of_tg(event 514). If so, signal decomposition is performed by control block 508 enabling signal decomposition block 510. If not, the method 500 ends and the subsystem 104 may perform other processes on the final decomposed signal, e.g., by the filter 112. In an exemplary embodiment, the delay 512 block causes a delay of the signal decomposition block 510. The signal decomposition block 510 loops back to block 502 and returns to the output of the just completed layer 110 to perform the next iteration. The decomposition block 510 is used to perform layer 1(110a) first, and then for subsequent iterations, the decomposition block 510 performs higher layers, e.g., from the second layer 110b to the nth layer 110 n. In an exemplary embodiment, the final decomposed signal from nth layer 110n is then stored in memory or sent to other subsystems for any or all of processing, transmission, and reassembly.

Referring to FIG. 6, by comparing the input signal with target A_tgRatio of performanceComparing (block 506), when the input | x (k) | is less than or equal to the target A_tgThen signal decomposition is stopped (block 510). In some exemplary embodiments, A_tgAnd the check duration 514 are design parameters that may be predefined. In other exemplary embodiments, A may be determined in real time_tgAnd a check duration 514.

In an exemplary embodiment, when the envelope amplitude value of any one of the output signals in a layer reaches the target A_tgThen signal decomposition is stopped (block 510). For example, in the case of symmetric decomposition, only one of the output signals needs to be aligned with target A_tgThe comparison is performed because the remaining signals have the same envelope amplitude. In another exemplary embodiment, when the envelope amplitudes of all output signals in a layer reach target A_tgThen signal decomposition is stopped (block 510). For example, in the case of asymmetric decomposition, all output signals may be summed with target A_tgA comparison is made to stop signal decomposition when the maximum amplitude of all output signals is less than a target value (block 510).

Fig. 7 illustrates a signal processing system 100 with reduced resources provided by an exemplary embodiment. By calculating a specific'd' value for each layer, it is possible to reduce resources by shutting down certain branches, e.g. one of the two branches in each module, compared to the signal processing system 100 shown in fig. 4.

To illustrate how to close certain branches, in equations (2.9) and (2.10), A is used_mThe threshold value'd' is replaced by/2.

And

in the system 100 of FIG. 4, the second layer 110b, module 1, may be used to calculate x_{21_1}And x_{21_2}. As can be seen from equations (2.11) and (2.12), x_{21_1}And x_{21_2}Having a common component

Only one calculation is needed. Continuing with the example, for the second layer 110b, at module 1, only x needs to be calculated_{21_1}And stores it in memory. Upon reconstruction, x_{21_2}Can pass through x_{21_1}According to common componentsAnd (6) recovering. As a result, the system 100 may save about half of the computing resources.

The system 100 shown in FIG. 7 illustrates a high-level block diagram of resource reduction. Only one of equations (2.11) and (2.12) needs to be calculated, where one output needs to be processed and saved to memory. This may therefore reduce hardware or software resources without impacting performance. In the system 100 of fig. 7, only one decomposed output signal needs to be calculated for each input signal for the second layer 110b through the nth layer 11 n. This is in contrast to the implementation of the system 100 shown in fig. 4, which has an exponential increase in the decomposed output signal at each layer 110.

Referring now to fig. 8, fig. 8 shows a graph 800 of simulation results for module 1 of the second tier 110b of the signal processing system 100. As shown in fig. 8, different dynamic ranges of signals within the same layer can be obtained by programming the threshold'd'. A diagram of the first branch 802 and the second branch 804 are also shown separately.

Fig. 9A shows a spectrogram 900 of a prior art system with only threshold decomposition. In other words, there is no initial constant envelope decomposition algorithm. Diagram 900 shows the original signal and the first and second outputs ('s 1' or's 2', respectively). Fig. 9B shows a spectral plot 910 of the original signal and the first layer 110a signal output ('s 1' or's 2', respectively) provided by an exemplary embodiment. In the illustrated symmetrical case, s1 and s2 typically overlap for purposes of illustration, and thus appear as the same signal.

As can be seen from the spectrogram 900, the bandwidths of s1 and s2 are narrower and more clearly defined than the original signal due to the constant amplitude envelopes of s1 and s 2. The bandwidth of the original signal is much wider.

FIG. 10A shows a Peak-to-Average Power Ratio (PAPR) graph 1000 of the performance of the prior art system described in FIG. 9A with threshold decomposition only. The X-axis is the PAPR of the 'original signal' (input signal 108 of fig. 1). The illustrated "raw signal" is used only as a reference to the X-axis. The vertical axis represents the probability that the PAPR is greater than the PAPR of the reference source signal. Plotted against the Y-axis are the output of the second branch's 2' of the threshold-only decomposition, the filtered output of the second branch's 2-filt' of the threshold-only decomposition, and the filtered output of the first branch's 1-filt' of the threshold-only decomposition.

Fig. 10B illustrates a Peak-to-Average Power Ratio (PAPR) graph 1010 of the output of the signal processing system 100 of fig. 1 provided by an exemplary embodiment. The X-axis is the PAPR of the 'original signal' (input signal 108 of fig. 1). The raw signal shown is used as a reference to the X-axis only. The vertical axis represents the probability that the PAPR is greater than the PAPR of the reference source signal. Plotted against the Y-axis are the outputs of's 1' and's 2'. In the illustrated symmetrical case,'s 1' and's 2' typically overlap and thus appear as the same signal. As shown in fig. 1010, PAPR of s1 and s2 is generally low, especially PAPR of original signal is high, compared to original signal.

Fig. 10C shows an error versus spectrum plot 1020 of the decomposition module 102. The original signal and the error due to the decomposition of the original signal by the decomposition module 102 are displayed in units of decibels (dB) and frequencies (MHz).

The following table summarizes the simulation results of the prior art threshold-only decomposition system and the performance of the decomposition module 102.

TABLE 1

In table 1, "quantization bit" means the number of bits used to represent a corresponding decomposed signal, "Average Error Vector Magnitude (EVM),%" means the offset from the original signal of a specific quantization bit set. In table 1, a blank entry means that the simulation has no exact quantized bit value, which exactly corresponds to a particular set of quantized bits.

In table 1, only the threshold decomposition yields two decomposition signals, denoted as's 1' and's 2'. Referring to FIG. 4 and Table 1, for decomposition module 102,'s 1' denotes x_{21_1}(k) 's 2' denotes x_{21_3}(k) 's 3' denotes x_{22_1}(k) 's 4' denotes x_{22_2}(k) In that respect In table 1, the decomposition module 102 also represents a case of symmetric decomposition. Therefore, the results shown in table 1 are only for the decomposed signals's 1' and's 3'. Note that in this symmetric decomposition case,'s 2' and's 4' (not shown) need not be represented with quantization bits, as they can then be recovered from's 1' and's 3', respectively, as described above with respect to equations (2.11) and (2.12).

As can be seen from table 1, at higher quantized bit values such as "float" and "10", the performance of the decomposition module 102 is comparable to a threshold-only decomposition. At lower quantized bit values, decomposition module 102 performs better operations than EVM with threshold-only decomposition. For example, in the case of quantization bit '6', the EVM percentage of the threshold-only decomposition is 2.94, and the EVM percentage of the decomposition module 102 is 1.7.

A similar conclusion to be drawn from table 1 is that to achieve a similar EVM, decomposition module 102 typically requires fewer quantized bits than just a threshold decomposition, especially if the quantized bit values are low.

Referring again to fig. 1, the subsystem 104 receives only the final decomposed signal, e.g., the final decomposed signal has a dynamic range and a constraint band that is smaller than the original signal 108. These characteristics of the resulting decomposed signal may improve the performance of the subsystem 104.

Fig. 11 illustrates a block diagram of the reconstruction module 106 provided by an exemplary embodiment. In an exemplary embodiment, the reconstruction module 106 is configured to reconstruct or reassemble the received signals from one or more of the subsystems 104 by, for example, reversing the functionality of the decomposition module 102. The reconstruction module 106 thus generates a restored signal 114. The reconstruction module 106 includes an addition module 110.

For example, in fig. 4, a plurality of decomposed signals may be output from the nth layer 110 n. In fig. 11, the summing module 1100 receives these multiple split signals as multiple input signals. For example, the adding module 1100 is configured to generate the restored signal 114 by adding a plurality of received input signals.

Fig. 12 shows a block diagram of the reconstruction module 106 provided by the exemplary embodiment in the case of symmetric decomposition. The reconstruction module 106 of fig. 12 is used in the case of reconstructing a decomposed signal output as a result of a symmetric decomposition, as is the case with the system 100 of fig. 7. As shown at the summing block 1102, the reconstruction block 106 need only receive exactly two input signals (the two input signals corresponding to the two primary output signals from the system 100 of fig. 7). The summing module 1102 sums the two input signals together. A scale factor 1104 is also applied to block 1102. in an exemplary embodiment, scale factor 1104 can be a predefined scale factor. In contrast to the summing module 1100 of fig. 11, which receives more than two input signals, the summing module 1102 of fig. 12 need not receive more than two primary input signals.

Referring to FIG. 1, in an exemplary embodiment, the number of layers 110 may vary. The decomposition is performed in sequence until the decomposed signal reaches the target amplitude A_tg. In an exemplary embodiment, the number of layers is determined by real-time automatic layer control. In an exemplary embodiment, when the input signal reaches | x | < A_tgFor a certain period of time T_tgAfter that, the signal decomposition is stopped. In another exemplary embodiment, the number of layers may be predefined.

It will be appreciated that 0 and A may be selected_mAny value in between is used as the threshold'd'. In an exemplary embodiment, the system 100 may be used to generate a symmetric or asymmetric decomposed signal.

Furthermore, the hybrid multi-layer architecture allows one or more filters 112, such as low pass filters or band pass filters, to be used in exemplary embodiments without significantly reducing PAPR.

In an exemplary embodiment, signal decomposition may improve the system dynamic range or power efficiency of the subsystem 104, for example, the lianear amplification using a non-LInear component (LINC) Power Amplifier (PA).

In an exemplary embodiment, decomposing the signal reduces bandwidth expansion. For example, the limited bandwidth may produce a decomposed signal with a PAPR of less than 3 dB.

In an exemplary embodiment, system 100 is implemented by a Personal Basic Service Set (PBSS) control point (PCP), Access Point (AP), or Station (STA), as understood in the art.

In an exemplary embodiment, at least one module of system 100 is implemented by an electronic component. For example, the decomposition module 102 may be implemented by decomposition module electronics. The electronic components may be provided as semiconductor circuits, for example forming part or all of an integrated circuit package. The circuit may be a digital circuit or an analog circuit. In some embodiments, the circuit is preconfigured according to a specified number of layers, a specified threshold'd', a specified time period for processing a particular source signal, or other specified threshold, and so forth. In other embodiments, the circuit is reconfigurable and reprogrammable through a control interface or user interface.

Some example embodiments apply to signal processing in millimeter wave (mm-wave) wireless communication systems. Some exemplary embodiments are applicable to signal processing in Wi-Fi (Wi-Fi, TM) communication systems, as described in the IEEE 802.11 family of standards. It will be readily appreciated that the exemplary embodiments can be applied to other wireless communication systems, as well as wired or optical systems, and other communication environments.

Some example embodiments apply to signal processing in single channel systems, multi-channel systems, beamforming, multi-channel systems, multiple-input-multiple-output (MIMO) systems, massive MIMO systems, multi-channel systems, or multi-carrier systems. Some example embodiments may be applied to wired or wireless systems, including 4G, intended to cover and encompass higher generation systems, including 5G.

Through the description of the above exemplary embodiments, the above exemplary embodiments may be implemented only by hardware, or may be implemented by software and a necessary general hardware platform. Based on such understanding, the technical solutions of the exemplary embodiments can be embodied in the form of software products. The software product may be stored in a non-volatile or non-transitory storage medium, which may be a compact disk read-only memory (CD-ROM), a USB flash drive, or a removable hard drive. The software product comprises a number of instructions enabling a computer device (personal computer, server or network device) to perform the method provided by the exemplary embodiments. Such execution may correspond to, for example, simulation of logical operations as described herein. According to an example embodiment, a software product may additionally or alternatively include a plurality of instructions that cause a computer apparatus to perform operations to configure or program a digital logic device.

The example apparatus and methods described herein may be implemented by one or more controllers according to example embodiments. The controller may comprise hardware, software, or a combination of hardware and software, depending on the particular application, component, or function. In some example embodiments, one or more controllers may include analog or digital components, and may include one or more processors, one or more non-transitory storage media, such as memory storing instructions executable by at least one of the one or more processors, one or more transceivers (or separate transmitters and receivers), one or more signal processors (such as one or both of an analog signal processor and a digital signal processor), and one or more analog circuit components.

In the described methods or block diagrams, the blocks may represent any or all of events, steps, functions, procedures, modules, messages, state-based operations, and the like. Although some of the examples above have been described as occurring in a particular order, those skilled in the art will appreciate that certain steps or processes may be performed in a different order, as long as the result of the change in order of any given step does not prevent or impair the occurrence of subsequent steps. Further, some of the messages or steps described above may be deleted or merged in other embodiments, and some of the messages or steps described above may be split into multiple sub-messages or sub-steps in other embodiments. Some or all of the steps may even be repeated as desired. Elements described as methods or steps are equally applicable to systems or subcomponents and vice versa. The words "transmit" or "receive," etc., are used interchangeably depending on the role of the particular device.

The embodiments discussed above are illustrative and not restrictive. Exemplary embodiments described as methods are equally applicable to systems and vice versa.

Some exemplary embodiments may vary, including combinations and subcombinations of any of the above embodiments. The exemplary embodiments shown above are only examples and are not intended to limit the scope of the present invention in any way. The innovative variations described herein will be apparent to those of ordinary skill in the art and are within the intended scope of the invention. In particular, features of one or more of the above-described embodiments may be selected to create alternative embodiments that include subcombinations of features that may not be explicitly described above. Furthermore, features of one or more of the above-described embodiments may be selected and combined to create alternative embodiments that include combinations of features that may not be explicitly described above. Features suitable for such combinations and sub-combinations will be apparent to those skilled in the art upon a review of the present disclosure as a whole. The subject matter described herein is intended to cover and embrace all suitable technical variations.

Accordingly, the specification and figures are to be regarded in an illustrative manner only and include any and all modifications, variations, combinations, or equivalents.