Description:TECHNICAL FIELD
The present disclosure relates to wireless networks, and more specifically relates to methods and baseband unit for switching signal transmission from a first frequency band to a second frequency band in an Open Radio Access Network (O-RAN) Radio Unit (O-RU).

BACKGROUND
Radio frequency (RF) power amplifiers (PAs) are inherently non-linear devices, and generate unwanted intermodulation distortion products (IMDs), which manifest themselves as spurious signals in an amplified RF output signal, separate and distinct from a RF input signal. A digital predistortion (DPD) has been widely used as a high-performance linearization technique for compensating nonlinearities, to obtain a linear signal, caused by the RF power amplifiers in radio transmitters. But, when a network device or a User Equipment (UE) moves to higher frequency bands in a fifth generation (5G) system, the network device or the UE requires more complex hardware architecture to obtain the linear signal that can increase power consumption.
That is, in the 5G system, power consumption can be very high due to power amplifiers. When the network device or the UE operates in a Frequency Range 2 (FR-2) band, the network device or the UE require mixers or local oscillators (LO) to obtain the linear signal. If compared to existing technologies, the power consumption of the 5G system is much higher. It can be in degrees of Kilo Watts for the FR-2 and this huge power consumption increases an operating expense (OPEX) very much for operators. A few examples of such power consumption are shown in FIG. 1A and FIG. 1B, in which FIG. 1A illustrates an instantaneous power level of a signal with high peak power (Ppeak) compared to its average power (Pavg) and FIG. 1B illustrates power efficiency as a function of output power for a power amplifier.
Some of the prior art references are given below for obtaining the linear signal:
US20210391832A1 discloses use of machine learning based approach to improve determination/generation of digital pre-distortion (DPD) parameters for power amplifiers. For DPD parameter generation, a feed-forward neural network algorithm is used to provide a desired (linear) output.
US20210328608A1 discloses a recurrent neural network (RNN) in a wireless device or system that may utilize feedback after processing of a compensated wireless transmission signal to determine how efficiently the DPD filter is compensating such wireless transmission signals.
WO2021179879A1 discloses a radio over fiber (RoF) system and a nonlinear compensation method to improve the linearization effect of the RoF system. The RoF system includes a predistortion module that is used to perform digital predistortion (DPD) on a baseband signal according to a temperature value of electric devices in a Remote Radio Unit (RRU).
A non-patent literature document entitled: “Neural Network DPD for Aggrandizing SM-VCSEL-SSMF-Based Radio over Fiber Link Performance” discloses a neural network (NN)-based digital predistortion (DPD) solution to overcome the signal impairments and nonlinearities in Analog Optical fronthauls using radio over fiber (RoF) systems.
Another non-patent literature document entitled: “Augmented Iterative Learning Control for Neural-Network-Based Joint Crest Factor Reduction and Digital Predistortion of Power” discloses an approach to realize a joint CFR and DPD model using neural networks (NN) to compensate PA non linearities and improve power efficiency.
While the prior arts cover various approaches for compensating PA non linearities and improving power efficiency in the 5G system, but none of the prior arts discloses implementation of inverse modelling method with augmented cascaded feed forward neural networks techniques in a baseband unit of an O-RU for compensating PA non linearities and improving power efficiency. In light of the above-stated discussion, there is a need to overcome the above stated disadvantages.

OBJECT OF THE DISCLOSURE
A principal object of the present disclosure is to provide methods and a baseband unit for switching signal transmission from a first frequency band to a second frequency band in an O-RAN Radio Unit (O-RU).
Another object of the present disclosure is to implement an Augmented Cascaded Feed Forward Neural Network (ACFFNN) technique in the baseband unit of the O-RU.
Another object of the present disclosure is to limit the use of mixers or oscillators (hardware) to reduce non-linear distortions in signals.

SUMMARY
Accordingly, the present disclosure provides a method of switching signal transmission from a first frequency band to a second frequency band in an O-RU ("system”). The method includes limiting, by a baseband unit, an input bandwidth of an input signal x(n) in a first frequency band, whereby the input signal is transformed into a modulated signal. Further, the method includes estimating, by the baseband unit, a Digital Pre-Distortion (DPD) coefficient in an Augmented Cascaded Feed Forward Neural Network (ACFFNN) based on the input signal and the modulated signal and the method includes removing, by the baseband unit, distortion in the modulated signal based on the DPD coefficient to obtain a linear signal capable of being transmitted in the second frequency band. Furthermore, the method includes transmitting, by the baseband unit, the linear signal in the second frequency band. The second frequency band is greater than the first frequency band.
The method can be used to compensate for direct current (DC) offset, in-phase and quadrature-phase (IQ) imbalance along with frequency multiplier, and power amplifier (PA) distortions in the O-RU.
The method is implemented in a Single Input Single Output (SISO) system and multiple-input and multiple-output (MIMO) system.
Accordingly, the present disclosure provides an O-RU (“system”) for switching signal transmission from a first frequency band to a second frequency band. The O-RU is configured to limit an input bandwidth of an input signal x(n) coming from a baseband in a first frequency band to form a modulated signal using a baseband unit. The input signal is transmitted to a frequency multiplier and a power amplifier in the O-RU, wherein limiting the input bandwidth reduces a distorted signal bandwidth. Further, the O-RU is configured to place the frequency multiplier in the Radio Frequency (RF) chain for frequency translation from the first frequency band to the second frequency band, wherein the RF chain comprises at least one nonlinear power amplifier, at least one filter, and mixers. Further, the O-RU is configured to input the input signal and the modulated signal of the at least one nonlinear power amplifier and the frequency multiplier, baseband equivalent input and baseband equivalent output of the at least one nonlinear power amplifier and the frequency multiplier in an ACFFNN. The O-RU is further configured to estimate a Digital Pre-Distortion (DPD) coefficient in the ACFFNN based on the input signal and the modulated signal. The O-RU is further configured to remove distortion in the modulated signal based on the DPD coefficient on the input signal frequency multiplier to obtain a linear signal capable of being transmitted in the second frequency band and transmit the linear signal in the second frequency band and obtain the fined distortion-free signal.
These and other aspects herein will be better appreciated and understood when considered in conjunction with the following description and the accompanying drawings. It should be understood, however, that the following descriptions are given by way of illustration and not of limitation. Many changes and modifications may be made within the scope of the invention herein without departing from the spirit thereof.

BRIEF DESCRIPTION OF FIGURES
The invention is illustrated in the accompanying drawings, throughout which like reference letters indicate corresponding parts in the drawings. The invention herein will be better understood from the following description with reference to the drawings, in which:
FIG. 1A illustrates an instantaneous power level of a signal with high peak power (Ppeak) compared to its average power (Pavg), according to prior art.
FIG. 1B illustrates power efficiency as a function of output power for a power amplifier, according to prior art.
FIG. 2A illustrates a block diagram of a baseband unit.
FIG. 2B illustrates an architecture of an Augmented Cascaded Feed Forward Neural Network (ACFFNN) in the baseband unit.
FIG. 2C illustrates a block diagram of an O-RU, having the baseband unit, for switching signal transmission from a first frequency band to a second frequency band.
FIG. 3 is a flow chart illustrating a method of switching signal transmission from a first frequency band to a second frequency band in the O-RU.
FIG. 4A to FIG. 4 C are graphs illustrating pre-distorter performance results using inverse modelling.
FIG. 5 illustrates a frequency multiplication.

DETAILED DESCRIPTION
In the following detailed description of the invention, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be obvious to a person skilled in the art that the invention may be practiced with or without these specific details. In other instances, well known methods, procedures and components have not been described in details so as not to unnecessarily obscure aspects of the invention.
Furthermore, it will be clear that the invention is not limited to these alternatives only. Numerous modifications, changes, variations, substitutions and equivalents will be apparent to those skilled in the art, without parting from the scope of the invention.
The accompanying drawings are used to help easily understand various technical features and it should be understood that the alternatives presented herein are not limited by the accompanying drawings. As such, the present disclosure should be construed to extend to any alterations, equivalents and substitutes in addition to those which are particularly set out in the accompanying drawings. Although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are generally only used to distinguish one element from another.
The present disclosure provides methods and a baseband unit for switching signal transmission from a first frequency band (i.e., FR1) to a second frequency band (i.e., FR2) in an Open Radio Access Network (O-RAN) Radio Unit (O-RU). The FR1 defines bands in a sub-6 GHz spectrum (although 7125 MHz is the maximum) and the FR2 defines bands in a mmWave spectrum. Because of the higher carrier frequencies in the FR2, the FR2 has a higher maximum bandwidth. Bandwidths include 5-100 MHz (FR1) and 50/100/200/400 MHz (FR2).
The deficiency in previous linearization techniques (as shown in FIG. 1A, FIG. 1B and background) can be solved by implementing Augmented Cascaded Feed Forward Neural Network (ACFFNN) techniques in the baseband unit of the O-RU to obtain the linear signal at a higher frequency range. The linear signal refers to an output produced by a power amplifier that is accurate copy of the input signal (generally at increased power levels). The linear signal is free from harmonic distortion or intermodulation distortion.
Unlike conventional techniques, the proposed technique includes purely baseband method that is solely implemented in the O-RU and does not need any command form an ORAN Distributed Unit (O-DU) and an ORAN-Centralized Unit (O-CU), which advantageously saves power. By using the proposed method, the baseband unit limits the use of mixers or oscillators (hardware) to reduce non-linear distortions in the signal. By using the ACFFNN technique, cascading and augmentation map all the possible crosstalk, interference, and nonlinearities of the cascaded stages of the frequency multiplier and the baseband unit of the O-RU gets a linear signal at higher frequency bands.
FIG. 2A illustrates a block diagram of a baseband unit (102). The baseband unit (102) includes a bandwidth limiting unit (102a), an ACFFNN-DPD network (102b), and a training and coefficient extraction unit (102c).
The bandwidth limiting unit (102a) limits an input bandwidth of an input signal x(n) (or X(n)) in a first frequency band, whereby the input signal is transformed into a modulated signal (e.g., baseband modulated signal or the like). In general, the input signal is generated from a baseband that contains current and past samples of Cartesian in-phase/ quadrature (I /Q) components along with envelope-dependent terms. Limiting the input bandwidth of the input signal x(n) is done dynamically. Alternatively, limiting the input bandwidth of the input signal x(n) is done based on quality of input signals and a frequency multiplication factor of a frequency multiplier (110) (as shown in FIG. 2c).
The ACFFNN-DPD (Augmented Cascaded Feed Forward Neural Network - Digital Pre-Distortion) network (102b) and the training and coefficient extraction unit (102c) estimate a DPD coefficient in an ACFFNN based on the input signal and the modulated signal. The DPD coefficient is estimated by unwrapping the input signal x(n), applying division on the input signal by a frequency multiplication factor of the frequency multiplier (110), combining a proposed signal b(n) and an input signal y(n). The proposed signal b(n) is obtained while applying bandwidth limiting on the modulated signal, and applying the ACFFNN on the combined signal having the proposed signal b(n) and the input signal y(n). The DPD coefficient estimates the coefficients of the ACFFNN of a nonlinear power amplifier (112) and the frequency multiplier (110) by receiving the baseband equivalent input y(n) and providing the baseband equivalent output x’(n) (or X’(n)) for the nonlinear power amplifier (112) and the frequency multiplier (110).
Further, the ACFFNN-DPD network (102b) and the training and coefficient extraction unit (102c) remove the distortion in the modulated signal based on the DPD coefficient to obtain a linear signal capable of being transmitted in a second frequency band. The distortion in the modulated signal is removed by receiving a trained network with an optimum error, and applying the trained network on the proposed signal (or pre-processed signal) b(n). The proposed signal b(n) is obtained while applying bandwidth limiting on the modulated signal, and linearizing a received pre-distorted output signal x’(n) to form the linear signal in the second-frequency band. The received pre-distorted output signal x’(n) is an output of the trained network. The trained network, herein is ACFFNN-DPD network (102b), is the network that is trained using a neural network from the input and output data and a trained neural network (e.g., ACFFNN) is a model having appropriate weights which is trained for minimal error.
Further, the ACFFNN-DPD network (102b) and the training and coefficient extraction unit (102c) transmit the linear signal in the second frequency band. The second frequency band is greater than the first frequency band.
In an example, as shown in FIG. 2A, x(n) represents the baseband modulated signal, b(n) represents the pre-processed signal, y(n) represents the captured signal, and x’(n) represents the pre-distorted output signal. The pre-processed signal b(n) is obtained by applying bandwidth limiting on the generated input signal. The output signal y(n) is collected from the Frequency multiplier output for defined samples in baseband. After training the model with optimum error, the proposed model is applied on the pre-processed signal to get a pre-distorted output signal, which linearizes the frequency multiplier and provides a linear signal at the second frequency band/FR-2/higher frequency bands called as pre-distorted output signal x’(n). The baseband modulated signal is expressed as shown in equation (1):
…. (1)
In order to remove an abrupt jump in a phase of the baseband modulated signal, a phase unwrapping and division by a frequency multiplication factor of the frequency multiplier (110) is applied on the baseband modulated signal. Hence the resultant signal is given as shown in equation (2):
…. (2)
The pre-processed signal b(n) and the captured signal y(n) are used for the pre-distorter training and coefficient extraction using the training and coefficient extraction unit (102c). The ACFFNN-DPD network (102b) is used for DPD coefficients estimation. In the proposed ACFFNN-DPD network (102b), an output from a previous layer is provided as input to next layers, hence it reduces number of neurons required and hidden layers required for a good convergence and reduces number of iterations. The ACFFNN-DPD network (102b) performs cascading and augmentation. The objective of doing cascading and augmentation here is to map all the possible crosstalk, interference, and nonlinearities of the cascaded stages of a frequency multiplier.
During the above operation, the frequency multiplier (110) generates memory effects that are compensated by using a delay element (not shown). Since the frequency multiplier (110) is a universal approximator, hence it will map all other hardware imperfections (such as power amplifier (PA) nonlinearity, modulator imperfection, cross-modulation, and cross-over interference due to the multiple-frequency (multiband) and multiple-input/output transmission schemes, and apply the ACFFNN for DPD coefficients estimation) in the RF chain also to provide a distortion free output. The data is divided into I and Q vectors to contain the present and past values of I and Q as shown in equation (3):
B(n)=[I_in (n),I_n (n-1).....I_in (n-m),Q_in (n),Q_in (n-1)....Q_in (n-m) |b_in (n)|,|b_in (n-1)|,....|b_in (n-M)| |b(n)|^3,|b_in (n-1)|^3,....|b_in (n-M)|^3 ] ….. (3)
where n represents the current training sample, and m is the memory depth required in the modelling. Similarly, the target output vector is defined using Iout and Qout data as shown in equation (4):
Y(n)=[I_out (n),Q_out (n)] ….(4)
Once the baseband unit (102) has the trained network with the optimum error, then the trained network is applied on the signal b(n) to obtain a pre-distorted output signal x’(n) which will linearize the frequency multiplier and the baseband unit (102) will get a linear signal at the FR-2 or higher frequency bands.
The baseband unit (102) filters an input transmission signal in accordance with the DPD coefficient data, before amplifying the higher transmission signal using the power amplifier (112).
FIG. 2B illustrates an architecture of the ACFFNN of the ACFFNN-DPD network (102b) present in the baseband unit. Typically, a cascade feed-forward neural network is a class of neural network which is like feed-forward networks but include a connection from the input and every previous layer to following layers. In a network which has three layers, the output layer is also connected directly with the input layer beside with hidden layer. As with feed-forward networks, a two or more-layer cascade-network can learn any finite input-output relationship arbitrarily well given enough hidden neurons. The cascade feed-forward neural network can be used for any kind of input to output mapping. The advantage of this method is that it accommodates the nonlinear relationship between input and output by not eliminating the linear relationship between the two.
As shown in FIG. 2B, a first layer is an input layer, second and third layers are first and second hidden layers. A fourth layer is an output layer while q and r are the numbers of neurons of two hidden layers. The net input in the first hidden layer is given as shown in equation (5):
…..(5)
wherein W1ij represents the synaptic weight between the ith input from the previous layer (input) to the jth neuron of the present layer (hidden layer 1), b1j denotes bias of the jth neuron in the first layer.
The net input in the output layer is given as shown in equation (6):
….(6)
wherein W3kl represents the synaptic weight between the kth input from the previous layer (second hidden layer) to the lth neuron of the present layer (output layer), W 3jl represents the synaptic weight between the jth input from the previous layer (first hidden layer) to the lth neuron of the present layer (output layer), and W3il represents the synaptic weight between the ith input from the previous layer (input layer) to the lth neuron of the present layer (output layer).
During the backward propagation, the performance index for ACFFNN is computed as shown in equation (7):
….(7)
where e is the error, Iout(n) and Qout(n) are the baseband output, Îout(n) and are output from the ACFFNN. Based on the error signal, backward computation is done to adjust the synaptic weights. Further, Levenberg-Marquardit (LM) technique is used which is an approximation to a Gauss-Newton`s methods. The E is minimized with respect to parameter u that depends on the synaptic weights.
From FIG. 2B, it can be observed that the input signal contains the current and past samples of the Cartesian I /Q components along with the envelope-dependent terms. The corresponding input signal represented as shown in equation (8):
…..(8)
where n represents the current training sample, and m is the memory depth for modeling. The delayed version of the signal is achieved by the Z-n. |xin(n)| is the amplitude of the current samples.
FIG. 2C illustrates a block diagram of the O-RU (100) (aka system (100)), having the baseband unit (102), for switching signal transmission from the first frequency band to the second frequency band. The O-RU (100) converts radio signals sent to and from an antenna to a digital signal that is transmitted over a fronthaul to a Distributed Unit (DU). The O-RU (100)operating in a lower frequency band can be software upgraded to a higher frequency band (mm Wave) using frequency multipliers.
The O-RU (100) includes a baseband unit (102), a Digital-to-Analog Converter (DAC) (104), a first mixer (106a), a driver power amplifier (PA) (108), the frequency multiplier (110), the PA (or nonlinear power amplifier) (112), a coupler (114), an antenna (116), an attenuator (118), a band-pass filter (BPF) (or Filter) (120), a second mixer (106b), an Analog-to-Digital Converter (ADC) (122) and a Local oscillator (LO) (124).
The baseband unit (102) limits the input bandwidth of the input signal x(n) coming from the baseband in the first frequency band to form the modulated signal, where limiting the input bandwidth reduces the distorted signal bandwidth. The input signal is transmitted to the frequency multiplier (110) and the power amplifier (112) in the O-RU (100) through the DAC (104), the first mixer (106a), and the driver power amplifier (PA) (108), where the DAC (104) converts the baseband digital signal to a baseband analog signal. The first mixer (106a) receives the baseband analog signal from the DAC (104), mixes with a radio frequency carrier received from the LO (124) and passes the mixed signal to the driver power amplifier (PA) (108). The driver power amplifier (PA) (108) is used to gain up the radio frequency carrier from a relatively low power to drive the PA or the nonlinear power amplifier (112) properly.
The frequency multiplier (110) performs the frequency translation from the first frequency band to the second frequency band in the RF chain and the power amplifier (112) converts the low-power radio-frequency signal into the higher-power signal, wherein the RF chain comprises the frequency multiplier (110), the nonlinear power amplifier (112), the band pass filter (120), and the mixers (106a, 106b). A distortion (unwanted alterations generated during amplification) is introduced in the power amplifier (112) that implies the variation in the waveform received at the output with respect to the applied input. The power amplifier (112) shares the higher-power signal to the antenna (116) through the coupler (114) without any losses, which is transmitted to the attenuator (118). The attenuator (118) can be an adjustable attenuator and a fixed attenuator, for example. The attenuator (118) controls the attenuation value of the processed baseband digital signal. The processed baseband digital signal from the attenuator (118) is fed to the ADC (122) through the BPF (120) and the second mixer (106b). The BPF (120) removes/filters the noise of processed baseband digital signal and passes the same to the second mixer (106b), where the second mixer (106b) receives the processed baseband digital signal mixed with the radio frequency carrier received from the LO (124). The ADC (122) converts the processed baseband signal to a baseband digital signal.
The baseband unit (102) receives the input signal and the modulated signal of the nonlinear power amplifier (112) and the frequency multiplier (110), the baseband digital signal (i.e., baseband equivalent input y(n)) and the baseband equivalent output of the nonlinear power amplifier (112) and the frequency multiplier (110) through the ADC (122) in the Augmented Cascaded Feed Forward Neural Network (ACFFNN) and estimates the Digital Pre-Distortion (DPD) coefficient in the ACFFNN based on the input signal and the modulated signal. The baseband unit (102) removes the distortion in the modulated signal based on the DPD coefficient on the input signal frequency multiplier to obtain the linear signal capable of being transmitted in the second frequency band and transmits the linear signal in the second frequency band and obtain the fined distortion-free signal. The functionality of the baseband unit (102) is described above in detail in conjunction with FIG. 2A and FIG. 2B.
FIG. 3 is a flow chart (300) illustrating a method of switching signal transmission from the first frequency band to the second frequency band in the O-RU (100). The operations (302-310) are performed by the baseband unit (102). For the sake of brevity, the operations and functions of the baseband unit (102) are not repeated again in the patent disclosure.
At Step (302), the method includes generating the input signal x(n) and applying the BW (bandwidth) limiting to obtain the new signal b(n). The bandwidth limiting is applied on the input signal to obtain the pre-processed signal and to remove the abrupt jump in the phase of the input signal.
At Step (304), the method includes transmitting the input signal x(n) through the frequency multiplier (110) and the PA (112). At Step (306), the method includes collecting the frequency multiplier output y(n) for the defined samples in the baseband. At Step (308), the method includes using the signal b(n) and y(n) to estimate the DPD coefficient with the neural network (ACFFNN). At Step (310), the method includes applying the DPD coefficient on the input signal to obtain x`(n) as input to the frequency multiplier (110) to obtain the linearized output (distortion free).
The proposed method is implemented in the baseband unit (102) to get the tunned signal from the baseband as input to the frequency multiplier (110) and the power amplifier (112). The method can be used to compensate for direct current (DC) offset, in-phase and quadrature-phase (IQ) imbalance along with frequency multiplier, and power amplifier (PA) distortions. The method is implemented in a Single Input Single Output (SISO) system and multiple-input and multiple-output (MIMO) system.
The proposed method reduces the cost of development of FR-2 products and saves the operation expenses (OPEX) for operators with less power consumption. The FR-2 measurement can be done using existing instruments, which will save upfront cost or can be used to create more test-benches. The proposed method can be used in FR-1 products and can be easily software upgraded to mmWave by using the frequency multipliers in FR-1 range products.
FIG. 4A to FIG. 4C are graphs 402, 404, 406 illustrating pre-distorter performance results. FIG. 4A to FIG. 4C show the efficiency of the Inverse modelling algorithm using Augmented Cascaded Feed Forward Neural Networks where the estimated and measured signal are matching.
FIG. 5 illustrates a frequency multiplication. The output of a frequency multiplier (110) is the nth harmonic of the input frequency signal. In case of PA non-linearity, the distortion seems 5 times while in case of frequency multiplier it is 10 times and using cascade multipliers will increase the non-linearity and intermixing of the components.
The various actions, acts, blocks, steps, or the like in the flow chart (300) may be performed in the order presented, in a different order or simultaneously. Further, in some implementations, some of the actions, acts, blocks, steps, or the like may be omitted, added, modified, skipped, or the like without departing from the scope of the invention.
The embodiments disclosed herein can be implemented using at least one software program running on at least one hardware device and performing network management functions to control the elements.
It will be apparent to those skilled in the art that other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention. While the foregoing written description of the invention enables one of ordinary skill to make and use what is considered presently to be the best mode thereof, those of ordinary skill will understand and appreciate the existence of variations, combinations, and equivalents of the specific embodiment, method, and examples herein. The invention should therefore not be limited by the above-described embodiment, method, and examples, but by all embodiments and methods within the scope of the invention. It is intended that the specification and examples be considered as exemplary, with the true scope of the invention being indicated by the claims.
The methods and processes described herein may have fewer or additional steps or states and the steps or states may be performed in a different order. Not all steps or states need to be reached. The methods and processes described herein may be embodied in, and fully or partially automated via, software code modules executed by one or more general purpose computers. The code modules may be stored in any type of computer-readable medium or other computer storage device. Some or all of the methods may alternatively be embodied in whole or in part in specialized computer hardware.
The results of the disclosed methods may be stored in any type of computer data repository, such as relational databases and flat file systems that use volatile and/or non-volatile memory (e.g., magnetic disk storage, optical storage, EEPROM and/or solid state RAM).
The various illustrative logical blocks, modules, routines, and algorithm steps described in connection with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. The described functionality can be implemented in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the disclosure.
Moreover, the various illustrative logical blocks and modules described in connection with the embodiments disclosed herein can be implemented or performed by a machine, such as a general purpose processor device, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components or any combination thereof designed to perform the functions described herein. A general-purpose processor device can be a microprocessor, but in the alternative, the processor device can be a controller, microcontroller, or state machine, combinations of the same, or the like. A processor device can include electrical circuitry configured to process computer-executable instructions. In another embodiment, a processor device includes an FPGA or other programmable device that performs logic operations without processing computer-executable instructions. A processor device can also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Although described herein primarily with respect to digital technology, a processor device may also include primarily analog components. A computing environment can include any type of computer system, including, but not limited to, a computer system based on a microprocessor, a mainframe computer, a digital signal processor, a portable computing device, a device controller, or a computational engine within an appliance, to name a few.
The elements of a method, process, routine, or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor device, or in a combination of the two. A software module can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of a non-transitory computer-readable storage medium. An exemplary storage medium can be coupled to the processor device such that the processor device can read information from, and write information to, the storage medium. In the alternative, the storage medium can be integral to the processor device. The processor device and the storage medium can reside in an ASIC. The ASIC can reside in a user terminal. In the alternative, the processor device and the storage medium can reside as discrete components in a user terminal.
Conditional language used herein, such as, among others, "can," "may," "might," "may," “e.g.,” and the like, unless specifically stated otherwise, or otherwise understood within the context as used, is generally intended to convey that certain alternatives include, while other alternatives do not include, certain features, elements and/or steps. Thus, such conditional language is not generally intended to imply that features, elements and/or steps are in any way required for one or more alternatives or that one or more alternatives necessarily include logic for deciding, with or without other input or prompting, whether these features, elements and/or steps are included or are to be performed in any particular alternative. The terms “comprising,” “including,” “having,” and the like are synonymous and are used inclusively, in an open-ended fashion, and do not exclude additional elements, features, acts, operations, and so forth. Also, the term “or” is used in its inclusive sense (and not in its exclusive sense) so that when used, for example, to connect a list of elements, the term “or” means one, some, or all of the elements in the list.
Disjunctive language such as the phrase “at least one of X, Y, Z,” unless specifically stated otherwise, is otherwise understood with the context as used in general to present that an item, term, etc., may be either X, Y, or Z, or any combination thereof (e.g., X, Y, and/or Z). Thus, such disjunctive language is not generally intended to, and should not, imply that certain alternatives require at least one of X, at least one of Y, or at least one of Z to each be present.
While the detailed description has shown, described, and pointed out novel features as applied to various alternatives, it can be understood that various omissions, substitutions, and changes in the form and details of the devices or algorithms illustrated can be made without departing from the scope of the disclosure. As can be recognized, certain alternatives described herein can be embodied within a form that does not provide all of the features and benefits set forth herein, as some features can be used or practiced separately from others.
, Claims:CLAIMS

We Claim:
1. A method of switching signal transmission from a first frequency band to a second frequency band in an Open Radio Access Network (O-RAN) Radio Unit (O-RU) (100), the method comprising:
limiting, by a baseband unit (102), an input bandwidth of an input signal x(n) in a first frequency band, whereby the input signal is transformed into a modulated signal;
estimating, by the baseband unit (102), a Digital pre-distortion (DPD) coefficient in an Augmented Cascaded Feed Forward Neural Network (ACFFNN) based on the input signal and the modulated signal;
removing, by the baseband unit (102), distortion in the modulated signal based on the DPD coefficient to obtain a linear signal capable of being transmitted in the second frequency band; and
transmitting, by the baseband unit (102), the linear signal in the second frequency band.

2. The method as claimed in claim 1, wherein the input signal x(n) comprises a baseband modulated signal.

3. The method as claimed in claim 1, wherein the DPD coefficient is estimated by:
unwrapping the input signal x(n);
applying division on the input signal by a frequency multiplication factor of a frequency multiplier (110);
combining a proposed signal b(n) and a baseband equivalent signal y(n), wherein the proposed signal b(n) is obtained while applying bandwidth limiting on the modulated signal; and
applying the ACFFNN on the combined signal having the proposed signal b(n) and the baseband equivalent signal y(n).

4. The method as claimed in claim 1, wherein the distortion in the modulated signal is removed by:
receiving a trained network with an optimum error;
applying the trained network on a proposed signal b(n), wherein the proposed signal b(n) is obtained while applying bandwidth limiting on the modulated signal; and
linearizing a received pre-distorted output signal x’(n) to form the linear signal in the second-frequency band, wherein the received pre-distorted output signal x’(n) is an output of the trained network.

5. The method as claimed in claim 1, wherein the DPD coefficient estimates the coefficients of the ACFFNN of a nonlinear power amplifier (112) and a frequency multiplier (110) by receiving a baseband equivalent input y(n) and providing the baseband equivalent output x’(n) for the nonlinear power amplifier (112) and the frequency multiplier (110).

6. The method as claimed in claim 1, wherein the second frequency band is greater than the first frequency band.

7. The method as claimed in claim 1, wherein limiting the input bandwidth of the input signal x(n) is done dynamically.

8. The method as claimed in claim 1, wherein limiting the input bandwidth of the input signal x(n) is done based on quality of input signals and a frequency multiplication factor of a frequency multiplier (110).

9. The method as claimed in claim 1, wherein limiting the input bandwidth of the input signal x(n) comprises:
generating one set of baseband modulated signals defined by ;
removing abrupt jump in a phase of a baseband modulated signal to generate a pre-processed signal;
combining the pre-processed signal b(n) and an input signal y(n) for a pre-distorter training and coefficient extraction;
mapping hardware imperfections with the extracted coefficient;
receiving a trained network with the optimum error, wherein the trained network is a network that is trained using a neural network from an input data and an output data, wherein the trained network is a model having appropriate weights which is trained for minimal error; and
applying the trained network on the pre-processed signal b(n) to get a pre-distorted output signal x’(n), wherein the pre-distorted output signal x’(n) linearizes the frequency multiplier (110) and a power amplifier (112) to get a linear signal at the second frequency band.

10. The method as claimed in claim 9, wherein removing the abrupt jump in the phase of baseband modulated signal comprises:
unwrapping a division by a frequency multiplication factor of the frequency multiplier (110) applied to the modulated signal, wherein the resultant signal is given as:

wherein, the baseband modulated signal is expressed as:

while applying division by the frequency multiplication factor of the frequency multiplier (110), where K = frequency multiplication factor and b (n)= Proposed signal.

11. The method as claimed in claim 9, wherein the hardware imperfections comprising power amplifier (PA) nonlinearity, modulator imperfection, cross-modulation, and cross-over interference due to the multiple-frequency (multiband) and multiple-input/output transmission schemes.

12. The method as claimed in claim 9, comprising:
filtering, by the baseband unit (102), an input transmission signal in accordance with the DPD coefficient, before amplifying the higher transmission signal using the power amplifier (112).

13. The method as claimed in claim 1, wherein the method is used to compensate for direct current (DC) offset, in-phase and quadrature-phase (IQ) imbalance along with frequency multiplier, and power amplifier (PA) distortions.

14. The method as claimed in claim 1, wherein the method is implemented in a Single Input Single Output (SISO) system and multiple-input and multiple-output (MIMO) system.

15. An Open Radio Access Network (O-RAN) Radio Unit (O-RU) (100) for switching signal transmission from a first frequency band to a second frequency band, wherein the O-RU (100) is configured to:
limit an input bandwidth of an input signal x(n) coming from a baseband in a first frequency band to form a modulated signal using a baseband unit (102), wherein the input signal is transmitted to a frequency multiplier (110) and a power amplifier (112) in the O-RU (100), wherein limiting the input bandwidth reduces a distorted signal bandwidth;
place the frequency multiplier (110) in a Radio Frequency (RF) chain for frequency translation from the first frequency band to the second frequency band, wherein the RF chain comprises at least one nonlinear power amplifier (112), at least one filter (120), and mixers (106a, 106b);
input the input signal and the modulated signal of the at least one nonlinear power amplifier (112) and the frequency multiplier (110), baseband equivalent input and baseband equivalent output of the at least one nonlinear power amplifier (112) and the frequency multiplier (110) in an Augmented Cascaded Feed Forward Neural Network (ACFFNN);
estimate a Digital Pre-Distortion (DPD) coefficient in the ACFFNN based on the input signal and the modulated signal;
remove distortion in the modulated signal based on the DPD coefficient on the input signal frequency multiplier to obtain a linear signal capable of being transmitted in the second frequency band; and
transmit the linear signal in the second frequency band and obtain the fined distortion-free signal.