Reduction of Power Fluctuation in ECMA-368 Ultra Wideband Communication Systems Using Multilayer Perceptron Neural Networks

Louliej, Abdelhamid; Jabrane, Younes; Said, Brahim Ait Es; Ouahman, Abdellah Ait

doi:10.1007/s11277-013-1096-x

Cited by 3 publications

(6 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since ANNs work with integer inputs, the ACE‐AGP signals must be decomposed into real and imaginary parts. The training process of an ANNT can be described as Use the ACE‐AGP algorithm to obtain x AGP from the original time‐domain signal x Org Split the x Org and x AGP signals into training x Org, tr , x AGP, tr and test sets x Org, ts , x AGP, ts . Decompose the x Org and x AGP signals into real and imaginary parts

((),, x_{Re}^{italicOrg}, x_{Im}^{Org, tr})

((),, x_{Re}^{italicAGP}, x_{Im}^{AGP, tr})

, respectively. Create two ANN models ANNT Re , ANNT Im for real and imaginary parts, respectively. Get

x_{Re}^{ANNT}

and

x_{Im}^{ANNT}

by training the models ANNT Re and ANNT Im with the pairs

((),, x_{Re}^{italicANNT}, x_{Re}^{italicAGP})

and

((),, x_{Im}^{italicOrg}, x_{Im}^{italicAGP})

, respectively. Obtain x ANNT with

x^{ANNT} = x_{Re}^{ANNT} + {jx}_{Im}^{ANNT}

. More details about ANNT method‐based PAPR reduction can be found in Louliej et al for interested readers. The major drawback of ANNT‐based PAPR reduction is its offline training time based on ACE‐AGP signals, which is not suitable for real‐time applications.…”

Section: Papr Reduction Methodsmentioning

confidence: 99%

“…Since ANNs work with integer inputs, the ACE-AGP signals must be decomposed into real and imaginary parts. The training process of an ANNT can be described as More details about ANNT method-based PAPR reduction can be found in Louliej et al 16 for interested readers. The major drawback of ANNT-based PAPR reduction is its offline training time based on ACE-AGP signals, which is not suitable for real-time applications.…”

Section: Time-domain Anns (Annt) Methodsmentioning

confidence: 99%

“…In Mishra et al, the ANNs are being successfully applied to reduce the high computational complexity of OFDM systems. Moreover, in Louliej et al, a hybridization between ACE‐AGP and ANNs is proposed for PAPR reduction in ECMA‐368 UWB systems. The main idea is to train a feedforward ANN with two layers and two neurons per layer with the radial basis function at the input layer and the linear function at the output layer using the low PAPR signals obtained via the ACE‐AGP method.…”

Section: Papr Reduction Methodsmentioning

confidence: 99%

“…In addition, the non‐linearities of high power amplifier generate out‐of‐band distortion and increase the bit error rate (BER). Many techniques have been used to reduce PAPR in OFDM signals, among them the selected mapping method (SLM), partial transmit sequences (PTS), active constellation extension (ACE), and time‐domain artificial neural networks (ANNT) are proposed for ECMA‐368 UWB systems.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

PAPR reduction in ECMA‐368 UWB communication systems using parametric discrete sliding norm transform

Mehallel

Abed

Boukrouche

et al. 2018

Int J Communication

View full text Add to dashboard Cite

Summary ECMA‐368 Standard adopts multiband orthogonal frequency division multiplexing (MB‐OFDM) for high rate ultra‐wideband (UWB) communication systems as a possible candidate for indoor wireless personnel area networks. One of the main drawbacks of MB‐OFDM modulation is the high peak‐to‐average power ratio (PAPR), which limits the power efficiency of the digital‐to‐analog converter and the high power amplifier due to nonlinear distortion. In this paper, a post inverse discrete Fourier transform nonlinear transformation called parametric discrete sliding norm transform (PDSNT) is proposed to reduce the PAPR in MB‐OFDM systems. The PDSNT method in addition to its much lower computational complexity than other schemes recently proposed for ECMA‐368 UWB systems introduces an efficient controlling parameter to obtain a good tradeoff between PAPR reduction and bit error rate (BER) performance degradation. Simulation results show that the proposed PDSNT‐based MB‐OFDM scheme can achieve better performance in PAPR reduction than conventional solutions with affordable BER performance degradation in UWB channel models.

show abstract

((),, x_{Re}^{italicOrg}, x_{Im}^{Org, tr})

((),, x_{Re}^{italicAGP}, x_{Im}^{AGP, tr})

, respectively. Create two ANN models ANNT Re , ANNT Im for real and imaginary parts, respectively. Get

x_{Re}^{ANNT}

and

x_{Im}^{ANNT}

by training the models ANNT Re and ANNT Im with the pairs

((),, x_{Re}^{italicANNT}, x_{Re}^{italicAGP})

and

((),, x_{Im}^{italicOrg}, x_{Im}^{italicAGP})

, respectively. Obtain x ANNT with

x^{ANNT} = x_{Re}^{ANNT} + {jx}_{Im}^{ANNT}

Section: Papr Reduction Methodsmentioning

confidence: 99%

Section: Time-domain Anns (Annt) Methodsmentioning

confidence: 99%

Section: Papr Reduction Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

PAPR reduction in ECMA‐368 UWB communication systems using parametric discrete sliding norm transform

Mehallel

Abed

Boukrouche

et al. 2018

Int J Communication

View full text Add to dashboard Cite

show abstract

“…In [11], a neural network (NN) technique, referred to as Multilayer Perceptrons (MLPs), to obtain signals with low envelope fluctuations has been developed. Indeed, NN have been widely applied in solving optimization problems [12,13,14]. In the case of the PAPR proposal in [11], the NN were trained with the Approximate Gradient Projection (AGP) from ACE [10] and thus the result is an NN that generates from the original signal another one with similar characteristics as ACE but without its complexity and in one shot.…”

Section: Introductionmentioning

confidence: 99%

Practical Guidelines for Approaching the Implementation of Neural Networks on FPGA for PAPR Reduction in Vehicular Networks

Louliej

Jabrane

Jiménez

et al. 2018

Sensors

Self Cite

View full text Add to dashboard Cite

Nowadays, the sensor community has become wireless, increasing their potential and applications. In particular, these emerging technologies are promising for vehicles’ communications (V2V) to dramatically reduce the number of fatal roadway accidents by providing early warnings. The ECMA-368 wireless communication standard has been developed and used in wireless sensor networks and it is also proposed to be used in vehicular networks. It adopts Multiband Orthogonal Frequency Division Multiplexing (MB-OFDM) technology to transmit data. However, the large power envelope fluctuation of OFDM signals limits the power efficiency of the High Power Amplifier (HPA) due to nonlinear distortion. This is especially important for mobile broadband wireless and sensors in vehicular networks. Many algorithms have been proposed for solving this drawback. However, complexity and implementations are usually an issue in real developments. In this paper, the implementation of a novel architecture based on multilayer perceptron artificial neural networks on a Field Programmable Gate Array (FPGA) chip is evaluated and some guidelines are drawn suitable for vehicular communications. The proposed implementation improves performance in terms of Peak to Average Power Ratio (PAPR) reduction, distortion and Bit Error Rate (BER) with much lower complexity. Two different chips have been used, namely, Xilinx and Altera and a comparison is also provided. As a conclusion, the proposed implementation allows a minimal consumption of the resources jointly with a higher maximum frequency, higher performance and lower complexity.

show abstract

Dimensioning an FPGA for Real-Time Implementation of State of the Art Neural Network-Based HPA Predistorter

et al. 2021

Self Cite

View full text Add to dashboard Cite

Orthogonal Frequency Division Multiplexing (OFDM) is one of the key modulations for current and novel broadband communications standards. For example, Multi-band Orthogonal Frequency Division Multiplexing (MB-OFDM) is an excellent choice for the ECMA-368 Ultra Wideband (UWB) wireless communication standard. Nevertheless, the high Peak to Average Power Ratio (PAPR) of MB-OFDM UWB signals reduces the power efficiency of the key element in mobile devices, the High Power Amplifier (HPA), due to non-linear distortion, known as the non-linear saturation of the HPA. In order to deal with this limiting problem, a new and efficient pre-distorter scheme using a Neural Networks (NN) is proposed and also implemented on Field Programmable Gate Array (FPGA). This solution based on the pre-distortion concept of HPA non-linearities offers a good trade-off between complexity and performance. Some tests and validation have been conducted on the two types of HPA: Travelling Wave Tube Amplifiers (TWTA) and Solid State Power Amplifiers (SSPA). The results show that the proposed pre-distorter design presents low complexity and low error rate. Indeed, the implemented architecture uses 10% of DSP (Digital Signal Processing) blocks and 1% of LUTs (Look up Table) in case of SSPA, whereas it only uses 1% of LUTs in case of TWTA. In addition, it allows us to conclude that advanced machine learning techniques can be efficiently implemented in hardware with the adequate design.

show abstract

Reduction of Power Fluctuation in ECMA-368 Ultra Wideband Communication Systems Using Multilayer Perceptron Neural Networks

Cited by 3 publications

References 22 publications

PAPR reduction in ECMA‐368 UWB communication systems using parametric discrete sliding norm transform

PAPR reduction in ECMA‐368 UWB communication systems using parametric discrete sliding norm transform

Practical Guidelines for Approaching the Implementation of Neural Networks on FPGA for PAPR Reduction in Vehicular Networks

Dimensioning an FPGA for Real-Time Implementation of State of the Art Neural Network-Based HPA Predistorter

Contact Info

Product

Resources

About