A Novel Companding Technique to Reduce High Peak to Average Power Ratio in OFDM Systems

Abdulwahid Mohammed,Amin Nassar,Tawfik Ismail,Hassan Mostafa

doi:10.1109/access.2021.3062820

Abdulwahid Mohammed, Amin Nassar + Show 2 more

Open Access

https://doi.org/10.1109/access.2021.3062820

Copy DOI

Abstract

The reduction of the high peak-to-average-power ratio (PAPR) is important to the efficiency of the orthogonal frequency division multiplexing (OFDM) technique. Excessive PAPR contributes to nonlinear clipping induced harmonic distortions that reduce system reliability. In this article, a new technique for decreasing the high PAPR in OFDM with minimum effects on the system performance is proposed. The technique uses the image adjust (IMADJS) function to reduce the high PAPR of transmitted OFDM signals by compressing large signals and expanding small signals. In comparison, the IMADJS strategy has the advantage of maintaining a constant average power level before and after companding. A comparative analysis is provided between the proposed (IMADJS) technique and well-known companding techniques such as μ-law, absolute exponential (AEXP), and the new error function (NERF). The comparison is based on PAPR, bit error rate (BER), power spectral density (PSD), and average power performance metrics. Simulation results confirm that the IMADJS technique significantly improved the drawbacks of the PAPR. Furthermore, the PAPR is reduced by 2.81dB. The IMADJS technique has less impact on the original power spectrum than on other companding schemes. The average out-of-band radiation of the IMADJS technique reaches about -50dB at the frequency of 0MHz. In contrast, the average of the original OFDM signal is that the out-of-band radiation reaches around -52dB. The AEXP and NERF companding techniques reach about -46dB, while the μ-law companding technique hits about -37dB.

Highlights

Orthogonal Frequency Division Multiplexing (OFDM) is a multi-carrier technique, which is effective and efficient in 4G and 5G mobile networks [1]–[3]
The results show that the new error function (NERF) companding technique provided better Peak-to-Average Power Ratio (PAPR) reduction and bit error rate (BER) performance than the μ−law method
The complexity calculations are considered under the following assumptions: 1) The complexity of addition is equal to the complexity of subtraction, 2) The complexity of multiplication matches the complexity of division, 3) Ignoring the complexity of abs, max, and compare functions, 4) Exponential function requires two multiplications and one addition, 5) erf(t) function requires 2t multiplications and 2t additions, 6) In absolute exponential (AEXP) companding technique, we assumed d = 2

Summary

INTRODUCTION

Orthogonal Frequency Division Multiplexing (OFDM) is a multi-carrier technique, which is effective and efficient in 4G and 5G mobile networks [1]–[3]. The observed instantaneous PAPR might reach as high values as N times of the average OFDM symbol amplitude [6] In this case, high power amplifier (HPA) and digital to analog converter (DAC) are required to overcome the large dynamic. A closed-form of BER expression is proposed and validated with the Monte-Carlo simulation This scheme reduced the PAPR at the expense of an increase in the average signal power, which causes a nonlinear distortion. It used a non-linear function to change the compressed signal to a uniform distribution This scheme essentially reduced the PAPR and kept the average signal-power constant. This causes a relatively high signal distortion and spectrum side-lobes generation, which increases the BER. Where, E{.} is the statistical expectation operator, max |x(n)|2 is the peak OFDM signal power and E{|x(n)|2} is the average power of the OFDM signal

MATHEMATICAL FORMULATION OF THE PROPOSED COMPANDING TECHNIQUE

COMPANDING ALGORITHMS AND COMPUTATIONAL COMPLEXITY ANALYSIS

SIMULATION RESULTS AND DISCUSSION

CONCLUSION