Probabilistic Amplitude Shaping for PAPR Reduction in High-Order Modulation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing PAPR management techniques, such as tone reservation and frequency domain spectral shaping, are ineffective for high-order modulation schemes like 16-QAM and 32-QAM, leading to challenges in reducing peak-to-average power ratio in 5G and 6G wireless communication systems, which affects amplifier efficiency and bit rate.

Innovation Solution

A distribution matcher processes input bit sequences to generate symbol sequences with predefined transition probabilities, reducing symbol-to-symbol transitions associated with higher power fluctuations, and a channel coder multiplies these symbols with parity bits to achieve lower PAPR, applicable to higher-order modulation schemes like QAM-16 or QAM-32.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If tone reservation and frequency domain spectral shaping techniques are used, then PAPR is reduced, but effectiveness deteriorates for high-order modulation schemes like 16-QAM and 32-QAM

Engineering Contradiction:
ImprovePAPRVSAvoideffectiveness for high-order modulation
Core Design Contradiction:
Loss of energyVSAdaptability or versatility

Solution Approach 1:

The invention changes the probability distribution parameter of symbol transitions from uniform to non-uniform, specifically designing a transition probability matrix that favors transitions between symbols with smaller amplitude differences. This parameter change in the statistical properties of the modulation sequence effectively reduces PAPR while maintaining compatibility with high-order modulation schemes like 16-QAM and 32-QAM

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The invention performs preliminary shaping of the modulation sequence by pre-defining transition probabilities before transmission. The transmitter pre-processes the bit sequence into a shaped modulation sequence that inherently has lower PAPR characteristics, avoiding the need for post-processing or reserved tones, thus making the technique effective for high-order modulations

Inventive Principle:
Principle #10Preliminary action

2Use of energy by moving object

If PAPR is reduced using conventional techniques, then amplifier efficiency improves, but bit rate capability deteriorates for high-order modulations

Engineering Contradiction:
Improveamplifier efficiencyVSAvoidbit rate
Core Design Contradiction:
Use of energy by moving objectVSProductivity

Solution Approach 1:

By changing the transition probability parameters to favor smaller amplitude transitions, the invention reduces the peak power requirements, allowing the amplifier to operate more efficiently in its linear region. Simultaneously, the shaped sequences maintain the full spectral efficiency of high-order modulations like 16-QAM and 32-QAM, preserving bit rate capability

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The preliminary shaping of modulation sequences ensures that the reduced PAPR and high bit rate capabilities are both achieved from the outset. The shaped sequences are generated before transmission with optimized transition probabilities that simultaneously satisfy both amplifier efficiency requirements and high data rate requirements for 5G and 6G systems

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS20240340209A1Probabilistic amplitude shaping applied to PAPR reduction
Publication Date: 2024.10.10 NOKIA TECHNOLOGIES OY
  • US20240340209A1 patent drawing
  • US20240340209A1 patent drawing
  • US20240340209A1 patent drawing

AI summary

According to an example aspect of the present disclosure, there is provided an apparatus comprising memory configured to store a set of probabilities of transitions between modulation symbols, the set of probabilities corresponding to a lower peak-to-average power ratio than a set of equal probabilities of transitions between modulation symbols, distribution matcher circuitry configured to process an input bit sequence into a symbol sequence, wherein frequencies of transitions between symbols in the symbol sequence conform to probabilities in the set of probabilities, a channel coder configured to generate parity bits of unity and negative unity from bit information obtained from the symbol sequence, grouping circuitry configured to assign symbols of the symbol sequence into groups each comprising two or more symbols, and multiplication circuitry configured to multiply each one of the groups with a corresponding one of the parity bits.