Fixed-weight coding method for uniformity processing in holographic data storage.

Area of Science:

• Optoelectronics
• Data Storage Technologies
• Information Science

Background:

• Holographic data storage offers high capacity and speed but suffers from non-uniform diffraction efficiency.
• This non-uniformity distorts reconstructed images, hindering deep learning decoding and increasing bit error rates.
• Existing decoding methods struggle with the energy distribution variations inherent in holographic media.

Purpose of the Study:

• To develop a robust method for improving data integrity in holographic storage systems.
• To mitigate the impact of non-uniform diffraction efficiency on data reconstruction and decoding.
• To enhance the performance of deep learning models for holographic data decoding.

Main Methods:

• Implemented a fixed-weight multi-level data page encoding method for uniform energy distribution.
• Introduced an adaptive homomorphic filtering technique to reduce energy distribution variance in reconstructed images.
• Trained a complex amplitude decoding convolutional neural network using the filtered dataset and encoding information.

Main Results:

• Achieved a uniform energy distribution during the data page encoding phase.
• Successfully reduced the variance in energy distribution of reconstructed holographic images.
• Reduced the bit error ratio of complex-amplitude data by over 60% compared to traditional methods.

Conclusions:

• The proposed fixed-weight encoding and adaptive filtering methods effectively address non-uniform energy distribution in holographic storage.
• This approach significantly enhances data decoding accuracy and supports the advancement of holographic storage technology.
• The methods show potential for broader applications beyond holographic data storage.