Optimization of the 24-Bit Fixed-Point Format for the Laplacian Source

The 32-bit floating-point (FP32) binary format, commonly used for data representation in computers, introduces high complexity, requiring powerful and expensive hardware processing energy consumption, hence being unsuitable implementation on sensor nodes, edge devices, other devices with limited resources. Therefore, it is often necessary to use formats of lower complexity than FP32. This paper proposes the usage 24-bit fixed-point format that will reduce two ways, by decreasing number bits fact has significantly less format. optimizes examines its performance Laplacian distribution, exploiting analogy between uniform quantization. Firstly, optimization quantizer performed deriving new closed-form formulas a very accurate calculation maximal amplitude. Then, optimized key parameter proposing adaptation procedures, aim obtain same as optimal wide range variance input data. It shown proposed achieves 18.425 dB higher while complex.