Variable FFT Engine for Base Station Multi-Frequency Allocation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing fast Fourier transform (FFT) hardware configurations in base stations require multiple FFT engines of varying sizes to support different communication bands, leading to inefficient hardware usage and increased costs due to unnecessary resource consumption.

Innovation Solution

A variable FFT apparatus and method that combines minimal hardware components, including storage units for twiddle factors and FFT processors, to perform fast Fourier transforms of varying sizes by selecting appropriate twiddle factors and aligning input data for efficient multi-size, multi-output processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple FFT hardware engines of varying sizes are provided to support different communication bands, then the system can support various FFT sizes (1024-point, 2048-point, etc.), but the hardware resources are wasted and system cost increases

Engineering Contradiction:
Improvesupport for various FFT sizesVSAvoidhardware resources
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The patent implements a universal FFT hardware engine that can perform multiple FFT sizes (1024-point, 2048-point, etc.) through software configuration rather than requiring separate dedicated hardware engines for each size. The single engine adapts to different FFT sizes by loading appropriate twiddle factors and configuring processing parameters, eliminating the need for multiple specialized hardware components while maintaining full support for various communication bands and FFT requirements

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent changes the operational parameters of the FFT engine (such as twiddle factors, buffer sizes, and processing configurations) to adapt to different FFT size requirements. By dynamically adjusting these parameters based on the required FFT size, the system achieves versatility across multiple communication bands without requiring physical hardware changes or additional dedicated engines for each configuration

Inventive Principle:
Principle #35Parameter changes

2Quantity of substance

If a variable fast Fourier transform hardware engine supporting 1024 points and 2048 points is used, then hardware resources are reduced, but one 2048-point engine is used to form a 1024-point engine resulting in unnecessary consumption of hardware resources

Engineering Contradiction:
Improvehardware resourcesVSAvoidhardware resource consumption
Core Design Contradiction:
Quantity of substanceVSLoss of energy

Solution Approach 1:

The patent implements a dynamic FFT engine that can adapt its operational mode based on the required FFT size. Rather than statically configuring the engine for a specific size or using a larger engine to simulate a smaller one, the system dynamically reconfigures its internal processing parameters, twiddle factors, and data flow to efficiently handle the exact FFT size required, eliminating unnecessary computational operations and hardware resource consumption

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS8510362B2Apparatus and method for variable fast fourier transform
Publication Date: 2013.08.13 SAMSUNG ELECTRONICS CO LTD
  • US8510362B2 patent drawing
  • US8510362B2 patent drawing
  • US8510362B2 patent drawing

AI summary

The present invention relates to an apparatus and method for variable fast Fourier transform. According to an embodiment of the present invention, two n-point fast Fourier transform (FFT) processors are used to generate two n-point FFT output data or one 2n-point FFT output data. The one 2n-point input data is alternately input to the two n-point FFT processors. Each of the two n-point FFT processors selects a twiddle factor for the n-point input data or the 2n-point input data and performs fast Fourier transform. A butterfly operation is performed on signals obtained by performing fast Fourier transform on the 2n-point input data signal, and the processed signals are aligned in an output order. According to this structure, it is possible to realize a fast Fourier transform hardware engine that selectively performs multi-frequency allocation in a base station system that supports the multi-frequency allocation.