Digital Filter Memory Allocation for Signal Value Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital filters face inefficiencies in memory usage due to the majority of memory cells remaining unused as most signal values are similar, leading to poor memory saving efficiency.
Innovation Solution
A digital filter design that includes a processor and memory for updating signal value and statistical information sets, allowing for the reduction of signal values and insertion of new values based on check results, enhancing memory efficiency by utilizing a signal value set and aggregated value set, and optionally including a difference value set for calculating output values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a memory provides 256 memory cells to save aggregated data amount for all possible signal values (0-255), then the filter can handle all signal values, but most memory cells remain unused resulting in poor memory saving efficiency
Solution Approach 1:
The patent segments the signal values into multiple groups or ranges, and only allocates memory cells for the signal values that actually appear in the input data. Instead of providing memory for all 256 possible signal values, the system dynamically creates memory entries only for observed signal values, thereby segmenting the memory allocation based on actual usage patterns and eliminating waste from unused memory cells.
2Adaptability or versatility
If the memory saves aggregated data amount for each signal value from 0 to 255, then the filter can generate output for any signal value, but the saving efficiency of memory is poor due to most memory cells being unused
Solution Approach 1:
The patent implements dynamic memory allocation where the memory structure adapts to the actual signal values present in the input data. The system dynamically creates, updates, and removes memory entries based on which signal values are observed, rather than using a static pre-allocated memory structure. This dynamic approach maintains adaptability to handle various signal values while optimizing memory efficiency by only allocating space for actually used signal values.
Data Source
AI summary
A digital filter for filtering signals includes a processer and a memory. The processer receives a plurality of input signal values and a plurality of input aggregated values corresponding to the plurality of input signal values. The memory saves a signal value set and a statistical information set. The processer generates a check result according to the plurality of input signal values and the signal value set. The processer updates the statistical information set of the memory according to the check result, and decides whether to insert at least one new signal value into the signal value set of the memory, wherein the processer generates an output value according to the signal value set and the statistical information set of the memory.