Floating-Point Counter Arrays for Long-Interval Waveform Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fixed point counters become unwieldy and inefficient when dealing with a large number of count values, requiring excessive data storage and consuming more power due to the need for a large number of bits to represent long times between states in waveforms.
Innovation Solution
The implementation of floating point counters that adjust precision dynamically, using a combination of mantissa and exponent values to specify times, reducing the number of bits required and enabling efficient generation and analysis of waveforms with varying pulse durations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If fixed point counters are used to represent large count values, then the precision and range of count values can be maintained, but the data storage requirements and circuit complexity increase significantly
Solution Approach 1:
The patent changes the numerical representation parameter from fixed-point to floating-point format. This allows the same circuit width to represent both small values with high precision and large values with appropriate precision, resolving the contradiction between precision and circuit complexity.
Solution Approach 2:
The patent introduces dynamic precision adjustment through floating-point representation, where the precision and range of count values can adapt based on the magnitude of the value being represented. This dynamic approach eliminates the need for consistently high precision across all count values, reducing circuit complexity.
2Duration of action of moving object
If fixed point counters are used to represent large count values, then the full range of count values can be captured, but the power consumption increases due to the large number of bits required
Solution Approach 1:
The patent changes the time representation parameter from fixed-point to floating-point format, enabling efficient representation of both short and extremely long time intervals between waveform states. This parameter change reduces the average number of bits required, thereby reducing power consumption while maintaining the ability to capture the full range of time durations.
Solution Approach 2:
The patent introduces dynamic bit allocation for time representation, where the number of bits used to represent time intervals adapts based on the actual duration being measured. Short intervals use fewer bits while extremely long intervals use more bits, optimizing power consumption across different operating conditions.
3Measurement precision
If fixed point counters are used with a large number of bits to represent long times, then the accuracy of time measurement is maintained, but the data storage requirements become excessive
Solution Approach 1:
The patent changes the time measurement representation from fixed-point to floating-point format, allowing accurate representation of both short and extremely long time intervals using a consistent, limited number of bits. This parameter change eliminates the need to allocate maximum bits for all time measurements, reducing overall data storage requirements while maintaining measurement accuracy across the full range of time values.
Data Source
AI summary
Aspects of the present disclosure relate to floating point timers and counters that are used in a variety of contexts. In some implementations, a floating point counter can be used to generate a wave form made up of a series of pulses with different pulse lengths. An array of these floating point counters can be used to implement a pool of delays. In other implementations, an array of floating point counters can be used to analyze waveforms on a number of different communication channels. Analysis of such waveforms may be useful in automotive applications, such as in wheel speed measurement for example, as well as other applications.