Selective Tone Event Detector for Digital Telephony
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing tone event detection systems in digital telephony face complexity and inefficiency due to the need for multiple independent tone detectors, often resulting in errors, especially when detecting tones with low-frequency components, which require extensive processing and memory resources.
Innovation Solution
A communication system with a tone event detector that includes a pre-processing unit, a bank of dependent tone detectors, and decision logic, which selectively enables detectors based on tone indication, reducing complexity and processing time by enabling detection only when a tone is present, and using a multistage detection process to identify tone characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple independent tone detectors are used to detect various tone events, then the detection capability for different modulation schemes is improved, but the system complexity increases significantly
Solution Approach 1:
The patent employs a single universal tone detector that can detect multiple types of tone events with different modulation schemes (AM, FM, phase modulation) and frequency components. The detector uses a bank of Goertzel filters that can be selectively activated based on the detected tone characteristics, allowing one detector to perform the work of multiple specialized detectors while maintaining low system complexity.
Solution Approach 2:
The tone detector dynamically adjusts its operation by selectively enabling different Goertzel filters based on the detected tone characteristics. The system transitions from a static configuration of multiple fixed detectors to a dynamic approach where a single detector adapts its filter bank activation based on real-time tone analysis, reducing complexity while maintaining versatility.
2Measurement precision
If Goertzel Filters are used to detect tone events, then the estimation of Fourier transform at specific frequencies is improved, but the processing time and memory cost increase significantly
Solution Approach 1:
Instead of continuously activating all Goertzel filters in the bank, the system activates only the necessary number of filters based on the detected tone characteristics. This partial action approach reduces processing time and memory usage while maintaining sufficient precision for accurate tone event detection, avoiding the excessive computation of analyzing all possible frequency components when fewer are actually present.
3Measurement precision
If Goertzel Filters are used to detect low-frequency tones, then the detection capability is improved, but the number of required samples increases significantly
Solution Approach 1:
The system performs preliminary tone indication detection using a simplified method before activating the full Goertzel filter bank for detailed analysis. This preliminary action quickly identifies the presence and basic characteristics of low-frequency tones, allowing the system to prepare the appropriate filters in advance and reduce the total number of samples required for accurate detection compared to continuously monitoring all frequencies.
4Adaptability or versatility
If multiple independent tone detectors are deployed, then the coverage of tone events is improved, but the error rate increases
Solution Approach 1:
The patent merges the functionality of multiple independent tone detectors into a single integrated tone detector that uses a unified decision-making process. By combining the detection results from multiple Goertzel filters within one detector and using a single set of decision logic, the system eliminates the errors that arise from multiple independent detectors making separate decisions, while maintaining comprehensive coverage of various tone events.
Data Source
AI summary
In one embodiment, a tone event detector first determines whether the presence of a tone is indicated on the input signal, and then, based on this determination, selectively determines whether a tone has been detected on the input signal. For example, in one embodiment, tone detection is performed only when the presence of a tone is first indicated, such that if the presence of a tone is not indicated, tone detection need not be performed. This may help reduce complexity of a tone event detector since a simplified method may be used to indicate the presence of a tone, and the more complex algorithms for tone detection may be enabled only when needed. Also, in one embodiment, detection of a tone includes generating one or more tone characteristics corresponding to the detected tone which may then be used to determine whether the detected tone corresponds to a valid tone event.