A Fractional Delay Method Based on Dynamic Length Frame Processing
A technology of fractional delay and frame processing, which is applied in the field of signal processing, can solve the problems of increased calculation amount and complex filter design process, etc., and achieve the effect of improving performance, good delay performance and improving delay accuracy
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Publication Date
- 2021-11-02
Smart Images

Figure 1 
Figure 2 
Figure 3
Abstract
Description
Technical field
[0001] The present invention belongs to the field of signal processing, particularly to a method for dynamic fractional delay length of the frame processing. Background technique
[0002] Precise digital signals in a communication delay, synchronization, signal processing, speech signal processing, have a wide field of applications. Especially in array signal processing, digital beam forming, the field of wideband analog signal source, time-domain analysis and adaptive time delay estimation target echo and other problems are related to the precise time delay signal.
[0003] Fractional digital delay currently used design methods are mainly two: (1) Direct time-domain implementation. This had mainly two forms dense sampling and interpolation in the time domain. Oversampling the received signal in the process, it is achieved by increasing the sampling frequency of the analog signal. It will increase the computational complexity and hardware cost of a substantial inc...
Examples
Embodiment 1
[0166] The simulation signal parameters are set to: set the sampling frequency fs to 2000 Hz, the sampling function is:
[0167] COS (2 * pi * fc1 * t) + sin (2 * pi * fc2 * t)
[0168] Among them, Fc1 = 21.786Hz, FC2 = 45.337 Hz, total sample point number L = 1999, acquisition data sequence s (n); delay time is set to t_delay = 8.3567.
[0169] According to (2), the single processing data segment length k = 1000; the linear decision window length τ = 31; the linear detection amplitude threshold g = 5; single frame length threshold A = 100.
[0170] According to the step (3), the sampling data sequence S (N) is segmented, the sliding step is b = 999, resulting in segmentation data sequence S j (L), J = 0, 1; L = 0, 1, ..., 999, J is the segmentation segmentation number, L for each segment data sequence sampling point number, at which time the number of data segments q = 2.
[0171] According to (4), set the current processing data sequence number j = 0, and t_delay is divided into...