Hybrid matrix recognition method in underdetermined blind source separation based on tensor regular decomposition
An underdetermined blind source separation and mixing matrix technology, applied in the field of communication, can solve problems such as difficult to meet, unsatisfactory performance, and affecting the recognition accuracy of the mixing matrix, so as to overcome the difficulty of extraction, solve the identification of the mixing matrix, and improve the recognition accuracy Effect
Patent Information
- Authority / Receiving Office
- CN · China
- Current Assignee / Owner
- Publication Date
- 2015-02-25
Smart Images
Figure 1 Figure 2 Figure 3
Abstract
Description
Technical field
[0001] The invention belongs to the field of communication technology, and particularly relates to a hybrid matrix identification method, which can be used in under-determined blind source separation of source signals in the fields of speech, communication, radar and biomedicine under time-frequency aliasing conditions. Background technique
[0002] Blind source separation BSS refers to the purpose of separating the source signal only through the observation signal received by the sensor under the condition of unknown transmission channel and source signal. This method has been widely used in speech signal processing, image processing, radar, Various fields such as communication and biomedicine. As a classic algorithm for blind source separation, independent component analysis ICA and its extended algorithms are mostly used to solve problems when the number of observation signals is equal to or greater than the number of source signals. This kind of blind source s...
Examples
Embodiment Construction
[0020] The present invention will be described in further detail below with reference to the accompanying drawings.
[0021] Refer to figure 1 , The implementation steps of the present invention are as follows:
[0022] Step 1: Sample the source signal at the receiving end to obtain the observation signal.
[0023] M sensors sample the source signal at equal intervals at time t to obtain the observation signal x i (t), where 1≤i≤M, t∈[1,2,...,N], N is the length of the sampled data.
[0024] Step 2: Calculate the fourth-order covariance matrix of the observed signal.
[0025] (2.1) Calculate the fourth moment of the observation signal:
[0026] m ^ i , j , k , l ( τ 1 , τ 2 , τ 3 ) = 1 T X t = 1 N x i ( t ) x j * ( t + τ 1 ) x k * ( t + τ 2 ) x l ( t + τ 3 ) ,
[0027] Among them, 1≤i,j,k,l≤M,τ 1 ,τ 2 ,τ 3 They are the delays ...