A Method for Separating Mixed Signals in Time-Frequency Domain

Abstract

The invention discloses a method for separating time-frequency domain signals. A system for wireless channel statistical multiplexing is composed of source signals, a wireless channel, a DSP separating system and output signals. The DSP separating system adjusts elements in a separating matrix W through independence judgment and the self-adaptation separating algorithm to enable the output signals to be independent and to restore the source signals, the swarm algorithm GBCA with direction guiding serves as the separating algorithm to search for the optimal separating matrix W, and blind source separating is achieved. The swarm algorithm GBCA (BCA based on gradient) with direction guiding is provided, in other words, direction guiding is added into the search process, a section of distance is advanced towards the gradient direction after iteration is finished every time, the randomness of BCA can be reduced, robustness can be improved, the separating property and convergence speed of the algorithm are further improved, and the algorithm is applied to blind source separating of the time-frequency mixed signals in wireless communication.

Description

technical field [0001] The invention belongs to the technical field of wireless communication, and specifically relates to a separation algorithm based on statistical domain separation of overlapping wireless communication mixed signals in the time and frequency domains. Algorithm improvement, referred to as GBCA algorithm. Background technique [0002] With the rapid development of modern society, how to more effectively develop and utilize increasingly tight but open wireless spectrum resources has become an important research topic. Radio spectrum is a natural resource widely used in today's information society. Due to the rapid development of the wireless service industry, people's demand for radio spectrum resources is getting higher and higher. For the effective problem of spectrum resources, three types of typical technologies have widely coexisted in recent years, including frequency division multiplexing (FDM), time division multiplexing (TDM) and code division mul...

AI Technical Summary

Problems solved by technology

[0004] The problem to be solved by the present invention is: the time-frequency domain mixed signal separation method proposed in the existing wireless communication system research has a slow convergence speed, takes a long time, and requires high settings, and the newly proposed bee colony algorithm BCA has strong randomness and robustness. Poor performance, which may lead to unstable separation performance, and there are problems in computing efficiency and resource occupation in terms of complexity

Images