Power line communication system impulse noise suppression method based on L1/2 norm regularization

Abstract

The invention discloses a power line communication system impulse noise suppression method based on L1 / 2 norm regularization. At a sending end, discrete time domain signals with cyclic prefixes is transmitted to a receiving end through a channel; at the receiving end, based on a compressed sensing theory, through introduction of regularization parameters and utilization of an iteration solving method, first the problem of estimation of impulse noise is converted into the problem of regularization of an L1 / 2 norm minimum value of the impulse noise, then the problem is converted into a form containing a two-step iteration result and the iteration solving method is utilized for solution, and thus estimation of an amplitude point of the impulse noise is relatively accurate; and by adoption of a threshold function to process the iterative result, a sampling point which is not affected by impulse noise interference has a relatively small estimation error.

Description

technical field [0001] The present invention relates to a power line communication system pulse noise suppression technology, in particular to a technology based on L 1 / 2 A norm-regularized impulsive noise suppression method for power line communication systems. Background technique [0002] Power line communication (Powerline Communication, PLC) is a communication method that uses a power line medium for carrier transmission. As a well-established infrastructure, power lines can greatly reduce communication costs by using them for data transmission. Existing typical power line communication system applications include: remote meter reading, real-time quotation, and intelligent energy management. In future application designs, including slow-scanning TV images, video and audio signals, they can all be transmitted through the power line network. However, since the power line was originally designed to transmit electric energy, the voltage on the line is high, the noise is ...

AI Technical Summary

Problems solved by technology

Some researchers proposed a kind of impulse noise L 0 The norm minimization method estimates and eliminates the impulsive noise, but directly solves L 0 The norm is NP (non-deterministic polynomial, non-deterministic polynomial) difficult problem

To solve directly solve L 0 The norm is a technical problem of NP-hard problems, and some researchers have proposed to solve L 0 The NP-hard problem of the norm is relaxed into a convex problem, that is, solving the L of the impulsive noise 1 Norm minimization problem, but the accuracy of this method’s estimation of impulse noise amplitude points is low; some researchers also proposed a method that uses L 0 Norm transformed into smooth L of continuous function 0 Norm (smoothed L 0 ,SL 0 ) minimization method estimates the impulsive noise, but since this method uses a continuous function for approximate solution, there will be estimation errors at the sampling points not disturbed by the impulsive noise

Images