Predistortion parameter processing method and device

Abstract

The invention provides a predistortion parameter processing method and a device; wherein the method comprises the following steps: an initial signal matrix is built by utilizing data which carries out predistortion treatment and power amplification feedback data after periodic filtering processing begins; the data which carries out predistortion treatment is a reference signal, and the data for power feedback is an input signal; QR decomposition processing is carried out to the built initial signal matrix, and the predistortion parameters are determined by carrying out backward iteration operation to the matrix which carries out the QR decomposition; by adopting the method and the device, the arithmetic workload and the arithmetic complexity can be effectively reduced; in addition, the determined predistortion parameters begin to be stored from Nth sampling point symbol, so as to well save the expenses of the system operation.

Description

technical field [0001] The present invention relates to the field of communication technology, in particular to a method and device for processing predistortion parameters. Background technique [0002] At present, with the large-scale construction of global 3G networks, operators are paying more and more attention to reducing construction costs and maintenance costs, and power amplifiers, as one of the most expensive devices in communication systems, have higher and higher requirements for efficiency. This makes the digital pre-distortion DPD (Digital Pre-Distortion) technology develop rapidly. Among them, the power amplifier is closely related to the filter: when the output of the filter is a linear function of the input, the filter is a linear filter, otherwise it is a nonlinear filter. If the parameters change with time, it is also called a time-sensitive filter. According to the characteristics of the RF power amplifier (PA), when the statistical characteristics of the...

AI Technical Summary

Problems solved by technology

[0005] Although the least squares method can also obtain the predistortion parameters by solving the inverse matrix of the normal equation, the amount of calculation is still relatively large. For this reason, the prior art proposes to obtain the predistortion parameters by RLS (Recursive Least-Squares, recursive least squares) parameter, namely: w m (n)=w m (n-1)+g m (n)·e(n), where the error signal is: This RLS is similar to the recursive formula of LMS, the difference is only the gain coefficient, that is while C m (n-1) is also obtained according to the recursive formula, that is, the autocorrelation matrix: Therefore, this method only needs to use the initial predistortion parameter w of the predistortion m (0), pre-distorted signal x m (0) and the intermediate variable C m (0) can obtain the predistortion parameters; although RLS can perform optimal filtering on a known set of data, it mainly performs recursive calculations on the correlation matrix of the input data, and the calculation is relatively complicated and the amount of calculation is large (O[N 3 ]), and in the measurement noise or non-stationary environment, there will be stability problems;

[0006] Therefore, a method for predistortion processing using the QRD-RLS (QR Decomposition Recursive Least Squares) algorithm is proposed in the prior art. This QRD-RLS is an improvement to the RLS algorithm. The correlation matrix of the data is recursive, but the input data matrix is ​​directly recursive, the amount of calculation is greatly reduced, and at the same time, it has good data stability, so it is more suitable for the DPD system; however, this algorithm still has its own defects. , that is, each output sample needs to be O[N 2 ] times of multiplication, but for a real-time system, the amount of multiplication is still very large, for example, only 1000 samples are taken, N=60, then 3.6 million multiplications are required, which is difficult to implement for existing chips

Images