Low-delay parallel digital down-conversion method and system

Abstract

The invention provides a low-delay parallel digital down-conversion method and system, and the method comprises the steps: obtaining an intermediate frequency and a sampling rate before a prompt signal with changed parameters is refreshed, and calculating a frequency offset compensation value and an initial phase of a current period in an eight-path parallel processing mode; performing iterative computation on the frequency offset compensation value and the initial phase through a coordinate rotation digital computation method to obtain waveform data; obtaining a calculation initial value according to the waveform data; when the prompt signal of the parameter change is refreshed, obtaining an output waveform of the current calculation according to the refreshed calculation initial value; performing eight-path parallel down-conversion on the input data subjected to serial-parallel conversion and the output waveform to obtain output data; and carrying out parallel-serial conversion on the output data and then outputting the output data. According to the scheme, fewer storage resources are required, the time delay is lower, and higher compensation precision can be realized.

Description

technical field [0001] The invention relates to the technical field of wireless communication, in particular to a low-delay parallel digital down-conversion method and system. Background technique [0002] In wireless communication systems, it is often necessary to compensate the frequency offset of the received signal, for example, the signal sampling rate is f s , the frequency offset to be compensated is f 0 , the received signal is X(K) (k=0,1...n), and the signal after frequency offset compensation is y(k) (x, y are complex numbers), then: [0003] y(k)=x(k)*(cos(f0 / fs*2*pi*(k-1))+1i*sin(f0 / fs*2*pi*(k-1))), where, k=1,2,3...n; [0004] Equivalently viewed as: y(k)=x(k)*(cos(delta)+1i*sin(delta)), where k=1,...n, and delta is between 0-2*pi. [0005] When eight channels are used in parallel, the mth channel (m=0,1,2...7) at time k: y_m(k)=x(8*k+m)*(cos(8*f0 / fs*2*pi* (k-1)+m*f0 / fs)+1i*sin(f8*0 / fs*2*pi*(k-1))+m*f0 / fs). [0006] In a broadband wireless system, when th...

AI Technical Summary

Problems solved by technology

[0007]However, in order to achieve high-precision frequency offset compensation, the look-up table method needs to store a relatively large table, which takes up more resources; the digital calculation method of coordinate rotation will increase the delay , and in order to achieve higher precision, a large number of iterations is required, which will also increase power consumption; the second look-up table method is a compromise solution, which can better balance storage space and power consumption, but after the second look-up table method is still It is necessary to do one more complex multiplication, which will also consume certain resources, and the performance is also affected by the size of the secondary lookup table, so it is difficult to achieve high precision

Images