非均匀采样率并发双频数字预失真技术

5G通信的高速发展对发射机支持多模式多频带并发有了新的需求。考虑非均匀采样率并发双频场景下发射机的线性化方案,针对低采样率频段失真频谱超过采样带宽,从而恶化建模精度这一问题,提出一种混叠消除的并发双频数字预失真(DPD)方案。该方案通过在高采样率下构造模型基函数,然后通过低通滤波器来滤除超出采样带宽的频谱分量,再降采样到实际的低采样率上,这样便消除了混叠频谱,能够保证足够的建模精度。为了验证提出的DPD方案,基于中心频点3.5 GHz的宽带Doherty功放搭建了实验测试平台,分别在3.42 GHz和3.57 GHz两个频带发送10 MHz和40 MHz的LTE信号,采用DPD方案后,两个频带邻近信道功率比(ACPR)指标分别改善11.3 dB和9.5 dB。

Nonuniform-Sampling-Rate Concurrent Dual-Band Digital Predistortion Technique

Along with the rapid development of the 5G communication, new requirements occur for the concurrent multi-mode multi-band transmitters. For the nonuniform-sampling-rate concurrent dual-band transmitter, the bandwidth of nonlinear distortion of the low-sampling-rate band is larger than the sampling rate. The aliasing effect will be caused and significantly deteriorate the modelling accuracy. An aliasing-cancelled dual-band digital predistortion (DPD) scheme is proposed in this paper. By constructing the basis function of the nonlinearity model at a high sampling rate, the aliasing spectra can be filter out using a digital low-pass filter. Then, the aliasing effect is cancelled after downsampling to the low sampling rate and the model accuracy is improved. To demonstrate the proposed DPD scheme, a testbench is set up with a 3.5 GHz wideband Doherty power amplifier (PA). The PA is operated at 3.42 GHz and 3.57 GHz with 10 MHz and 40 MHz bandwidth LTE signals. The adjacent channel power ratio (ACPR) performances are improved by 11.3 dB and 9.5 dB, respectively.