With the increasing demand for higher data rate, increasing throughput requires wider bandwidth. Due to the nonlinear effect of power amplifier, severe nonlinear distortion effects will appear onboard satellite. Digital predistortion (DPD) scheme is usually employed to compensate for the distortions and memory effects introduced by traveling wave tube amplifier (TWTA) and output multiplexing filter (OMUX). In conventional predistorter, the signal loss of band-limited feedback signal output through OMUX is usually ignored. Actually, it will affect the linear effect of TWTA. In order to solve the problem as to improve wideband nonlinear satellite downlink, this paper introduces a novel spectral extrapolation method based on deep neural network to recover the band-limited feedback signal. On this basis, an advanced orthogonal matching pursuit algorithm is adopted in the nonlinear TWTA model construction to further reduce the DPD complexity. The proposed setup effectively compensates the distortions and is well suited for systems that generate data bits on satellites. KEYWORDS digital predistortion, nonlinear, output channel filter, satellite downlinks 1 | INTRODUCTION Increasing data rates over satellite channels have motivated the quest for efficient transmission of spectrum, while sharing a single high power amplifier with multiple carriers. Compensation for nonlinear distortion with memory is a key technology for the highly efficient operation of communication systems. The main contributor to nonlinearity in satellite is the traveling wave tube amplifier (TWTA) when it is driven close to saturation. In addition, with the increase in demand for transmitting broadband signals, the memory effects of output signals through TWTA and output multiplexing filter (OMUX) become more prominent. The nonlinear distortion caused by TWTA manifests itself in two major effects.One is that spectral spreading or regrowth causes spillage into adjacent frequency bands and may violate the stringent spectrum limits set by the regulatory commissions. Another effect is in-band distortion causes constellation warping and clustering. Unless appropriate methods are taken, both two effects are detrimental and can severely degrade system performance .1,2 Digital predistortion is a transmitter-based method of effectively mitigating nonlinear distortion. There are two categories of approaches, data predistortion and signal predistortion. Data predistortion is more suitable in "bent pipe" satellites as the uplink can be corrected at the gateway. A data predistortion method using Volterra inverse is proposed in Piazza et al 3 for multicarrier satellite operation. For multicarrier data predistortion, Beidas et al 4 utilizes stochastic approximation to introduce another data predistortion method with enhanced performance. However, in adaptive technology, most of the available signal predistortion methods are still based on inversions that minimize the equalization mean squared error (MSE), using various models such as Volterra or Wiener. 5 Signal p...