Receiver-Based Algorithm for Bidirectional Nonlinear Satcom Channels

Bidirectional satellite communication between earth terminals on the same frequency at the same time is an emerging means of doubling spectral efficiency on satellite channels. With a linear satellite repeater, cancellation of self-interference can be done with standard echo cancelling methods. However, satellite amplifiers are normally operated in a highly-nonlinear regime, for maximum power output, and this complicates the interference removal process. Our research studies two algorithms for `inverting’ the nonlinearity at the receiving terminal, making the resulting channel roughly linear, so echo cancellation again becomes attractive. These methods are 1) a MAP detection approach which selects the most-probable choice among candidate input sequences, given the observed amplifier output; and 2) a particle filtering approach which approximates the MMSE estimate of the input signal, given the output sequence. These approaches are compared at different SNR’s and different amplifier backoff’s, and it is concluded that the particle filtering approach is more robust in the presence of downlink noise,while also offering reasonable receiver complexity.