Early Stopping for Interleaver Recovering of Turbo Codes

Abstract

Parameter recovering of channel codes is important in applications such as cognitive radio. The main task for that of a turbo code is to recover the interleaver. The existing optimal algorithm recovers interleaver parameters incrementally one by one. This algorithm continues till the end even if it has failed in the halfway. And there would be lots of wasted computation, as well as incorrectly recovered parameters that will badly deteriorate turbo decoding performance. To address such drawbacks, this paper proposes an early stopping method for the algorithm. Thresholds needed for the method are set through theoretical analysis. Simulations show that the proposed method is able to stop the algorithm in time after it has failed, while no significant degradation in the correct recovering probability is observed.