Abstract
This paper examines the application of sparse estimation techniques for the estimation of a discrete-time equivalent multipath channel in the aeronautical telemetry context. The sensing matrix comprises samples of SOQPSK-TG (a continuous phase modulation) based on the pilot bit sequence currently defined in the aeronautical telemetry standard. Representative algorithms from the three broad classes of sparse estimators were examined side-by-side using computer simulations to estimate the post-equalizer bit error rate (BER). Ideal and non-ideal frequency offset synchronization were assumed in the simulations. The results show that the performance of the matching pursuit algorithms seemed to be better suited to this application in the sense that no additional steps were required and the post-equalizer BER of the best matching pursuit algorithm was slightly better than that of the other sparse estimation techniques. In the case of both ideal and non-ideal frequency offset synchronization, the post-equalizer BER achieved by the generalized orthogonal matching pursuit (GOMP) algorithm was approximately 1.5 dB better than that obtained using the nonsparse-constrained maximum likelihood channel estimate.
Funding Information
  • Test Resource Management Center's Test and Evaluation Science and Technology Program
  • U.S. Army Program Executive Office for Simulation, Training and Instrumentation (W900KK-13-C-0026 (PAQ))

This publication has 29 references indexed in Scilit: