Font Size: a A A

Study On Adaptive Interference Cancellation Over L-DACS1 Based On High Order Statistic

Posted on:2017-11-11Degree:MasterType:Thesis
Country:ChinaCandidate:Y HeFull Text:PDF
GTID:2322330503465984Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
L-band digital aeronautical communication systems(L-DACS) are new air-to-ground wireless datalinks proposed by the FAA and EUROCONTROL to supplement existing civil aviation technologies. L-DACS comprise two candidates: L-DACS1 and L-DACS2. L-DACS1 is a frequency-division duplex configuration utilizing OFDM modulation techniques, whereas L-DACS2 is a time-division duplex configuration utilizing GMSK modulation schemes. L-DACS1 outperforms L-DACS2 in such aspects as spectral efficiency, spectrum extensibility and interference robustness. We focus on the DME effect on the performance of the L-DACS1 onboard receiver. To improve the spectrum efficiency, a gap between two adjacent DME channels that has a bandwidth of approximately 500 kHz is intended to be used by L-DASC1. Because the power level of DME exceeds that of L-DACS1 when this inlay deployment concept is adopted, the DME interference has a severe impact on the performance of L-DACS1 and therefore must be mitigated. Several strategies have been proposed for DME interference cancellation, including the use of time windowing, notch filtering, pulse blanking and the LMMSE canceller. However, time windowing causes inter-symbol interference and notch filtering/pulse blanking induces inter-carrier interference due to time and frequency domain impairment, respectively. The use of an LMMSE canceller is more effective than the preceding three techniques at minimizing the mean square error values of the L-DACS1 and DME signals, in which the second-order statistics of both are used. However, preliminary studies show that the amplitude of the envelope of an OFDM-like signal(such as L-DACS1) can be approximated by a Gaussian distribution, which enables us to model it as a stationary colored Gaussian process. Second-order statistics-based adaptive filters, including the LMMSE canceller, are sensitive to colored Gaussian noise, which significantly degrades the performance of the LMMSE canceller. Therefore, we develop an adaptive interference-cancellation scheme named CE-LMS that uses high-order statistics. By first estimating the DME signal precisely and subtracting it, CE-LMS can reconstruct L-DACS1 signals without the effect of asymptotically Gaussian L-DACS1 signals, the main work and innovations are as follows.(1) We investigate modeling the coexistence of the L-DACS1 signal and DME interference as a deterministic signal(DME) plus colored Gaussian noise(L-DACS1), and we develop an adaptive interference-cancellation scheme named CE-LMS that uses high-order statistics. By using a three-order, cumulant-based cost function and a variable step-size adjustment mechanism, CE-LMS can effectively eliminate the pulse-like DME interference over L-DACS1. Then we propose the implementation of transversal and lattice filter-based CE-LMS algorithm and derive the the Cramer-Rao bound.(2) The simulation focuses on the L-DACS1 forward link transmission from ground stations to airborne stations where the victim L-DACS1 receiver is exposed to the DME interference from the ground station. In this simulation, the evaluation involves the interference-cancellation ratio, the bit error rate, the baseband power spectrum, root mean square error, error variance, normalizad coefficient error and channel capacity of the CE-LMS algorithm, time windowing, pulse blanking and the cumulant-based MMSE algorithm. The numerical results demonstrate that, compared with time windowing, pulse blanking and the cumulant-based MMSE algorithm, the CE-LMS algorithm can achieve a better cancellation ratio and baseband power spectrum, higher normalized channel capacity, lower bit error rate, lower root mean square error, lower error variance and comparable convergence rate, with only a moderate increase in computational complexity.These results can serve as guidelines for L-band interference-mitigation research and will promote the application and deployment of L-DACS1.
Keywords/Search Tags:L-DACS1, DME, high-order statistics, interference cancellation, adaptive filtering
PDF Full Text Request
Related items