The problem of improving the frequency response of linear arrays is addressed in this thesis. The improvement is measured in the level of rejection of the undesired sources. A new, simple, and practical method is suggested. It is based on adjusting the sampling time in selected channels. Interference cancellation for finite bandwidth signals is accomplished only at, or close to, the carrier frequency. Wideband cancellation requires additional hardware and computational complexity. Conventional way of the wideband interference cancellation utilizes tapped delay line adaptive filters in each channel.
The Time Adjustable Sampling (TAS) method achieves a part of the tapped delay line effect by adjusting the sampling time in some selected array channels. Signal beamforming is combined with TAS. The beamforming is done using complex weights. The combination of the complex weights and TAS provides improvement over the usage of the complex weights only. Hardware and computational requirements in implementing the TAS method are significantly lower than for the tapped delay line structure of the same dimension.
Different configurations are investigated. The comparison with a conventional method is given. It was found that TAS significantly improves wideband linear prediction. A computer simulation confirmed results obtained by the model analysis.
|