dc.description.abstract |
Detection and correction of faulty arrays is one of the important research areas in
beamforming. If some elements in the array fail to work, the main objective is to steer
the main beam in the desired direction and place the nulls in the direction of
interferers. This research area has gained direct applications such as sonar, radar
satellite, mobile communications etc.
The work presented in this dissertation is mainly divided into two parts. In first
part, our contribution is to develop an efficient method for the detection of faulty
sensors. Specifically, the symmetrical structure of linear array is proposed, which has
two advantages. Firstly instead of finding all damaged patterns, only half patterns are
needed and, secondly, we require scanning the region from 0 0 to 90 0 instead of 0 0 to
180 0 , which obviously reduces the computational complexity to half. The basic tools
designed to detect the location of faulty sensors are nature inspired heuristic
computing and compressed sensing technique. These techniques are firefly
algorithm, cultural algorithm with differential evolution, harmony search algorithm,
cuckoo search algorithm and the compressed sensing techniques such as Parallel
coordinate decent algorithm, separable surrogate functional algorithm and Iterative
reweighted least square algorithms. Furthermore, the compressed sensing
techniques are hybridized with evolutionary algorithm.
The second part of this dissertation, the correction of faulty arrays is formulated in
a unique way and five approaches are proposed to correct the faulty pattern. In the
first approach, our contribution regarding correction of faulty sensors has achieved
better null depth level (NDL) due to symmetrical element failure (SEF) technique. The
symmetric element failure maintains the null depth almost close to that of the original
array. The null depth of all nulls, especially the first one, has been achieved with the
help of SEF technique. Null placement and sidelobe suppression have been
achieved by hybridizing genetic algorithm with pattern search. In the secondvii
approach, the symmetrical structure is used for the correction of faulty beams in
failed array antenna. The corrected pattern has been achieved by a cultural algorithm
with differential evolution using a proper fitness function. In the third approach, a
cuckoo search algorithm (CSA) is developed based on SEF technique along with
distance adjustment between the antenna elements for the correction of faulty
beams. The proposed SEF technique along with distance adjustment d n based on
CSA provides better results in terms of SLL and nulls in the direction of known
interferers. In the fourth approach, a CSA is designed for the correction of single
element failure. This time the proposed technique has used a new fitness function for
the suppression of SLL and nulls in the direction of known interferers. In the fifth
approach, using the advantage of symmetrical structure of linear array, a simple
method has been developed for the reconstruction of faulty beams. The method
recovers the failed element signal from its symmetrical counterpart element by taking
its conjugate.
The simulation results of detection and correction of faulty arrays are presented
in comparison with the available techniques. In case of detection, our approach is
computationally efficient to detect the complete as well as partial faulty elements. In
case of correction we achieved better NDL which is of great importance in
beamforming and nulls placement back to their original positions after the failure of
elements. |
en_US |