Katzis, Konstantinos

The Uniform hexagonal array (UHA) sidelobe level (SLL) reduction is a crucial consideration for avoiding co-channel interference and improving Carrier-to-interference Ratio (CIR) performance in HAP wireless networks using the TVWS spectrum. The SLL reduction in UHA has been researched by applying evolutionary algorithms such as genetic algorithm (GA) and Particle Swarm Optimization (PSO). However, such algorithms have the drawbacks such as early convergence and the inability to reach globally. The beam pointing approaches for UHA antennas using Modified PSO (MPSO), for reducing the SLL on finding the optimal current weights is presented in this article. The UHA antenna array with 19 cells fitted with 169 elements in the HAPs cellular structure is considered to simulate the proposed approach. Then, the MPSO approach is employed and compared with the standard PSO algorithm and Uniform Weighting (UW) scheme. As per the simulation findings, the proposed approach drastically decreased sidelobes, with a reduction of up to -6.84dB and -13.3dB when compared to PSO and UW respectively. In other words, the proposed approach outperforms the UW and the PSO algorithms. Moreover, the CIR performance of the proposed approach has been demonstrated in terms of average outage probability and the proposed approach provides a better result.

Aiming at the problem of downlink power allocation in cognitive high-altitude platform wireless networks exploiting TV white space spectrum, it is mathematically formulated as a constrained optimization problem, and then an improved immune clonal optimization algorithm is proposed. The mathematical optimization model, algorithm realization process, and key technologies for power allocation are given, and coding, cloning, and mutation suitable operator for algorithm solving are designed. The findings of the simulation experiment indicate that, under the constraints of total transmit power, bit error rate, and interference tolerable to the main user, the algorithm can obtain a greater total data throughput, faster convergence speed, and better power allocation can be obtained. Finally, the proposed algorithm outperforms the Particle Swarm Optimization algorithm.

The coverage and capacity of exploiting TV White Spaces (TVWS) from High Altitude Platforms (HAPs) with multiple antenna payloads using uniform rectangular phased array (URA) antennas are explored in this article. A scenario is suggested in which a HAP and TVWS base station are installed on the HAP within a HAP coverage area with a radius of 100 km and a cell radius of 10.5 km. The formation of HAP cells with various reuse patterns has been established. An analysis of coverage, capacity and power profile function for TVWS link from 25 × 25 URA HAP arrays is presented. To minimize the sidelobe levels in the antenna array synthesis, window function has been assumed. Power profile within a cell using URA is highlighted. Channel to Interference (CIR) performance contour, coverage probability and channel capacity have been determined. The channel capacity for the theoretical Shannon bound and practical truncated Shannon bound are determined and compared. Results show that a maximum channel capacity of 53 Mbps can be attained from the proposed model.

The issue of spectrum allocation in cognitive high altitude platform wireless networks using TV White Space (TVWS) spectrum is presented in this paper. The mathematical model of spectrum allocation is given, and this model is converted into a constrained optimization problem with the goal of maximizing network benefits. Then, a spectrum allocation optimization approach is proposed that is based on improved immune clonal selection algorithm. According to the characteristics of the problem, the design, coding, cloning, mutation, hypermutation and selection operators suitable for the algorithm's solution are presented. Finally, a simulation experiment was carried out on this algorithm. The proposed algorithm was validated and proved that it converges better than Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). Moreover, experiments demonstrate that the suggested approach maximizes network benefits or rewards more effectively when compared with GA and PSO algorithms and results proved the effectiveness of the algorithm.