Karasal, hava ve uzay tabanlı haberleşme sistemleri arasındaki girişimin minimizasyonu

Abstract

Mevcut haberleşme sistemleri ve yeni geliştirilen sistemler için, spektrumun kısıtlı bir kaynak olması nedeniyle çeşitli hizmetlere tahsis edilen frekans bandlarının ortak kullanımına ihtiyaç duyulmaktadır. Ancak bu durum sistemler arasında band içi girişim yaşanmasına neden olmaktadır. Sağlıklı iletişimin sağlanabilmesi adına bu istenmeyen işaret iletiminin kabul edilebilir seviyede tutulması haberleşme için oldukça önemli bir problemdir. Bu nedenle girişimin azaltılması yönünde çeşitli yöntemler geliştirilmektedir. Teorik ve deneysel çalışmalar yapılmaktadır. Bu çalışmada, karasal, hava ve uzay tabanlı haberleşme sistemleri arasında olası haberleşme senaryolarından bahsedilerek, girişimin minimizasyonu ele alınmıştır. Çözüm yöntemi olarak girişim problemindeki parametreler için optimum değerlerin elde edilebilmesi amacıyla Parçacık Sürü Optimizasyonu (PSO) ve Sürekli Genetik Algoritma (SGA) yöntemleri kullanılarak eniyileme işlemi gerçekleştirilmiştir. Çok boyutlu problemleri için kesin çözüme yakın optimum çözümler sunmaları nedeniyle, bu algoritmalar ile girişimin minimizasyonu hedeflenmiştir. Öncelikle popülasyon temelli yöntemler olan PSO ve SGA ayrıntılı olarak anlatılmış, süreçleri tanıtılmış, avantajlarından ve dezavantajlarından bahsedilerek, bu iki eniyileme tekniği kıyaslanmıştır. Literatürde yer alan, 11 GHz'de radyo-yayın istasyonunun uydu yer istasyonuna yaptığı girişim, 1885 - 1980 MHz bant aralığında yüksek irtifa platformu ile hücresel sistem arasındaki girişim, 47.9 - 48.2 GHz'de HAPS kullanıcı terminalleri ve uydu alıcısı arasındaki girişim ve 1800/2100 MHz'de yayın yapan uçak kabin içi ağ ile karasal hücresel ağ arasındaki girişim ele alınmıştır. Aynı frekans bandında yayın yapan sistemlere ait girişim denklemleri kullanılarak, algoritmaların amaç fonksiyonları oluşturulmuştur. Tezde detaylı olarak anlatılan girişim senaryoları ve optimizasyon için tercih edilen sezgisel algoritmalar kullanılarak kullanıcı arayüzüne sahip MATLAB tabanlı bir yazılım geliştirilmiştir. Elde edilen sonuçlar, istenilen değerlere yakınsayarak, optimizasyon yöntemleri ile eniyilemenin gerçekleştirilerek sistemler arasındaki girişimin mininize edilebildiğini göstermiştir.

The radio-frequency spectrum is a scarce resource and the requests from both existing and new technologies are increasing for access to spectrum. Therefore, it is often necessary to share the frequency band for efficient utilization to limited radio spectrum. Each transmitter has an influence on each receiver, ultimately interference become the serious issue for aeronautical, terrestrial and satellite communication systems that operate in the same frequency band.As a result of the sharing of frequency bands due to the limited spectrum, undesirable interference power inevitably arises. For the operation of any radio system, interference is an important problem that must be solved; the level of the unwanted or interfering signals must be taken into account and must be kept very low levels. Hence interference mitigation approaches are extremely required to enable frequency sharing. Communication systems must overcome this problem to maintain communication successfully. Therefore, a variety of approaches such as optimization methods have been applied for reducing the interference signal power. First of all the main principles of the selected optimization methods are presented in this study.By inspiring from the events in nature, many optimization techniques named as heuristic methods have been developed. In this thesis, Partical Swarm Optimization(PSO) which is based on swarm intelligence and Continuous Genetic Algorithm(CGA) which is based on the principles of genetics have been described in detail and used for minimization of interference. Particle Swarm Optimization comes from the research on the behavior of the bird movements. Here the approach of the swarm is likened to seek a solution to a problem. This method is prefered for multi-parameter and multi-variable optimization problems. Basically, the algorithm consists of three steps. In the first step positions and velocities of the particles are defined, in the next step velocities of the particles are updated with respect to the most optimum position and in the last step new locations of the particles are reached. The position of each particle is changed according to its fitness (pbest) and the position relative to the other particles (gbest).Like the PSO, CGA consists of a population of solutions, here referred to as individuals rather than particles. Genetic Algorithm is a stochastic optimization method which is based on the theory of natural selection. In this technique, via using of the principles of genetics which are the crossover, mutation and selection, optimization continues to generate new individuals from existing individuals. Genetic Algorithm has two main variants such as binary GA (BGA) and continuous GA (CGA). In general the objective function is better represented and better results are achieved by CGA; therefore, Continuous Genetic Algorithm in this study is used to solve the interference mitigation problems.As mentioned in this thesis, PSO and SGA illustrate the similarities and differences in variety of approaches. PSO in some ways resembles SGA. Both are population-based and create new solutions from existing solutions. Like SGA, PSO includes random components to avoid local minima. However, the algorithms differ in the approaches to sharing information to create and improve new solutions. SGA uses the selection, crossover, and mutation operators. The number and type of operators used are different. Compared to SGA, PSO is more advantageous on many points including its simplicity and easy implementation. PSO is easier to adapt, requires fewer parameters and converge faster to global optimum results.After brief information on the optimization methods, in-band interference which occur due to the fact that frequency sharing between different systems is examined and interference has been minimized. Therefore, two optimization techniques are used and adopted into interference scenarios occurred between terrestrial, satellite and aeronautical systems.First interference model is between fixed-satellite and terrestrial Radio Relay Services on 11 GHz. Satellite communication systems with very small aperture terminal (VSAT) earth-stations are operated in 6/4 – 14/11, 12 GHz. However, as mentioned previously, due to the need for effective usage of the available frequency spectrum, for example, 14/11 GHz has been allocated to both satellite and terrestrial communications services, which naturally results in problem associated with interference. The second scenario is that interference effect into cellular system from high altitude platform station (HAPS) which is used as a base station to provide IMT-2000 service. As envisaged, one of the most important issues on the sharing problems between these systems is interference minimization because of the fact that HAPS system should not cause harmful interference to the cellular IMT-2000 system.Moreover, the interference from the HUTs to the geostationary earth orbit (GEO) satellite receiver is presented. As HAPS utilizes the frequency bands previously allocated for conventional systems, both systems share the frequency band, from 47.9 GHz to 48.2 GHz. Mitigation of the interference level is performed via PSO and CGA.The last scenario is the interference between aircraft in-cabin and terrestrial cellular systems. Four different cases are examined. In case 1 the interference caused by the ground base transceiver station to the aircraft mobile station are studied in case of a UMTS network. In case 2 one on-board mobile terminal interferes or communicates with the terrestrial base station is examined. In case 3 the potential interference from the aircraft base transceiver station to the ground mobile station is examined. Finally, case 4 examines that one on-board mobile station interferes with the ground base station due to the fact that its emission could have impact on the terrestrial uplink. Interference scenarios between the systems shared the same frequency band have been referred in each section and interference formulas in cost functions of the optimization algorithms are evaluated. Then, PSO and CGA are applied. In this study, optimization methods used are used to minimize the interference between aeronautical, terrestrial and satellite systems that operate in the same frequency band. For this purpose, convergence to the desired value of the unwanted recieved signal is obtained via MATLAB based software for the chosen interference scenarios from literature. The results presented in this thesis show that optimal values of the parameters in the interference calculations are able to be obtained by using PSO and CGA. Findings demonstrate the applicability of PSO and CGA to the minimization of interference problems. In continuation of this study, different interference scenarios can be examined and both of the PSO and CGA can be used. Other optimization methods can also be investigated. Finally performance of the methods can be compared.