Çevrimsel durağanlık tabanlı uzay-zaman blok kod zamanlaması kestirimi
Cyclostationarity based space-time block code timing estimation
- Tez No: 517288
- Danışmanlar: PROF. DR. HAKAN ALİ ÇIRPAN
- Tez Türü: Yüksek Lisans
- Konular: Elektrik ve Elektronik Mühendisliği, Electrical and Electronics Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 2018
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Elektronik ve Haberleşme Mühendisliği Ana Bilim Dalı
- Bilim Dalı: Telekomünikasyon Mühendisliği Bilim Dalı
- Sayfa Sayısı: 61
Özet
Telsiz haberleşme sistemlerinde güvenli bir iletişimin sağlanabilmesi için verici tarafından gönderilen mesajların alıcıda hatasız ya da minimum hata oranı ile elde edilebilmesi şarttır. Bu amaçla gönderilen sinyale ilişkin iletim parametrelerinin alıcıda ön bilgi olarak yer alması ya da alıcı tarafından, gönderilen işaretten yararlanılarak kestirimi zorunludur. Literatürde gönderilen sinyale ilişkin taşıyıcı frekansı, bandgenişliği, taşıyıcı frekans kayması, simge zamanlaması, modülasyon türü ve çoklu erişim tekniği gibi parametrelerin alıcıda elde edilmesi için geliştirilmiş birçok yöntemler ve algoritmalar bulunmaktadır. Bu yöntemler ve algoritmalar işbirlikli ve işbirliksiz(gözü kapalı) olarak ikiye ayrılmaktadır. Pratikte kısıtlı bandgenişliği ve analiz edilecek sinyalin önceden bilinmeyen tipte olması gibi sebeplerden dolayı verici ile alıcının işbirliği içerisinde bulunamadığı senaryolarla sıklıkla karşılaşılmaktadır. İşbirliksiz yöntemler bu senaryolar için çözüm getirerek literatürde önemli bir yere sahip olmuşlardır. Özellikle çağdaş haberleşme sistemlerinin kullanıldığı askeri ve sivil uygulamalarda bu problemlerle karşılaşıldığından işbirliksiz yöntemler güncelliğini korumaktadır. Çağdaş haberleşme sistemlerinde kullanılan Çok Girişli Çok Çıkışlı (Multiple Input Multiple Output, MIMO) sistemler verici ve alıcı kısımlarında çok sayıda anten kullanarak iletişimin kalitesini artırmakta ve daha yüksek hızlarda iletişime olanak tanımaktadırlar. MIMO sistemler bu avantajlarından ve birçok uygulama için makul seviyedeki karmaşıklıklarından ötürü yeni nesil haberleşme sistemlerinde yaygın olarak kullanılmaktadırlar. İşaretlerin iletiminde kullanılan Uzay-Zaman Kodlama (Space-Time Coding, STC) teknikleri sayesinde MIMO sistemler Tek Girişli-Tek Çıkışlı (SISO) sistemlere nazaran daha fazla kanal sığası sunmaktadırlar. MIMO sistemlerde STC işlemi yukarıda belirtilen parametrelere ek olarak uzay-zaman kodlarına ilişkin bazı yeni parametrelerin de alıcı tarafından elde edilebilmesini gerektirir. Nispeten düşük karmaşıklığı sebebiyle pratikte daha yaygın olarak kullanılan Uzay-Zaman Blok Kodları (Space-Time Block Codes, STBC) alıcıda kod blok zamanlaması bilgisinin elde edilmesini zorunlu kılar. Aksi halde alıcı, STBC çözme işlemini gerçekleştirirken yanlış kod başlangıç anından başlayabilir ve bu durum büyük simge hatalarının oluşması ve muhtemelen iletişimin son bulmasıyla sonuçlanacaktır. Bu çalışmada pratikte yaygın olarak kullanılan bir uzay-zaman blok kodu olan Alamouti Kodu (AL) için çevrimsel durağanlık tabanlı bir blok zamanlaması kestirim algoritması önerilmiştir. Önerilen algoritma vericide yapılan kodlama işleminden kaynaklanan, verici anten sinyalleri arasındaki çapraz ilişkilere ait çevrimsel durağanlık özelliklerini kullanarak blok zamanlamasını kestirmektedir. Bununla birlikte gönderilen sinyale ilişkin kod tipi ve alınan işaretten başka hiçbir ek bilgiye ihtiyaç duymamaktadır, bir başka deyişle gözü kapalı bir şekilde işlemektedir. Önerilen algoritmanın İşaret-Gürültü Oranı'na (Signal-to-Noise Ratio, SNR) bağlı performansı bilgisayar benzetimleri yardımıyla ölçülmüştür. Benzetim sonuçları önerilen algoritmanın literatürde var olan algoritmalara göre daha iyi performans sergilediğini ve düşük SNR rejimine karşı daha dayanıklı olduğunu ortaya koymaktadır. Yüksek performansı ve kanal matrisi, modülasyon tipi, gürültü varyansı gibi parametrelere ve eğitici dizilere (Training Sequences, TS) ihtiyaç duymaması sebebiyle önerilen algoritma literatürdeki algoritmalar arasında önemli bir yere sahip olmuştur.
Özet (Çeviri)
In wireless communication systems, in order to provide a reliable communication, it is necessary that the messages sent by the transmitter are regenerated at the receiver with no or minimum error rate. For this reason, it is mandatory for the receiver to have the transmission parameters as an a-priori information or to regenerate them using the received signal. By using the obtained parameters, the receiver can combat channel impairments like carrier frequency and phase offsets, oscillator clock mismatch and fading. In literature, there exists numerous methods and algorithms aiming at the regeneration of these parameters belonging to the transmitted signal e.g. carrier frequency, bandwidth, carrier phase and frequency offset, symbol timing, modulation type and multiple access techniques. These methods and algorithms can be grouped into two categories as cooperative and non-cooperative(blind) methods. Cooperative methods have been used widely in practice where there is a sufficient amount of bandwidth for training or pilot signals. This group of methods does not have the adaptation problem of non-cooperative solutions, which is one of the strongest reasons that remains their usage in practice. However, in practice, there also exists application scenarios where the transmitter and the receiver cannot cooperate due to the bandwidth concerns or the unknown signal types to be analyzed. This type of scenarios can be handled by using non-cooperative methods that requires only the received signal to obtain the crucial transmission parameters to combat channel impairments. Non-cooperative methods have gained a significant acceptance by providing solutions for these scenarios and remained their popularity by taking place in contemporary communication systems in both civil and military applications. Multiple Input-Multiple Output (MIMO) systems employed in contemporary communication systems boost the reliability of communications and provides opportunities for higher transmission rate by employing multiple antennas at the transmitter and at the receiver side. By generating ideally independent channels, MIMO systems provide receivers multiple diverse signals that has different level of qualities in signal strength. The receiver provides diversity by combining the signals from each receive antenna, resulting a better signal strength. Thanks to these advantages and their reasonable complexities for a variety of applications, MIMO systems are widely employed in modern wireless communication systems and will be employed in the next-generation wireless communication systems. Space-Time Coding (STC) techniques used in MIMO signal transmissions opens a door for more channel capacity than Single-Input Single Output (SISO) systems do. STCs use space-time redundancy of modulated symbols and coding techniques to provide receivers diversity gain and coding gain which result in a lower bit errors in decoding. STC operation in MIMO systems require receivers to have space-time code related parameters in addition to the ones declared above. Space-Time Block Codes (STBC) are employed widely in practical MIMO systems due to their reasonable complexities and they constrain the receiver to obtain the code block timing information which is defined as the current time instant that the receiver starts STBC decoding operation. Otherwise, the receiver may start decoding at a false code block start time instant and this would result in a great amount symbol decoding errors and possibly an outage. In this work, a novel cyclostationarity based block timing estimation algorithm is proposed for the Alamouti code which is a popular space-time block code in practice. STBC coding operation generates time correlations between the transmitted signals. By inspecting these correlations, one can observe that the transmitted signals STBC have cyclostationarity feature which has a fundamental period equal to the STBC code length. The cyclostationarity feature presents different characteristics for different code types like the popular Alamouti code analyzed in this work. For a given STBC, the cyclostationarity characteristics are a function of the lag parameter and the cyclic frequency parameter. In addition to the code type, these cyclostationarity characteristics also depend on the received code block timing. In this thesis, this dependency is exploited for the estimation of the code block timing parameter. However, this dependency cannot be used directly for the estimation of the code block timing parameter. This is due to the fact that the received signal at each receive antenna is actually a linear combination of the transmit signals, namely the transmit signals are multiplied by fading coefficients and added to the white noise. To overcome this problem, the proposed algorithm uses a blind MIMO channel estimation and compensation method to estimate a noisy version of the transmit signal vector. This estimated vector is then used to estimate the code block timing parameter. However, this vector cannot be used directly to estimate the cyclostationarity characteristics due to the permutation and phase ambiguities encountered in the recovered transmit signal vector after the blind channel estimation. Because of these restrictions and some mathematical and statistical intractabilities in estimating the cyclostationarity characteristics, one cannot derive an optimal decision rule for the code block timing parameter estimation problem. As an alternative, a test statistics based approach is used. First, a test statistic which is not affected by the permutation and phase ambiguities and is based on the estimated cross correlation functions of the estimated transmit signals, is derived. This test statistic is then used with a simple Euclidean distance based decision rule to estimate the code block timing information. Since the Alamouti code has a block length of two symbol periods, this decision rule is equal to deciding whether the receiver is at the beginning or in the middle of the code block to start decoding. Therefore the problem actually is a binary hypothesis testing problem and the proposed algorithm decides one of the two hypotheses to obtain the block timing parameter. To sum up, the proposed algorithm involves a blind channel estimator and Alamouti code block time decision algorithm to address the code block timing estimation problem and it is based on a test statistic which depends on the cyclostationarity characteristics of the transmit signal and and is used with a simple Euclidean distance criterion based decision rule to estimate the code block timing parameter. The performance of the proposed algorithm with respect to the Signal-to-Noise Ratio (SNR) levels is measured by computer simulations. The average probability of correct block timing decision is chosen as the performance criterion. First, the performance of the proposed algorithm is compared with a second-order statistics based, influential algorithm in the literature. For a fair comparison, channel estimation is also introduced in the compared algorithm. Results expose that the proposed algorithm outperforms the existing algorithm and is more robust to the low SNR regime. In the second stage, the performance of the proposed algorithm with respect to the number of antennas and the number of the observed symbols. According to the results, the performance of the proposed algorithm increases with increasing number of received antennas due to the better channel estimation. Furthermore, an increased number of observed symbols increases the estimation accuracy of the test statistics which leads to an increased overall performance of the proposed algorithm as expected. In the final stage, the performance of the proposed algorithm with respect to the symbol timing errors is investigated. Here, symbol timing errors represent the transmitter-receiver clock mismatch. For a various number of symbol timing offsets, the curves for average probability of correct block timing decision is obtained. Simulation results point to a critical fact that the proposed algorithm is robust to the symbol timing offsets, which makes the algorithm more robust for the practical scenarios including the receiver clock drifts and multipath delays. Tu sum up the simulation results, the proposed algorithm outperforms the existing algorithm by a wide margin. Furthermore, it performs better especially in the low SNR regime. The performance of the algorithm increases with number of receive antennas and the number of observed symbols. Finally, it is robust to the symbol timing offsets. The proposed method requires no essential information other than the code type and the received signal, namely, it operates in a blind way. There is no requirement of an a-priori knowledge on the channel matrix, the employed modulation type, noise variance and pre-arranged training sequences. These qualities increase the attractiveness of the algorithm for practical scenarios including the MIMO signal identification which has become a popular research field in the last decade. The proposed algorithm can be used in intelligency and spectrum management applications where the receivers need to detect, analyze, classify and decode the signals in the spectrum. These applications include recovering the licensed signal parameters to check if the transmission fits to the regulations like bandwidth or power limitations, analyzing the unknown threat signals in military and intelligency applications and decoding terrestrial mobile radio signals where there is not sufficient bandwidth to cooperate with the transmitter. Therefore, the proposed algorithm has become a good alternative to existing methods in numerous influential application areas. As the future work, the robustness of the proposed algorithm will be increased by addressing the carrier frequency offset problem which is frequently encountered in practice. The structure of the algorithm is highly convenient that most of the existing carrier frequency estimators can be integrated easily. The second stage of the future work will be extending the scope of the algorithm to include the other STBCs in the literature. Since different codes have different cyclostationarity characteristics, new test statistics have to be found for each new code to be incorporated. Moreover, the problem becomes more challenging for the codes that require a high number of transmit antennas. This is due to the fact that the increasing effect of permutation and phase ambiguities. Thus, new statistics have to be found so that they are not effected by these ambiguities. The next work will be on addressing this problem.
Benzer Tezler
- Time series forecasting via computational intelligence methods
Zaman serileri tahminlemede bilgi işlemsel zeka uygulamaları
ATAKAN ŞAHİN
Yüksek Lisans
İngilizce
2016
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiKontrol ve Otomasyon Mühendisliği Ana Bilim Dalı
YRD. DOÇ. DR. TUFAN KUMBASAR
- Wireless channel modeling based on extreme values theory for ultra-reliable low latency communications
Ultra güvenilir düşük gecikmeli haberleşme için uç değer teorisine dayalı kablosuz kanal modelleme
NILOOFAR MEHRNIA
Doktora
İngilizce
2022
Elektrik ve Elektronik MühendisliğiKoç ÜniversitesiElektrik ve Elektronik Mühendisliği Ana Bilim Dalı
PROF. DR. SİNEM ÇÖLERİ
- Compressive sensing of cyclostationary propeller noise
Çevrimsel durağan pervane gürültüsü için sıkıştırmalı algılama
UMUT FIRAT
Doktora
İngilizce
2023
Elektrik ve Elektronik Mühendisliğiİstanbul Teknik ÜniversitesiElektronik ve Haberleşme Mühendisliği Ana Bilim Dalı
PROF. DR. TAYFUN AKGÜL
- Direction finding for coherent, cyclostationary signals via a uniform circular array
Düzgün dairesel dizi kullanarak evreuyumlu, çevrimsel durağan işaretlerin yönlerinin bulunması
BURCU ATALAY ÇETİNKAYA
Yüksek Lisans
İngilizce
2009
Elektrik ve Elektronik MühendisliğiOrta Doğu Teknik ÜniversitesiElektrik-Elektronik Mühendisliği Ana Bilim Dalı
DR. ARZU TUNCAY KOÇ
- Bilinmeyen bir işaret kaynağından alınan doğrusal modülasyonlu işaretlerin simge hızının ve modülasyon türünün kör olarak kestirimi
Blind symbol rate and modulation type estimations of linear modulated signals received from an unknown signal
AHMET GÜNER
Doktora
Türkçe
2013
Elektrik ve Elektronik MühendisliğiKaradeniz Teknik ÜniversitesiElektrik-Elektronik Mühendisliği Ana Bilim Dalı
DOÇ. DR. İSMAİL KAYA