Tam-çift yönlü iletişim sistemleri: Başarım analizi ve röle uygulamaları
Full-duplex communication systems: Performance analysis and applications of relays
- Tez No: 338219
- Danışmanlar: YRD. DOÇ. DR. GÜNEŞ ZEYNEP KARABULUT KURT
- Tez Türü: Yüksek Lisans
- Konular: Elektrik ve Elektronik Mühendisliği, Electrical and Electronics Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 2013
- 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: 97
Özet
Bu çalışmada ilk olarak tam-çift yönlü (full-duplex, FD) radyolardaki öz-girişim (self-interference, SI) kanalından gelen SI işaretinin etkisini ortadan kaldırabilmek amacıyla temelband bir alıcı yapısı önerilmiştir. SI kanalının istatiksel dağılımını belirlemek için yazılım tabanlı radyo birimlerinden (universal software radio peripheral, USRP) oluşan bir test ortamı kurulmuştur. Gerçekleştirilen ölçümler sonucunda dar bandlı işaretleşmede bile SI kanalının zaman yayılımlı olabileceği belirlenmiştir. Bu bilgi ışığında noktadan noktaya iletişimde kullanılmak üzere sayısal SI işaretlerini ortadan kaldırabilmek amacıyla temelband bir FD alıcı yapısı önerilmiştir. Ses iletimindeki yankıyı ortadan kaldırmaya yarayan yapı temel alınarak FD alıcı için Nyquist'den daha hızlı örnekleme işlemine dayanan kademeli girişim dengeleyici tasarlanmıştır. Tam-çift yönlü aktarmalı (full-duplex relaying, FDR) sistem ideal SI iptali durumunda veri iletim hızını iki katına çıkarmaktadır ve FDR noktadan noktaya iletişim yapılan sistemlerle aynı başarımı yakalamaktadır. FDR sisteminin başarımını daha iyi inceleyebilmek amacıyla sistemin band genişliği, antenlerin yerleşimlerindeki hata ve işaret seviyelerindeki farklılıklardan dolayı ideal olarak gerçekleştirilemeyen SI iptalinin derecesi parametrik olarak tanımlanmıştır. SI işaretinin iptalinin ideal olarak yapılamadığı durumlarda FDR sistemin başarımının SI iptal katsayısına bağlı olduğu belirlenmiş ve SI iptal katsayısının hangi değerlerinde yarı-çift yönlü aktarmalı (half-duplex relaying, HDR) sisteme göre daha iyi veri iletim hızı sağladığı bulunmuştur. Tam-çift yönlü fiziksel katman ağ kodlamalı (FD-PNC) sistemlerde FDR sisteminde olduğu gibi FD tekniği sayesinde iletim süresi yarıya düşer ve ideal SI iptali durumunda veri iletim hızı iki katına çıkar. Bu motivasyonla FD-PNC sistemlerin bit hata başarımlarına ilişkin SI iptal katsayısına bağlı, sıkı alt ve üst teorik limitler bulunarak benzetim çalışmalarıyla teorik sonuçlar desteklenmiştir. FD-PNC sistemin bit hata oranları için teorik alt ve üst limit ifadeleri elde edilmiştir. Tez kapsamında yapılan çalışmalar sonucunda yapılan katkılar üçe ayrılmaktadır. Öncelikle, USRP yardımıyla ölçülen SI kanalının zaman yayılımlı karakteristiğine uygun, aşırı örnekleme yapısına dayanan kademeli girişim dengeleyici içeren bir temelband alıcı yapısı önerilmiştir. Ardından SI iptal işleminin her zaman ideal olarak gerçekleştirilemediği durumlar göz önüne alınarak, iletim süresini HDR sisteme göre yarıya düşüren FDR sisteminin SI iptal katsayısının hangi değerlerinde daha iyi başarım verdiğini gösteren üst sınır ifadesi bulunmuştur. Ayrıca, FD-PNC sistemlerin bit hata başarımlarına ilişkin SI iptal katsayısına bağlı, sıkı alt ve üst teorik limitler bulunarak benzetim çalışmalarıyla teorik sonuçlar desteklenmiştir.
Özet (Çeviri)
Wireless communication systems use half-duplex (HD) transmission technique, which can enable only the transmission or the reception of data at a specified time interval or frequency band. However, taking into account that the ever increasing demand of data traffic, efficient usage of radio sources is crucial. Full-duplex (FD) transmission technique is developed as a method to increase the spectral efficiency at the same frequency channel during the same time interval. Requiring a single frequency band for transmission, FD technique can ideally double the spectral efficiency. It can also prevent the hidden node problem and help security of physical layer applications. Although FD transmission technique have these advantages, it has not yet been used in communication systems practically due to the generated self-interference (SI) by the local transmit antenna. The SI signal can be more powerful (15-100 dB) than the transmission signal at the receiving antenna. In order to eliminate the effects of SI, active and passive cancellation techniques are proposed in the literature. In order to design an operational FD system, using only passive cancellation techniques would not be sufficient to eliminate the effects of the SI. Passive and active cancellation techniques are jointly applied in FD systems to remove destructive effects of the SI on the system's error performance. The recent studies are related to the cancellation of SI signals on FD communication. With this motivation, in the first part of thesis, in order to design the baseband receiver model in accordance with the properties of SI channel, we use NI USRP-2921 software defined radio kits for estimating SI channel. The distance between transmit and receive antennas is set to $10$ cm and the radio is operated on FD mode. Narrow-band signals with 200 kHz bandwidth are used at the 2.4 GHz with an oversampling factor of 8. From test results, we observe that the channel is time dispersive with an exponential distribution, even for narrow-band signals. % Here, the SI channel can also be modeled as $|h_{i,i}[k]|=\beta e^{-\lambda k}$.\\ We demonstrate via measurements that the SI channel in FD systems can be time dispersive and we propose a new fractionally spaced SI canceler to improve the error performance that eliminates residual interference, time dispersiveness of the channel. There are two different sampling rates are used as baud and fractional rate. The channel estimation block operates at the fractionally sampled space in order to determine the SI channel accurately. A time division multiplexing based channel estimation protocol can be implemented between nodes so that in the first time slot node the first node transmits pilot symbols for estimating SI and transmission channel. This procedure can be repeated for the other node in the following time slot. Once the SI channel is estimated, a tapped delay line, composed of unit delays, is implemented in order to replicate SI. Following this, SI signal is transferred to the baud space for cancellation. After the cancellation process, a channel equalizer is designed and implemented for the transmission channel. Then the received signals are fed into the decision rule block in order to obtain received symbols correctly. The results of studies are; if the received signal is sampled at baud sampling rate of lower sampling than the local channel's characteristics, the SI channel components can not be estimated completely, hence remaining local channel coefficients cause residual interference, the bit error rate (BER) performance gets deteriorate and an error floor is encountered. We formulate this residual interference and verify the presented error performance analysis with simulations considering different oversampling ratios at the receiver for $M$-QAM modulations. We also highlight the importance of accurate channel estimates with simulation and analytical results. We verify our analytical results via simulations, showing that the proposed SI canceler can critically enhance the performance of FD systems. In the FD radios, it needs to be emphasized that the oversampling rate must be chosen equal the length of local channel to discard the SI completely. In order to combat wireless fading channel characteristics, relaying techniques are proposed in the literature. If source and destination nodes could not communicate with each other because of absence of line of sight (LOS) between them, they can communicate with each other by using a node, which is named as relay node. This transmission technique is termed as relaying transmission. On the other hand, if there are LOS between source and destination nodes, the received signals which are radiated source and relay nodes, are combined to improve the system's BER performance. This technique is termed as cooperative communication. Relaying/cooperative systems enhance the system's BER performance against the wireless channel characteristics. Relay node can use different relaying techniques like as amplify-and-forward (AF), decode-and-forward (DF). Relay node only amplifies the received signal and forward it to the destination node in AF relaying technique. In case of DF relaying, relay node decodes the received signal and sends the re-encoded signal to the destination node. If additive white Gaussian noise is very dominant, the sensitivity of noise propagation will increase. On the other hand, the error propagation problem can occur due to decoding and re-encoding the receiving signal at relay node. We examine relaying systems with FD operating nodes in the second part of thesis. Classical HD relaying systems need two phases, FD relaying systems could reduce $50 \%$ the required time interval for transmitting only one signal. By this motivation, data signaling rates and error performances of HD and FD relaying systems in flat fading channels are investigated in cases of ideal and non-ideal SI cancellation. Because of the bandwidth of the system, antenna placement errors and difference between desired and SI signals levels are caused to non-ideal SI cancellation process. Single relay system is considered for AF and DF techniques. Channel models are selected according to Nakagami-\textit{m} distribution since it provides a wider fading interval according to \textit{m}-parameter. The last part of thesis consists of two way relay channel (TWRC) systems in case of absence of LOS between end nodes. Two end nodes tend to exchange their signals to each other through a relay node in TWRC. Considering two end nodes wants to communicate with each other through a relay node in the classical HD relay systems. The communication process occurs in four time slots. In the first time slot, end node send its own signal to relay node and the relay node sends the received signal radiating from the first end by using amplify and forward (AF) or decode and forward (DF) technique to the second end node. Later, second end node sends its signal to relay at third time slot and relay node and the relay node sends the received signal radiating from first end by using AF or DF technique to the second end node. Because of the disadvantage of necessary transmission time interval, instead of HD relay systems, physical layer network coding (PNC) system is proposed in the literature. The transmission process occurs within two time slot between two end nodes for PNC system. In the firs phase, two end nodes send their signals to relay node which is named as multi access (MA) phase, and relay sends its signal to two nodes, referred to as broadcast (BC) phase. Relay transmits a function of the collided symbols from end nodes. Having the $a \; priori$ knowledge about their own transmitted symbols, end nodes can detect the unknown symbol. Information exchange through PNC hence takes two time slots in TWRC. The use of FD nodes in TWRC is proposed for the first time in the pioneering work of Rankov and Wittneben, where authors investigate the information theoretical rate region without considering bit level transmission issues such as modulation and coding techniques. The studies of TWRC with FD nodes explore the information-theoretic standpoints. We propose a new physical layer network code for TWRC where the transmission takes a single time slot via combining the MA and BC phases by making use of FD transmission at all nodes. In the proposed network coding technique, named as full duplex physical layer network coding (FD-PNC), relay uses DF relaying technique. The maximum likelihood (ML) based joint detection (JD) is used at the relay in order to eliminate the interference, which occurs due to the reception of two symbols from distinct sources at the same time slot. We adopt the max-log approximation to the ML detection rule, which is frequently preferred in order to reduce the computational complexity of the receiver. The proposed FD-PNC system can save $50\%$ of the time needed by the classical PNC approaches. Note that the FD-PNC saves $66.7\%$ of the time when compared to the classical network coding, where the MA phase takes two time slots (requiring 3 time slots in total). %On the other hand, similar to FD relay communication, FD TWRC communication utilizes only one time slot. Hence, the capacity of the system is doubled with FD TWRC system in the ideal SI cancelation situation. Contributions result from the study made in this thesis concept are splitted into three categories. Firstly, the baseband receiver structure which contains fractionally spaced equalizer and is appropriate for the time dispersiveness characteristic of measured SI channel is proposed. Secondly, by taking into account of the fact that SI cancellation process can not be ideally realized. Upper bound expression of FD relaying system decreases the transmission time to half compared to the HD relaying system which give more successful results related to the selection of value of SI cancellation coefficient is derived. A hybrid relaying system that is able to choose between FD or HD transmission according to the value of SI cancellation parameter is proposed to maximize the data rate of the system. Finally, we obtained the theoretical BER expressions of FD-PNC in the case of non-ideal SI cancellation. We obtain tight upper and lower bounds for theoretical BER expressions. We confirm our analytical results via simulations. %lower and upper bound expressions are obtained theoretically for FD-PNC system.
Benzer Tezler
- Çok atlamalı çift-yönlü röleli ağlar
Multi-hop bi-directional relay networks
BARIŞ KÖKSAL
Yüksek Lisans
Türkçe
2012
Elektrik ve Elektronik Mühendisliğiİstanbul Teknik ÜniversitesiElektronik ve Haberleşme Mühendisliği Ana Bilim Dalı
PROF. DR. ÜMİT AYGÖLÜ
- Anten seçimli fiziksel katman ağ kodlama
Physical layer network coding with antenna selection
MUSTAFA YAĞIMLI
Yüksek Lisans
Türkçe
2012
Elektrik ve Elektronik Mühendisliğiİstanbul Teknik ÜniversitesiElektronik ve Haberleşme Mühendisliği Ana Bilim Dalı
PROF. DR. İBRAHİM ALTUNBAŞ
- Array processing and optimization techniques of beamforming and resource allocation for enhanced spectral efficiency in 5g and beyond systems
5g ve ötesı sıstemlerının hüzmeleme ve spektral verımlılığının vektör íşleme ve optımızasyon ıle gerçekleştırılmesı
NANN WIN MOE THET NANN WIN MOE THET
Doktora
İngilizce
2021
Elektrik ve Elektronik Mühendisliğiİstanbul Medipol ÜniversitesiElektrik-Elektronik Mühendisliği ve Siber Sistemler Ana Bilim Dalı
Assoc. Prof. Dr. MEHMET KEMAL ÖZDEMİR
- Novel data partitioning and scheduling schemes for dynamic federated vehicular cloud
Dinamik federe araç bulutu için yeni bir görev yükü paylaşımı ve iş planlaması şemaları
WISEBORN MANFE DANQUAH
Doktora
İngilizce
2022
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiBilgisayar Mühendisliği Ana Bilim Dalı
PROF. DR. DENİZ TURGAY ALTILAR
- Genetic algorithm based ARINC-664 worst case delay optimization using network calculus
Genetik algoritma ve network calculus ile ARINC-664 uçtan uca gecikme optimizasyonu
EYÜP CAN AKPOLAT
Yüksek Lisans
İngilizce
2022
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiElektronik ve Haberleşme Mühendisliği Ana Bilim Dalı
PROF. DR. HAKAN ALİ ÇIRPAN