Differential Six-Port Transceiver Design and Analysis from a Wireless Communication System Perspective

Full text

(1)LiU-ITN-TEK-A--12/044--SE. Differential Six-Port Transceiver Design and Analysis from a Wireless Communication System Perspective Muhammad Umar Umair Yasir 2012-06-12. Department of Science and Technology Linköping University SE-601 74 Norrköping , Sw eden. Institutionen för teknik och naturvetenskap Linköpings universitet 601 74 Norrköping.

(2) LiU-ITN-TEK-A--12/044--SE. Differential Six-Port Transceiver Design and Analysis from a Wireless Communication System Perspective Examensarbete utfört i Elektroteknik vid Tekniska högskolan vid Linköpings universitet. Muhammad Umar Umair Yasir Handledare Magnus Karlsson Examinator Adriana Serban Norrköping 2012-06-12.

(3) Upphovsrätt Detta dokument hålls tillgängligt på Internet – eller dess framtida ersättare – under en längre tid från publiceringsdatum under förutsättning att inga extraordinära omständigheter uppstår. Tillgång till dokumentet innebär tillstånd för var och en att läsa, ladda ner, skriva ut enstaka kopior för enskilt bruk och att använda det oförändrat för ickekommersiell forskning och för undervisning. Överföring av upphovsrätten vid en senare tidpunkt kan inte upphäva detta tillstånd. All annan användning av dokumentet kräver upphovsmannens medgivande. För att garantera äktheten, säkerheten och tillgängligheten finns det lösningar av teknisk och administrativ art. Upphovsmannens ideella rätt innefattar rätt att bli nämnd som upphovsman i den omfattning som god sed kräver vid användning av dokumentet på ovan beskrivna sätt samt skydd mot att dokumentet ändras eller presenteras i sådan form eller i sådant sammanhang som är kränkande för upphovsmannens litterära eller konstnärliga anseende eller egenart. För ytterligare information om Linköping University Electronic Press se förlagets hemsida http://www.ep.liu.se/ Copyright The publishers will keep this document online on the Internet - or its possible replacement - for a considerable time from the date of publication barring exceptional circumstances. The online availability of the document implies a permanent permission for anyone to read, to download, to print out single copies for your own use and to use it unchanged for any non-commercial research and educational purpose. Subsequent transfers of copyright cannot revoke this permission. All other uses of the document are conditional on the consent of the copyright owner. The publisher has taken technical and administrative measures to assure authenticity, security and accessibility. According to intellectual property law the author has the right to be mentioned when his/her work is accessed as described above and to be protected against infringement. For additional information about the Linköping University Electronic Press and its procedures for publication and for assurance of document integrity, please refer to its WWW home page: http://www.ep.liu.se/. © Muhammad Umar, Umair Yasir.

(4) Differential Six-Port Transceiver Design and Analysis from a Wireless Communication System Perspective. Muhammad Umar Umair Yasir. i.

(5) ii.

(6) iii.

(7) Abstract In modern telecommunication there is the demand of high data rates using wideband component design. FCC has introduced the UWB spectrum for high speed data communication. UWB systems have attracted the attention of researchers. Six-port transmitters and receivers are strong candidates for UWB systems and research is being done on six-port modulators and demodulators. In this work an effort is made to compare the performance of conventional singleended six-port transmitter and receiver with differential six-port transmitters and receivers. In this thesis, single ended and differential six-port correlators are designed on 7.5 GHz using Agilent Inc. EDA tool ADS and their performance is evaluated. A new wide-band differential six-port correlator is implemented using rat-race couplers and double-sided parallel strip-line phase inverter. The designed six-port correlators are used for 8-PSK modulation and demodulation. For transmitter-receiver system, mixed analog-DSP designing is used. The integral components of the system are evaluated individually and behavioral modeling is used to evaluate the complete transmitter-receiver system. The single-ended and differential systems are evaluated for noisefigure, dynamic range, bit error rate and data rate.. iv.

(8) Acknowledgements We would like to recognize the effort and support of following persons who have helped and enabled us to successfully complete the thesis work: Our examiner Dr. Adriana Serban for her support, guidance and giving us the opportunity to work with her. Our Supervisor Dr. Magnus Karlsson for his support in the thesis, especially in PCB fabrication and measurements. Mr. Gustav Knutsson for his help in PCB fabrication. All WNE degree students specially Ionut-Alexandru Apolozan. Never ending support of our beloved families who have always encouraged and supported us throughout the life.. v.

(9) List of Abbreviations ADS ASK BLC BPF DSP DSPSL EDA EM EMI FCC FSK HPF I/Q IC IF LNA LO LPF NF PA PCB PSK QAM RF SMA SNR UWB VCO VGA VNA WPD. Advanced Design System Amplitude Shift Keying Branch Line Coupler Band Pass Filter Digital Signal Processing Double Sided Parallel Strip Line Electronic Design Automation Electromagnetic Electromagnetic Interference Federal Communications Commission Frequency Shift Keying High Pass Filter In-Phase and Quadrature Phase Integrated Circuit Intermediate Frequency Low Noise Amplifier Local Oscillator Low Pass Filter Noise Figure Power Amplifier Printed Circuit Board Phase Shift Keying Quadrature Amplitude Modulation Radio Frequency Subminiature version A Signal to Noise Ratio Ultra Wide Band Voltage Controlled Oscillator Variable Gain Amplifier Vector Network Analyzer Wilkinson Power Divider. vi.

(10) vii.

(11) Table of Contents. ABSTRACT ............................................................................................................................................................. IV ACKNOWLEDGEMENTS .......................................................................................................................................... V LIST OF ABBREVIATIONS ....................................................................................................................................... VI TABLE OF CONTENTS...........................................................................................................................................VIII 1. INTRODUCTION ............................................................................................................................................ 1 1.1 1.2 1.3 1.4. 2. THE UWB COMMUNICATION TECHNOLOGY ...............................................................................................................2 MOTIVATION AND OBJECTIVE OF THE THESIS .............................................................................................................3 METHOD ............................................................................................................................................................4 CONTRIBUTIONS...................................................................................................................................................4. THEORETICAL BACKGROUND ........................................................................................................................ 5 2.1 MODULATION SCHEMES ........................................................................................................................................5 2.1.1 Amplitude modulation ...........................................................................................................................6 2.1.2 Phase modulation ..................................................................................................................................6 2.1.3 Frequency modulation ...........................................................................................................................9 2.2 TRANSCEIVER ARCHITECTURES .................................................................................................................................9 2.2.1 Transmitter designs ...............................................................................................................................9 2.2.2 Receiver designs ...................................................................................................................................12 2.3 DIFFERENTIAL SIGNALING .....................................................................................................................................14 2.3.1 Two wire signaling ...............................................................................................................................15 2.3.2 Voltages and currents in differential signaling ....................................................................................16 2.3.3 Differential impedance ........................................................................................................................17 2.3.4 Mixed-mode S-Parameters ..................................................................................................................17 2.3.5 PCB structures for differential signaling ..............................................................................................18. 3. SIX-PORT CORRELATOR............................................................................................................................... 19 3.1 IDEAL SIX-PORT CIRCUIT .......................................................................................................................................19 3.1.1 Wilkinson Power Divider ......................................................................................................................20 3.1.2 Quadrature Branch line coupler ...........................................................................................................21 o 3.1.3 180 Branch Line Coupler .....................................................................................................................22 3.2 MODULATION USING SIX-PORT CORRELATOR ...........................................................................................................23 3.3 DEMODULATION USING SIX-PORT CORRELATOR .......................................................................................................25. 4. DESIGN AND IMPLEMENTATION OF SIX-PORT CORRELATOR ...................................................................... 29 4.1 SINGLE-ENDED DESIGNS .......................................................................................................................................30 4.1.1 Classical single-ended design ...............................................................................................................30 4.1.2 Single-ended design with matching stubs............................................................................................35 4.2 DIFFERENTIAL DESIGNS ........................................................................................................................................38 4.2.1 Classical differential design .................................................................................................................38 4.2.2 Wideband differential design ...............................................................................................................44 O 4.3 WIDEBAND 180 COUPLER DESIGN .......................................................................................................................50. 5. SIX-PORT MODULATOR AND DEMODULATOR DESIGN ............................................................................... 54 5.1. SIX-PORT MODULATOR IMPLEMENTATION ...............................................................................................................54. viii.

(12) 5.1.1 Variable port impedances ....................................................................................................................55 5.1.2 8-PSK modulation using mixed analog-DSP designing .........................................................................55 5.2 SIX-PORT DEMODULATOR IMPLEMENTATION............................................................................................................59 5.2.1 Diode modeling ....................................................................................................................................60 5.2.2 Notch filter ...........................................................................................................................................61 5.2.3 Digital judgment circuit .......................................................................................................................61 5.3 SIX-PORT TRANSMITTER-RECEIVER SYSTEM ..............................................................................................................61 6. DESIGNED SIX-PORT TRANSCEIVER SYSTEM EVALUATION .......................................................................... 64 6.1 6.2 6.3 6.4. 7. CONCLUSION & FUTURE WORK .................................................................................................................. 70 7.1 7.2. 8. NOISE FIGURE COMPARISON ................................................................................................................................64 BER AND DYNAMIC RANGE COMPARISON ..............................................................................................................65 BER AND DATA RATE COMPARISON .......................................................................................................................67 MODULATED SIGNAL CONSTELLATION DIAGRAMS AND POWER SPECTRUM COMPARISON ................................................67. CONCLUSION .....................................................................................................................................................70 FUTURE WORK ..................................................................................................................................................70. REFERENCES ............................................................................................................................................... 72. ix.

(13) 1 INTRODUCTION Wireless communication is ubiquitous nowadays. We can find its presence everywhere around us. From indoor communication applications to long range communication systems which not only connect the world together but can also communicate with the Rovers sent to Mars, wireless communication has redefined the possibilities. There are numerous applications and standards varying in data rate from some Kbits/s to Gbits/s and coverage distance from some meters to hundreds of kilometers. The trend is to replace wired communication with wireless devices which can communicate with similar efficiency and reliability. Advancements in the research and design of wireless devices and techniques are going on for improving the reliability and robustness in the communication and to introduce new innovations to the everyday life. It all dates back to the prediction and description of Electromagnetic waves by James Clerk Maxwell in 1873 [1]. Maxwell, on mathematical grounds, suggested that a varying Electric field produces a varying Magnetic field and same in vice versa. Heinrich Hertz carried out a set of experiments in 1887-1891 and validated the theory presented by Maxwell [2]. Guglielmo Marconi, an Italian inventor, carried out first demonstration of wireless communication in 1895. In 1940s, during the World War 2, research and development of RADAR (RAdio Detection And Ranging) attracted much attention to the field of Wireless communication specifically Microwave communication. Also, parallel advancements in the field of Solid state physics such as invention of first Silicon transistor at Texas instruments in 1954 and its use as a switch in electronic devices paved the way for development of complex digital communication devices. The cellular mobile communication started in the late 1970s and early 1980s. The first commercial analog cellular system was launched by NTT (Nippon Telephone & Telegraph) in Tokyo, Japan in December 1979. In 1981 NMT (Nordic Mobile Telephone) introduced cellular system in Nordic countries. Cellular mobile system started to evolve and expand rapidly. The advent of 2G (2nd Generation) systems in 1991 came with Digital version of cellular systems which offered more spectral efficiency and security. The second generation got the name of GSM (Global System of Mobiles).The consequent introductions of various data transmission technologies and standards such as GPRS (General Packet Radio Service) and EDGE (Enhanced Data rates for GSM Evolution) to GSM started a new era of wireless communication and spread the use of cell phones dramatically. In addition to these WANs (Wide Area Networks), short range wireless communication networks or WPANs (Wireless Personal Area Networks) got evolved tremendously as well. Bluetooth was introduced in 1999. It is a short range (1-100 meters) wireless alternative for communication wires/cables. The technology’s latest version 3.0 + HS incorporates with 802.11 (Wi-Fi) and can serve up to data rates as high as 24 Mbit/sec. Before that version 2.0 + EDR was able to provide 3 Mbit/sec data rate [3]. On the other hand IEEE (Institute of Electrical and Electronic Engineers) developed specifications for 802.11 in 1997 [4]. 802.11 is a WLAN (Wireless Local Area Network) specification and is popularly known as Wi-Fi. The technology 1.

(14) now has different versions namely 802.11 a, b, e, g and n with the latest in practice i.e. 802.11 n can reach up to net data rates of 300 Mbit/sec. Following is a summary of various short range communication standards and technologies: IEEE 802.11 a/b/g/n (Wi-Fi)  Bluetooth, IEEE 802.15.1  ZigBee, IEEE 802.15.4  UWB (Ultra wideband) These technologies in terms of coverage distance and data rates are depicted in Figure 1.1. Figure 1.1 Different wireless communication technologies in terms of distance and data rates.. The last mentioned i.e. UWB is the focus of this thesis and will now be discussed in detail.. 1.1 The UWB communication technology Ultra-Wideband is a recent inclusion to the short range wireless communication technologies and it was first authorized by FCC (Federal Communications Commission) of USA in 2002 [5]. FCC allocated a wideband i.e. 3.1-10.6 GHz and with emission limitation on power spectral density of -41 dBm/MHz. This limitation was implemented to reduce interference with other communication technologies. A larger bandwidth and lower power offers a massive increase in data rates compared to other narrowband short range standards such as Bluetooth and 802.11 WiFi. The interrelation between channel capacity with bandwidth is described by Shannon equation:C = B × log2(1 + SNR) Where C = channel capacity or data rate or throughput B = channel bandwidth SNR = Signal to noise ratio or 2.

(15) This means that for higher channel capacity we can either allocate more bandwidth or increase the signal to noise ratio. Increasing SNR in wireless communication implies that we increase the transmitting power which in turn can increase interference with other communication systems so it is not desirable. Also its relation is logarithmic with the channel capacity. Then other option is increasing bandwidth, which the UWB has employed. The suggested bandwidth by FCC of 3.1-10.6 GHz is roughly 7 GHz. The first half of this band overlaps with the unlicensed 5.725 – 5.875 GHz ISM (Industrial, Scientific and Medical) band. Therefore in Europe, Asia and Japan there are further requirements of LDC (Low Duty Cycle) and DAA (Detect and Avoid) techniques in the 3.1 – 4.8 GHz band to avoid interference with existing technologies. EC (European Commission) has limited the UWB bandwidth for devices without requirements for DAA mitigation techniques to 6-8.5 GHz. NICT (National Institute of Information and Communications Technology) in Japan has divided the UWB band into separate bandwidths of 3.4 - 4.8 GHz and 7.25 – 10.25 GHz. In the first band interference mitigation techniques are required and further the allowed average transmission power is reduced to -70 dBm/MHz. In China, the approved bands for UWB operation are 4.2 - 4.8 GHz and 6 – 9 GHz [6]. UWB communication can be categorized in two ways when it comes to transmission of channel signaling i.e. (i) carrier free UWB communication (ii) carrier based UWB communication. Carrier free UWB, also known as impulse radio UWB, the data is sent in the form of short or low duty cycle pulses utilizing the whole allocated band. Carrier based UWB is further divided into (a) single carrier (b) multi carrier. Example of single carrier is the use of DSSS (Direct sequence spread spectrum) technique. Multi carrier UWB involves OFDM (Orthogonal Frequency Division Multiplexing) technique in which signal is sent using multiple modulated orthogonal carriers [7].. 1.2 Motivation and Objective of the Thesis Wireless communication in UWB requires new components designed and optimized for this frequency band. A lot of research and design work has been going on designing LNA (Low Noise Amplifier), Antenna (transmit/receive), Mixer, Band Pass filter, Power Amplifier, Six port Direct carrier modulator and other components in the wireless communication system hierarchy, for UWB band. Several studies have been carried out in Linköping University prior to this [6], [8] and [9]. All of the above mentioned works were concentrated on communication system architectures employing Six-port direct carrier modulator. Six-port systems are made up of passive microwave components and are simpler than conventional heterodyne radio transceivers involving IF (Intermediate Frequency) components. The main advantages of Six-port correlator are of large frequency bandwidth and less power consumption [10] [11]. Both these aspects make it a suitable choice for short range, high data rate UWB systems. As mentioned above, research in [8] was the study on implementing differential six-port transceiver. Differential six-port promises the advantages of signal integrity, reduced common mode noise, crosstalk suppression, compactness of the system and increased dynamic range [12]. 3.

(16) The goal of this thesis is to implement 8-PSK (Phase shift keying) modulation/demodulation on both single-ended and differential six port designs with a central frequency of 7.5 GHz. The objective is the system level implementation of both designs and their comparison from a wireless communication system perspective.. 1.3 Method The EDA (Electronic Design Automation) software tool used for the simulation tasks is ADS (Advanced Design Systems) version 2011 from Agilent Inc. The bandwidth chosen among the total UWB bandwidth is 6 – 9 GHz. This bandwidth is not subjected to the requirement of interference mitigation techniques by any country’s laws. Individual components of Six-port modulator/demodulator i.e. Wilkinson power divider and Branch line coupler are designed and optimized for 7.5 GHz centre frequency. They are then combined together to make the six-port correlator and the frequency, phase and amplitude responses were analyzed. The layout components are generated for both single ended and differential designs and EM (Electromagnetic) simulation are done using the ADS Momentum tool. The next part is the 8PSK signal modulation/demodulation. DSP (Digital Signal Processing) blocks in ADS are used to produce/recover the baseband signals. A communication system as a whole is simulated and system level parameters are analyzed.. 1.4 Contributions The work is done by two persons, Muhammad Umar and Umair Yasir. The individual contributions are mentioned in Table 1.1. Table 1.1 Individual contributions in work. Muhammad Umar      . Umair Yasir . Design of Six-port Correlators on schematic and layout levels Optimization of the designs for modulation Development of mixed analog-DSP 8PSK modulation scheme with port-5 signal digital control technique Demodulation of the 8-PSK signal on single-ended design PCB fabrication and measurement Report writing Ch 2,4,5.     . 4. Design of Six-port Correlators on schematic and layout levels Study of ADS DSP blocks DSP processing of the signal for mixed analog-DSP simulations Study of voltage-controlled switch to replace the ideal switches in the modulator PCB fabrication and measurement Report writing Ch: 1,3,6,7.

(17) 2 THEORETICAL BACKGROUND Telecommunication systems are an important constituent of life nowadays. A variety of telecommunication systems are being used depending on the need of the situation. The simplest scenario is a one way transmission and reception of information (depicted in Figure 2.1) where a source wants to send the information to a sink through a channel [13]. To make the information message appropriate for the channel the sender and detectors are used. The sender interprets the source’s message to the form appropriate for the channel and detector interprets the message in the channel back to original form. The task is to design this sender and detector to optimize the speed, efficiency and cost.. Source. Sender. Channel. Detector. Sink. Figure 2.1 A simple telecommunication model. Starting from the simple one way communication scenario the telecommunication technology is advancing towards more and more complex architectures exploiting the sophisticated signal processing techniques. The modern communication systems use several additional techniques including source coding, channel coding interleaving, multiplexing and frequency spreading [14]. All these efforts are being made to make the sender and detector (in Figure 2.1) more efficient.. 2.1 Modulation schemes The appropriate utilization of a communication channel requires shift of the information signal frequency into other frequency band suitable for transmission over the channel. For example a radio system operates by converting audio signal of 20 Hz – 20 kHz to radio signal of 30 kHz and upward. This process of shifting the range of the frequency to higher frequencies appropriate for transmission over the channel is called modulation [15]. Usually modulation is performed by varying the characteristics of a higher frequency sinusoidal wave according to the modulating signal (modulating wave). On the receiver side the reverse process of the modulation is performed known as demodulation. Modulation types are classified as analog modulation and digital modulation depending on the type of the modulating signal.. 5.

(18) 2.1.1 Amplitude modulation In amplitude modulation the amplitude of the carrier wave is varied according to the modulating wave. In case of digital baseband data the modulating signal is input symbols in form of distinctive voltage levels. So the modulated output wave also has distinct amplitude levels as input symbols with the frequency of carrier wave. An amplitude modulated wave can be expressed as: (2.1) SRF is the output modulated signal, Am(t) is the amplitude of the baseband signal and 𝛚c = 2πf is the angular frequency of the carrier wave. If the baseband signal is digital then this type of modulation is called Amplitude shift keying (ASK). If the input signal is an ordinary bit stream with levels of 0 and 1, it will control the modulated wave like a switch as shown in Figure 2.2. This type of modulation is also called onoff keying (OOK) [16]. The demodulation of amplitude modulated signal can be done by passing the signal from a low pass filtering circuit e.g. envelop detector, with removes the high frequency components of carrier wave and baseband signal is retrieved.. Modulating bits Carrier wave. ASK Modulated wave. Figure 2.2 Carrier wave, modulating signal and ASK modulated signal. 2.1.2 Phase modulation In phase modulation the phase of the carrier signal is varied according to the modulating wave. In other words the information is inserted in the phase of the modulated wave. This type of modulator is implemented using a multiplier.. 6.

(19) (2.2) Where AC is the amplitude of carrier and m(t) is modulating signal. In case of the digital baseband signal where the baseband signal is represented by unique voltage levels the output modulated wave takes discrete phase shifts. This is called phase shift keying (PSK). For a single bit stream of 0s and 1s the simplest phase modulation is done by transmitting the carrier with phase of 0o and 180o to represent a binary 0 and 1 respectively. This kind of modulation is called binary phase shift keying (BPSK). Figure 2.3 illustrates the BPSK modulated wave.. Modulating bits Carrier wave. PSK Modulated wave. FSK Modulated wave. Figure 2.3 Modulating signal, carrier wave, BPSK and BFSK signals. In order to utilize bandwidth more efficiently higher order PSK techniques are employed called M-PSK. M represents the number of symbols carried by the modulated wave in form of identical phase shifts. For this purpose the baseband data stream is divided into two or more parallel data steams which modulate two orthogonal carrier waves which are called in-phase wave and quadrature-phase wave. These two waves are then added together to get M-PSK modulated signal. QPSK, 8-PSK and 16-PSK use 4, 8 and 16 phase shifts respectively. Each phase represents an identical symbol in baseband. Figure 2.4 shows the QPSK and 8-PSK modulation points in signal space diagram (constellation diagram).. 7.

(20) Q. Q 1. 1. 90O. -1. 1. 45O. -1. I. -1. 1. I. -1. (a). (b) O. Figure 2.4 (a) QPSK and (b) 8-PSK modulated constellation diagram. Phase difference between any two symbols is 90 for O QPSK and 45 for 8-PSK. An M-PSK modulated wave is represented by the equation:. [. ]. (2.3). where i represents the symbol number. The data capacity per symbol increases with increased order modulation but on the cost of increased probability of bit error rate because the Euclidean distance between the two symbols in signal space decreases. In digital modulation a combination of ASK and PSK can be used to increase the number of symbols in signal space, increasing data rate [16]. This type of modulation is called Quadrature Amplitude Modulation (QAM). In QAM two orthogonal carrier waves are amplitude modulated and added together to get QAM signal. Figure 2.5 represents constellation diagrams for two types of QAM symbols. A QAM signal can be represented by the equation: [. ]. (2.4). Where XI and XQ are the baseband signals respectively called in-phase and quadrature-phase components of the baseband signal.. 8.

(21) Q. Q. 1. 1. -1. 1. -1. I. -1. 1. -1. (a). (b). Figure 2.5 (a) 4-QAM and (b) 16-QAM modulated constellation diagram. 2.1.3 Frequency modulation Frequency modulation exploits the frequency shifting of the carrier wave according to the modulating signal. The information to be transmitted is inserted in the frequency of the carrier wave. Figure 2.3 shows a frequency modulated wave. For digital modulation the discrete frequency values are used to represent the baseband symbols. It is called frequency shift keying. The frequency shift between the two values must be as small as possible to save bandwidth. The minimum shift which can be used is 1/2Tb, (Tb is the bit interval). FSK implemented using this criterion is called minimum-shift keying or fast-frequency shift keying [16].. 2.2 Transceiver architectures The word transceiver is a combination of “transmitter” and “receiver”. Transceiver is a device which can behave like transmitter as well as receiver. Efficient transceiver design in wireless communication is crucial [17]. The design should be capable of supporting high data rates with minimum errors maintaining communication over a long distance. A transceiver is supposed to be able to combat channel noises, attenuation and fading.. 2.2.1 Transmitter designs The task of the transmitter is to mix the information signal with a higher frequency carrier to produce a high power modulated signal in appropriate frequency band. The output power varies from few mW up to several kW [17]. The generalized output wave equation can be written as:. 9.

(22) (2.5a) (2.5b) Usually two types of transmitter architectures are used. One technique is to modulate directly at the transmission frequency and second is to do modulation at some lower frequency called intermediate frequency (IF) and then upconvert the signal to some higher frequency for transmission [6]. The former one is called Homodyne and latter one is called heterodyne transmitter as shown in Figure 2.6.. LPF. BPF & Matching network. I Baseband processing circuit. +. PA. Q LPF. sin. cos Self-modulation. VCO. LO-leakage. (a) LPF. BPF & Matching network. I BPF. BPF Baseband processing circuit. +. PA. Q LPF. sin. cos IF VCO. VCO. (b) Figure 2.6. (a) Homodyne transmitter, (b) Heterodyne transmitter architecture. In Figure 2.6a the VCO (LO) frequency is exactly equal to the carrier frequency; the modulation and upconversion are directly done at a single stage simultaneously. While in Figure 2.6b a relatively low frequency (f1) carrier is used for modulation and a second carrier with relatively. 10.

(23) higher frequency (f2) is used to further upconvert the modulated wave to a higher frequency (f1+f2). In both type of designs the signal is amplified by a power amplifier (PA). Quadrature Imbalance For a homodyne transmitter it is difficult to produce LO signals perfectly at quadrature phase at high frequency. The inaccuracy in the phase shift and mismatches in I- and Q-paths distorts the resultant constellation [6]. In heterodyne transmitter the modulation is performed at relatively low frequency reducing the quadrature errors. LO leakage and Self Modulation LO leakage is prominent when there is poor isolation between LO and RF ports of mixer causing LO signal to escape towards the antenna. In homodyne transmitters the LO leakage signal cannot be eliminated by the band-pass filter as it is exactly at the signal frequency [6]. While in heterodyne transmitter LO leakage is not a notable problem. LO leakage causes unnecessary power dissipation and constellation offset. Self modulation is caused when the modulated signal is reflected by PA backwards and escapes from mixer to VCO, disturbing the VCO spectrum. This problem is prominent in heterodyne transmitters as the VCO is not at the carrier frequency [6]. We can summarize the advantages and disadvantages for both types of transmitter structures as [18]:. Transmitter Type. Homodyne. Heterodyne. Advantages.  Low cost  High integratibilty  Simple structure  Reliable performance  No LO leakage. Disadvantages  Quadrature Imbalance  LO leakage  Self modulation    . Expensive Larger in size Additional filtering Increased power dissipation. Table 2.1 Comparison summary of transmitter architectures. 11.

(24) 2.2.2 Receiver designs The tasks of a RF receiver are to down-convert the signal frequency and demodulate it to get the original information back. The receiver has to be able to detect a very low-power signal in a noisy environment in the presence of other unwanted frequencies. Receiver front-end designs are much complex than transmitter designs due to above mentioned requirements [6]. A simplest receiver is a tuned radio receiver which contains a BPF, a LNA to amplify the signal power, demodulator, LPF and a PA. Depending on the scenarios a variety of receiver architectures are implemented including homodyne (direct conversion or zero-IF), heterodyne, super-heterodyne and low IF receivers [6]. Figure 2.7 presents the structures for homodyne and heterodyne receivers.. LPF I VGA. BPF. Baseband processing circuit. LNA. VGA. Q sin. cos. LPF. VCO. (a). LPF Image reject BPF. BPF. I VGA. BPF. Baseband processing circuit. LNA VGA. Q sin. cos. LPF. VCO. IF VCO. (b) Figure 2.7. Receiver architectures (a) Homodyne (direct-conversion receiver) (b) Heterodyne receiver. Homodyne receiver converts the RF signal to baseband in a single stage as presented in Figure 2.7a. The signal is first filtered then amplified by low-noise amplifier (LNA) to increase its power to some detectable level. After amplification the signal is down-converted and demodulated to baseband signal in a single stage. The signal is then low-pass filtered to remove 12.

(25) high frequency components and variable gain power amplifier (VGA) adjusts the signal power to appropriate level for analog to digital conversion. Heterodyne receivers are more popular than homodyne. Heterodyne receivers works in two stages as illustrated in Figure 2.7b. First the signal is filtered then amplified by LNA, after amplification it is further filtered to suppress image frequencies. In first stage it is downconverted to a lower frequency and in second stage it is demodulated to get the baseband signal. The signal is further filtered and amplified by VGA. A receiver can be evaluated in terms of its noise figure, sensitivity, selectivity and dynamic range [17][19]. High performance components and careful design approach must be applied to design a high fidelity receiver. The main RF receiver design aspects are: Quadrature Imbalance The imperfect quadrature phase shift and difference in amplitudes of LO signals cause distorts the demodulated signal. This problem becomes prominent in homodyne receivers. In addition, mismatches in I- and Q- signal paths also corrupts the signal constellation. In heterodyne receivers the demodulation is performed at relatively low frequency reducing the quadrature imbalance problem. LO leakage LO leakage is caused by the poor isolation between LO and RF ports of the mixer. In receivers the LO signal escapes from the mixer toward the RF port get passed through the parasitic in the LNA. It may get radiated by the antenna causing unwanted radiations by the receiver or it may get back to the input of the LNA getting amplified and fed to the mixer. Mixing with original LO signal it causes self-mixing producing DC components at the output of the mixer. In homodyne receiver designs this DC component is superimposed on the baseband signals distorting them [6]. Image frequencies In heterodyne receivers the carrier frequency is first down-converted to an IF frequency. During mixing the frequencies at a distance of 2fIF from the carrier are also shifted to IF band causing interference with the original signal. To eliminate image frequency an image rejection filter is used to filter out the image frequencies prior to mixing. In homodyne receivers image frequencies are not a problem. Sensitivity Sensitivity is defined as minimum signal level required at the antenna to obtain a defined signal to noise ratio (SNR) at the receiver output [17]. Wireless communication range depends upon the smallest level of the signal a receiver can process. Homodyne receives exhibit relatively good sensitivity compared to heterodyne [17]. Selectivity. 13.

(26) Selectivity is the ability of the receiver to receive a particular band while rejecting the adjacent bands [17]. In multiband radio communication several radio transmissions are done on adjacent bands with a guard-band. The band-pass filters in the receivers affect the selectivity most. Heterodyne receivers have relatively good selectivity than homodyne due to increased filtering [17]. Noise figure Noise figure is defined as the ratio of input SNR to output SNR. The noise figure of the whole receiver system is dominated by the noise figure of first active component. For that reason the amplifier in the beginning of RF receiver is optimized for the low noise figure hence called Low noise amplifier (LNA). The receiver sensitivity depends on the receiver’s noise figure. Comparing homodyne with heterodyne receiver, the advantages and disadvantages can be summarized as [18][17]:. Receiver Type. Advantages. Homodyne.  Low cost  High integratable  No image frequency problem  Reduced filtering  Better sensitivity. Heterodyne.  Reliable performance  No LO leakage problem  Better selectivity. Disadvantages.  Quadrature Imbalance  LO leakage  Low selectivity.  Expensive and bulky  Additional filtering  Increased power dissipation. Table 2.2 Comparison summary of receiver architectures. 2.3 Differential signaling Typically the electronic systems share a single conductor for current return path between transmitter and receiver called ground. This use of a single reference conductor (ground) is called single ended signaling. The IC package pins have resistance and parasitic causing shifts in the ground plane. One receiver may be acting as a transmitter for other receivers adding further shifts in the ground plane as explained in Figure 2.8a. Moreover noise exists between each two points on the ground conductor. If the reference voltage on the ground conductor is shifted too 14.

(27) much the single ended signaling no longer works. The noise produced by unnecessary voltage drop on the impedance of ground connection on signal return path is called ground bounce [20].. Tx. Rx. Tx. Current path. Current to next device. Gnd. Gnd. Package connection resistance. ZA. Shared Ground. Package connection resistance. current return path ZB. (a). Tx. Signal wire. Rx. Tx. Current to next wire. Signal return. Gnd. ZA. Shared Ground. Gnd. Package connection resistance. Package connection resistance. current return wire. ZB. (b) Figure 2.8 Signal transmission systems between two devices with package resistances ZA and ZB (a) Single-ended system with shared ground as current return path (b) two-wire system.. 2.3.1 Two wire signaling Two wire signaling can solve the ground shift problem at cost of one extra wire used as signal return path instead of common ground as shown in Figure 2.8b. In high frequency circuits the wires or PCB traces have coupling with the system chassis or other conducting materials. It causes the current induced in the chassis which can be modeled as added parasitic. The transmitted signal current finds this parasitic as an option for the return path. The current that returns from the parasitic is called stray current [20]. In high speed circuits the stray current may cause malfunctioning of the system.. 15.

(28) The solution of the above mentioned problem is the transmission of the signals mutually opposite on the both conductor wires (or traces). This type of signaling is called differential signaling. If both the conductors have the identical coupling with the reference (or the chassis) both wires induce opposite signals cancelling the effect of each other. If both wires do not have the same coupling or the signals are not perfectly complementary, some amount of current will flow in the reference called common mode current.. 2.3.2 Voltages and currents in differential signaling Let denote the instantaneous voltages on the two wires as v1 and v2 with respect to an arbitrary reference. The difference in the instantaneous voltages v1 and v2 is called differential voltage vd and the average of the instantaneous voltages is called common-mode voltage vc [20]. (2.6a). (2.6b). Common-mode voltage causes production of common mode current. Common-mode current is not cancelled by noise cancellation property of the differential signaling. Moreover it contributes a lot in electromagnetic interference (EMI) radiations [21]. Another decomposition of the differential signaling is the even-mode and odd-mode voltages. An odd-mode voltage in a conductor is one whose opposite exists in the second conductor. Odd mode voltage vo is half of differential voltage vd. Even-mode voltage is one which is same on both conductor wires. It is same as common mode voltage vc.. (2.7a). (2.7b). Voltage on one conductor is the sum of even- and odd-mode voltages and on the other it is the difference. Current on one conductor is the sum of differential- and common-mode currents and on the other it is the difference [20][21]. (2.8a). 16.

(29) (2.8b) (2.8c) (2.8d). 2.3.3 Differential impedance Similar to voltage and currents, impedance for differential signaling is also categorized as differential, common-mode, even and odd impedances. The definitions are as follows [22]: Differential impedance is the impedance seen into a transmission line when exited in differential mode. Common-mode impedance is seen into a transmission line when excited with same signals on both conductors. Odd-mode impedance is the impedance of single conductor of transmission line while the other is excited with opposite signal. Odd-mode impedance is half of differential impedance. Even-mode impedance is the impedance of single conductor of transmission line while the other excited with the same signal. Even-mode impedance is the double of common-mode impedance. 2.3.4 Mixed-mode S-Parameters S-parameters or scattering parameters are used to define the response of a microwave network at RF frequencies. S-parameters for port n are defined by the normalized input and output power waves (an and bn respectively) when all the ports are terminated in matched conditions. S-parameters for a single-ended two port network can be defined by a 2×2 matrix. For to define a two port differential network a 4×4 matrix is required as there is a pair of signals on each port [23]. The differential signaling contains both differential- and common-mode signaling so single ended S-parameters cannot be used to fully define the behavior of transmission in differential signaling. For that purpose mixed-mode s-parameters are used. The mixed-mode s-parameters for a differential two-port network are defined by [24]:. [. ]. [. ][. 17. ]. (2.9).

(30) Where adn and acn are the normalized incident power waves for differential- and common-mode respectively and bdn and bcn are the normalized reflected power waves for differential and common-mode respectively for the port n. The S-matrix provides information for various transmission behaviors as explained as follows: Sdd:. differential-mode S-parameters. Scc:. common-mode S-parameters. Scd:. conversion from common- to differential-mode. Sdc:. conversion from differential- to common-mode. With mixed-mode S-parameters the details of wave propagation in differential signaling and mode conversion are represented. More details on mode conversion can be found in [23].. 2.3.5 PCB structures for differential signaling Different PCB configurations for differential signaling are in use including edge-coupled differential microstrip, edge-coupled stripline, broadside-coupled stripline and double-sided parallel stripline (DSPSL) [20][25]. Only double-sided parallel stripline structure is discussed here. It is like double sided microstrip structure without ground plane with strips on the both sides of PCB exactly on top and bottom of each other as shown in Figure 2.9c.. (a). +. +. _. _. (b). (c). Figure 2.9 Cross sections of (a) microstrip and (b) double-sided parallel strip-lines. (a-c) showing conversion of a microstrip line into differential strip line. The design of DSPSL is related to design of simple microstrip line. The characteristic impedance of a DSPSL of width w on a substrate of height 2h is double the characteristic impedance of a microstrip line with same width on the same material substrate with height h [26]. It means if we join two microstrip PCBs back-to-back without ground planes as shown in Figure 2.9, we get a DSPSL with the characteristic impedance double of microstrip line. If we place a conductor sheet of infinite size between these two PCBs it will not disturb the field distribution for the transmission lines but it will convert the DSPSL to two identical microstrip lines with half of the impedance that of DSPSL [26]. This technique can be used to convert the differential DSPSL structure to single-ended structure, especially in the case of measuring the differential structure with single-ended laboratory equipment.. 18.

(31) 3 SIX-PORT CORRELATOR The six-port correlator was first used as laboratory instrument for measurement of reflection coefficients and S-parameters of components from 1972-1994 [27] [28] [29]. It was first demonstrated in 1994 by Ji Li, R.G.Bosisio and Ke Wu that six-port correlator can be used in radio receiver operated at millimeter-wave frequencies [30]. They used six-port correlator to demodulate the digitally modulated signal at microwave/millimeter-wave frequencies. Since then intensive research work is going on designing radio systems employing six-port modulator/demodulator. Different types of modulation schemes such as 16-QAM, 64-QAM and QPSK have been successfully implemented. The six-port promises wide bandwidth so its use in UWB applications has been the focus of the research [31].. 3.1 Ideal six-port circuit Six-port correlator, also known as six-port junction or network is made up of passive microwave components such as Wilkinson power divider and Quadrature 90o branch line couplers joined together through transmission lines. Six-port correlator is the fundamental component of the direct-carrier six-port modulator/demodulator circuit which is an alternative radio transceiver architecture approach for broadband wireless communication systems. Based on different combinations and arrangements of WPD (Wilkinson power divider) and BLCs (branch line couplers), various configurations of six-port correlator are available. One of the most commonly used configurations consists of one WPD and three BLCs and is shown in Figure 3.1 [32]. This configuration has been used in this thesis as well. λ/4. Wilkinson Power Divider. Port 3. λ/4. Port 4. 50 Ω. Port 1 Port 2. Quadrature Branch Line Coupler. Port 5 Port 6. Figure 3.1 Six-port Correlator--made up of one WPD and three Quadrature BLCs [33]. 19.

(32) The individual components i.e. WPD and BLC will now be discussed in detail.. 3.1.1 Wilkinson Power Divider Wilkinson power divider is a three-port network made up of transmission lines and is used to divide the power of a signal equally in two with introduction of 90o phase shift in both. The Sparameters matrix of WPD is given below:( )[ √. ]. (3.1). An ideal WPD would exactly divide the input power provided at port 1 into two equal power outputs at ports 2 and 3. WPD with transmission line lengths is shown in Figure 3.2. λ/4. ZO. Port 2. ZO. Port 1. 2ZO. Port 3 ZO. λ/4. Figure 3.2. Wilkinson Power Divider with corresponding lengths and impedances of transmission lines. If Vn is the incident voltage wave at port n, and Vn is the reflected wave from same port, then ideal WPD has the following characteristics:-. if V1 = A cos(t) V1 = 0 (ideal matching, no reflections) then V2- = V3- =. √. 20. (cos(t) – 90o).

(33) 3.1.2 Quadrature Branch line coupler The Quadrature BLC is a four-port network made up of transmission lines and divides the input power at port 1 into two equal but mutually 90o shifted output powers at ports 2 and 3. Port 4 is isolated from port 1. The S-parameters matrix is given below:-. [. √. ]. (3.2). Quadrature BLC is shown in Figure 3.3. λ/4. Port 1. ZO. ZO. λ/4. Port 4 Figure 3.3. ZO. Port 2. ZO. ZO. ZO. Port 3. Quadrature Branch Line Coupler with corresponding lengths and impedances of transmission lines. In terms of incident and reflected voltage signals, we can write the characteristics as:if V1 = A cos(t) is the input voltage at port 1 and port 4 is terminated with Zo i.e. V4 = 0 V1 = 0, (ideal matching, no reflections) then V2- =. (cos(t) – 90o). √. V3- = -. (cos(t) ). √. ±180o phase shift. Port 4 is usually terminated with Zo, characteristic impedance of the transmission line. But as can be seen in the S-parameter matrix, if a signal is applied to this port, it is divided in two and experiences a phase shift of -90o at port 2 and -180o at port 3. So in terms of individual voltage components at ports 2 and 3, resulting from the voltage signals at ports 1 and 4, we have V2 = V2- = V3 = V3- =. √. √. 21. V1 V1 -. √. √. V4 V4.

(34) 3.1.3 180o Branch Line Coupler 180o Branch Line Coupler is also called Ring Coupler or Rat-race coupler. It has a different orientation of transmission lines and their impedances than that of Quadrature BLC. It consists of ring of transmission line with the total length of six quarter wave-lengths and characteristic impedance of √ Zo. Four ports are connected with the ring in such a way that three mutual distances are equal i.e. one quarter wave-length while one mutual distance is three quarter wavelengths. Ring coupler is shown in Figure 3.4. Port2 ZO Port1. /4. 3 /4. Port3. Port4. Figure 3.4 Ring Coupler with corresponding lengths and impedances of transmission lines. Based on the selection of input and output ports, it can give two equal outputs of either the same phase or 180o apart. The S-parameters matrix is given below:-. √. [. ]. (3.3). If input is given at port 1, two equal amplitude signals with same phase are at ports 2 and 3, while port 4 is isolated. If 2 is the input port, we get two equal amplitude but 180o phase apart signals at ports 1 and 4 while port 3 is isolated. Both configurations are shown in Figure 3.5. o. Output 0. Input. Input. Output 0o. Output 0o. Isolated Output 180o. Isolated. Figure 3.5 Ring coupler with different setting of input port and the corresponding outputs.. 22.

(35) 3.2 Modulation using Six-port Correlator For signal modulation using the Six-port correlator (depicted in Figure 3.1) first the S-parameters of six-port are analyzed. Using WPD and BLC S-parameters matrices, S-parameters of six-port correlator can be calculated as [6]:. (3.4) [. ]. So S-parameters matrix gives the following general relationships:(3.5a). (3.5b). (3.5c). (3.5d). (3.5e). (3.5f) The six-port modulator circuit with respective incident and reflected power waves and variable impedance terminations at the ports is shown in Figure 3.6.. 23.

(36) i = 3,4,5,6. λ/4 λ/4. 3. Γ3. Z i1 Z i2 Z i1 Z i2. Γ4 4. a1 LO in. b1. 50 Ω. 1. 2. b2 5. Γ5 Γ6. 6. RF out a2. Z i1 Z i2 Z i1 Z i2. Figure 3.6 Six port Correlator being used a modulator with variable impedance terminations. ai and bi are input and output waves respectively from port i.. The six-port shown above is a common configuration of the Six-port correlator used as a modulator. Port 1 is used as input port for LO (Local Oscillator) signal and ports 3 to 6 are terminated with variable impedances controlled by the baseband signals at these ports. Port 2 is the output port for RF modulated signal. In Figure 3.6 Assume that ports 1 and 2 are perfectly matched i.e. there is no reflection; port 1 is the input port for LO signal and port 2 is the output RF signal port. Assuming perfect matching we get, b1 = 0, a2 = 0. Applying the LO signal at port 1, ( √. ). , as. √. (3.6). and for the rest of ports, (3.7) where ai is the incident power wave and bi is the reflected wave at ports i where i = 3,4,5 and 6. Then applying equations 3.5a - 3.5e, we obtain the output RF signal at port 2 (3.8). 24.

(37) In equation 3.8, the reflection coefficients Г3, Г4, Г5 and Г6 are realized by applying the baseband signals to the variable impedances at these ports. These variable impedances can be any type of switch which can take two or more values based on the state of the baseband signal. For higher order modulation such as 16-QAM or 64-QAM more values or states of the impedances are required such as 8 for 64-QAM [31]. For lower order modulation such as QPSK, only 2 states (either short or open) for the impedance termination are enough [34]. In equation (3.8) Г3, Г4 represent the in-phase component (ГI) and Г5, Г6 represent the quadrature component (ГQ). If for a particular case such as in QPSK we assume Г3 = Г4 and Г5 = Г6 then equation 3.8 can be written in voltage signal form as [. ]. (3.9). In time domain, equation 3.9 can be written as [6]:(. √. (. )). (3.10). Equation (3.10) describes an RF modulated signal. The amplitude and phase of this signal is dependent on the variations of the reflection coefficients of the in-phase and quadrature phase components. Equation 3.10 can also be written in simplified form as.  Where A(t) is the modulated amplitude and. (3.11). (t) is the modulated phase of the RF signal.. 3.3 Demodulation using Six-port Correlator The demodulation of the RF signal is achieved by mixing the received RF signal with the LO signal which is equal in frequency to the LO signal at the transmitter. Schottky diodes are used for the mixing purpose. The configuration is shown in Figure 3.7 [35].. 25.

(38) ai =Γibi. bi. i = 3,4,5,6. (o)2 3. LPF ̶. LPF. +. I. (o) 2 4. a1 LO in. b1 1. 50 Ω. Six-port Correlator 2. RF in a2. b2 (o) 2 5. LPF ̶. LPF. +. Q. (o) 2 6. Figure 3.7 Six-port Demodulator with Diodes, LPFs and Differential Amplifiers. The squared terms after the diode stage are then low pass filtered to reject the higher frequency components and these low frequency signals are then fed into differential amplifier. The LO signal is fed into port 1 and RF signal is fed at port 2. We assume perfect matching at these ports so that (3.12a).  . where. √. and. (3.12b) (3.12c). (3.13a). (3.13b). (3.14a) (3.14b). 26.

(39) Now according to the S-parameters matrix in (3.4), we deduce the following signals at the ports 3, 4, 5 and 6 [. . (. )]. (3.15a). [. . (. )]. (3.15b). (. )]. (3.15c). (. )]. (3.15d). [. [. . . After squaring of the these signals by the diodes, they are passed to LPF (Low Pass Filter) to reject the higher order harmonics such as 2LO(t), 2c(t) and LO(t)+ c(t). We have for V3 [6]:. . . (3.16a). . . (3.16b). . . (3.16c). . . (3.16d). Similarly at other ports,. Now to be able to recover the baseband signal from the resulted signals above, the first two unwanted component terms are to be eliminated from each signal. In order to achieve this, differential amplifier is used. Secondly the LO signal at the receiver end should be exactly equal to the LO signal at the transmitter i.e. the carrier signal such that. ct - LOt = 0 Differential amplifiers give the output baseband signals i.e. 27.

(40) {. }. {. }. (. ). (3.17a). {. }. {. }. (. ). (3.17b). In this way both the in-phase and quadrature components of the baseband signal can be recovered.. 28.

(41) 4 Design and implementation of six-port correlator The task is to design single-ended and differential six-port correlator for UWB with centre frequency of 7.5 GHz with minimum amplitude and phase imbalance for maximum bandwidth. Various structures of six-port correlators are designed, simulated and fabricated. First, the components of six-port are designed as standalone structures. These designs are simulated to evaluate the performance, then optimized and integrated to make the six-port correlator. The sixport structures are then optimized and fabricated. Advance Design System (ADS) from Agilent Inc. is used for circuit simulations and evaluations. Rogers 4350B substrate is used for PCB fabrication of the designs. Substrate specifications are mentioned in Table 4.1. Single-ended designs are fabricated for 50 Ω and differential designs for 100 Ω port impedances. 50 Ω SMA female connectors for ports are used in PCB fabrication. To measure the differential designs with available single-ended vector network analyzer (VNA) a single-ended to differential conversion mechanism is used and a wideband 180O coupler is designed and fabricated to be used as the converter. The design and evaluation process is divided into three parts. First, the six-port designs are simulated in ADS on schematic level to have a look on the ideal results. Second, the schematic structure is converted to layout structure and simulated using ADS momentum field-solver. In third step the layout design is modified to be suitable for fabrication and then fabricated for practical evaluation. Table 4.1 Rogers4350B substrate specifications. Relative dielectric constant. 3.66. Substrate thickness. 254 µm. Conductor thickness. 35 µm. Metal conductivity. 5.8×107 S/m. Loss tangent. 0.004. Surface roughness. 0.001 mm. 29.

(42) 4.1 Single-ended designs Initially single-ended designs are simulated, fabricated and analyzed for the performance. The designs are optimized for 50 Ω port impedances and for best performance on 7.5 GHz. Two type of single-ended structures are designed: i) classical design ii) design with matching stubs.. 4.1.1 Classical single-ended design This six-port design has been frequently used in the research [36][37]. It contains a Wilkinson divider with three quadrature couplers. This is a simple and compact design. The layout of the design is shown in Figure 4.1. The design dimensions are 20 mm × 25 mm. P3. P4. 25 mm. P7 P1 (LO). P2 (RF). P6. P5 20 mm. Figure 4.1 Layout design of single-ended classical six-port correlator. Port 1 is referred as LO (local oscillator) and port 2 as RF (radio frequency) port. Port 3 and 4 are used as in-phase and port 5 and 6 are used as quadrature-phase ports. Port 7 is not used and terminated by 50 Ω termination. From fabrication point this design could not be fabrication with available equipment because of smaller size. To make the design feasible for fabrication, extra transmission-line lengths are added to the interconnects, increasing the design size to 31 mm × 34 mm. The electrical length of added transmission-line on each port is λ/4 (6 mm). The modified design layout is shown in Figure 4.2 and fabricated design in Figure 4.3.. 30.

(43) P3. P4. 34 mm. P7. P1 (LO). P2 (RF). P6. P5. 31 mm. Figure 4.2 Layout design of single-ended classical six-port correlator modified for fabrication. The design dimensions are 31 mm × 34 mm. P4. P3. 50 Ω. Termination. P1 P2 P6. P5. Figure 4.3 Manufactured Single-ended classical six-port correlator prototype. The simulated and measured S-parameter results for this design are shown in Figure 4.4 to 4.11.. 31.

(44) 0. Input Reflection (dB). -10. -20. -30 dB(S(1,1)) dB(S(1,1)) dB(S(1,1)) dB(S(1,1)) dB(S(2,2)) -40 dB(S(2,2)) dB(S(2,2)) dB(S(2,2)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) -50 dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)) -60. Measured S11 Measured S22 Simulated S11 Simulated S22. 6.0. 7.0. 7.5. 8.0. 8.5. 9.0. Frequency (GHz). 0. -20 -20 -20 -20 -30 -30 -30 -30 -40 -40 -40 -40 -50 -50 -50 -50 -60 -60 -60 6.0 -60 6.0 6.0 6.0. 6.5. Figure 4.4 Measured and simulated input reflection coefficients. -10 -10 -10 -10. Isolation between P1 and P2 (dB). (dB) (dB) (dB) (dB) Reflection Reflection Reflection Reflection Input Input Input Input. 0 0 0 0. 6.5 6.5 6.5 6.5. -10. -20 -30 -40. 7.0 7.5 7.0 7.5 dB(S(1,1)) 7.0 7.5 dB(S(1,1)) -50 7.0 Frequency 7.5 (dB). 8.0 8.0 8.0 8.0. Frequency (dB) dB(S(2,2)) Frequency dB(S(2,2)) Frequency (dB) (dB) -60 dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1))6.0. 8.5 8.5 8.5 8.5. 9.0. 9.0 Simulated 9.0 9.0. Measured 6.5. dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)). 7.0. 7.5. 8.0. 8.5. 9.0. Frequency (GHz). Figure 4.5 Measured and simulated isolations between Port 1 and Port 2 Forward Transmission from Port 1 (dB). 0 0. -20 -20 -30 -30 -40 -40 -50 -50 -60 -60 6.0 6.0. 6.5 6.5. 7.0 7.0. 0. -3. -6. -9. -12. dB(S(1,1)) dB(S(1,1)) dB(S(1,1)) dB(S(2,2)) dB(S(2,2)) dB(S(2,2)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)) dB(S(1,1)). 7.5 -15 8.0 7.5 8.0 0 6.0 0 Frequency (dB) 0 Frequency (dB). 8.5 8.5. 9.0 9.0. dB(S(2,2)) 6.5 7.0. 7.5. Frequency (GHz) dB(_2_SixportSE..S(1,1)). Measured S31 Measured S41 Simulated S31 Simulated S41 8.0. 8.5. 9.0. -10Measured and simulated S-parameters for transmission from Port 1 to Port 3 and 4 Figure 4.6-10 -10. (dB) (dB) (dB) (dB) n Input Reflection Reflection Reflection Input Input. (dB) Reflection(dB) Input InputReflection. -10 -10. dB(_2_SixportSE..S(2,2)). -20 -20 -20. 0 -30 -30 -30 -10 -40 -40 -40 -20 -50 -50. 32.

(45) Forward Transmission from Port 1 (dB). 0. -3. -6. dB(S(1,1)) dB(S(1,1)) dB(S(1,1)) dB(S(1,1)) dB(S(2,2)) dB(S(2,2)) dB(S(2,2)) dB(S(2,2)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)). -9. -12. -15. 6.0. 6.5. 7.0. 0 0 0 0. Measured S61 Measured S51 Simulated S61 Simulated S51. 7.5. 8.0. 8.5. 9.0. Frequency (GHz). -10 -10 -20 -20 -20 -20 -30 -30 -30 -30. Forward Transmission from Port 2 (dB). (dB) (dB) (dB) (dB) Reflection Reflection Reflection Input Reflection Input Input Input. (dB) (dB) (dB) (dB) Reflection Reflection Reflection Reflection Input Input Input Input. -10 Measured and simulated S-parameters for transmission from Port 1 to Port 5 and 6 Figure 4.7 -10. -40 -40 -40 -40 -50 -50 -50 -50. 0 -60 0 -60 0 -60 -60 6.0 6.0 -10 6.0 -10 6.0 -10 -20 -20 -20. 0 -30 -30 -30. 0. -3. -6. dB(S(1,1)) dB(S(1,1)) dB(S(1,1)) dB(S(2,2)) dB(S(2,2)) dB(S(2,2)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)) dB(S(1,1)). Measured S32 Measured S42 Simulated S32 Simulated S42. dB(S(2,2)) -9. -12. 6.5 7.0 7.5 6.5 7.0 7.5 dB(_2_SixportSE..S(1,1)) 6.5 7.0 7.5 6.5 7.0 Frequency 7.5 (dB). Frequency dB(_2_SixportSE..S(2,2)) Frequency (dB) (dB). 8.0 8.0 8.0 8.0. 8.5 8.5 8.5 8.5. 9.0 9.0 9.0 9.0. Frequency (dB). -15 6.0. 6.5. 7.0. -10 -40 -40 -40. 7.5. 8.0. 8.5. 9.0. Frequency (GHz). Figure 4.8 Measured and simulated S-parameters for transmission from Port 2 to Port 3 and 4 -20 -50 -50 -50. -50. -60. Forward Transmission from Port 2 (dB). -30 -60 -60 -60 6.0 6.0 -40 6.0. (dB) (dB) (dB) (dB) Reflection Reflection Reflection Input Reflection Input Input Input. 0 0 0 -10 -10 -10 -20 -20 -20. 6.0. 0 -30 -30 -30 -10 -40 -40 -40. 0. -3. -6. dB(S(1,1)) dB(S(1,1)) dB(S(1,1)) 7.0 7.5 7.0 7.5 dB(S(2,2)) 7.0 7.5 dB(S(2,2)) dB(S(2,2)) Frequency (dB) Frequency dB(_2_SixportSE..S(1,1)) Frequency (dB) (dB) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)) dB(S(1,1)) 6.5 6.5 6.5. 6.5. 7.0. 7.5. 8.0 8.0 8.0. 8.5 8.5 8.5. Measured S52 Simulated S62. Simulated S52 8.0. dB(S(2,2)). 9.0 9.0. 9.0 Measured S62. 8.5. 9.0. Frequency (dB). -9. dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(2,2)). -12. -15 6.0. 6.5. 7.0. 7.5. 8.0. 8.5. 9.0. Frequency (GHz). Figure 4.9 -20 Measured and simulated S-parameters for transmission from Port 2 to Port 5 and 6 -50 -50 -50. -30 -60 -60 -60 6.0 6.0 -40 6.0. -50. 6.5 6.5 6.5. 7.0 7.0 7.0. 7.5 7.5 7.5. 33. Frequency (dB) Frequency Frequency (dB) (dB). 8.0 8.0 8.0. 8.5 8.5 8.5. 9.0 9.0 9.0.

(46) Simulated Phase Difference (Degree). dB(S(1,1)) dB(S(1,1)) 300 dB(S(2,2)) dB(S(2,2)) dB(S(1,1)) dB(_2_SixportSE..S(1,1)) dB(S(1,1)) dB(_2_SixportSE..S(1,1)) dB(S(2,2)) 200 dB(_2_SixportSE..S(2,2)) dB(S(2,2)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) 100 dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)). 0. -10 -100 0 -20 -10 -20 -10 -30 -20 -30 -20 -40 -30 -40 -30 -50 -40 -50 -40 -60 -50 -60 6.0 -50 6.0 -60 -60 6.0. 6.0. 0. S61 – S51. 0. -100 6.5. 7.0. -20 -100. 7.5. 8.0. 8.5. 9.0. Frequency (GHz). Figure 4.10 Simulated phase difference between Port 4 and 3, and between Port 5 and 6, when Port 1 or Port 2 is used as input port 6.5. 7.0. 7.5. 8.0. 8.5. 9.0. 6.5. 7.0 dB(S(1,1)). 7.5. 8.0. 8.5. 9.0. 8.0 8.0. 8.5 S31 – S419.0 8.5 S – S 9.0. Frequency 400 (dB) Frequency (dB). 7.0 7.5 dB(S(2,2)) 300 7.0 7.5 dB(S(1,1)) Frequency (dB) dB(_2_SixportSE..S(1,1)) Frequency 200 (dB) dB(S(2,2)) 100 dB(_2_SixportSE..S(2,2)) dB(S(1,1)) dB(_2_SixportSE..S(1,1))0 dB(S(2,2)) dB(_2_SixportSE..S(2,2)) -100 dB(S(1,1)) dB(_2_SixportSE..S(1,1)) 6.5 6.5. 61. 51. -200. dB(S(2,2)) dB(_2_SixportSE..S(2,2)) -300 dB(_2_SixportSE..S(1,1)). -10 0. (dB) Reflection (dB) Reflection (dB) (dB) Input Reflection Input Reflection Input Input. S61 – S51 S32 – S42. 6.0. Measured Phase Difference (Degree). (dB) (dB) (dB) (dB) Reflection Reflection Reflection Reflection Input Input Input Input. 0. S31 – S41. S32 – S42 S62 – S52. -400. dB(_2_SixportSE..S(2,2)) 6.0. 6.5. 7.0. 7.5. 8.0. 8.5. 9.0. Frequency (GHz). -30 -20 -10 0 -40 -30 -20 -10 -50 -40 -30 -20 -60 -50 -40 6.0 -30 -60 -50 -40 6.0 -60 -50 6.0. Figure 4.11. Measured phase difference between Port 4 and 3, and between Port 5 and 6, when Port 1 or Port 2 is used as input port. Figure 4.4 shows the input reflection coefficients for Port 1 and 2. The simulated results shows 6.5 input reflection 7.0 7.5 9.0 frequency but measured results are not in good the less than8.0-20 dB 8.5 at the centre agreement with the simulated results. Measured results have input reflections coefficients higher Frequency (dB) than -10 dB for the centre frequency. It can also be noticed in Figure 4.4 the isolation between 6.5 7.0 7.5 8.0 8.5 9.0 Port 1 and Port 2 is in simulated results is -50 dB for the centre frequency and for the extreme Frequency (dB) sides of selected band it goes to approximately -9dB but again the measured results are not that 6.5 7.0 7.5 8.0 8.5 9.0 good as simulated. The best isolation is measured to be -22 dB on 8.5 GHz. Frequency (dB). -60 6.0. 6.5. 7.0. 7.5. Frequency (dB). 8.0. 8.5. 9.0. 34.

(47) The forward transmission response is depicted in Figure 4.5 and 4.6. Simulated results shows the values around -6 dB for transmission to all ports. The measured results are deviated from the simulated results specially the transmission to Port 4 and 5 which have the transmission loss of approximately -9 dB at centre frequency . Transmission to Port 3 and 6 have relatively better measured response and are close to simulated values. A same behavior is visible in Figure 4.7 and 4.8 for transmission to same ports but now Port 2 as input port. The simulated results hover around -6 dB for centre frequency (7.5 GHz) and the measured results have more loss and float around -9 dB at centre frequency. Figure 4.9 and 4.10 shows the phase response of simulated and measured values respectively. The phase difference between Port 3 and 4, and Port 6 and 5 for transmission when Port 1 or 2 is used as input port is close to 90O for both simulated and measured results at centre frequency. The phase difference deviate smoothly as frequency is changed from the centre frequency. The measured results have deviations from the simulated because of bad etching, substrate errors, SMA connectors and soldering. The main problem encountered in this project is uneven etching of copper on PCB changing the width of transmission lines. The transmission line with different widths exhibits different impedance causing high input reflection coefficients as in this case.. 4.1.2 Single-ended design with matching stubs The classical design presented in previous topic has compact size but exhibits a narrow band response. The Wilkinson divider has a relatively uniform response on a larger bandwidth than the quadrature couplers. To make the quadrature couplers wideband, matching networks with open-circuited stubs are applied on the ports. The idea is to exploit the fact that coupling depends upon the admittance of the ports [38]. The matching network is presented in Figure 4.11.. λ/2 Zstub. Zline. λ/2. Zo. Figure 4.11 Matching network applied on the coupler ports to broadband the response. The optimized values for Zstub and Zline are 50 Ω and 80 Ω respectively. The open-circuited stub is folded inward to save the space on PCB.. 35.

(48) To increase the bandwidth of the six-port correlator, the simple quadrature couplers are replaced with these optimized quadrature couplers with matching networks. The layout design is shown in Figure 4.12. The design dimensions are 52 mm × 47 mm. The port assignment is same as specified for the previous design.. P3. 47 mm. P4. P7. P1 (LO). P2 (RF). P5 P6 52 mm Figure 4.12 Layout design for single-ended six-port correlator with matching networks. The simulated results for this six-port correlator design are shown in Figure 4.12 to 4.15. Input Reflection (dB). 0. dB(S(1,1)) dB(S(1,1)) dB(S(2,2)) dB(S(2,2)). -10. -20. -30. Simulated S11 Simulated S22 -40. dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) 6.0 dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(2,2)). 6.3. 6.6. 6.9. 7.2. 7.5. 7.8. 8.1. 8.4. 8.7. 9.0. Frequency (GHz). Figure 4.12 Simulated input reflection coefficients on Port 1 and 2 0 0. (dB) Reflection(dB) put Reflection Input. -10 -10 -20 -20 -30 -30 -40 -40. 36.

(49) Forward transmission from Port 1 (dB). -3. dB(S(1,1)) dB(S(1,1)) -9 dB(S(2,2)) dB(S(2,2)) dB(S(1,1)) dB(_2_SixportSE..S(1,1)) dB(S(1,1)) dB(_2_SixportSE..S(1,1)) -12 dB(S(2,2)) dB(_2_SixportSE..S(2,2)) dB(S(2,2)) dB(_2_SixportSE..S(2,2)) dB(_2_SixportSE..S(1,1)) dB(_2_SixportSE..S(1,1)) -15 dB(_2_SixportSE..S(2,2)) 6.0 6.5 dB(_2_SixportSE..S(2,2)). 6.0 6.0. (dB) (dB) (dB) (dB) Reflection Reflection Reflection Reflection Input Input Input Input. 0 0 -10 -100 0 -20 -20 -10 -10 -30 -30 -20 -20 -40 -40 -30 -30 -50 -50 -40 -40 -60 -60 6.0 -50 -50 6.0. -60 -60. 6.0 6.0. Simulated S41 Simulated S51 Simulated S61 7.0. 7.5. 8.0. 8.5. 9.0. Figure 4.13 Simulated S-parameters for transmission from Port 1 to Port 3, 4, 5 and 6. -3. Forward transmission from Port 2 (dB). (dB) (dB) (dB) (dB) Reflection Reflection Reflection Reflection Input Input Input Input. -10 -100 0 -20 -20 -10 -10 -30 -30 -20 -20 -40 -40 -30 -30 -50 -50 -40 -40 -60 -60 6.0 -50 -50 6.0. Simulated S31. Frequency (GHz). -6. dB(S(1,1)) dB(S(1,1)) dB(S(2,2)) -9 dB(S(2,2)) dB(S(1,1)) 6.5 7.0 7.5 dB(_2_SixportSE..S(1,1)) dB(S(1,1)) 6.5 7.0 7.5 dB(_2_SixportSE..S(1,1)) Frequency (dB) dB(S(2,2)) -12 dB(_2_SixportSE..S(2,2)) dB(S(2,2)) Frequency (dB) dB(_2_SixportSE..S(2,2)) 6.5 7.0 7.5 dB(_2_SixportSE..S(1,1)) 6.5 7.0 7.5 dB(_2_SixportSE..S(1,1)) Frequency (dB) -15 dB(_2_SixportSE..S(2,2)) Frequency (dB) 6.0 6.5 dB(_2_SixportSE..S(2,2)). Simulated S32 8.0 8.0. 8.5 9.0 Simulated S42 8.5 9.0. Simulated S52 S62 8.5Simulated 9.0 8.5 9.0. 8.0 8.0 7.0. 7.5. 8.0. 8.5. 9.0. Frequency (GHz). Figure 4.14 Simulated S-parameters for transmission from Port 2 to Port 3, 4, 5 and 6. 6.5 6.5. Simulated Phase Difference (Degree). 0 0. -60 -60. -6. 400 300 200 100. dB(S(1,1)) 0. 7.0 -100 7.0. 7.5 7.5. dB(S(2,2)). 8.0 8.5 8.0dB(S(1,1)) 8.5. 9.0. 9.0 dB(_2_SixportSE..S(1,1)) dB(S(1,1)) Frequency (dB) -200 Frequency (dB) dB(S(2,2))S31 – S41 dB(_2_SixportSE..S(2,2)) dB(S(2,2)) 6.5 7.0 -300 7.5 8.0 8.5 9.0 6.5 7.0 7.5 dB(_2_SixportSE..S(1,1)) 8.0 8.5 S61 – S51 9.0 dB(S(1,1)) dB(_2_SixportSE..S(1,1)) Frequency (dB) -400 Frequency (dB) 0 6.0 dB(S(2,2)) dB(_2_SixportSE..S(2,2)) 6.3 6.6 6.9 7.2 7.5 7.8 dB(_2_SixportSE..S(2,2)) Frequency (GHz) dB(_2_SixportSE..S(1,1)). S32 – S42 S62 – S52 8.1. 8.4. 8.7. 9.0. -10. (dB) (dB) ction ut Reflection. -10 0 -20 -10 -30 -20 -40. 0 Simulated phase difference between Port 4 and 3, and between Port 5 and 6, when Port 1 or Port 2 is used as dB(_2_SixportSE..S(2,2)) input port -20. (dB) (dB) put Reflection Input Reflection. Figure 4.16 0. -10 -30 -20 -40 -30 -50 -40. 37.

No results found