<p>The thesis is largely built on a collection of published and submitted papers where the main focus is to analyze and optimize single-carrier adaptive coded modulation systems with and without antenna diversity. Multidimensional trellis codes are used as component codes. The majority of the analysis is done with both estimation and prediction errors being incorporated. Both channel estimation and prediction are performed using a pilot-symbol-assisted modulation scheme. Thus, known pilot symbols (overhead information) must be transmitted; which consumes power and also degrades system spectral efficiency. Both power consumption and pilot insertion frequency are optimized such that they are kept at necessary values to maximize system throughput without sacrificing the error rate performance. The results show that efficient and reliable system performance can be achieved over a wide range of the considered average channel quality. Going from a single-input single-output system to both spatially uncorrelated and correlated single-input multiple-ouput (SIMO) systems, and further to an uncorrelated multiple-input multiple-output (MIMO) diversity system, is the evolution of the thesis. In the SIMO case, maximum ratio combining is used to combine the incoming signals, whereas the signals are space-time combined in the MIMO diversity system. The multiple-input single-output system comes out as a special case of a MIMO system. Besides the spatially uncorrelated antenna array, the effect of spatial correlation is also considered in the SIMO case. In this case, only prediction error is considered and channel estimation is assumed to be perfect. At first, the impact of spatial correlation in a predicted system originally designed to operate on uncorrelated channels is quanitifed. Then, a maximum a posteriori (MAP)-optimal “space-time predictor” is derived to take spatial correlation into account. As expected, the results show that the throughput is still lower than the uncorrelated system, but the degradation is decreased when the MAP-optimal space-time predictor is used. Thus, by exploiting the correlation properly, the degradation can be reduced. By numerical examples, we demonstrate the potential effect of limiting the predictor complexity, of fixing the pilot spacing, as well as of assuming perfect estimation. The two first simplifications imply lower system complexity and feedback rate, whereas the last assumption is usually made to ease the mathematical analysis. The numerical examples indicate that all the simplifications can be done without serious impact on the predicted system performance.</p>

In the nineties, two major events brought about a revolution in the field of digital communications: the invention of turbo codes and the development of multiple-input multiple-output (MIMO) wireless systems. The turbo codes are error-correcting codes which enable communication systems to operate close to the channel capacity with a reasonable complexity. Multiple-antenna transceivers, a.k.a. MIMO wireless systems, mitigate the effect of multipath fading that occurs in most terrestrial wireless communications. MIMO transmissions are more reliable than single-input single-output transmissions. Moreover, the data rate achievable by MIMO systems is also much higher. This thesis combines these two recent breakthroughs in digital communications. Iterative decoding is a key factor in the success of turbo codes. This principle has proved to be so powerful that it has soon been applied to other schemes such as iterative channel equalization, also known as turbo equalization. In the first part of this thesis, we derived a low-complexity iterative equalizer for frequency-selective MIMO channels. Its robustness against channel estimation errors was also addressed. The second part of this thesis is devoted to the estimation of the channel impulse response and the noise variance for coded transmissions over MIMO channels. We proposed several new iterative estimators based on the expectation-maximization algorithm, the expectation-conditionalmaximization algorithm and the minimum mean square error criterion. Finally, we derived lower bounds on the mean square error of channel estimators. In particular, the Cramer-Rao bound for the estimation of frequency-selective MIMO channels were computed. We mainly investigated the case where no training sequence are transmitted and the case where the receiver has a partial knowledge of the transmitted sequence.

High data rate applications are the trend in today's wireless technology. W-CDMA standard was designed to support such high data rates of up to 3.84 Mcps. The main purpose of this research was to analyze the feasibility of a fixed-point implementation of the W-CDMA downlink receiver algorithm on a general-purpose digital signal processor (StarCore SC140 by Motorola). The very large instruction word architecture of SC140 core is utilized to generate optimal implementation, to meet the real time timing requirements of the algorithm. The other main aim of this work was to study and evaluate the performance of the W-CDMA downlink structure with incorporated space-time transmit diversity. The effect of the channel estimation algorithm used was extensively studied too.

The thesis is largely built on a collection of published and submitted papers where the main focus is to analyze and optimize single-carrier adaptive coded modulation systems with and without antenna diversity. Multidimensional trellis codes are used as component codes. The majority of the analysis is done with both estimation and prediction errors being incorporated. Both channel estimation and prediction are performed using a pilot-symbol-assisted modulation scheme. Thus, known pilot symbols (overhead information) must be transmitted; which consumes power and also degrades system spectral efficiency. Both power consumption and pilot insertion frequency are optimized such that they are kept at necessary values to maximize system throughput without sacrificing the error rate performance. The results show that efficient and reliable system performance can be achieved over a wide range of the considered average channel quality. Going from a single-input single-output system to both spatially uncorrelated and correlated single-input multiple-ouput (SIMO) systems, and further to an uncorrelated multiple-input multiple-output (MIMO) diversity system, is the evolution of the thesis. In the SIMO case, maximum ratio combining is used to combine the incoming signals, whereas the signals are space-time combined in the MIMO diversity system. The multiple-input single-output system comes out as a special case of a MIMO system. Besides the spatially uncorrelated antenna array, the effect of spatial correlation is also considered in the SIMO case. In this case, only prediction error is considered and channel estimation is assumed to be perfect. At first, the impact of spatial correlation in a predicted system originally designed to operate on uncorrelated channels is quanitifed. Then, a maximum a posteriori (MAP)-optimal “space-time predictor” is derived to take spatial correlation into account. As expected, the results show that the throughput is still lower than the uncorrelated system, but the degradation is decreased when the MAP-optimal space-time predictor is used. Thus, by exploiting the correlation properly, the degradation can be reduced. By numerical examples, we demonstrate the potential effect of limiting the predictor complexity, of fixing the pilot spacing, as well as of assuming perfect estimation. The two first simplifications imply lower system complexity and feedback rate, whereas the last assumption is usually made to ease the mathematical analysis. The numerical examples indicate that all the simplifications can be done without serious impact on the predicted system performance.

Orthogonal Frequency Division Multiplexing (OFDM) is a multicarrier modulation technique which is robust against multipath fading and very easy to implement in transmitters and receivers using the inverse fast Fourier transform and the fast Fourier transform. A guard interval using cyclic prefix is inserted in each OFDM symbol to avoid the inter-symbol interference. This guard interval should be at least equal to, or longer than the maximum delay spread of the channel to combat against inter-symbol interference properly.<p> In coherent detection, channel estimation is required for the data detection of OFDM systems to equalize the channel effects. One of the popular techniques is to insert pilot tones (reference signals) in OFDM symbols. In conventional method, pilot tones are inserted into every OFDM symbols. Channel capacity is wasted due to the transmission of a large number of pilot tones. To overcome this transmission loss, incoherent data detection is introduced in OFDM systems, where it is not needed to estimate the channel at first. We use differential modulation based incoherent detection in this thesis for the data detection of OFDM systems. Data can be encoded in the relative phase of consecutive OFDM symbols (inter-frame modulation) or in the relative phase of an OFDM symbol in adjacent subcarriers (in-frame modulation). We use higher order differential modulation for in-frame modulation to compare the improvement of bit error rate. It should be noted that the single differential modulation scheme uses only one pilot tone, whereas the double differential uses two pilot tones and so on. Thus overhead due to the extra pilot tones in conventional methods are minimized and the detection delay is reduced. It has been observed that the single differential scheme works better in low SNRs (Signal to Noise Ratios) with low channel taps and the double differential works better at higher SNRs. Simulation results show that higher order differential modulation schemes don¡¯t have any further advantages. For inter-frame modulation, we use single differential modulation where only one OFDM symbol is used as a reference symbol. Except the reference symbol, no other overhead is required. We also perform channel estimation using differential modulation. Channel estimation using differential modulation is very easy and channel coefficients can be estimated very accurately without increasing any computational complexity. Our simulation results show that the mean square channel estimation error is about ¡¼10¡½^(-2) at an SNR of 30 dB for double differential in-frame modulation scheme, whereas channel estimation error is about ¡¼10¡½^(-4) for single differential inter-frame modulation. Incoherent data detection using classical DPSK (Differential Phase Shift Keying) causes an SNR loss of approximately 3 dB compared to coherent detection. But in our method, differential detection can estimate the channel coefficients very accurately and our estimated channel can be used in simple coherent detection to improve the system performance and minimize the SNR loss that happens in conventional method.

In recent years, the state-of-the-art in field programmable gate array (FPGA) technology has been advancing rapidly. Consequently, the use of FPGAs is being considered in many applications which have traditionally relied upon application-specific integrated circuits (ASICs). FPGA-based designs have a number of advantages over ASIC-based designs, including lower up-front engineering design costs, shorter time-to-market, and the ability to reconfigure devices in the field. However, ASICs have a major advantage in terms of computational resources. As a result, expensive high performance ASIC algorithms must be redesigned to fit the limited resources available in an FPGA. <p> Concurrently, coaxial cable television and internet networks have been undergoing significant upgrades that have largely been driven by a sharp increase in the use of interactive applications. This has intensified demand for the so-called upstream channels, which allow customers to transmit data into the network. The format and protocol of the upstream channels are defined by a set of standards, known as DOCSIS 3.0, which govern the flow of data through the network. <p> Critical to DOCSIS 3.0 compliance is the upstream demodulator, which is responsible for the physical layer reception from all customers. Although upstream demodulators have typically been implemented as ASICs, the design of an FPGA-based upstream demodulator is an intriguing possibility, as FPGA-based demodulators could potentially be upgraded in the field to support future DOCSIS standards. Furthermore, the lower non-recurring engineering costs associated with FPGA-based designs could provide an opportunity for smaller companies to compete in this market. <p> The upstream demodulator must contain complicated synchronization circuitry to detect, measure, and correct for channel distortions. Unfortunately, many of the synchronization algorithms described in the open literature are not suitable for either upstream cable channels or FPGA implementation. In this thesis, computationally inexpensive and robust synchronization algorithms are explored. In particular, algorithms for frequency recovery and equalization are developed. <p> The many data-aided feedforward frequency offset estimators analyzed in the literature have not considered intersymbol interference (ISI) caused by micro-reflections in the channel. It is shown in this thesis that many prominent frequency offset estimation algorithms become biased in the presence of ISI. A novel high-performance frequency offset estimator which is suitable for implementation in an FPGA is derived from first principles. Additionally, a rule is developed for predicting whether a frequency offset estimator will become biased in the presence of ISI. This rule is used to establish a channel excitation sequence which ensures the proposed frequency offset estimator is unbiased. <p> Adaptive equalizers that compensate for the ISI take a relatively long time to converge, necessitating a lengthy training sequence. The convergence time is reduced using a two step technique to seed the equalizer. First, the ISI equivalent model of the channel is estimated in response to a specific short excitation sequence. Then, the estimated channel response is inverted with a novel algorithm to initialize the equalizer. It is shown that the proposed technique, while inexpensive to implement in an FPGA, can decrease the length of the required equalizer training sequence by up to 70 symbols. <p> It is shown that a preamble segment consisting of repeated 11-symbol Barker sequences which is well-suited to timing recovery can also be used effectively for frequency recovery and channel estimation. By performing these three functions sequentially using a single set of preamble symbols, the overall length of the preamble may be further reduced.

In recent years, the state-of-the-art in field programmable gate array (FPGA) technology has been advancing rapidly. Consequently, the use of FPGAs is being considered in many applications which have traditionally relied upon application-specific integrated circuits (ASICs). FPGA-based designs have a number of advantages over ASIC-based designs, including lower up-front engineering design costs, shorter time-to-market, and the ability to reconfigure devices in the field. However, ASICs have a major advantage in terms of computational resources. As a result, expensive high performance ASIC algorithms must be redesigned to fit the limited resources available in an FPGA. <p> Concurrently, coaxial cable television and internet networks have been undergoing significant upgrades that have largely been driven by a sharp increase in the use of interactive applications. This has intensified demand for the so-called upstream channels, which allow customers to transmit data into the network. The format and protocol of the upstream channels are defined by a set of standards, known as DOCSIS 3.0, which govern the flow of data through the network. <p> Critical to DOCSIS 3.0 compliance is the upstream demodulator, which is responsible for the physical layer reception from all customers. Although upstream demodulators have typically been implemented as ASICs, the design of an FPGA-based upstream demodulator is an intriguing possibility, as FPGA-based demodulators could potentially be upgraded in the field to support future DOCSIS standards. Furthermore, the lower non-recurring engineering costs associated with FPGA-based designs could provide an opportunity for smaller companies to compete in this market. <p> The upstream demodulator must contain complicated synchronization circuitry to detect, measure, and correct for channel distortions. Unfortunately, many of the synchronization algorithms described in the open literature are not suitable for either upstream cable channels or FPGA implementation. In this thesis, computationally inexpensive and robust synchronization algorithms are explored. In particular, algorithms for frequency recovery and equalization are developed. <p> The many data-aided feedforward frequency offset estimators analyzed in the literature have not considered intersymbol interference (ISI) caused by micro-reflections in the channel. It is shown in this thesis that many prominent frequency offset estimation algorithms become biased in the presence of ISI. A novel high-performance frequency offset estimator which is suitable for implementation in an FPGA is derived from first principles. Additionally, a rule is developed for predicting whether a frequency offset estimator will become biased in the presence of ISI. This rule is used to establish a channel excitation sequence which ensures the proposed frequency offset estimator is unbiased. <p> Adaptive equalizers that compensate for the ISI take a relatively long time to converge, necessitating a lengthy training sequence. The convergence time is reduced using a two step technique to seed the equalizer. First, the ISI equivalent model of the channel is estimated in response to a specific short excitation sequence. Then, the estimated channel response is inverted with a novel algorithm to initialize the equalizer. It is shown that the proposed technique, while inexpensive to implement in an FPGA, can decrease the length of the required equalizer training sequence by up to 70 symbols. <p> It is shown that a preamble segment consisting of repeated 11-symbol Barker sequences which is well-suited to timing recovery can also be used effectively for frequency recovery and channel estimation. By performing these three functions sequentially using a single set of preamble symbols, the overall length of the preamble may be further reduced.

In this thesis, a single carrier ATSC DTV baseband transmitter, part of the receiver(including channel estimator and channel equalizer), were modeled. Since multi-pathinduced ISI (inter symbol interference) is the most significant impact on theperformance of single carrier DTV reception, modeling and implementation of singlecarrier channel estimator and channel equalizer have been the focus of the thesis. Westarted with the investigation of channel estimation methods. Afterwards, severalchannel estimators and equalizers were modeled and the performance of each channelequalization methods in different scenarios was evaluated. Our results show that thefrequency domain equalizer can achieve low computing cost and handle long delaypaths. Another important issue to be considered in block equalization is Inter-BlockInterference (IBI). The impact of IBI was investigated via behavior modeling. In lastpart of our thesis, two methods for IBI cancellation are compared and the proposal forhardware implementation was given.

Orthogonal Frequency Division Multiplexing (OFDM) is a multicarrier modulation technique which is robust against multipath fading and very easy to implement in transmitters and receivers using the inverse fast Fourier transform and the fast Fourier transform. A guard interval using cyclic prefix is inserted in each OFDM symbol to avoid the inter-symbol interference. This guard interval should be at least equal to, or longer than the maximum delay spread of the channel to combat against inter-symbol interference properly.<p> In coherent detection, channel estimation is required for the data detection of OFDM systems to equalize the channel effects. One of the popular techniques is to insert pilot tones (reference signals) in OFDM symbols. In conventional method, pilot tones are inserted into every OFDM symbols. Channel capacity is wasted due to the transmission of a large number of pilot tones. To overcome this transmission loss, incoherent data detection is introduced in OFDM systems, where it is not needed to estimate the channel at first. We use differential modulation based incoherent detection in this thesis for the data detection of OFDM systems. Data can be encoded in the relative phase of consecutive OFDM symbols (inter-frame modulation) or in the relative phase of an OFDM symbol in adjacent subcarriers (in-frame modulation). We use higher order differential modulation for in-frame modulation to compare the improvement of bit error rate. It should be noted that the single differential modulation scheme uses only one pilot tone, whereas the double differential uses two pilot tones and so on. Thus overhead due to the extra pilot tones in conventional methods are minimized and the detection delay is reduced. It has been observed that the single differential scheme works better in low SNRs (Signal to Noise Ratios) with low channel taps and the double differential works better at higher SNRs. Simulation results show that higher order differential modulation schemes don¡¯t have any further advantages. For inter-frame modulation, we use single differential modulation where only one OFDM symbol is used as a reference symbol. Except the reference symbol, no other overhead is required. We also perform channel estimation using differential modulation. Channel estimation using differential modulation is very easy and channel coefficients can be estimated very accurately without increasing any computational complexity. Our simulation results show that the mean square channel estimation error is about ¡¼10¡½^(-2) at an SNR of 30 dB for double differential in-frame modulation scheme, whereas channel estimation error is about ¡¼10¡½^(-4) for single differential inter-frame modulation. Incoherent data detection using classical DPSK (Differential Phase Shift Keying) causes an SNR loss of approximately 3 dB compared to coherent detection. But in our method, differential detection can estimate the channel coefficients very accurately and our estimated channel can be used in simple coherent detection to improve the system performance and minimize the SNR loss that happens in conventional method.

Cooperative communications have received much attention recently, due to its ability to attain cooperation diversity. But when two nodes communicate via relays, it is difficult to get the perfect channel information, so relays must estimate their forward channel and backward channel in order to amplify the data to the destination. We investigate the effect of channel estimation error, and design the LMMSE estimator to estimate the channels, and also we consider the multi-relays to assist the whole system for training and data transmission. We propose the SNR gap ratio, outage probability, and the BER simulations for the analysis. Simulation shows that when using multi-relays, it can mitigate the effect of channel estimation errors in all of the amplify-and-forward (AF) scenarios.