<p>Among new emerging digital communication systems, there is a clear trend of an increasing number of services using high capacity broad band connections, e.g. transfer of images, video and high quality sound. This makes it necessary to find bandwidth efficient modulation formats and efficient channel equalization solutions at the receiver. A modulation format, with possibilities for both relatively simple equalizer structures and bandwidth efficient solutions is Orthogonal Frequency Division Multiplexing (OFDM). The symbol stream is divided into parallel symbol streams, which are modulated on to separate subchannels. The frequency spectra of the subchannels overlap, but the orthogonality of the subchannels are maintained in the time domain at the sampling instant.</p><p>To obtain orthogonality between subchannels in OFDM system, one of the assumptions which are made, is exact knowledge of the carrier frequency at the receiver. In the case of a carrier frequency offset (CFO) between transmitter and receiver, the orthogonality between subchannels are lost. With a CFO some of the signal power will be transferred into interference power, i.e. noise, reducing the system performance. All digital transmission systems suffers from performance degradation in the case of a CFO and it is thus important to minimize the CFO, i.e. perform carrier frequency synchronization. The CFO generated interference is special for the OFDM systems and makes them more vulnerable to CFO than single carrier (SC) systems. In addition, the symbol length is increased in OFDM systems compared to SC systems, reducing the amount of CFO tolerated before phase slipping occurs (i.e. erroneous decisions due to CFO generated phase errors). Extra care should be taken in the case of OFDM systems to synchronize the carrier frequency at the receiver with the carrier of the transmitter.</p><p>The main topics of this work have been:</p><p>• Carrier synchronization in bandwidth efficient OFDM systems on stationary channels. To obtain maximum bandwidth efficiency, neither pilots, guard intervals, repeated sequences or other redundant signalling is used in the proposed methods. Four new OFDM carrier frequency acquisition algorithms are proposed, with performance investigated by simulations. The frequency estimator of Kay is investigated for use in Decision Directed (DD) carrier frequency tracking, with OFDM and non-constant amplitude modulation.</p><p>• Consequences of non perfect carrier frequency tracking and time varying transmission channels in OFDM systems. Both OFDM systems using QAM with rectangular pulses and OFDM systems using O-QAM with finite length pulses have been investigated.</p><p>Degradation due to non-perfect tracking on stationary channels has been calculated and performance requirements for the developed tracking algorithms are found.</p><p>For flat Rayleigh fading channels, degradation due to non-perfect tracking and doppler spread are calculated.</p>

<p>We address the problem of improving the throughput and the availability of digital communications in the High Frequency (HF, 3-30 MHz) band. In standardized military waveforms, the data is protected by an error-correcting code (ECC), and the code bits are shuffled by an interleaver and mapped onto a signal constellation for modulation onto a single carrier. Training sequences are multiplexed into the stream of transmitted symbols to aid the receiver in tracking the channel variations. The channel imposes severe time-varying intersymbol interference (ISI) as well as additive noise. Conventional receivers for such a system would first perform adaptive equalization (to mitigate the ISI) and symbol demapping, deinterleave the received code bits, and finally perform decoding, where the redundancy of the ECC is used to make high-quality decisions on the transmitted data bits even when bit errors have been introduced by the channel. Such a receiver is suboptimal because the equalizer does not make use of the redundancy introduced by the ECC, and is outperformed by an iterative scheme called turbo equalization. In turbo equalization, a.k.a. iterative equalization and decoding, soft information on the code bits is fed back from the decoder to the equalizer in an iterative fashion, and by performing the equalization and decoding tasks several times the bit error rates become significantly smaller than for a conventional “single-pass” receiver. Since we are dealing with an unknown time-varying channel, we must also perform channel estimation. We include channel estimation in the iterative loop of the turbo equalizer, using soft information fed back from the decoder as “training sequences” between the ordinary transmitted training sequences. Then, the receiver performs iterative channel estimation, equalization, and decoding, which can also be called adaptive turbo equalization. We have proposed a receiver using adaptive turbo equalization, and performed simulations using the MIL-STD-188-110 waveform at 2400 bps, transmitted over an ITU-R poor channel (a commonly used channel to test HF modems). We find that the proposed receiver outperforms a conventional receiver by 2-3 dB in terms of required signal-to-noise ratio to achieve a certain bit error rate. In this dissertation, we give an introduction to the fields of HF communications and standardized HF waveforms, channel modelling, and turbo equalization. We present an analysis of measured channel data to motivate our research in turbo equalization. We then present our research contributions to the field of turbo equalization: A low-complexity soft-in soft-out equalizer for time-varying channels, a comparative study of channel estimation algorithms using soft information as the input signal, and an investigation of adaptive turbo equalization using a technique known as EXIT charts. Finally, we present our main practical result, which is our proposed receiver and the corresponding simulation results. </p>

We address the problem of improving the throughput and the availability of digital communications in the High Frequency (HF, 3-30 MHz) band. In standardized military waveforms, the data is protected by an error-correcting code (ECC), and the code bits are shuffled by an interleaver and mapped onto a signal constellation for modulation onto a single carrier. Training sequences are multiplexed into the stream of transmitted symbols to aid the receiver in tracking the channel variations. The channel imposes severe time-varying intersymbol interference (ISI) as well as additive noise. Conventional receivers for such a system would first perform adaptive equalization (to mitigate the ISI) and symbol demapping, deinterleave the received code bits, and finally perform decoding, where the redundancy of the ECC is used to make high-quality decisions on the transmitted data bits even when bit errors have been introduced by the channel. Such a receiver is suboptimal because the equalizer does not make use of the redundancy introduced by the ECC, and is outperformed by an iterative scheme called turbo equalization. In turbo equalization, a.k.a. iterative equalization and decoding, soft information on the code bits is fed back from the decoder to the equalizer in an iterative fashion, and by performing the equalization and decoding tasks several times the bit error rates become significantly smaller than for a conventional “single-pass” receiver. Since we are dealing with an unknown time-varying channel, we must also perform channel estimation. We include channel estimation in the iterative loop of the turbo equalizer, using soft information fed back from the decoder as “training sequences” between the ordinary transmitted training sequences. Then, the receiver performs iterative channel estimation, equalization, and decoding, which can also be called adaptive turbo equalization. We have proposed a receiver using adaptive turbo equalization, and performed simulations using the MIL-STD-188-110 waveform at 2400 bps, transmitted over an ITU-R poor channel (a commonly used channel to test HF modems). We find that the proposed receiver outperforms a conventional receiver by 2-3 dB in terms of required signal-to-noise ratio to achieve a certain bit error rate. In this dissertation, we give an introduction to the fields of HF communications and standardized HF waveforms, channel modelling, and turbo equalization. We present an analysis of measured channel data to motivate our research in turbo equalization. We then present our research contributions to the field of turbo equalization: A low-complexity soft-in soft-out equalizer for time-varying channels, a comparative study of channel estimation algorithms using soft information as the input signal, and an investigation of adaptive turbo equalization using a technique known as EXIT charts. Finally, we present our main practical result, which is our proposed receiver and the corresponding simulation results.

Among new emerging digital communication systems, there is a clear trend of an increasing number of services using high capacity broad band connections, e.g. transfer of images, video and high quality sound. This makes it necessary to find bandwidth efficient modulation formats and efficient channel equalization solutions at the receiver. A modulation format, with possibilities for both relatively simple equalizer structures and bandwidth efficient solutions is Orthogonal Frequency Division Multiplexing (OFDM). The symbol stream is divided into parallel symbol streams, which are modulated on to separate subchannels. The frequency spectra of the subchannels overlap, but the orthogonality of the subchannels are maintained in the time domain at the sampling instant. To obtain orthogonality between subchannels in OFDM system, one of the assumptions which are made, is exact knowledge of the carrier frequency at the receiver. In the case of a carrier frequency offset (CFO) between transmitter and receiver, the orthogonality between subchannels are lost. With a CFO some of the signal power will be transferred into interference power, i.e. noise, reducing the system performance. All digital transmission systems suffers from performance degradation in the case of a CFO and it is thus important to minimize the CFO, i.e. perform carrier frequency synchronization. The CFO generated interference is special for the OFDM systems and makes them more vulnerable to CFO than single carrier (SC) systems. In addition, the symbol length is increased in OFDM systems compared to SC systems, reducing the amount of CFO tolerated before phase slipping occurs (i.e. erroneous decisions due to CFO generated phase errors). Extra care should be taken in the case of OFDM systems to synchronize the carrier frequency at the receiver with the carrier of the transmitter. The main topics of this work have been: • Carrier synchronization in bandwidth efficient OFDM systems on stationary channels. To obtain maximum bandwidth efficiency, neither pilots, guard intervals, repeated sequences or other redundant signalling is used in the proposed methods. Four new OFDM carrier frequency acquisition algorithms are proposed, with performance investigated by simulations. The frequency estimator of Kay is investigated for use in Decision Directed (DD) carrier frequency tracking, with OFDM and non-constant amplitude modulation. • Consequences of non perfect carrier frequency tracking and time varying transmission channels in OFDM systems. Both OFDM systems using QAM with rectangular pulses and OFDM systems using O-QAM with finite length pulses have been investigated. Degradation due to non-perfect tracking on stationary channels has been calculated and performance requirements for the developed tracking algorithms are found. For flat Rayleigh fading channels, degradation due to non-perfect tracking and doppler spread are calculated.

<p>In this dissertation, two filter banks optimization problems are studied. The first problem is the optimization of filter banks used in a subband coder under a bit constraint. In the second problem, a multiple input multiple output communication system is optimized under a power constraint. Three different cases on the filter lengths are considered: unconstrained length filter banks, transforms, and finite impulse response filter banks with arbitrary given filter lengths. </p><p>In source coding and multiple input multiple output communication systems, transforms and filter banks are used to decompose the source in order to generate samples that are partly decorrelated. Then, they are more suitable for source coding or transmission over a channel than the original source samples. Most transformers and filter banks that are studies in the literature have the perfect reconstruction property. In this dissertation, the perfect reconstruction condition is relaxed, so that the transforms and filter banks are allowed to belong to larger sets, which contain perfect reconstruction transforms and filter banks as subsets. </p><p>Jointly optimal analysis and synthesis filter banks and transforms are proposed under the bit and power constraints for all the three filter length cases. For a given number of bits used in the quantizers or for a given channel with a maximum allowable input power, the analysis and synthesis transforms and filter banks are jointly optimized such that the mean square error between the original and decoded signal is minimized. Analytical expressions are obtained for unconstrained length filter banks and transforms, and an iterative numerical algorithm is proposed in the finite impulse response filter bank case. </p><p>The channel in the communication problem is modelled as a known multiple input multiple output transfer matrix with signal independent additive vector noise having known second order statistics. A pre- and postprocessor containing modulation is introduced in the unconstrained length filter bank system with a power constraint. It is shown that the performance of this system is the same as the performance of the power constrained transform coder system when the dimensions of the latter system approach infinity.</p><p>In the source coding problem, the results are obtained with different quantization models. In the simplest model, the subband quantizers are modelled as additive white signal independent noise sources. The proposed unconstrained length filter banks, and it is shown that the proposed transform has better performance than the Karhunen-Loève transform. Also, the proposed transform coder has the same performance as a transform coder using a reduced rank Karhunen- Loève analysis transform with jointy optimal bit allocation and Wiener synthesis transform. The proposed finite impulse response filter banks have at least as good theoretical rate distortion performance as the perfect reconstruction filter banks and the finite impulse response Wiener filter banks used in the comparison. </p><p>A practical coding system is introduced where the coding of the subband signals is performed by uniform threshold quantizers using the centroids as representation levels. It is shown that there is a mismatch between the theoretical and practical results. Three methods for removing this mismatch are introduced. In the two first methods, the filter banks them selves are unchanged, but the coding method of the subband signals is changed. In the first of these two methods, quantizers are derived such that the additive coding noise and subband signals are uncorrelated. Subtractive dithering is the second method used for coding of the subband signals. In the third method, a signal dependent colored noise model is introduced, and this model is used to redesign the filter banks. In all three methods, good correspondence is achieved between the theoretical and practical results, and comparable or better practical rate distortion performance is achieved by the proposed methods compared to systems using perfect reconstruction filter banks and finite impulse response Wiener synthesis filter banks. </p><p>Finally, conditions for when finite impulse response filter banks are optimal are derived. </p>

<p>This thesis covers a wide field of applications, with an emphasis on applications of reflectance spectroscopy for diagnostic purposes. Reflectance spectroscopy in the visible part of the spectrum has been proved to be a valuable tool in a variety of applications including e. g. port-wine stain diagnostics, diagnostics of liver pathology, neonatal jaundice and age determination of bruises for forensic applications.</p>

In this dissertation, two filter banks optimization problems are studied. The first problem is the optimization of filter banks used in a subband coder under a bit constraint. In the second problem, a multiple input multiple output communication system is optimized under a power constraint. Three different cases on the filter lengths are considered: unconstrained length filter banks, transforms, and finite impulse response filter banks with arbitrary given filter lengths. In source coding and multiple input multiple output communication systems, transforms and filter banks are used to decompose the source in order to generate samples that are partly decorrelated. Then, they are more suitable for source coding or transmission over a channel than the original source samples. Most transformers and filter banks that are studies in the literature have the perfect reconstruction property. In this dissertation, the perfect reconstruction condition is relaxed, so that the transforms and filter banks are allowed to belong to larger sets, which contain perfect reconstruction transforms and filter banks as subsets. Jointly optimal analysis and synthesis filter banks and transforms are proposed under the bit and power constraints for all the three filter length cases. For a given number of bits used in the quantizers or for a given channel with a maximum allowable input power, the analysis and synthesis transforms and filter banks are jointly optimized such that the mean square error between the original and decoded signal is minimized. Analytical expressions are obtained for unconstrained length filter banks and transforms, and an iterative numerical algorithm is proposed in the finite impulse response filter bank case. The channel in the communication problem is modelled as a known multiple input multiple output transfer matrix with signal independent additive vector noise having known second order statistics. A pre- and postprocessor containing modulation is introduced in the unconstrained length filter bank system with a power constraint. It is shown that the performance of this system is the same as the performance of the power constrained transform coder system when the dimensions of the latter system approach infinity. In the source coding problem, the results are obtained with different quantization models. In the simplest model, the subband quantizers are modelled as additive white signal independent noise sources. The proposed unconstrained length filter banks, and it is shown that the proposed transform has better performance than the Karhunen-Loève transform. Also, the proposed transform coder has the same performance as a transform coder using a reduced rank Karhunen- Loève analysis transform with jointy optimal bit allocation and Wiener synthesis transform. The proposed finite impulse response filter banks have at least as good theoretical rate distortion performance as the perfect reconstruction filter banks and the finite impulse response Wiener filter banks used in the comparison. A practical coding system is introduced where the coding of the subband signals is performed by uniform threshold quantizers using the centroids as representation levels. It is shown that there is a mismatch between the theoretical and practical results. Three methods for removing this mismatch are introduced. In the two first methods, the filter banks them selves are unchanged, but the coding method of the subband signals is changed. In the first of these two methods, quantizers are derived such that the additive coding noise and subband signals are uncorrelated. Subtractive dithering is the second method used for coding of the subband signals. In the third method, a signal dependent colored noise model is introduced, and this model is used to redesign the filter banks. In all three methods, good correspondence is achieved between the theoretical and practical results, and comparable or better practical rate distortion performance is achieved by the proposed methods compared to systems using perfect reconstruction filter banks and finite impulse response Wiener synthesis filter banks. Finally, conditions for when finite impulse response filter banks are optimal are derived.

This thesis covers a wide field of applications, with an emphasis on applications of reflectance spectroscopy for diagnostic purposes. Reflectance spectroscopy in the visible part of the spectrum has been proved to be a valuable tool in a variety of applications including e. g. port-wine stain diagnostics, diagnostics of liver pathology, neonatal jaundice and age determination of bruises for forensic applications.

No description available.

No description available.