Advanced modulation and demodulation techniques for wireless communications.

摘要

Many wireless communications systems operate over channels exhibiting fading and interference. The increasing demand for wireless applications continually challenges system engineers to design more efficient modulation and demodulation techniques for communication in these channel conditions. In addition, battery life and size requirements limit the complexity of the transmitter/receiver pairs. Noncoherent demodulation simplifies the receiver structure at the cost of performance. Block demodulation is a noncoherent technique which recovers some of the performance cost of noncoherent reception. However, for communication channels exhibiting fading and interference, noncoherent ML block demodulation generally requires the knowledge of received signal strength and fading levels. Furthermore, for some applications, ML block demodulation is computationally prohibitive and even intractable. This dissertation presents block demodulation methods based on computationally simple and efficient data clustering algorithms. These demodulation schemes do not require knowledge of the received signal strength, fading levels, or channel phase offset. Numerical results show that these block demodulation techniques provide significant improvement over classical differential demodulation techniques.;For coded systems, some of these block demodulation techniques provide rehability information that enhance performance of the decoder. Block coded systems can use this information to erase unreliable symbols prior to decoding. This dissertation presents the performance of noncoherent block demodulation with erasure generation for various phase and amplitude-phase modulation techniques with Reed-Solomon coding. Results show that these techniques significantly improve performance over errors-only decoding.;This dissertation also investigates the concept of multiresolution modulation. Systems employing multiresolution modulation transmit multiple data streams of possibly different rates and reliability requirements in parallel over a single channel. Multiresolution signaling schemes based on hybrids of frequency shift keying, amplitude shift keying, and phase shift keying provide flexibility in achieving different reliability requirements for two simultaneous data streams. This dissertation presents several such signaling schemes along with practical demodulation techniques that are compatible with frequency-hop spread-spectrum transmission. This dissertation also presents several approaches to coding and decoding of these multiresolution schemes. Numerical results show that these hybrid multiresolution techniques are capable of simultaneously conveying two independent data streams with different reliability requirements.