SCALE TIME OFFSET ROBUST MODULATION (STORM) FOR CODE DIVISION MULTIACCESS MULTIUSER DETECTION AND MITIGATION
Open Access
- Author:
- Jenkins, David M
- Graduate Program:
- Electrical Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- September 23, 2008
- Committee Members:
- Randy Keith Young, Committee Chair/Co-Chair
John F Doherty, Committee Chair/Co-Chair
Timothy Joseph Kane, Committee Member
Allan Gunnar Sonsteby, Committee Member - Keywords:
- spread spectrum
code division multiaccess
time scale correlation
multiuser detection and mitigation
autoambiguity - Abstract:
- This thesis introduces and analyzes Scale-Time Offset Robust Modulation (STORM). STORM is a new waveform design for modulating information for transmission. In this thesis, STORM supplements a code division multiaccess (CDMA) communications system to increase the capacity of a CDMA network. Since current commercial CDMA systems have dedicated spectrum, the capacity is primarily limited by the interference of other CDMA users; thus, power control and multiuser interference mitigation are the focus of CDMA capacity enhancements. Although extensive research in these areas continues, significant, robust, practical capacity enhancements have been difficult to realize for existing commercial CDMA systems such as Wideband-CDMA (WCDMA). STORM offers potential to practically enhance the capacity of CDMA systems by robustly realizing more rapid power control and more efficient multiuser interference mitigation. Additionally, this thesis proposes the practical implementation of STORM into the WCDMA standard/processing. With the objective of increasing capacity for a given bandwidth, three analyses are emphasized in this thesis: power control, multiuser detection, and robust information extraction in degraded channels. The analysis of the state-of-the-art implementations in practical, highly dynamic channels leads to performance bounds that STORM may extend. Existing power control implementations degrade significantly in rapidly varying channels with many simultaneous users. Existing multiuser detection requires that the users with the dominant signal energy in the receiver be estimated and tracked; this requirement is difficult to implement in highly dynamic environments and existing approaches are computationally intensive which may lead to significant undesirable processing/energy demands. Scale Time Offset Robust Modulation (STORM) has been proposed to supplement transmitted reference signaling by adding time scale to the reference signal. With this additional piece of information, STORM enables rapid synchronization using a low rate processor and enables non-coherent addition of multipath energy. To analyze these features, STORM is evaluated in both a nonlinear time-varying dispersive acoustic channel using a physically based channel model, and a electromagnetic multipath spherical scattering environment with a finite difference time domain simulation. STORM is analyzed and its performance is quantified for practical insertion into a WCDMA architecture. STORM is shown to increase capacity by enhancing multiuser detection and facilitating robust performance in scattering and dispersive environments. STORM is incorporated into WCDMA uplink and downlink by modifying the the in-phase Gold scrambling code and using the modified code to replace the quadrature scrambling code. STORM detector performance is evaluated with this modification. A WCDMA multiuser environment is created with eighty mobile users. STORM delivers a 6 dB STORM detector signal-to-noise ratio for each user. Successive interference cancellation (SIC) performance has been known to increase when the power ordering of users is estimated between iterations, and to date, this performance increase has not been practically realized due to computational complexity. STORM practical implementation in WCDMA delivers the ability to detect the user with the highest signal power with significantly fewer calculations improving the feasibility of SIC with power ordering between iterations. For this reason, the STORM enhanced SIC performance increase offers a potentially significant increase in capacity.