Date of Award
2014
Document Type
Thesis
Degree Name
M.S. in Engineering Science
Department
Electrical Engineering
First Advisor
Lei Cao
Second Advisor
John N. Daigle
Third Advisor
Ramanarayanan Viswanathan
Relational Format
dissertation/thesis
Abstract
This thesis examines the robust behavior of quantized compressed sensing measurements during transmission through an additive white gaussian noise wireless channel. The poor rate-distortion performance that accompanies compressed sensing after applying quantization has led to several works in quantized compressed sensing. However, most of these works have less consideration of the effect of transmission channel on the resulting bit stream of the quantized compressed sensing measurements. For an additive white gaussian noise wireless channel model, the quantizer and bit energy signal-to-noise ratio determines the degree of the channel effect. This thesis explores the effect of quantization and channel noise during the transmission of quantized compressed sensing image over additive white gaussian noise wireless channel. Based on the effect, an optimal resource allocation algorithm is generated to maximize the compressed sensing performance. This was achieved by deriving mathematical expressions that estimates the total distortion of a quantizer and determining the resource (i.e bit and power) allocation that minimizes the mean square error. This procedure is carried out using three quantizers (i.e Uniform scalar quantizer, Cumulative Distribution function based quantizer, and Lloyd-maxx quantizer). Simulations are formed that confirms our claim of deteriorating performance after considering channel effect, and significant improvement in the performance of compressed sensing particularly under extreme channel conditions based on the proposed optimal bit and power allocation algorithm.
Recommended Citation
Olanigan, Saheed, "Optimized Image Compressed Sensing And Transmission Through Wireless Channels" (2014). Electronic Theses and Dissertations. 523.
https://egrove.olemiss.edu/etd/523
Concentration/Emphasis
Emphasis: Electrical Engineering