Joint coding and security: Secure arithmetic coding and secure MIMO communications

Recently, due to high demand for privacy and data protection as well as efficiency and faithfulness in communication, it is now a very natural and indispensable issue to consider the combination of security and source/channel coding. In this dissertation, two possible combinations of security and coding are studied. First, joint encryption and source coding are investigated. Second, secure multiple input multiple output (MIMO) communication based on information-theoretic approach with channel coding is explored.;First, a modified arithmetic coding that offers both encryption and compression is presented. The system utilizes a binary arithmetic coder in which the overall length within the range [0,1) allocated to each symbol is preserved, but the traditional assumption that a single contiguous interval is used for each symbol is removed. A key known to both the encoder and decoder is used to describe where the intervals are "split" prior to encoding each new symbol. The repeated splitting has the effect of both scrambling the intervals and altering their lengths. Additionally, a series of permutations are applied at the input and the output of the encoder. The overall system provides simultaneous encryption and compression, with negligible coding efficiency penalty relative to a traditional arithmetic coder. Then, the idea of binary secure arithmetic code is extended to an adaptive, context-based M-ary arithmetic coder.;Next, secure communications in a MIMO system in which the eavesdropper's channel is inherently superior to that of the legitimate receiver is described. Both the legitimate receiver and the eavesdropper are assumed to have equal knowledge and comparable computational capabilities and there is no "secret key" known only to the legitimate receiver. The combination of artificial noise added at the transmitter, parallelization of the legitimate channel using the singular value decomposition, and low density parity check codes can overcome the superiority of the eavesdropper's channel and enable secure communications. Next, the algorithm of data or artificial noise transmission is investigated to maximize the secrecy capacity based on the signal to noise ratio (SNR) difference between both receivers. Results showing the trade-offs among data rate, the SNR difference, and the effective SNR are presented.