الفهرس 
ContentsOnly 14 pages are availabe for public view
 |  | from | 79 |  |  |
| | | from | 79 | | |
Abstract
Due to the increasing demand for wireless communications, both millimeter-wave band and spectral efficiency of transmission schemes have gained a great concern recently. Also, achieving secure wireless communications is of high importance. Recently, physical layer security has been intensively investigated as an extra layer of protection for wireless communications. A new technique called Directional modulation (DM) has emerged as an implementation to physical layer security. A special type of DM suitable for millimeter-wave wireless communications is antenna subset modulation (ASM) in which the small wavelength nature in millimeter-wave band is exploited to equip a transmitter with a large antenna array. By randomly choosing few elements of the array for transmitting a symbol, secure communication in a single direction is obtained. Using only a few portion of the antenna array in transmission makes this large array underutilized. The unused portion of the antenna array can be exploited to increase the spectral efficiency of ASM. This work provides a technique for broadcasting multi-beams antenna subset modulation (MASM) in which the large antenna array is partitioned to simultaneously transmit data streams to different, spaced spatially receivers. First, the statistical model for randomized MASM has been driven, and the symbol error rate of MASM using QPSK and secrecy capacity are obtained. After that, a method for selecting antenna subsets with minimum side lobe levels for ASM is introduced and generalized for MASM. A new variable constraint is applied to the binary optimization problem, which was originally relaxed to a convex one, to control the localization of optimum solution within the antenna array. Simulation results show that ASM can be used efficiently for multi-directional broadcasting communication, while maintaining its inherent security. Also, simulation results are used to show the superiority of the proposed optimization algorithm to traditional array thinning optimization algorithms like simulated annealing algorithm (SA).