This thesis focuses on a user-friendly software-defined radio (SDR) development workflow for prototyping, research and education in wireless communications and networks. Specifically, a Simulink interface to the Universal Software Radio Peripheral 2(USRP2) SDR platform is devised in order to enable over-the-air data transmission and reception using a Simulink signal source and sink, in addition to controlling a subset of the hardware resources of the USRP2 platform. Using the USRP2 as the RF front end, this interface will use Simulink for software radio development and signal processing libraries of the digital baseband component of the communication transceiver design. This combination of hardware and software will enable the rapid design, implementation, and verification of digital communications systems in simulation, while allowing the user to easily test the system with near real time over-the-air transmission. The use of Simulink and MATLAB for communication transceiver development will provide streaming access to the USRP2 without the steep learning curve associated with current workflows. These widely available software packages and the USRP2 will make digital communication system prototyping both affordable yet highly versatile, enabling researchers and industry engineers to conduct studies into new wireless communications and networking architectures including cognitive radio. Furthermore, the interface will allow users to become familiar with tools used in industry while learning communications and networking concepts.

Due to the ever increasing dependency on existing wireless technologies and the growing usage of sophisticated wireless devices, the demand for bandwidth is rising exponentially. Also, the Federal Communications Commission (FCC) has reserved a considerable amount of spectrum for licensed users. As a result, the unlicensed spectrum usage is constrained to the overcrowded unlicensed spectrum. Various spectral management surveys have indicated inefficient spectrum utilization in the licensed spectral bands. The congested unlicensed spectrum and inefficiently used licensed frequency bands calls for an approach to use the available spectrum opportunistically. Therefore, the concept of "Spectrum Pooling", which is based on Dynamic Spectrum Access (DSA), was proposed to make the unused sections of licensed spectrum available to the unlicensed users. In Spectrum Pooling, an empty section of licensed spectrum is borrowed by a secondary user for certain period of time without interfering with the licensed user. Orthogonal Frequency Division Multiplexing (OFDM) is a transmission scheme that is a candidate for Spectrum Pooling since it is capable of forming an adaptive spectral shape that allows coexistence of licensed and unlicensed users while attemting to minimize any interference. Subcarriers in the OFDM signal can be deactivated to generate Non-Contiguous OFDM (NC-OFDM). Even though NC-OFDM allows efficient use of available spectrum, it causes out of band (OOB) radiation, which adversely affects the performance of adjacent user. This thesis presents two novel techniques for combat the effects of OOB radiation generated by NC-OFDM. The proposed techniques employ a filtering-based approach combined with the technique of windowing in order to suppress the unwanted sidelobes by around 35dB-40dB. The attenuation is achieved without affecting other transmission parameters of the secondary user significantly.

The purpose of this paper is to study the possibilities of the OFDM system adaptive modulation. The study presents mechanism for acquiring the necessary data of every frequency subchannel using pilot signals and application of these data for system adjustment. The paper introduces the water-filling principle for subchannel power allocation on the basis of their SNR and adaptive modulation mechanisms for variety of OFDM usage. These mechanisms are implemented into a MATLAB model. The code is then transformed into graphical representation and using USRP2, the signal is transmitted by real world channel.