| CODIPHY or Composing On-Demand Intelligent Physical Layers aims to solve two fundamental problems in practical cognitive radio networks: Collaboration between two radio physical layers (PHY) with varying capabilities to agree on a common communication protocol, using an ontology based description of the internal structure of the radio subsystems and secondly, provide a method to compose a functioning radio pipeline from a set of pre-compiled components, using the high-level representation provided by the ontology, to target heterogeneous platforms. CODIPHY isolates the various domains of radio engineering, but still allows sharing of domain knowledge to achieve the common goal of radio adaptation.;CODIPHY goes beyond the concept of "knobs" in cognitive radios, to decompose the radio pipeline and then build it back again to implement a different wireless protocol. To automate this process through collaborative learning is the goal of this thesis. Instead of solving the generic problem of collaboration for all waveforms and protocols, we focus on the most common family of waveform used in modern wireless systems and cognitive radio networks, Orthogonal Frequency Division Multiplexing (OFDM), and validate the methodology of CODIPHY by prototype implementations on heterogeneous radio platforms using HDLs and high level programming languages.;CODIPHY is the culmination of experiences gathered from radio prototyping for cognitive radio networks and is greatly influenced by MAC-PHY crosslayer research, where the radio is treated as a mutable entity rather than fixed in function. In this thesis, we present the steps required to realize the concept of CODIPHY, which range from the implementation of cognitive radio prototypes on FPGA to its application in design and evaluation of novel crosslayer protocols. These steps provide valuable insights to the requirements and formulation of CODIPHY. Therefore, CODIPHY is a multi-disciplinary effort that facilitates knowledge sharing among disparate areas of research. |
