Spectrum Section Preallocation for Cooperative Sensing and Transmission in Cognitive Radio Ad Hoc Networks

A mobile device with embedded cognitive radios (CRs) is integrated into a novel low-cost massive antenna configuration multihop or device-to-device (D2D) wireless network that temporarily uses any portion of the spectrum for transmissions. However, the number of spectral bands (SBs) that a CR can sense within a specific time is limited. Furthermore, CR node transmissions interfere with one another. This paper proposes a method both for cooperative sensing and transmission with preallocated disjoining of spectrum sections (SSs) from neighboring links to increase the likelihood of matching available SBs. This minimizes sensing and negotiating overheads and reduces interference from a viewpoint of high-level graph model and mathematical formulation. The multiple-constraint routing with the resource allocation for cooperative sensing problem is proven to be NP-hard. The resource allocation method is compared with existing SB postallocation methods (i.e., random SB allocation and fixed SB allocation) and preallocation methods (i.e., link-based SS allocation and multiple parallel SB allocation). The proposed methods handle the resource preallocation mechanism with a centralized solution but handle the sensed results with a decentralized solution to reduce the negotiating and selecting time. We evaluate the performance of the system capacity, end-to-end transmission delay, fairness index, and standard deviation, thereby verifying the quality of the proposed method in terms of interferences, the fading effect, and other environment-dependent conditions. The results of the proposed method are more favorable than those of other methods.