Two fast growing technologies, MIMO and cognitive radio (CR), can both effectively combat the transmission interference among links and thus increase the network throughput. MIMO exploits spatial degree of freedom (DoF) through spatial multiplexing and interference cancellation within the same frequency channel, while CR exploits all available frequency channels for transmissions. We consider an ad hoc network where each node is equipped with an array of cognitive radios. A radio can tune to a different channel and transmit independently, or transmit together with other radios on the same channel using MIMO mode. Additionally, different frequency and spatial channels could have different conditions. There is a big challenge for nodes to distributively coordinate in selecting a transmission channel and/or a spatial DoF taking advantage of this unprecedented flexibility and diversity of channels for a higher network performance. In this work, we mathematically model the opportunities and constraints for such a network with the objective of maximizing the weighted network throughput. We propose a centralized algorithm as our comparison benchmark, and a distributed algorithm to flexibly assign spectrum channel or spatial DoF exploiting the multiuser diversity, channel diversity and spatial diversity for a higher performance in a practical network. The algorithm further supports different transmission priorities, reduces transmission delay and ensures fair transmissions among nodes by providing all nodes with certain transmission probability. The performance of our algorithms are studied through extensive simulations and the results demonstrate that our algorithm is very effective and can significantly increase the network throughput while reducing the delay.

The rapid advances of MIMO to date have mainly stayed at the physical layer. Such fruits have not fully benefited MIMO research at the network layer mainly due to the computational complexity associated with the matrix-based model that MIMO involves. Recently, there have been some efforts to simplify link layer model for MIMO so as to facilitate research at the upper layers. These models only require simple numeric computations on MIMO's degrees-of-freedom (DoFs) to characterize spatial multiplexing (SM) and interference cancellation (IC). Thus, these models are much simpler than the original matrix-based model from the communications world. However, achievable DoF regions of these DoF-based models are not analyzed. In this paper, we re-visit this important problem of MIMO modeling. Based on accounting of how DoFs are consumed for SM and IC, we develop a tractable link layer model for multi-hop MIMO networks. We show that under common assumptions of DoF-based models and additional assumption of no dependency cycle, this model includes all the feasible solutions by the matrix-based model under SM and IC for any network topology. This work offers an important building block for theoretical research on multi-hop MIMO networks.

The rapid advances of MIMO to date have mainly stayed at the physical layer. Such fruits have not been fully benefited at the network layer mainly due to the computational complexity associated with the matrix-based model that MIMO involves. Recently, there are some efforts to simplify link layer model for MIMO so as to ease research for the upper layers. These models only require numeric computations on MIMO's degrees-of-freedom (DoFs) for spatial multiplexing (SM) and interference cancellation (IC) to obtain a feasible rate region. Thus, these models are much simpler than the original matrix-based model from the communications world. However, none of these DoF-based models is shown to achieve the same rate region as that by the matrix-based model. In this paper, we re-visit this important problem of MIMO modeling. Based on accurate accounting of how DoFs are consumed, we develop a simple link layer model for multi-hop MIMO networks. We show that this model is optimal in the sense of achieving the same rate region as that by the matrix-based model under SM and IC for any network topology. This work offers an important building block for theoretical research on multi-hop MIMO networks.

We study a multicast communication problem in a multi-hop ad hoc network where each node is equipped with a cognitive radio (CR). The goal is to minimize the required network-wide resource to support a set of multicast sessions, with a given bit rate requirement for each multicast session. The unique characteristics and complexity associated with CR distinguish this problem from existing multicast problems for ad hoc networks. In this paper, we formulate this problem via a cross-layer approach by taking consideration of scheduling and routing jointly. Although the problem formulation is in the form of a mixed-integer linear program, we develop a polynomial-time algorithm that offers highly competitive solutions. By comparing the solution values with a lower bound, we show that the proposed algorithm can provide a solution that is close to the optimum.

Cognitive radio (CR) and multiple-input multiple-output (MIMO) are two independent physical layer technologies that have made significant impact on wireless networking. CR operates on the channel/band level to exploit white space across spectrum dimension while MIMO operates within the same channel to improve spectral efficiency within the same band. In this paper, we explore MIMO-empowered CR network, which we call {\rm CRN}^{{\rm MIMO}}, to achieve the ultimate flexibility and efficiency in dynamic spectrum access and spectrum utilization. Given that CR and MIMO handle interference at different levels (across channels vs. within a channel), we are interested in how to jointly optimize both so as to maximize user throughput in a multihop network. To answer this question, we develop a tractable mathematical model for {\rm CRN}^{{\rm MIMO}}, which captures the essence of channel assignment (for CR) and degree-of-freedom (DoF) allocation (for MIMO) within a channel. Based on this mathematical model, we use numerical results to show how channel assignment in CRN and DoF allocation in MIMO can be jointly optimized to maximize throughput. More important, for a {\rm CRN}^{{\rm MIMO}} with A_{{\rm MIMO}} antennas at each node, we show that joint optimization of CR and MIMO offers more than A_{MIMO}-fold throughput increase than a CRN (without MIMO).