Abstract
The microwave ( \mu Wave) spectrum or low-frequency spectrum has become overcrowded due to the ever-increasing demand in bandwidth-intensive applications, such as video streaming and gaming. Owing to the fact that the spectral efficiency of \mu Wave links is approaching its fundamental limits, there is a growing consensus in both industry and academia that millimeter wave (mmWave) will play an important role in increasing data rate and scarcity of spectrum. To redress the predicament and to ensure adequate availability of bandwidth, increasing attentions have been paid to the use of mmWave spectrum for future broadband wireless networks. The mmWave spectrum offers wider bandwidth compared to the \mu Wave spectrum and a promising solution to overcome the scarcity of spectrum while providing gigabits per second (Gb/s) of data rates. To fully exploit the high potential rates of mmWave in mobile networks, a number of technical problems must be addressed. The mmWave signal transmission suffers from the issues of blockage and deafness. The former challenge is characterized by the loss of Quality of Service (QoS) in non-line-of-sight (NLOS) conditions, while the latter is characterized by the misalignment between the main lobe of transmitter and the receiver beam. Two strategies can be adopted to deal with the issues:1) relaying, use of another path with the help of relay(s) providing line-of-sight (LOS) link(s) and b) fall back, in the case of service interruption, switch back to the \mu Wave spectrum-based transmission switching between mmWave and \mu Wave frequency bands can support higher rates and QoS requirements. However, a tradeoff exists in employing any of the aforementioned strategies. In this paper, we study the joint spectrum allocation and relay selection problem to maximize the weighted sum rate relay selection (using resource block and time slot) along with a fallback approach for spectrum allocation in an HMMC network. A fallback approach is introduced to enable the hybrid transmission of \mu Wave and mmWave, and an iterative bipartite relay-spectrum selection algorithm is proposed to solve the problem. The performance evaluation results show that the proposed fallback approach outperforms the conventional \mu Wave or mmWave transmission schemes in terms of data rate, transmit power, and the number of users.