The advent of mobile Internet has led to phenomenal growth of the mobile data traffic over the past few years. Spectrum sharing is seen by national regulators, in both Europe and USA, as a viable solution for allocating additional spectrum to mobile broadband networks in a timely fashion, since technologies that are capable to implement it already exist. Licensed shared access (LSA) systems are built on the idea of cognitive radio networks where a primary incumbent network offers its surplus spectrum to the secondary licensee network(s). In contrast to cognitive radio networks, where quality of service (QoS) for the secondary network is not promised, the licensee operators in LSA get a guaranteed QoS when they access incumbent spectrum dynamically. In this chapter, we discuss LSA spectrum management framework and propose novel algorithms. The spectrum is made available to the licensee mobile network operators (MNOs) at specific time instants, in a specific service area, and for a specific time period; and the MNOs dynamically access the spectrum. The chapter comprises two LSA spectrum management scenarios. First, we propose a decentralized spectrum allocation algorithm, which aims at providing resources to the MNOs in such a way that they can utilize them to serve their users and the spectrum allocation is fair when more than one MNOs share the same incumbent. In the second scenario, we consider Virtual Network Operators (VNO) that construct networks from a pool of shared resources and provide flexibility in how services are offered to the end-users and enable dynamic use and release of resources according to the demand. An auction-based approach to the LSA framework is proposed, with VNOs that share not only spectrum but also infrastructure, i.e. cloud-based massive-MIMO antennas. Further, we investigate how our approach performs in terms of the shared resources as a function of minimum required average user rate and antenna price, where VNOs can lease spectrum in blocks of 5 MHz, according to LSA. The results show that given certain price of infrastructure, the proposed auction-based allocation of resources is optimal in terms of the number of VNOs that can be served by the available spectrum, regardless of the user rate.