A method and apparatus for optimal dynamic spectrum access (DSA) by a Secondary User (SU) in a Primary User (PU) LTE network is proposed. The cumulative hazard function from Survival Analysis is used to determine the probability of an SU successfully transmitting over a certain number of Physical Resource Blocks (PRBs) after the SU request arrives during an idle period and then the SU request for transmission is granted if the above probability is more than a set threshold. Using this principle, the proposed method optimally allocates PRBs from the set of idle LTE channels such that the probability of interference to the PU on each channel is below a set threshold while the number of PRBs assigned to the SU is maximized.
Long Term Evolution (LTE) systems do not share their idle Physical Resource Blocks (PRBs) with other users. This is a waste of scarce spectrum resources. When the idle PRBs of LTE system are used by a Secondary User (SU), then the LTE spectrum utilization goes up. Thus, this invention solves the problem of spectrum scarcity by increasing spectrum utilization. It provides guaranteed Quality of Service (QoS) to the Primary User (PU) LTE systems in terms of limiting the probability of interference to the LTE system below a set threshold. Hence it can ease some of the concerns that the LTE operators with sharing their spectrum.
This invention uses a non-parametric estimation of the cumulative hazard function from survival analysis to compute the probability of successful transmission of an SU in a PRB. Using this information an optimal PRB allocation scheme is proposed. Using this scheme, an SU can maximize the number of PRBs it can use to transmit while keeping the probability of successful transmission above a set threshold on each LTE channel (or equivalently keeping the probability of interference to the PU below a set threshold). This does not require any changes to the PU LTE system. Thus, it can be implemented in an SU transparent to the PU LTE system. The method is very simple, hence can easily be implemented in a practical dynamic spectrum access (DSA) system.
There has not been any scheme proposed to exploit the idle PRBs of an LTE system using the DSA paradigm. This invention can be implemented in an SU without having to make any changes to the PU LTE system. This invention provides an optimal allocation of PRBs while providing a guaranteed QoS to the PU LTE system in terms of keeping interference below a set threshold. So, the novelty of this invention is that it is an optimal scheme that maximizes allocation of PRBs to SUs while keeping interference to the PU LTE system below a set threshold.
Additionally, the communication industry is currently working on having LTE systems in a three-tier spectrum sharing architecture in the Citizens Broadband Radio Service (CBRS) band. But this sharing will be semi-dynamic and will be at a very coarse time scale (hours or even days). This invention provides dynamic spectrum access at the time scale of PRBs (0.5 ms) and hence can produce high spectrum utilization. This invention is optimal in the sense that it provides maximum number of PRBs to the SU while still limiting the probability of interference to the PU LTE system.
PU systems such as licensed LTE operators have been reluctant to share their spectrum with others because of their concern that it could lead to deterioration of performance of their own customers. This invention alleviates that concern by limiting the interference to the PU system. In addition, it provides an optimal method of allocating PRBs to the SU, thus leading to maximum spectrum utilization under the required constraints. This invention does not require any changes to be made to the PU LTE system, i.e., our DSA system will be transparent to the PU system. This also means our DSA system can work with legacy LTE systems. Thus, our solution will be attractive to both the PU and SU system operator.