Gbenga Salami

A Comparison Between the Centralized and Distributed Approaches for Spectrum Management

There is a growing demand for spectrum to accommodate future wireless services and applications. Given the rigidity of current allocations, several spectrum occupancy studies have indicated a low utilization over both space and time. Hence, to satisfy the demands of applications it can be inferred that dynamic spectrum usage is a required necessity. Centralized Dynamic Spectrum Allocation (DSA) and Distributed Dynamic Spectrum Selection (DSS) are two paradigms that aim to address this problem, whereby we use DSS (distributed) as an umbrella term for a range of terminologies for decentralized access, such as Opportunistic Spectrum Access and Dynamic Spectrum Access. This paper presents a survey on these methods, whereby we introduce, discuss, and classify several proposed architectures, techniques and solutions. Corresponding challenges from a technical point of view are also investigated, as are some of the remaining open issues. The final and perhaps most significant contribution of this work is to provide a baseline for systematically comparing the two approaches, revealing the pros and cons of DSA (centralized) and DSS (distributed) as methods of realizing spectrum sharing.

Nonpool based spectrum sharing for two UMTS operators in the UMTS extension band

This paper investigates spectrum sharing (in the form of code sharing) between two Universal Mobile Telecommunication System (UMTS) operators in the UMTS extension band (2500-2690 MHz) with equal and unequal number of proprietary carriers, respectively. The paper proposes a dynamic spectrum allocation (DSA) algorithm to address the problem of spectrum sharing between two operators on a non-pool basis. It also investigates the impact of adjacent channel interference (ACI) on the spectrum sharing gain. Additionally, an architecture that enables spectrum sharing to take place between two or more UMTS operators is presented. The simulated performance of the proposed DSA algorithm shows that under peak-hour loading, up to 32% increase in capacity can be obtained when compared to currently used fixed spectrum allocation (FSA).

Dynamic Spectrum Sharing Algorithm Between Two UMTS Operators in the UMTS Extension Band

This paper proposes a fast scale spectrum sharing algorithm to exploit instantaneous spectrum use during the peak hour between two Universal Mobile Telecommunication System (UMTS) operators. The algorithm is based on a joint call admission framework between the two UMTS operators. The process is both coordinated and cooperative. An intelligent network agent is used. A simulation platform has been developed in order to evaluate the performance of the algorithm. The performance of the fast-scale Dynamic Spectrum Allocation (DSA) algorithm is compared with Fixed Spectrum Allocation (FSA) when no spectrum sharing takes place. The impact of uniform and non-uniform traffic on the performance of the algorithm has been investigated. The non-uniform traffic case includes hotspot cells which have higher call volume. The results show a decrease in the spectrum sharing gain from 5% (Uniform traffic) to 2% (Non-Uniform traffic) due to these hotspot cells.

LTE indoor small cell capacity and coverage comparison

LTE indoor small cell solutions form an important part of the heterogeneous network ecosystem. The benefits are mainly due to spectral re-use over small distances, giving significant capacity and coverage benefits. This paper investigates the coverage and capacity gain achievable from two types of LTE indoor small cell product namely: 2 × 21 dBm and 2 × 24 dBm. Simulation results show that significant coverage gains up to 25% can be achieved on the high power product compared to the 2 × 21 dBm. The gain results are sensitive to propagation assumptions such as models and path loss exponents. Furthermore, the results also indicate that capacity gains up to 26% can be achieved on the 2 × 24 dBm, due to higher order modulation and coding (MCS) obtained in medium and cell edge conditions compared to the other unit. In addition, simulation results show the sensitivity of the throughput and capacity gain to macro layer reference signal received power (RSRP), as well as inter-cell interference from small cell cluster. The outcomes from this paper will be useful to guide operators/ vendors in the choice of an appropriate unit based on trade-offs between hardware complexities, cost, deployment scenarios.

Spectrum sharing and cognitive radio

Spectrum sharing has been the subject of a 3 year research programme organised by the UKs Virtual Centre of Excellence in Mobile and Personal Communications. Cognitive radio was identified as a key enabling technology to allow spectrum to be shared efficiently between terminals and networks. The project has harnessed the skills of 4 UK Universities working together to understand the potential benefits for its industrial members of cognitive radio technology. This paper discusses some of the work that has been carried out by the Mobile VCE researchers in the field of cognitive radio together with some of the key results and conclusions. The central work has developed algorithms that control the dynamic allocation of radio resources between cooperating network operators as well as spectrum access protocols that allow suitably equipped terminals to sense and use free spectrum. Simulation work has shown that a gain in the efficiency of spectrum use is feasible for both types of spectrum sharing.

On the performance evaluation of spectrum sharing algorithms between two UMTS operators

This paper proposes two new algorithms that address spectrum sharing between two Universal Mobile Telecommunication System (UMTS) operators on a pool and non-pooled basis in the UMTS extension band (2500MHz–2690MHz). The proposed pool based algorithm exploits the de-correlation in the traffic pattern based on the expected call requests. The non-pool based algorithm takes advantage of the opportunities available on the primary system during capacity limited situations on the secondary system. The simulated performance of the two Dynamic Spectrum Allocation (DSA) algorithms under peak hour loading shows improvement of 4.5% and 11% respectively over Fixed Spectrum Allocation (FSA). It is shown in the pool based algorithm that the DSA gain increases, as the traffic correlation coefficient between the two operators decrease. The results also show that the call setup delay associated with scheme depends on whether the operators share the Radio Network Controller (RNC).

A framework for UMTS inter-operator spectrum sharing in the UMTS extension band

This paper proposes a framework for spectrum sharing between multiple Universal Mobile Telecommunication System (UMTS) operators in the UMTS extension band. An algorithm is proposed, and the performance of the algorithm is investigated under uniform and non-uniform traffic conditions. The impact of call setup messages on the overall performance of the algorithm show that DSA gains in the region of 7% and 2% can be obtained under uniform and non-uniform traffic conditions.

The impact of queuing and call setup delays on UMTS spectrum sharing algorithm

This paper proposes spectrum sharing algorithms between two Universal Mobile Telecommunication System (UMTS) operators in order to improve underutilization of the spectrum. Furthermore, it investigates the call setup delays present in the proposed Fixed Spectrum Allocation (FSA) and Dynamic Spectrum Allocation (DSA) schemes, and the associated messages involved. The spectrum efficiency gains of the proposed algorithms are obtained by simulations. The impact of the queuing and call setup messages on the overall algorithm performance is also discussed. It is shown that due to the proposed techniques, spectrum sharing gain of about three percent (3%) is possible at an operating point of 98% satisfaction ratio, when the impact of the call setup messages is considered.

A two stage genetically inspired algorithm for spectrum sharing between two UMTS operators

This paper proposes a two stage algorithm to address spectrum sharing between two Universal Mobile Telecommunication System (UMTS) operators. The two stage algorithm uses both genetic algorithm and load balancing techniques. The first stage uses genetic algorithm as a solution to optimize the allocation when the correlation of traffic is low. The second stage uses load balancing scheme in the highly correlated traffic region. The simulation result shows that significant spectrum sharing gains up to 26 percent and 20 percent respectively, can be obtained on both networks using the proposed algorithm.