A Stochastic Geometric Approach Towards The Energy Efficiency Analysis Of Cache Enabled Networks
Omar Magdy Sleem
A Stochastic Geometric Approach Towards The Energy Efficiency Analysis Of Cache Enabled Networks Omar Magdy Sleem - p. ill. 21 cm.
Supervisor: Amr El-Sherif,
Laila Hesham,
Tamer ElBatt.
Thesis (M.A.)—Nile University, Egypt, 2018 .
"Includes bibliographical references"
Contents:
1.1 Motivation
1.2 Related Work
1.3 Contributions
2. Baseline System Model
2.1 System Model
2.1.1 Network Model
2.1.2 Content Characteristics
2.1.3 Serving Scenarios
2.2 Problem Formulation
2.2.1 Hit Probability Analysis
2.2.2 Rate Analysis
2.2.3 Power Analysis
2.2.4 Energy Efficiency
2.3 Simulation results
2.3.1 Simulation Setup
2.3.2 The effect of SBS density on EE
2.3.3 The effect of user density on EE
2.3.4 Optimal SBS density vs User Density
2.3.5 Maximum Energy Efficiency vs User Density
2.3.6 Optimal SBS density vs Cache Size
2.3.7 Observations from the numerical results
3. Poisson Cluster Process Network
3.1 System Model
3.1.1 Propagation Model
3.1.2 Content Characteristics
3.2 Hit Probability Analysis
3.3 Spectral Efficiency analysis
3.3.1 Distance between the users and any SBS
3.3.2 Distance between the user and any cluster center
3.3.3 Distance between the user and serving SBS
3.3.4 Distance between the user and intra-cluster interferers
3.3.5 Laplace transform of intra-cluster interference
3.3.6 Laplace transform of inter-cluster interference
3.3.7 Average Spectral Efficiency per User
3.4 Total Average Rate Per SBS
3.4.1 Users Number Distribution in Case of TCP
3.5 Power Analysis
3.6 Energy Efficiency Per SBS
3.7 Simulation Results
3.7.1 Simulation Setup
3.7.2 Validation of lemma 3
3.7.3 The effect of SBS density on EE
3.7.4 Per Cluster User Density vs Optimal SBS Density
3.7.5 Per Cluster User Density vs Maximum EE
3.7.6 Cache Size Vs Optimal SBS Density
Abstract:
thousand-fold increase in traffic and demand. Such rapid and exponential evolution
of the emerging networks introduces several crucial aspects that have to be thoroughly
captured. One of such is the critical need for providing reliable coverage, which can
be achieved by deploying small BSs (SBSs), i.e., pico-BSs and/or femto-BSs to handle
the vast number of connections. In addition, it has been suggested to provide such
SBSs with cache storage in order to reduce congestion on backhaul connections. Not
only can the SBSs enhance the overall network coverage, but also have the ability to
coordinate with neighboring SBSs and fetch data from their caches.
In this thesis, we investigate the impact of densification of cache-enabled small
base stations (SBSs) in multi-tier cellular network. In the proposed model, a SBS
has the ability to coordinate with neighboring SBSs and fetch data from their caches.
We focus on the effect of SBSs’ density on the network’s energy efficiency. To this
end, tools from stochastic geometry theory is used to model the SBSs and users distributions,
which enables us to find closed-form expressions for the network’s energy
efficiency as a function of the SBSs density together with the cache size at each SBS.
We first exploit a uniform Poisson point process (PPP) to model the distribution of
the SBSs. For more practical scenarios, we further extend the model to a Thomas
cluster point process (TCP) in order to account for the inherent clustering properties
iii
of the network. The optimal SBSs density that maximizes the energy efficiency is
characterized, and the relation between this optimal density and the cache size at the
SBS is highlighted. In addition, the effect of the users’ density in the network on the
energy efficiency and the optimal SBS density is investigated.
Text in English, abstracts in English.
Banking & Finance
Dissertation, Academic
332
A Stochastic Geometric Approach Towards The Energy Efficiency Analysis Of Cache Enabled Networks Omar Magdy Sleem - p. ill. 21 cm.
Supervisor: Amr El-Sherif,
Laila Hesham,
Tamer ElBatt.
Thesis (M.A.)—Nile University, Egypt, 2018 .
"Includes bibliographical references"
Contents:
1.1 Motivation
1.2 Related Work
1.3 Contributions
2. Baseline System Model
2.1 System Model
2.1.1 Network Model
2.1.2 Content Characteristics
2.1.3 Serving Scenarios
2.2 Problem Formulation
2.2.1 Hit Probability Analysis
2.2.2 Rate Analysis
2.2.3 Power Analysis
2.2.4 Energy Efficiency
2.3 Simulation results
2.3.1 Simulation Setup
2.3.2 The effect of SBS density on EE
2.3.3 The effect of user density on EE
2.3.4 Optimal SBS density vs User Density
2.3.5 Maximum Energy Efficiency vs User Density
2.3.6 Optimal SBS density vs Cache Size
2.3.7 Observations from the numerical results
3. Poisson Cluster Process Network
3.1 System Model
3.1.1 Propagation Model
3.1.2 Content Characteristics
3.2 Hit Probability Analysis
3.3 Spectral Efficiency analysis
3.3.1 Distance between the users and any SBS
3.3.2 Distance between the user and any cluster center
3.3.3 Distance between the user and serving SBS
3.3.4 Distance between the user and intra-cluster interferers
3.3.5 Laplace transform of intra-cluster interference
3.3.6 Laplace transform of inter-cluster interference
3.3.7 Average Spectral Efficiency per User
3.4 Total Average Rate Per SBS
3.4.1 Users Number Distribution in Case of TCP
3.5 Power Analysis
3.6 Energy Efficiency Per SBS
3.7 Simulation Results
3.7.1 Simulation Setup
3.7.2 Validation of lemma 3
3.7.3 The effect of SBS density on EE
3.7.4 Per Cluster User Density vs Optimal SBS Density
3.7.5 Per Cluster User Density vs Maximum EE
3.7.6 Cache Size Vs Optimal SBS Density
Abstract:
thousand-fold increase in traffic and demand. Such rapid and exponential evolution
of the emerging networks introduces several crucial aspects that have to be thoroughly
captured. One of such is the critical need for providing reliable coverage, which can
be achieved by deploying small BSs (SBSs), i.e., pico-BSs and/or femto-BSs to handle
the vast number of connections. In addition, it has been suggested to provide such
SBSs with cache storage in order to reduce congestion on backhaul connections. Not
only can the SBSs enhance the overall network coverage, but also have the ability to
coordinate with neighboring SBSs and fetch data from their caches.
In this thesis, we investigate the impact of densification of cache-enabled small
base stations (SBSs) in multi-tier cellular network. In the proposed model, a SBS
has the ability to coordinate with neighboring SBSs and fetch data from their caches.
We focus on the effect of SBSs’ density on the network’s energy efficiency. To this
end, tools from stochastic geometry theory is used to model the SBSs and users distributions,
which enables us to find closed-form expressions for the network’s energy
efficiency as a function of the SBSs density together with the cache size at each SBS.
We first exploit a uniform Poisson point process (PPP) to model the distribution of
the SBSs. For more practical scenarios, we further extend the model to a Thomas
cluster point process (TCP) in order to account for the inherent clustering properties
iii
of the network. The optimal SBSs density that maximizes the energy efficiency is
characterized, and the relation between this optimal density and the cache size at the
SBS is highlighted. In addition, the effect of the users’ density in the network on the
energy efficiency and the optimal SBS density is investigated.
Text in English, abstracts in English.
Banking & Finance
Dissertation, Academic
332