Downlink Joint Resource Allocation with Adaptive Modulation and Coding

Get Complete Project Material File(s) Now! »

Problem Statement

The LTE-based femtocell network is launched as “release 9”; it submerges in a macro-cell one resulting a hybrid two-tier network known as Heterogeneous Network (HetNet). Therefore, significant challenges confront the femtocells deployment. The major issue in such context is related to the resource allocation and interference management. In fact, two interference types are emerged in such HetNet: cross-tier interference (femto-macro interference) and co-tier interference (femto-femto interference).
In fact, LTE-based system is characterized by a wide licensed bandwidth to achieve high peak data rate. Although that, allocating channel resource, especially for the urban envi-ronment and in peak hours, is a major problematic due to the spectrum resource scarcity. Moreover, the femtocells operate in the same licensed spectrum of the macrocell. Especially, with increasing of higher data rate services requirements, scarce available spectrum must be carefully exploited to avoid severe transmission interference and enhance spectral eﬃciency.
Thus, the available radio spectrum must be eﬃciently distributed among femtocells and macrocell to provide satisfactory user experience and increase overall system capacity while mitigating the interference. Furthermore, other substantial objectives need special atten-tion for a successful application of LTE-femtocell new technology. Such that are increase overall system data rate, increase spectrum spatial reuse as well as enhance power and spectral eﬃciency. Therefore, to reach such objectives, physical layer considerations can be exploited in addition to the resource allocation issue considered as MAC layer scheduling task. Particularly, power control and Adaptive Modulation and Coding (AMC) mecha-nisms are performed. Furthermore, signal processing tools are manipulated as helpful way in such context.
Due to the power control mechanisms, sharing spectrum becomes a promising way to increase spectral eﬃciency while mitigate interference. On the other hand, Adaptive Mod-ulation and Coding technique, consisting to adjust data rate to the link quality, contributes to increase spectral reuse and overall system throughput and capacity. Both mechanisms are performed respecting to a Signal-to-Interference-plus-Noise Ratio (SINR) threshold as a channel-quality key indicator. Thus, there are three main resources to be simultaneously assigned by the scheduling task for diﬀerent users: channel, power and Modulation and Coding Scheme (MCS).
In this thesis, we tackle these topics following two steps: first, an optimization joint resource allocation model is proposed for OFDMA-based Downlink femtocell connection; then, for SC-FDMA-based Uplink femtocell connection regarding its specifications. Second, per-formance improvements of the LTE-based transmission are provided by introducing the wavelet signal processing tool.
Indeed, The link layer scheduling algorithms are deeply exploited to enhance interference mitigation by intelligently allocating resources among users. However, some limitations of the OFDMA and SC-FDMA schemes need to be overcome to allow the expansion of the required services that operators need to provide. Hence, physical layer improvements are necessary and combined to reach the objectives of the next generation service requirements.
To answer to the above issues, we present in what follows the results of our work. Three contributions are studied in order to provide eﬀective solutions for the challenges and practical deployment of heterogeneous networks.

Thesis Contributions

We are interested in developing the advanced mobile networks, especially LTE-based fem-tocell networks. Two major orientations are adopted based on: 1) data link layer, in terms of resource access and interference management and 2) physical layer, in terms of link adaptation and enhancement of the signal transmission reliability.
The main contributions proposed in this thesis can be summarized as follows:
We have proposed two new approaches for the downlink (forward) and uplink (reverse) transmissions respectively, for LTE-femtocell mobile network. Commonly, the architec-ture for these approaches is based on previously proposed architectures in the literature that considers clustered network architecture as a hybrid centralized/distributed way. It combines the high performance (centralized) and the low computational complexity (dis-tributed) advantageous eﬀects.
Furthermore, we considered the Quality of Service (QoS) factor required by each user. Rel-atively to this factor, we assigns a priority value specific for each user and thus, we can distinguish between two user type categories: High Priority (HP) users and Best Eﬀort (BE) users. Consequently, the scheduler first serves HP users then the BE ones. The user diﬀerentiation allows enhancing user experience under both low and heavy loaded networks.
LTE spectrum resource is shared between all users in active transmission mode in the net-work (the co-channel assignment is adopted).

OFDMA-based Downlink AMC-QRAP

The first contribution focuses on the downlink LTE-based femtocell network where the multi-user access technology used is the OFDMA (adopted by the 3GPP LTE system). In this case, users attempt to access the time-frequency grid region composed from several physical resource blocks. In the downlink case, there is no constraint on the manner to allo-cate the resource blocks for each user, but the interference on each allocated resource block must be mitigated. Our objective is to allocate resource blocks based on the link adapta-tion strategy in both senses of: 1) power control and 2) adaptive modulation and coding on each resource block. In fact, the link status is qualified by the Signal-to-Interference-plus-Noise Ratio (SINR) parameter for each resource block. Thus, each femtocell base station transmits enough power to serve its own users while minimizing interference to neighboring users. It respects a minimum required SINR level for reliable signal reception and adaptively it assigns higher transmit power to the cell-edge users compared to less-interfered closed-center users. On other hand, the modulation and channel coding schemes are combined together to provide a rating amount as additional resources to be exploited. Indeed, the femtocell adaptively assigns modulation and coding scheme (MCS) for users in accordance to the channel quality, respecting a target Block Error Rate (BLER). For high SINR, high-order MCSs are used in order to maximize spectral eﬃciency; otherwise, the robust low-order MCSs are performed to maintain transmission link and enhance bit error protection at the reception. This contribution is referenced in [4] and the forthcoming publication [5].

SC-FDMA-based Uplink AMC-QRAP

In order to provide a complete framework about the LTE-femtocell networks, we are mo-tivated to perform the above mentioned resource allocation problem for the uplink trans-mission taking into account many specifications. The 3GPP LTE standard adopts for the uplink connection the Single Carrier-Frequency Division Multiple Access (SC-FDMA) as an alternative scheme of the OFDMA. The main reason behind this is the low Peak-to-Average-Power Ratio (PAPR) value characterizing the SC-FDMA scheme. This is a critical feature for the mobile terminal in term of energy eﬃciency and components cost. There-fore, in our second contribution, we proposed a SC-FDMA uplink optimization resource allocation problem under the constraint of contiguous allocated physical resource blocks for each user. Unlike most previous uplink works considering constant power and fixed MCS over all resource blocks allocated to a user, we propose to adopt a joint power control and adaptive modulation and coding mechanisms over the resource blocks depending on the channel status. Moreover, we introduced a spectrum sensing technique that we detailed in the third contribution, in order to estimate the interference power values over the occupied resource blocks and reduce the complexity of the optimization problem. This contribution is the object of the reference [6].

Physical Layer LTE Enhancements

In the second part of our research, we have tackled the physical layer “signal processing” techniques as basic LTE enhancement approach. Thus, two contributions based on the “wavelet transform” (WT) signal processing have been proposed for the mentioned issues.

Wavelet-based OFDM Multicarrier Transmission Approach

As a third contribution in our thesis, we aims to develop the signal transmission mode of the OFDM-based modulation technique. As well known, the OFDM is the fundamental basis for both OFDMA-based downlink and SC-FDMA-based uplink transmissions. However, it suﬀers from several limitations in terms of PAPR, spectral eﬃciency, synchronization cost and the inter-carrier interference due to the lack of orthogonality. Fortunately, the wavelet transform oﬀers an orthogonal basis guarantying the robustness against inter-carrier interference and promising to significantly increase the transmission data rate. Therefore, we proved that with substituting the Fourier transform by the wavelet transform, the mentioned OFDM limitations can be overcome. Furthermore, the performance in term of bit error rate is significantly enhanced. Thus, we are motivated to suggest the “wavelet-based OFDMA” as an alternative transmission mode for both downlink and uplink of the LTE and beyond wireless communication standards. This contribution is the object of the reference [7].

Wavelet-based Edge Detection for Spectrum Sensing

Finally, we proposed an enhanced spectrum sensing approach based on wavelet transform tool that accurately delimits the occupied resource blocks in the whole spectrum. In fact, the wavelet transform is an attractive tool that exhibits excellent ability to analyze signal singularities within a discontinuous structure by zooming on localized signal features. No-tably, the wideband received cellular signal is characterized by its irregular structure in the frequency domain. Therefore, the wavelet transform presents at these discontinuities posi-tions maxima peak modulus. Additionally, in order to improve the detection performance, the white noise eﬀect can be reduced by introducing the wavelet transform of the signal autocorrelation instead of the raw signal. Moreover, we proposed an automatic method to pick these local maxima based on the windowing and characterized by its precise results. In the second contribution of our thesis, in order to identify the allocated resource blocks of the localized SC-FDMA uplink signal for each user, we suggest applying the wavelet transform-based spectrum sensing as an accurate and promised approach. As already men-tioned, it helped detect more precisely the used blocks and reduced the complexity of the optimization problem. This contribution appears under the reference [8].

Thesis Outline

In what follows we present the organization of our thesis :
We begin in Chapter 2 by introducing some fundamental notions of the wireless mobile communications. Then, we describe the LTE standard basically adopted in this thesis, presenting its main properties and the physical layer characteristics. The concept of the femtocells is exhibited while highlighting the main benefits and the challenges fronting the deployment of this technology. Finally, a relevant bibliography on the diﬀerent contribu-tions of this thesis dealing with previous related works on these topics is presented.
In Chapter 3, we present our first contribution in joint power and channel allocation with Adaptive Modulation and Coding (AMC) for Downlink OFDMA LTE-based femto-cell networks. Especially, we first describe the system in terms of network and propagation models, and we present some fundamental preliminaries to facilitate the problem formula-tion. Then, we explain the interest of the Adaptive Modulation and Coding strategy. After that, we formulate the joint resource allocation problem as a multi-objective non-linear op-timization problem considering the HP and BE users. Our problem is then transformed to single-objective problem dealing with the linear form. Thus, it is resolved as a linear opti-mization problem with the Linear Programming (LP) tools. To confirm our contribution benefits, we consider several performance metrics and finally we present the performance evaluation comparing with three previous works in the literature.
We address in Chapter 4, the SC-FDMA uplink scenario and present our second con-tribution in the resource allocation problem addressed in the previous Chapter taking into account the uplink specifications and the SC-FDMA constraints. First, the SC-FDMA scheme is described, furthermore, we prove its deployment alternatively to the OFDMA for the uplink. Next, we present the system model before formulating our uplink optimiza-tion resource allocation problem. We adopt the spectrum sensing technique in order to alleviate the additional uplink constraints and interference inputs. Our proposal for the uplink is then presented and described. Performance evaluation with extensive simulation is performed.
The physical layer signal processing tools to enhance the data rate as well as the per-formance of the LTE system will be tackled in Chapter 5. Initially, we describe the wavelet transform as an alternative way of the Fourier transform presenting its essential character-istics. Then, our third contribution in the wavelet-based OFDM multi-carrier as a basis modulation scheme is presented. After that, the fourth contribution in the spectrum sens-ing enhancement based on the wavelet transform is introduced. We complete this chapter by applying these signal processing promising techniques to our adopted LTE standard in order to enhance its performance in terms of data rate, spectral and energy eﬃciencies and quality of service (QoS).
The cell sizes have been decreasing to meet the speedily increasing subscribers’ amount and the high data rate and bandwidth requirement, in many fields from TV broadcasting to satellite communications and to mobile voice and data communications.
In what follows, We find very essential to describe our considered based system and care-fully pay attention to its specific properties.
Therefore, we first depict in Section 2.2 the LTE standard characteristics adopted in the femtocell network. This latter is also presented in Section 2.3 while tackling its benefits as well as the challenges that it faces. Next, we present some fundamental notions related to the cellular networks that are helpful throughout the thesis in Section 2.4. Then, in Sec-tion 2.5 we exhibit the concept of the link adaptation and show how it can be introduced in wireless communication systems in order to enhance throughput and performance. In Section 2.6 We make a particular attention to the literature works dealing with our dif-ferent contributions including the adaptive modulation and coding (AMC)-based resource allocation approaches for the downlink and the uplink respectively, in addition to the wavelet-based physical layer improvements. We conclude this chapter by the Section 2.7.

READ Overview of synchronous languages

Long Term Evolution (LTE) System

LTE Design Goals

The LTE system is designed by the way to oﬀer enhancements obviously observed in the application layer building on the physical layer specifications. The high throughput is the major benefit ahead; in addition the spectrum flexibility, coverage improvement, very low latency, increased spectral-and energy-eﬃciency and other benefits are also provided. We describe some of that in what follows.
LTE system provides high data rate services due to the large bandwidth occupation, high order digital modulation utilization, up to 6 bits/symbol for the 64-QAM, link adaptation based on the channel status and adopting of the Multiple Input Multiple Output (MIMO) transmission scheme. These techniques and others are introduced oﬀering a remarkable increasing of the data rate. Indeed, the peak data rate oﬀered by the LTE in 20 MHz, reaches 100 Mbps (Mega bits per second) in Downlink transmission while the Uplink oﬀers 50 Mbps.
The LTE system supports scalable bandwidths, 1.4 to 20 MHz as shown in Table 2.2. Moreover, the Orthoghonal Frequency Division Multiplexing (OFDMA) and the (Single Carrier-Frequency Division Multiple Access (SC-FDMA) techniques are the basis of the LTE system transmission (will be discussed later). These techniques are characterized by the high bandwidth flexibility and dynamically assignment to users; thus, they are adapted to the nomadic and mobile applications. In addition, LTE system supports both Frequency Division Duplexing (FDD) and Time Division Duplexing (TDD) to alternate between downlink and uplink traﬃcs. This duality also oﬀers spectrum flexibility.
Latency is an important communication factor reflecting the QoS. Users’ consideration latency is defined as the time taken by the data packet to be transmitted from the UE’s data buﬀer and the serving gateway of the core network and vice versa. Comparing to the older telecommunication system (e.g. GPRS or EDGE) where the round-trip latencies of the data networks are ranged in the 600 − 700 ms, the LTE round-trip latencies are in the 50 ms range.
Full performance services still provided for up to 5 km, and services are maintained with slight degradation for distance between 5 Km and 30 Km. The coverage improvement in addition to high performance transmission enhances the overall system capacity.

Physical Layer Specifications

The LTE system requirements diﬀer between the downlink (forward) and uplink (reversed) transmissions since the front-ends of the link: base station and UE terminal diﬀer in sev-eral ways (e.g. power consumption, equipment deployment and sizes, BS sophisticated infrastructure)-Figure.

Downlink Direction Characteristics

The basic of the transmission in downlink is the Orthogonal Frequency Division Multi-plexing (OFDM) technique. A large number of orthogonal, overlapping, narrowband sub-channels (sub-carriers) are transmitted in parallel and divide the available transmission bandwidth. Doing that, the multipath fading eﬀect is combated and so, the Inter-Symbol Interference (ISI) problem is mitigated. Indeed, for each sub-channel the channel status is considered as flat. This technique is based on the Fast Fourier Transform (FFT) of the information data. In order to mitigate the residue interference and the Inter-Carrier Inter-ference (ICI), a Cyclic Prefix (CP) is added to the head of modulated stream. The CP is a number of symbols copied from the tail, so we obtain the OFDM symbol. Consequently, the fading eﬀect of the channel is overcoming without need to complex equalizers and ex-pansive components at the receiver front-end. In downlink, where the receiver is the User Equipment (UE), this fact is very suitable since the power consumption and the terminal cost are notably reduced.
For multi-user access, the Orthogonal Frequency Division Multiple Access (OFDMA) is adopted as a basis of the LTE multi-access Downlink transmission. The media access is the time-frequency domain as represented in Figure 2.2. The smallest unit to allocate is called a Resource Element (RE) represented by one subcarrier during one OFDM period. Each user is allocated a determined number of Resource Blocks (RBs). In LTE system, each RB is composed from 6 or 7 OFDM symbols forming one slot and carried upon “12” subcarriers spaced by “15KHz”. The number of OFDM symbols in each slot depends on the CP length. Thus, the allocation granularity is “180 KHz” in frequency domain and a one slot of “0.5 msec” in time domain. Two CP types are considered: normal and extended, depending on the channel delay spread. For highly spreading environments, the extended CP is used and one slot corresponds to “6” OFDM symbols instead of “7” OFDM symbols for normal CP.
The OFDMA is a multi-carrier transmission scheme that enhances the spectral eﬃciency. However, as will be described later, this scheme presents some limitations. Especially, the OFDM technique suﬀers from high level Peak-to-Average Power Ratio (PAPR). This is due to the fact that each OFDM sub-carrier is independently modulated and as known each modulation scheme presents a diﬀerent symbol power. Indeed, a high PAPR im-poses high power consumption at the transmitter front-end and sophisticated non-linear amplifiers. This fact does not meet with the small size constraint of the UE in the Uplink transmission. Therefore, for the uplink, an alternative transmission technique is adopted by the LTE standard in order to overcome the high PAPR eﬀect, however presenting other In the uplink, the User Equipment (UE) transmits data by using a pre-determined number of RBs in the same manner described for the downlink transmission in Figure 2.2. The diﬀerence is by the multi-access modulation scheme. The Single Carrier-Frequency Division Multiple Access (SC-FDMA) is the multiple access scheme currently adopted for the uplink in the 3GPP LTE system. The reason behind is the need to the highly ’power-eﬃcient’ transmission for the UE, enabling improved coverage and reduced equipment complexity. In the LTE standard, the SC-FDMA is based on the Discrete Fourier Transform (DFT)-precoded OFDM that presents smaller PAPR than conventional OFDM. Thus, during each Resource Element (RE) a DFT-precoded OFDM symbol is transmitted instead of the OFDM symbol. The SC-FDMA has a similar performance as conventional OFDMA and it oﬀers the same degree of multipath mitigation. This scheme is actually a hybrid format that combines the low peak to average ratio provided by single-carrier systems with the multi-path interference resilience and flexible sub-carrier frequency allocation that OFDM provides.

Duplex schemes and framing

The duplexing is the mode adopted to diﬀerentiate between the downlink and uplink traﬃcs. The scheduler at the base station controls both transmissions. The FDD is operated on a paired spectrum and the TDD on an unpaired spectrum. In addition, FDD deployment prepares the way for the 3G services while the TDD is matched for the evolution to the Time Division-Synchronous Code Division Multiple Access (TD-SCDMA).
The 3GPP LTE standard is designed in manner to support both duplex schemes: the Frequency Division Duplex (FDD) and the Time Division Duplex (TDD) also called TD-LTE [9]. The duplex duality oﬀers spectrum flexibility according to the spectrum allocation and it simplifies the implementation of diﬀerent standards with diﬀerent duplex modes [10]. Although the physical layer processing is closely similar for FDD and TDD, these two schemes mainly diﬀer by the transmitted frame structure. Thus, we consider two frame types: “type 1” for FDD duplex mode and “type 2” for TDD duplex mode as displays in Figure 2.3.
• FDD frame structure: One radio FDD frame is transmitted during 10 ms and it is composed to 10 subframes of 1 ms duration each one. Each subframe is divided to two slots of 0.5 ms. The uplink and the downlink are transmitted simultaneously but each transmission over a specified frequency band with respect to a set of FDD scheme configurations. This duplex scheme is complex due to the synchronization requirements.
• TDD frame structure: As in the FDD case, there is 10 subframes or 20 slots that constitutes the 10 ms total frame. For the TDD scheme, the guard period is widely necessary in order to avoid overlapping between the Downlink and Uplink transmis-sions. Thus, a ’special subframe’ represents this guard period. This subframe is divided to three parts: a Downlink part (DwPTS), a guard part (GP) and an Uplink part (UpPTS), as mentioned in Figure 2.3. Since the Downlink and Uplink share the same frame, the two transmissions occur simultaneously within the active cell. Generally the uplink and downlink traﬃcs are not symmetrical depending on the data services and demands.

Femto-Small Cells Networks

Alternatively to be served by the central far-away macrocell base station, the subscriber can be easily connected to a small femtoCell base station also referred to the eNodeB that it independently deployed whenever and wherever he desired. The eNodeB also known as femtocell access point (FAP) is a low-power access point, based on mobile technology, providing wireless connections to customers suﬀering from coverage issues in confined en-vironments or spectrum capacity problems in high density urban areas. Thus, the FAP acts as a cellular base station designed for use in indoor environments [11]. Nowadays, the major constructors develop modules that can be directly plugged onto existing Wi-Fi hotspots and act as a standard LTE base station in an indoor location. All FAPs are connected to a femtocell gateway where the traﬃc is transmitted through the operator’s core network [12]. The limited coverage area of a FAP allows a smaller number of users to take full advantage of the available spectrum.
Femtocells can serve simultaneously up to 10 or 20 users and deliver higher data rate connections depending on the transmission technology.

Table of contents :

1 Introduction
1.1 Problem Statement
1.2 Thesis Contributions
1.2.1 Adaptive Modulation and Coding for QoS-based Femtocell Resource Allocation with Power Control: AMC-QRAP Approach
1.2.1.1 OFDMA-based Downlink AMC-QRAP
1.2.1.2 SC-FDMA-based Uplink AMC-QRAP
1.2.2 Physical Layer LTE Enhancements
1.2.2.1 Wavelet-based OFDM Multicarrier Transmission Approach
1.2.2.2 Wavelet-based Edge Detection for Spectrum Sensing
1.3 Thesis Outline
2 State of The Art
2.1 Introduction
2.2 Long Term Evolution (LTE) System
2.2.1 LTE Design Goals
2.2.2 LTE Physical Layer Specifications
2.2.2.1 Downlink Direction Characteristics
2.2.2.2 Uplink Direction Characteristics
2.2.2.3 Duplex schemes and framing
2.3 Femto-Small Cells Networks
2.3.1 Femtocells’ benefits
2.3.2 Femtocells’ Challenges
2.4 Fundamental Wireless Communication Notions
2.5 Link Adaptation Issue
2.6 Literature Review
2.6.1 Downlink resource allocation approaches
2.6.2 Uplink resource allocation approaches
2.6.3 Wavelet-based signal processing enhancements
2.6.3.1 Alternative wavelet-based OFDM approaches
2.6.3.2 Spectrum Sensing techniques
2.7 Conclusion
3 Downlink Joint Resource Allocation with Adaptive Modulation and Coding
3.1 Introduction
3.2 System Description and Notations
3.2.1 Network Model
3.2.2 Propagation Model
3.2.3 Notations
3.3 Adaptive Modulation and Coding Concept
3.3.1 Definition
3.3.2 Modulation & Coding Scheme and Link Quality
3.3.3 Fixed Modulation and Coding (FMC) vs. Adaptive Modulation and Coding (AMC)
3.4 Downlink OFDMA AMC-based Joint Resource Allocation Proposal
3.4.1 Problem Formulation
3.4.2 Problem Resolution
3.5 Performance Metrics
3.5.1 Throughput Satisfaction Rate (TSR)
3.5.2 Spectrum Spatial Reuse (SSR)
3.5.3 Rate of rejected users
3.5.4 Average channel efficiency
3.5.5 Transmission power
3.6 Performance Evaluation
3.7 Conclusion
4 Uplink Joint Resource Allocation with Adaptive Modulation and Coding
4.1 Introduction
4.2 SC-FDMA Transmission Mode
4.2.1 What is SC-FDMA and Why using it?
4.2.2 SC-FDMA v.s. OFDMA
4.2.2.1 Block Diagram and Symbol Transmission
4.2.2.2 SC-FDMA and OFDMA PAPR Comparison
4.3 System Description
4.3.1 System and Transmission Model
4.3.2 Notations
4.4 Uplink AMC-based Joint Resource Allocation Proposal
4.4.1 Problem Formulation
4.4.1.1 Uplink Interference Scenarios
4.4.1.2 Spectrum Sensing Phase
4.4.1.3 Resource Allocation Phase
4.4.2 Problem Resolution
4.5 Performance Metrics
4.5.1 Throughput Satisfaction Rate (TSR)
4.5.2 Rate of rejected users
4.5.3 Spectrum Spatial Reuse (SSR)
4.5.4 Transmission power
4.5.5 Fairness
4.6 Performance Evaluation
4.7 Conclusion
5 Wavelet-based LTE Physical Layer Enhancements
5.1 Introduction
5.2 Wavelet Signal Processing Tool
5.2.1 Fourier Transform: Analysis and Limitations
5.2.2 Short Term Fourier Transform- STFT
5.2.3 Wavelet Transform: Multi-Resolution Analysis
5.2.3.1 Definition and Characteristics
5.2.3.2 Types of the Wavelet Transform
5.2.3.3 Wavelet in Communications and Application fields
5.3 Wavelet-based OFDM Multicarrier Approach
5.3.1 Fourier-based OFDM Limitations
5.3.2 Wavelet Orthogonal basis for Multicarrier Transmission
5.3.3 Wavelet-based OFDM Alternative System
5.3.4 How the wavelet alleviates the Fourier-based OFDM problems?
5.3.5 Simulation Results and Comparison
5.4 Automatic Wavelet-based Edge Detection for Spectrum Sensing
5.4.1 Edge Detection Wavelet Property
5.4.2 Wavelet-based Spectrum Sensing Approach
5.4.2.1 System Model
5.4.2.2 Approach Description
5.4.2.3 Automatic Local Maxima Detection
5.5 Wavelet Applications for the LTE Mobile System
5.5.1 5th Generation Roadmap
5.5.2 Wavelet-based Downlink Enhancement
5.5.3 Wavelet-based Uplink Enhancement
5.6 Conclusion
6 Conclusion and Future Works
6.1 Conclusion
6.2 Future works and perspectives
7 List of Publications
Bibliography