Project topics and materials
Get Complete Project Material File(s) Now! »
Load Estimation in presence of a White Gaussian Noise
We assume now that the channel is affected by a White Gaussian Noise. In order to analyze the noise effect on the accuracy of our algorithm, same observations are used to reflect the estimated load versus the real one. The averaged MSE value is represented in respect to SNR in Fig. 2.7, with a fixed signal length of 100 OFDM symbols. We can notice that the precision of the algorithm is affected by a high noise level; however an acceptable error margin can still exist with a SNR around 3 dB.
Load Estimation with higher Symbol Length
We have analyzed the effect of signal length (i.e. the number of OFDM symbols) at the input in an Ideal free channel. Different realizations have been performed in order to reflect the averaged MSE with increased number of OFDM symbols duration 100*ts (the symbol duration (ts) = 3.2 μs), 200*ts, 300*ts, 400*ts and 1000*ts as can be shown in Fig. 2.8 with a SNR = 10 dB. As we can notice, the averaged MSE value decreases with the highest number of OFDM symbols, since the precision of the estimated load increases for a higher message length where the observations results are more accurate.
Load Estimation in presence of a Multipath Fading
In multipath fading, the radio signal propagates from the transmitter to the receiver via different multiple paths due to the obstacles and reflectors existing in the wireless channel. These multipaths are caused by mechanisms of reflection, diffraction, and scattering.
When the user is significantly far from the base station, the LOS signal path does not exist, and reception happens mainly from the indirect signal paths. These multiple paths have different propagation lengths and will cause amplitude and phase fluctuations and time delay in the received signal. Thus, the main effect of multipath propagation can be described in terms of fading and delay spread.
Small scale fading is also called Rayleigh fading because if the multiple reflective paths are large in number and there is no line-of-sight signal component, the envelope of the received signal is statistically described by Rayleigh distribution. When there is a dominant non-fading signal component present, such as a line-of-sight propagation path, the small-scale fading envelope is described by Rician distribution and, thus, is referred to as Rician fading.
In our simulation the channel under consideration can be modeled as a multipath fading channel in which the impulse response may follow distributions like Rayleigh distribution (in which there is no Line of Sight (LOS) ray between transmitter and receiver). The Rayleigh distribution follow the below Probability Density Function (PDF): ( ; ) = . − 2/(2 2) (2.22).
Attempt to estimate Channels Load in presence of Rayleigh attenuation model
In this section, we tried to estimate the load of the WiFi 802.11n overlapped channels in presence of a Rayleigh fading model, without considering the attenuation equal to 1 or normalized as per the previous simulation sections.
However, our attempt did not reach a successful result since we get an infinite number of possible solutions of attenuation values. Despite this non-finite result, we will represent here below our approach in details, just for the reference of future related investigations, aiming to calculate the load and the attenuation separately.
As per most real cases, the Rayleigh fading model is normally viewed as a suitable approach to take when analyzing and predicting radio wave propagation performance for areas such as wireless communications in urban environment where there are many reflections and refractions from buildings, obstacles etc…
Attenuation System Model
As mentioned previously, we consider that the attenuation is the general Rayleigh multipath fading channel model. The Rayleigh fading model uses a statistical approach to analyze the propagation and can be used in a number of environments. In probability theory and statistics, a Rayleigh random variable x follows the Probability Density Function (PDF) presented below [32].
( ; ′) = . − 2/(2 ′2) (2.22).
LTE Physical Layer Structure
The LTE radio interface or LTE-Uu interface is the interface between the UE and the Evolved NodeB or eNodeB. The eNodeB is the termination point for the radio protocols and related functionalities in LTE network, considered as a base station that controls all radio related functionalities of the Evolved Packet System (EPS) (Evolved UTRAN (E-UTRAN) and Evolved Packet Core (EPC)). It is the connecting layer between UE and EPC. All of the radio protocols from UE terminate at eNodeB. eNodeB is the essential part of mobility management in EPS. It also performs encryption/decryption of User Plane/Control Plane data and Internet Protocol (IP) header compression/decompression as well to decrease the sending of redundant data in IP header [37]. In addition to these basic functionalities, there are other Radio Resource Management (RRM) functionalities handled by the eNodeB. As a terminating node for the radio protocols, it sets Radio Resource Connection (RRC) and performs radio resource allocations to the users with QoS based prioritization [37].
Comparing it with UTRAN, it can be seen that the eNodeB performs the functionalities of both NodeB and Radio Network Controller (RNC). This simplifies the network structure and also in a way reduces the latency in the network as well. The eNodeBs are connected to their neighboring eNodeBs with the X2 interface. This connection becomes useful during the handover scenarios [37].
LTE System and Channel Assignment Model
In LTE systems, Orthogonal Frequency Division Multiple Access (OFDMA) is the multiple access technique used in the downlink. However, since it presents a high Peak-to-average Power Ratio, it is not possible to use OFDMA on the uplink. For the uplink, Single Carrier Frequency Division Multiple Access (SC-FDMA) is used [38]. The main difference between an OFDM and OFDMA system is the fact that in the OFDM, users are allocated on the time domain only while using an OFDMA system the user would be allocated by both time and frequency. This is useful for LTE since it makes possible to exploit. It is not possible to use OFDMA on the uplink since, as told before, it presents a high Peak-to-average Power Ratio. SC-FDMA presents the benefit of a single carrier multiplexing of having a lower Peak-to-average Power Ratio. On SC-FDMA, before applying the IFFT the symbols are pre-coded by a Discrete Fourier Transform (DFT). This way each subcarrier after the IFFT will contain part of each symbol..
With OFDMA and SC-FDMA the radio spectrum is divided up into 15 KHz subcarriers. These are allocated to UEs in groups of 12 known as Resource Blocks (RBs) or Physical Resource Block (PRBs) during one 0.5 ms slot [39]. The total number of those PRBs depends on the LTE bandwidth. A RB is assigned to a user to cater its service demand that can range from few kilobits per second (Kbps) to some megabits per second (Mbps) [40]. The 12 OFDM subcarriers of the RB that are adjacent to each other and each of these sub-time slot utilizes 6 OFDM symbols when normal CP is used and 7 OFDM symbols when extended CP is used. In RB assignment, the Channel State Information (CSI) plays a significant role [40] and this information is acquired periodically by an eNodeB from its connected users. Based on this information, an eNodeB decides upon the Modulation and Coding Scheme (MCS) and the number of radio blocks that it needs to allocate to its connected users. However, in LTE downlink, if a user has been assigned to more than one RB, all these RBs must have the same MCS. This increases the complexity of the radio resource allocation problem [40].
Main Drivers of WiFi Offload in 5G Systems
Mobile data offloading transfers cellular users to WiFi networks to relieve the cellular system from the pressure of the ever-increasing data traffic load. This approach was designed many years, since the third generation (3G) networks in years 2010 and 2011 [49], till the fourth generation (4G) and LTE-A networks, and still under analysis in current research areas of HetNets in the roadmap of 5G technology, even though mobile network operators and providers have deployed many small cells or femtocells solutions to cope with the forecasted needed capacity in licensed and unlicensed mode. To have a preliminary forecast about the data traffic in the upcoming years in mobile data traffic, we represent here below a rough forecasted measures according to Cisco’s annual Global Mobile Data Traffic Forecast Update (2017 – 2022) [50]:
• By 2022, there will be more than 12 billion mobile-ready devices and IoT connections, up from about 9 billion in 2017.
• By 2022, mobile networks will support more than 8 billion personal mobile devices and 4 billion IoT connections.
• The average mobile network speeds globally will increase more than three-fold from 8.7Mbps in 2017 to 28.5Mbps by 2022.
• By 2022, mobile video will represent 79 percent of global mobile data traffic, up from 59 percent in 2017.
• By 2022, 79 percent of the world’s mobile data traffic will be video, up from 59 percent in 2017.
• Mobile offload exceeded cellular traffic by a ton in 2017; 54 percent of total mobile data traffic was offloaded onto the fixed-line network through WiFi or femtocell in 2017.
• In 2017, 4G already carried 72 percent of the total mobile traffic and represented the largest share of mobile data traffic by network type. It will continue to grow faster than other networks, however the percentage share will go down slightly to 71 percent of all mobile data traffic by 2022.
Table of contents :
Contents
Acknowledgment
List of Abbreviations
List of Figures
List of Tables
General Introduction
CHAPTER 1 – WiFi Access Medium
1.1 IEEE 802.11 Standards
1.2 WiFi Physical and MAC Layers
CHAPTER 2- Proposed Algorithm for Load Measurement of the WiFi Physical Channels .
2.1 The basis of the estimated Load measurements
2.2 State of the Art
2.3 Channels Observations Model
2.4 Estimated Load times Attenuation in respect to different Observations .
2.5 Simulations Results
2.6 Attempt to estimate Channels Load in presence of Rayleigh attenuation model
2.7 Conclusion and potential Use Cases
CHAPTER 3 – LTE Capacity Management and Resources Allocation
3.1 LTE Physical Layer Structure
3.2 LTE System and Channel Assignment Model
3.3 LTE Schedulers
CHAPTER 4- WiFi Network Planning for LTE Offload
4.1 Main Drivers of WiFi Offload in 5G Systems
4.2 State of the Art
4.3 Overlay LTE/WiFi Network Model
4.4 Heavy Users Definition
WiFi Integration with LTE towards 5G Networks
4.5 WiFi Dimensioning Method
4.6 Simulations results
4.7 Conclusion
CHAPTER 5 – Profit sharing in case of WiFi LTE coexistence
5.1 Shapley Value: Definition and Properties
5.2 Revenue Sharing
5.3 Cost Sharing
5.4 Profit Sharing
5.5 Simulations Results
5.6 Conclusion
CHAPTER 6- Conclusion and Future Directions
6.1 Summary of Contribution
6.2 Future Directions
French Summary
Publications
References
