Sharif Digital Repository

The significant increase in the penetration of mobile phones and the significant growth of traffic generated in these networks has posed many challenges in the field of cost management, and the expansion of the existing network for mobile operators and has encouraged operators to change the current network architecture. On the other hand, the standard of the next mobile communication, i.e. the fifth generation, has a completely different architecture from the current architecture. Many of the logical and physical entities of the current network have undergone fundamental changes in the new network, and at first glance it does not seem that there is a direct transition from the previous... 

Nowadays sharing data between users and organizations can be difficult due to privacy and legal reasons. Therefore, real-world data is not fully exploited by machine learning methods. A novel model training method called Federated Learning was proposed to alleviate the aforementioned problem by enabling users to jointly train learning models without sharing data. Federated learning utilization faces many challenges among which high communication cost, statistical heterogeneity and system heterogeneity are most important. This research deals with these challenges by proposing two methods of Reduced Clustering Federated Learning (RCFL) and Weighted Federated Distillation (WFD). Moreover, we... 

Optimisation is ubiquitous and essential in almost all areas of engineering and computer science for the design and analysis of efficient systems and algorithms. Indeed, many engineering problems can be cast as optimisation problems, in which a decision must be made upon certain control parameters in order to maximise revenue or minimise an incurred cost in designing a system. Whereas deterministic optimization problems are formulated with known parameters, real-world problems almost invariably include parameters which are unknown at the time a decision is made. Stochastic optimisation is the approach for modelling optimization problems that involve uncertainty, and it has attracted a lot of... 

Increasing needs for wireless positioning systems, especially in modern wireless networks, have emerged new trends in research on local positioning systems and algorithms. In this dissertation, several geometric arrangement of positioning systems, different types of wireless measurements used for positioning, and estimation of desirable coordinate parameters based on them are investigated. The set of constrained linearized equations, by introducing intermediate variables are developed and compared with Cramer-Rao bounds. Performance bounds of ToA, TDoA and round-trip ToA systems are obtained and their sensitivity to the location changes of the target and reference stations is examined... 

In this work we have investigated the methods of providing quality of service in cognitive radio networks. At first, we have explained the various algorithms of providing BER quality of service in wireless networks. Especially we have explained those algorithms which are known as the distributed power control algorithms. Then we have introduced methods of using adaptive modulation and we have used this kind of modulation to provide delay quality of service in cognitive radio networks. We have done this by maximizing the sum of effective capacities of secondary links under the power constraints and also the interference constraints set by the primary receivers. Finally, we have introduced... 

In this thesis, we consider multiple-antenna K-user interference channels with backhaul collaboration in one side (among the transmitters or among the receivers) and investigate the trade-off between the rate in the channel versus the communication load in the backhaul. In this investigation, we focus on a first order approximation result, where the rate of the wireless channel is measured by the degrees of freedom (DoF) per user, and the load of the backhaul is measured by the entropy of backhaul messages per user normalized by log of transmit power, at high power regimes. This trade-off is fully characterized for the case of even values of K, and approximately characterized for the case of... 

In this thesis, we study the observability and stability problems of networked control systems (the control systems which communicate through communication channels) to answer two important questions about different system models. The first is that what is the constraints implied by the stability or observability of systems? (necessary condition on the observability and stability). The second is that under which constraints can we present a design for system to guarantee the observability or stability of systems? (sufficient condition on the observability and stability). Clearly, an efficient design is the one that presents a tight sufficient condition.One of system models addressed in this... 

Extensive studies have been performed on the genetic variations involved in common neurodegenerative diseases such as Alzheimer's, macular degeneration, and Parkinson's. In most cases, no specific gene has been identified pointing to a distinct pathogenic pathway, therefore, this study mainly aims to find common genes among aforementioned diseases according to determination of a specific pathogenic protein pathway.In this study, we reached a deep understanding of the function of nervous system and the discovery of causative agents of the diseases by applying the sources of information from genome datasets in bioinformatics analysis. The utilized database comprises the classification of... 

In recent years, machine learning methods have been widely used in various applications. In particular, using machine learning techniques in the healthcare area has had significant benefits for both physicians and patients. Since we live in the big data era, and at any given moment we leave a lot of information on websites, mobile phones, etc., deep learning methods are more important than ever.One of the most important applications of machine learning in medicine area is assisting cardiologists for cardiac arrhythmias detection. Some of these arrhythmias may only occur once a week, so a portable ECG recorder should be placed on the chest of people to record their electrocardiogram signal.... 

T lymphocytes or T cells are a type of white blood cell that play an important role in the immune system. Memory T cells are a subset of them that are able to reactivate when being re-exposed to the pathogen. Because of the properties of these cells, they are an attractive choice for immunotherapy. Initial memory subgroups were shown to be more persistently effective in immunotherapies. However little is known about development and differentiation of these subgroups. With the discovery of a new subset of T cells, called T memory stem cells (TSCM), They are considered to develop through four stages, naive T cells (TN), T memory stem cells (TSCM), central memory T cells (TCM) and effector... 

One of the cornerstones of population genetics is finding ancestral relations between people of a population. Mutations and recombinations are two major signals that can be exploited to infer the population ancestral relationships. Next Generation Sequencing (NGS) has paved the way for achieving this endeavor by providing massive data including single nucleotide polymorphisms, insertions, and deletions as well as structural variations between individuals. In this thesis, the goal is to infer the ancestral structure of a given gene from thousands of NGS datasets. We also attempt to decode times of important events including mutations and recombinations. To this end, we first split a gene into... 

Satisfying the growing demands for mobile services and radio communications makes the future wireless networks to fully reuse resources such as available frequency bands, which in turn leads to profound increase in inter-cell interferences (ICI). Forcing higher interference may degrade the performance of systems in serving the users especially in cell-edges. Necessity of mitigating ICI along with higher data rate requirements push the communication engineers to advanced technologies utilization. As a result of higher complexity based on promising technologies, resource management would play even a major role in next generation mobile networks. Analyzing these technical challenges that have... 

Channel equalization is a crucial part of the OFDM communications protocols, which in turn requires channel estimation. Pilot-based methods are one the most popular channel estimation approaches in OFDM systems. The pilot signals are generally classified as orthogonal and nonorthogonal pilots. Orthogonal pilots can better estimate the channels and are widely used in communications systems. The number of orthogonal pilots is proportional to the number of transmitters. Therefore, utilizing orthogonal pilots are not amenable for the increasing number of users in communications systems and the emerging of new technologies such as Massive-MIMO employing a large number of antennas. Consequently,... 

Multiuser detection provides the first comprehensive of the subject of multiuser digital communication, dealing with demodulation of mutually interfering digital information streams. Multiple access spread time codes which are more resistant to various channel impairments, is a motivation to design a multiple access system using spread time systems via circulant codes, to detect users bit stream without multiuser interference. So first of all we introduce multiple access systems and spread time codes. To have an overview of desired system we present a matrix representation of OFDM system. With allocating multiuser circulant coded to users in spread time system, in AWGN channel , users’ bit... 

Multiple-input multiple-output (MIMO) systems have been considered a lot in the recent decade. Use of these systems results in higher data rate and higher reliability in communication systems. Use of intermediate nodes which are called relay in cooperative networks result in higher reliability and higher coverage range in wireless networks. Finally use of multiple antennas in cooperative networks results in higher data rate, higher reliability and higher coverage range in wireless networks. A drawback of multi-antenna systems is that more antennas require more RF chains. More RF chains result in higher computational complexity and higher hardware cost. A remedy for this drawback is the... 

In this thesis, we derive an achievable Degrees of Freedom (DoF) region for a MIMO channel that models uplink transmission in a cellular network of two interfering cells with two users in each cell, under two different genie-aided cognitive scenarios; a cognitive cell together with a primary cell, and a cognitive transmitter with a primary transmitter in each cell. Most of the previous works dealt with networks with only two transmitters. Here, we study the effect of increase in the number of transmitters on nullifying and aligning the interference. In each scenario, an outer bound on the DoF region is derived. Then, we evaluate how cognition can improve the maximum achievable DoF by... 

As artificial intelligence (AI) applications continue to proliferate, the demand for deep neural network (DNN) models is on the rise. Many applications employing DNNs are sensitive to processing time or require significant processing power for their inputs. One current method for providing DNN-based services is through edge computing ser- vices. These services have less processing power compared to cloud services but can offer greater accessibility. While DNN models deployed at the edge offer promising prospects for delivering AI services with minimal latency, their potential for collabora- tion remains largely unexplored. In this thesis, we propose that DNN service providers pool their...