core. The second stage is the programming of network elements, while the third stage relates to the configuration of radio parameters.
2.2.3 Self-Optimization
Mobile networks are dynamic in nature. This pertains to traffic characteristics, the volume and variability of data exchanged between network elements, the joining of new users, the leave of others, and the movement of users among network cells. This results in variations of network performance as well as the level of service that users are experiencing. Therefore, self-optimization aims to maintain an optimal performance level for all network elements, through analysis of data measured and exchanged by network elements.
2.2.4 Self-Healing
The larger the network size, the more likely that failures will occur. The objective of self-healing is to continuously monitor the network in order to automatically detect and recover from unexpected possible failures. In future networks, it is expected that self-healing enables the network to predict faults and automatically take the necessary measures to avoid service degradation and disruptions.
2.2.5 Key Performance Indicators
KPIs are simple indicators that represent network performance. Here, we present examples of some important indicators:
Channel Quality Indicator: This represents the connection quality to all users in a cell. Obstacles and multipath fading are major factors that impact channel quality.
Handover Rate Indicator: This represents the mobility pattern of network users. It indicates the signaling traffic on the backbone network units which affects the overall network performance.
Cell Load Indicator: This represents the amount of load on a cell, in terms of users, traffic load, or a cost function.
Quality of Experience (QoE): This represents the satisfaction level of all users in the network or within each cell. Such indicator would characterize the QoS level users are experiencing.
2.2.6 SON Functions
It is important to discuss the fundamental optimization tasks of SONs. In this section, we present some important tasks:
Coverage: Coverage optimization is a process through which a network tries to cover an intended area with minimal number of base stations and transmit power levels.
Capacity: Capacity optimization refers to the process of providing users with the best possible QoS using minimal radio resources. This would imply radio frequency assignment and interference mitigation techniques.
Mobility: Mobility optimization deals with the process of ensuring transparent user movement between cells and at the same time minimizing the number of unnecessary handover requests.
Load Balancing: This refers to the process of distributing the load among network base stations, trying to maximize the QoE in the network and minimize the overhead on core network elements.
2.3 Cognitive Networks
Nowadays, communication networks are getting more complex and their configuration and management to achieve performance goals have become a challenging task. This is due to the following:
The significant increase in the number of network users.
The increase of the number of required networking elements at the network core.
The huge number of mobile applications.
The diversity of traffic.
The idea of cognitive networks is to improve the performance of networks and reduce the effort required for their configuration and management. Unlike current technologies, in which networking elements are unable to make intelligent decisions, the elements of a cognitive network have the ability to learn and dynamically self-adjust as response to changing channel and network conditions. Cognitive network elements utilize the principles of logic and learning in order to improve performance. Decisions are made to improve the overall network performance, rather than the performance of individual network elements. Thus, cognitive networks achieve the goal of intelligent, self-adjustment, and improved network performance, by intelligently finding optimal values of many adjustable parameters. They are required to learn the relationships among network parameters of the entire protocol stack.
As we indicated, a cognitive network should provide better performance to users. The cognition can be used to improve: utilization of network resources, QoS, security, access, control, or any other issue related to network management.
It must be emphasized that cognition is not only related to wireless networks, but also the idea applies to the management of network infrastructure and the various network elements [3]. To stimulate transition to cognitive networks, their performance must outweigh all additional complexities that they require. The question is how to measure the cost of a cognitive network. Such cost would primarily depend on the communications required to apply cognition, the architecture complexity, maintenance cost, and the operational complexity. For example, in wired networks, user’s behavior is clear and easily predictable, and therefore, it may not be interesting for some people to employ cognition with this type of networks. On the contrary, wireless networks often include heterogeneous elements and have characteristics that cannot be easily predicted, making them the best candidates to adopt the cognition concept.
Cognitive networks should use different measures, tools, and patterns as inputs to the decision-making processes. Then, they come up with results in the form of procedures or commands that can be implemented in modifiable network elements. It is important to note that the cognitive network must adapt to changes in the environment in which it operates and anticipate problems before they occur. Their architecture must be flexible, scalable and be supportive of future improvements and extensions.
Several research studies have been discussing the architecture and functionalities of cognitive networks. There is a need to rethink about network intelligence from being dependent on resource management to understanding the needs of network users and then transferring intelligence also to the elements of the network.
The central mechanism of the cognitive network is the cognitive process. This process implements real learning and decides the appropriate responses and actions based on observations in the network. The operation of the cognitive process mainly depends on whether its implementation is central or distributive as well as on the amount of state network information.
2.4 Introduction to Machine Learning
ML is a subset of AI. The aim of ML is to develop algorithms that can learn from data and solve specific problems in some context as human do [4]. ML has been proving its ability to overcome the challenges and complexities of mathematical formulation and solution of complex problems, including wired and wireless networking problems that require effective methods to quickly respond to dynamical changes of channels as well as the increasing diversification of services. Dynamic ML algorithms are able to process data and learn from it. They are replacement of complex algorithms which are written in a fixed way to conduct specific tasks.
The basic concept of ML is through training data that is used as input to the learning algorithm. The learning algorithm then produces a new set of rules, based on inferences from data, which
