Mobile Computing and Wireless Communication - Essay Example

Chapter Introduction We have come a long way since the days of the valve and ENIAC computer. The irrepressible Integrated Circuit (IC) family happens to be the key ingredient in this revolution. Recent advances in mobile computing [1] and wireless communication are the focus of study for this paper. The mobile instruments are now coming in a range of shapes and sizes with features adding up day by day, thanks to the continuing trend towards integration of more and more electronic circuits. This trend of miniaturizing Integrated Circuits has led to the availability of low-power embedded processors and radios as well, further strengthening the communication ability of the man on the move. This exploratory approach towards making things small yet sophisticated promises much more for the future. While analyzing the evolution [2] of mobile technology, it can be safely assumed that the size and power consumption patterns are on a decline while the bandwidths available for communication will continue to increase. Such trends lead us to believe that mobile communication is bound to play an increasingly important role in everyday life through a variety of new applications often referred to as "ubiquitous or pervasive" computing. The emergence of powerful portable computing devices, along with the advances in wireless communication technologies, has made mobile computing a reality. Sensor Networks: One typical application for mobile communication is the sensor network, used for gathering information about the surroundings of the mobile equipment. Such a network consists of small integrated devices scattered over a specified area in order to collect and share information. The goal in undertaking such exercise is to discretely observe, augment, analyze or control an environment in an automated manner. Mobile networks got further enhancement with the help of wireless communication and sophisticated sensors. Technologically advanced mobile networks in turn will enhance the environments in which we live and work. As our dependence increases on these devices, it is important to design the devices in such a way that they appear to be self organized whenever we have to deal with such devices, and do not need to be planned every time, ahead of use. Existing wireless networks for mobile phones etc. work on the basis of fixed cell based infrastructure. Coverage is provided by base stations which manage the radio resources from a central location thus integrating the services. Depending upon the area being covered by the cell, certain amount of bandwidth is allocated for mobile subscribers. These cells make use of location-based routing protocols for transfer of information packets from sources to destination. Depending upon the bandwidth allocated, an upper limit is fixed for the number of subscribers that one cell can support. Sensor networks, in particular require to support a large number of 'subscribers', resulting in an increase in density per cell. Such networks are often used for short-range temporary communication. Therefore, implementing an infrastructure supporting such a large sensor network becomes very expensive. Mobile ad hoc networks [3, 4] on the other hand allow direct communication between wireless devices operating within the nearby areas and short distances of each other. Such devices prove to be very useful in disaster prone areas. In case of disaster, whole communication infrastructure may be destroyed. In such cases the intercommunicating ability of mobile ad-hoc networks within short range proves to be very useful. For communicating between two mobile devices which are located beyond range are the method of forward messaging messages via intermediate neighbors is used. Decentralization of an ad hoc communication network also eliminates the need for a fixed infrastructure, and results in greater cost reductions. The most widespread notion of a mobile ad hoc network is a network formed without any central administration which consists of mobile nodes that use a wireless interface to send data in packet form. The issue of routing information packets between any pair of nodes becomes a challenging task because the nodes can move randomly within the network. 1.1 Incentive Though the mobile devices work on Integrated Circuits and hybrid technology, which consume lesser energy, but A wireless sensor node, an all important connecting link, consumes much more energy. Therefore continuance of the energy supply is very critical for the life of a wireless sensor node. In order to maximize the network life, usage of the sensor node therefore, needs to be optimized. Today, many people carry numerous portable devices, such as laptops, mobile phones, PDAs and mp3 players, for use in their professional and private lives. Ad hoc networks allow such computing devices to be used in new embedded applications with the help of ad hoc multi-hop networks. Even a decade ago the human being could not have imagined such an intelligent use of technology. The embedded systems, making all this possible, are collectively referred to as sensor networks [5, 6, and 7]. A sensor is a small device that gathers information from a variety of sources by interacting with its surroundings and then analyzes the information to automate a task. Sensors can be used in a broad variety of applications, like emergency services, roadside assistance, billing, navigation, tracking, and so on in different fields like medicine, agriculture, industry etc. to alert them by generating an appropriate message. Sensors often prove to be very helpful in automated systems, where some of the minor glitches can be fixed by the sensor itself, without the need for manual intervention. Therefore sensors are supposed to have a two fold task. One, using these devices, real time information can be gathered about the functioning of a system/ network and in case a human intervention is required to set right some problem, the sensor can send across alerting signals. Secondly, sensor keeps an eye continuously on the functioning of the system, and applies corrective measures on its own, whenever it finds need for some minor adjustments. Such an optimization would have been impractical to accomplish manually. Sensor networks can be used in a variety of application in our day-to-day functioning. For example, we may want to create database based on the sensor networks to track the health and wellbeing of people. The sensor network will remain functional throughout the daily routine of the human beings and will keep passing on the information to the network as they conduct their routine works. This way we will come to know about their heart rate, blood pressure, and blood sugar levels, during different times of the day, while performing different sets of tasks. This way we will come to know about any potential problems before they acquire serious proportions. There are numerous other applications of such mobile networks, the challenge is to develop a communication network that can efficiently operate and provide dependable information in versatile, mobile environments without failure. Another very crucial application for sensors and ad hoc networks is to provide alternative communication systems during the times of natural disaster, or terrorist attack when the existing infrastructure has failed. Sensors like radars, weather forecasting gadgets etc. can also provide early warnings of natural disasters, so that we can make all possible arrangements for preventing loss of lives and property.. In the emerging scenario, role of ad hoc networks appears to be much more promising with their contribution being solicited in intelligent and expert systems. Nanotechnology, the emerging field for robotics and automation may prove to be a happy hunting ground for ad hoc networks. 1.2 Problem Statement Modern IC based devices consumer very less power, as compared to their version of vacuum tube valves and transistors. A single battery cell often proves to be sufficient for powering a set of sensors for the duration of their lifetime. As the sensors are required to be very handy and portable, therefore the limitations of the power cell impose size and cost constraints on the device. The simplest radio attenuation model [8] suggests that power consumption increases with the square of the broadcast range. Limiting the communication range can also save significant power. Unfortunately, it can also make the infrastructure prohibitively expensive and impractical due to the physical constraints that are imposed by the medium. The solution is to allow sensors to organize themselves into ad hoc networks that can adapt to mobility. Many existing routing protocols are designed to scale in networks of a few hundred nodes. They rely on state concerning all links in the network or links on a route between a source and a destination. However, short-range communication can often pose problems, since connectivity becomes transient, which effectively defeats the purpose of setting up a dependable network. This makes routing messages over multiple hops a more difficult goal to achieve. Traditional routing algorithms [9, 10] work on the basis of a routing table, that pairs each destination with a corresponding next-hop and link cost. This way a shortest path is identified [11] by exchanging topological data. Further movement of packets depends upon the route table. Packets are forwarded by consulting the route table for the next-hop. With this approach a shortest path is established between the source and destination, thus saving the time and energy. Continuous mobility causes the next-hop node to be disconnected while connecting on to the next node. This way the nodes keep moving in and out of range. This sometimes results in breaking of the established path. The network again tries to reconstruct the path, which results in extra network traffic as new route table is constructed by the system. In case of high recurrence of such breakdowns, the overhead costs of routing can dominate the traffic load and cause congestion and consumption of precious energy. Moreover, in large-scale networks with a large number of sensor nodes, multiple nodes should be deployed, not only to increase the manageability of the network, but also to reduce the energy dissipation at each node. In this thesis, we will investigate problems that are related with locating multiple nodes in the sensor network area. Then, we investigate the use of multi-hop communication links and compare the amount of energy gain upon alternative routes using analytical techniques. 1.3 Hypothesis Ad hoc networks, also known as infrastructure-less networks have no fixed routers; all nodes are capable of movement and can be connected dynamically in an arbitrary manner. For routing the packets and communication through different nodes and hops, traditionally the routing algorithms works by maintaining a comprehensive table of nodes and connectiions consisting of final as well as interim destinations together with the next hop addresses. As soon as the message arrives at a router, the routing algorithm looks for the next destination in the route table and accordingly forwards it to the associated next hop. Such table-driven routing protocols maintain consistent and up-to-date routing information from each node to every other node in the network. The exchange of information between switches to determine the shortest path from source to destination is recorded in the form of route tables. Such an information recored in a route table remains accurate till there's a change in the network topology. An experiment was conducted to measure the distribution of nodes leaving the range over time. The experiment sampled 100 nodes moving randomly in a space 100m100m at a speed of 1ms1 over a period of 1000 seconds. The results are illustrated in Figure 1.1 for broadcast ranges of up to 20m. These results corroborate [17] other findings on connectivity, behavior, and network characteristics. It is my hypothesis that the routing algorithm will chose the best routing path based on the lowest energy cost amongst the available alternative paths. This can be accomplished by incurring the cost of energy according to the proposed solution from the source nodes to the destination, rather than the shortest path with each topological change. I propose to study routing cost and power consumption in different scalable solutions. 1.4 Goals The primary goal of this paper is to improve the understanding of mobile ad hoc networks by completing the following tasks: Study of an ad hoc routing algorithm that is designed to minimize the overall power consumption by selecting the lowest energy route cost of the alternative routes. Development of a simulation toolkit for producing continuous motion interleaved with data communication at high resolution, allowing different ad hoc networks to be evaluated in a controlled yet realistic environment. Definition of metrics to quantify route overhead, by taking into account the attenuation caused by the range. The metrics will allow different solutions to be compared objectively with respect to energy. Presentation of results to demonstrate that the algorithm will select the lowest overhead energy cost. Analysis of existing ad hoc routing algorithms to demonstrate why it will conserve energy when used in highly dynamic environments. 1.5 Outline The remainder of this paper is organized into chapters exploring the concepts behind ad hoc routing algorithms for mobile wireless networks. A new study of routing algorithms is proposed to reduce power consumption and to improve scalability by taking the unique properties of highly dynamic mobile networks into consideration. Chapter 2 provides background information on the physical organization of existing data networks. The subjects covered in this chapter include packet switching, and distributed routing algorithms for networks with stable topologies. Chapter 3 covers Mobile Ad Hoc Networks MANET the history behind the evolution of wireless communication from packet radio to cellular and ad hoc networks, focusing on issues related to packet forwarding and routing. In addition this chapter will also include a comparison of MANETs with other networks like GPRS, GSM, CDMA, VANETs, Bluetooth etc. Chapter 4 covers Ad hoc Routing Algorithms, Table-Driven Routing, Tagging-Driven Forward, leading to lower power consumption and greater reach of ad hoc network. Chapter 5 covers the model power and experimental hypotheses. Read More