Unmanned Aerial Vehicles



In recent times, there is an increasing impetus to research about other possible non - military applications for Unmanned Aerial Vehicles, UAVs. This growing interest has is not only shared by scientists but governments as well, in the United Kingdom at least. The Welsh Assembly Government (WAG) is currently interested and driving any academic research in this regard. This investigation is a direct testimony to that.

Against this backdrop, this dissertation building on the already established UAV concept, considers how key aspects of learning objectives achieved during the taught part of the programme (PART ONE) can be combined with the aspirations driving current UAV research trend and by so doing take up some of the challenges prevalent in our society today.


Communication is perhaps the singular identifiable area that has been transformed in leaps and bounds at a phenomenal pace within the last half century. Even though technology has improved communication within remote circumstances, its role in transforming the world into a ‘global village' even in (near) real-time is a unique wonder in itself.

As it were, communication is one of the key elements vital for the renowned impact of UAVs till date. As such, the nature, operation and effectiveness of the communication component is vital for its successful utilization in military and non-military applications. Within the same context, there is increasing research into communicating approaches that can be utilized with the now evolving wireless technology which is driving domestic, business as well as industrial solutions at present and for the foreseeable future. As a result, any efforts aimed at seeking other UAV applications must take these developments into consideration.

In undertaking this investigation, this dissertation will focus on the challenge of disaster response. Despite the rising impact of modern technologies in minimizing the occurrence of disasters, there still seems to be the inability to achieve commensurate success in the search and rescue of trapped victims in the aftermath of a disaster, a situation that accounts for the loss of lives many of which are neither replaceable nor quantifiable. Any promising outcomes, even if interim would imply a stepping stone in the attainment of what can be considered as a priceless solution to the enhancing disaster response thereby reducing the loss of life and property.

Development in the 21st century is driven by wireless communication which has given rise to concepts like wireless sensor networks (WSN), mobile ad-hoc wireless networks. These technologies hold the promise of the future, with the capability to establish networks at anytime, anywhere. These networks don't rely on extraneous hardware, which makes them an ideal candidate for rescue and emergency operations [Datta et al, 2006] 2711a. Even though the wireless concept is not entirely new, the advancements attained following the introduction of technologies such as Bluetooth, IEEE 802.11 and hyperlan are helping enable eventual commercial MANET deployments outside the military domain [Jayakumar and Gopinath, 2007].


Having explored the opportunity to undertake an industry-driven or sponsored research without success, this work is undertaken inspired by two secondary but ultimately fulfilling factors. The first is the opportunity to expand my knowledge of wireless protocols; an area very little is known about but yet is set to drive technological solutions and applications in the 21st century and beyond. The second motivation for embarking on this work was for the opportunity to explore how one's quest for knowledge of science and technology could be utilized for the service of common good. The project offered the chance to adopt a different approach to on-going efforts to support and promote various humanitarian efforts of Non Governmental Organizations such as Oxfam GB in offering relief and support to victims of disasters.


This work is aimed at investigating other possible non-military applications of Unmanned Aerial Vehicles especially for civil and commercial purposes. In accomplishing this, the following are the key objectives;

* Identify how existing communication technologies and infrastructure approaches (UAVs inclusive) can be explored, adapted and / or enhanced to aid disaster response

* Demonstrate a Master level understands of the taught contents of the program.

* That the research falls within the scope of the modules covered in the programme and that it reflects appreciable knowledge application to everyday life


This research is by and large an investigation and as such its approach is somewhat peculiar. There are two key elements covered in the study; unmanned aerial vehicles and sensor technology; the networks in which they can operate as well as the communication patterns they utilize. The concept of the former has been developed and already finds a variety of military applications. As such, a considerable amount of this research shall focus on the latter.

Indeed, the key aspect to attaining success is to establish a communication pattern that can remotely link-up the different aspects in real-time or as near to it as possible. In other words, finding the appropriate routing protocol is crux of the matter. In addressing this, thorough analysis of several protocols was undertaken using secondary data. The data was for the most part was obtained from a variety of peer reviewed researches on a range of protocols undertaken by different authors. The protocols were examined in order to ascertain choice performance characteristics over a wide range of operational conditions in a bid to meet the expected performance criteria required for the desired outcome.


Although the thought out procedures for undertaking this investigation were painstakingly adhered to, a number of factors identified beforehand influenced the approach and possibly the extent to which the outcome(s) of the investigation can be built upon. One of these was the inability to liaise with disaster management organizations especially in the world's most vulnerable areas in order to obtain primary data as well as other vital information. Secondly, the inability to explore real-life operational conditions made so by logistic constrains in simulating sensor nodes of commensurate magnitude. Lastly, this work takes cognizance of its inability to develop a scope that covers much of the learning objectives of the management modules.


The remainder of the report is presented as follows. Chapter two gives a brief and concise overview of the area of proposed application; disaster management. It offers background information that supports the choice topic. This is followed in chapter three with a review of some information / communication technologies existent today and what role they play in disaster management. The penultimate chapter is an in-depth study on the proposed applications. It focuses on the two main component parts; UAVs and sensor networks with greater focus on the protocols associated with the workability of the latter. A summary of the analysis is presented in chapter five which concludes the work.


An Overview

Discussing the concept of disaster management presupposes that there is an understanding of term disaster and other associated terms such as hazard, emergency, risks, vulnerability e.t.c. As such, this discourse shall begin with a clarification of some of these terminologies.


Emergencies: - In many instances, the term disaster and emergency are wrongly interchanged. Similarly, hazards are sometimes regarded as disasters; however all these terminologies are all clearly distinct. According to the World Health Organization, WHO

“An emergency is a situation where local resources and responses suffice in addressing the sudden occurrence of an incident or event”.

Usually, outside help is not required. On the other hand, disasters cannot be dealt with locally and require external assistance.

Hazards: - Hazards are often generally mistaken as disaster themselves but in the disaster management context, they could best be referred to as necessary disaster causes. While earthquakes, tsunamis, floods, cyclones e.t.c. are generally considered as hazards, there are certain circumstances with which they must be associated with before they in fact turn out to be disasters. Simply put, a hazard is that which could cause a potential disaster. Within the context of this discourse, a hazard as defined the WHO is

“A dangerous phenomenon, substance, human activity or condition that may cause loss of life, injury or other health impact, property damage, loss of livelihoods and services, social and economic disruption, or environmental damage”.

Understandably, there two kinds of hazards giving rise to two kinds of disasters, natural and man-made;

Natural hazards as previously illustrated are things like earthquakes, volcanic eruptions, landslides, tsunamis, floods and drought e.t.c. They refer to any naturally occurring physical events. They are caused by changes in the atmosphere, the earth's exterior, the sea or other body of water. They can happen in either of two ways; quickly or gradual build up. On the other hand, manmade hazards are due to human carelessness. Manmade hazards are connected with industries or energy generation facilities. Other examples in addition to oil spillage include explosions, leakage of toxic waste, pollution, dam failure, wars or civil strife etc. (Dey and Singh, 2006)

Disasters: - there a number of definitions of for disasters put forth by different schools of thought. One popular definition given by Quarentelly and cited by Thirunavukarasu (2009) is

“A crises situation that exceeds local capabilities”

Nowadays, the spate and impact of disasters has broader dimensions giving rise to a truly reflective definition as put forth by WHO: -

“A serious disruption in the functioning of a community or society causing widespread material, economic and social or environmental losses which affect the ability of those affected to cope using their own resources”.

Following from the above definition, it can be seen that regardless of the origin, whether natural or human origin, hazards are regarded as disasters only when it results in damage of property, injury or loss of lives. To buttress this definition, a high magnitude earthquake in an uninhabited desert cannot be regarded as a disaster. In other words, a disaster happens when a hazard impacts on a vulnerable population causing damage, casualties or disruption. (Dey and Singh, 2006). If the disaster magnitude is enormous, it is referred to as a ‘catastrophe'


Evidently, not all hazards lead to disasters. In actual fact, preventing disasters is becoming increasing possible owing to advancements in scientific knowledge and technological know-how notably sensing technology, which is the core focus of this work. Efforts in this regard form the basis for increasing appreciation and growing involvement disaster management.

According to Quarentelly, Disaster management can be defined as

“Actions taken in anticipation of or in response to unexpected events that adversely affects people or resources, threatening the continued operation of the organization”.

In order to fit our context, such unexpected events would refer to natural or artificial hazards. Disaster management is the discipline of dealing with and avoiding risks. It involves preparing for a disaster before it breaks out, disaster response (e.g. emergency evacuation, quarantine, mass decontamination, etc.), as well as supporting, and rebuilding society after natural or human-made disasters have occurred.


There exist different classification approaches for breaking down the disaster management process. Nevertheless, disaster management as it is practised today necessarily consist of four components namely Mitigation, Preparedness, Response and Recovery (Hollister & Yodmani, 2001). Some schools of thought would further classify Recovery into Relief, Reconstruction and Rehabilitation.


Mitigation process refers to efforts aimed at preventing hazards developing into disasters altogether, or reducing their impact on occurrence. Mitigation focuses on long-term measures for minimizing or eliminating risk. According to Quarentelly, Mitigation is considered genuine only if it is actualized prior a disaster occurring, nevertheless, any risk eliminating or reducing strategy overtime is referred to as mitigation. Mitigation is the most cost-efficient method for reducing the impact of hazards.


Preparedness is a state of being ready to react promptly and effectively in the event of an emergency. This is the phase that tinkers action plans to be carried out on the outbreak of a disaster. Being prepared implies that a plan of action exists such that it is clear as to what to do before a disaster or emergency occurs.

Preparedness procedures to be undertaken depend on the evaluation of hazard severity and vulnerability. This is also the basis for deciding mitigation strategy. Some imminent, potential disasters allow reasonable time to prepare for, others allow no time at all. Preparedness for either of these requires a plan. It is essential to identify the available resources, and the most effective ways to utilize them.


The response phase comprises of actions carried out during and immediately following the breakout a disaster. They are designed to provide immediate and urgent assistance to victims of the event and lessen the likelihood of secondary damage.

The response phase includes the mobilization of essential disaster response services and first responders to the disaster site. This would usually include a range of core emergency services, such as firefighters, police and ambulance crews. They may be supported by a number of secondary emergency services, such as specialist rescue teams.

The five basic stages of response to a disaster are

(i) Notification/ Warning,

(ii) Immediate Public Safety,

(iii) Property Security,

(iv) Public Welfare, and

(v) Restoration.


Recovery is the last phase of disaster management process. Recovery continues pending all systems return to normal, or near normal. The objective of this phase is to return the affected area to its prior state. This is dissimilar to the response phase in that recovery efforts in the former are concerned with issues and decisions that must be made after immediate needs are addressed.

There are two aspects of recovery, Short term and long term recovery. The first seeks to ensure that vital life support systems are restored to minimum operating standards. On the other hand, Long-term recovery may continue for a number of years, as the community slowly returns to pre-emergency or better conditions.

As 1212a points out, they are no two distinct recovery phases. Long-term recovery is simply refers those efforts, which are still remaining operational long after a disaster. It encompasses everything from total redevelopment of the disaster area to mitigation efforts aimed at preventing future occurrences


The consequences of one disaster whenever and wherever it occurs are in many respects priceless. Indeed, in many cases, the losses are irreplaceable. This point is further driven home looking at the statistics. According to CRED[2009], there were 354 natural disasters in 2008 in which 236,000 people died with 211,628,186 affected. This compares to the average for 2000-2007 of 397 natural disasters, with 66,812 deaths and a total of 231,588,104 people affected.

Without doubt, the Asian continent is the most vulnerable when it comes to natural disasters. This fact is supported with UN Environment protection group 0612a. In 2008, there were 141 events including, Cyclone Nargis in Burma which left nearly 140,000 people dead (or missing, presumed dead) and an earthquake in Sichuan, China killing 87,476 which made up 93% of the global disaster toll. Notably, 2008 was the deadliest year since 2004, the year of the Indian Ocean tsunami. [CRED, 2009]

Now more than ever before, there is increasing attention and commensurate efforts in matters associated with disaster management spread across all its components. Unsurprisingly, disaster management efforts are championed by leading international organizations such as the United Nations UN, International Telecommunications Union ITU, International Red Cross e.t.c. The UN remains the fulcrum for all the recent developments in disaster management especially in the area of disaster risk reduction which are largely concerned with preparedness and mitigation efforts.

Following the spate of disaster occurrences, she declared 1990-2000, the International Decade for Natural Disaster Reduction. Following from that, the second Wednesday in October was set aside as International Day for Natural Disaster Reduction. This seeks to highlight the urgent need for increased prevention activities to reduce loss of life, damage to property, infrastructure and environment, and the social and economic disruption caused by natural disasters. [CRED, 2009]

Asides the United Nations, there are hundreds of governments, establishments and organizations which support and engage in the disaster management cause. The greatest majority of these are possibly found in the area of recovery offering relief materials and resources to victims of disasters in different parts of the world. Another aspect of the cycle that is also receiving global attention is the area response. Modern day response to disasters is ineffective without state-of-the-art equipment and devices. The tremendous progress made in this regard is a fair reflection of works the International Telecommunications Union, some of which would be highlighted in the next section and indeed is the core focus of this dissertation.


A Literature Review

There exists a rich variety of modern infrastructure and technologies in use today spread across different areas of human endeavour. A number of these cutting-edge innovations are as a result of research that was originally driven by military concerns. Amongst many other resultant discoveries, Unmanned Aerial Vehicles, UAVs and Sensors are two key elements from those researches which are considered in this investigation. However, this section shall review particular applications as it relates to disaster management.

In undertaking this review, a relational approach is adopted in order to offer a better background understanding. This approach highlights different applications associated with different aspects of disaster management. As earlier pointed out, the disaster management process ideally consists of a number of aspects, of which four are core. These include; Preparedness, Mitigation, Response and Recovery [Hollister and Yodmani, 2001]. This discourse shall be limited to this four.


3.1.1 Mitigation; - this is the first and most important approach in reducing disasters, but it is equally a sound financial strategy. Azongi (2007) highlights the point made by Mutashubirwa, who opines that every dollar spent on disaster mitigation, saves up to 3 or 5 dollars worth of economic losses whenever a disaster or catastrophe strikes.

Radio communications provides a number of applications in disaster management. One emerging application is one remote sensing. Remote sensing is without doubt one area that has over the years become increasingly vital and useful. It offers many societal benefits over a broad range. Nevertheless, efforts in recent times have understandably been directed towards the application of radio communications, in general, for the purpose of disaster prediction, detection and mitigation (VonDeak, 2007). These efforts are coordinated by a specialized agency of the United Nations UN, International Telecommunications Union, ITU.

According to Short (2009), the term "remote sensing" was coined in the mid-1950's by Ms Evelyn Pruitt, a Geographer and Oceanographer, with the U.S. Office of Naval Research (ONR) outside Washington, D.C. to take into account new views from space gotten by the early meteorological satellites which were obviously more "remote" to their targets. This was used to distinguish the aerial photos from airplanes obtained up until then and was the medium for recording images of the Earth's surface. 3.1 is a pictorial illustration of a remote sensing system on an orbiting satellite, showing what it is looking at and how it gathers its data

In its literal sense, remote sensing is defined as the measurement or attainment of information of some characteristic of an entity or phenomenon, with the aid of a recording device that is not physically linked or in direct contact with the entity or phenomenon being observed e.g., the use at a distance (such as from aircraft, spacecraft, or ship) of any gadget or apparatus and its consequent use for gathering information relating to the environment, such as measurements of force fields, electromagnetic radiation, or acoustic energy. The technique employs such devices as the camera, lasers, and radio frequency receivers, radar systems, sonar, seismographs, gravimeters, magnetometers, and scintillation counters.

Sensing systems are usually of two kinds; space-based or terrestrial. Regardless, any sensing system is classified by ITU under radio regulations (article one) into two distinct categories, active and passive. The former obtains information by transmission and reception of radio waves while in the latter; it is by reception of radio waves of natural origin. Largely as a result of breakthroughs in sensing technologies, there have been advancements in earth observation systems and telemetry such as the use stream gauging stations e.t.c. These have subsequently improved and increased its applications in disaster management.

Two noteworthy applications are modelling and simulation. Remotely sensed images can be used to generate digital terrain models for areas susceptible to flooding. Undertaking this introduces the possibility for simulation of potentially disastrous conditions and the consequent identification of vulnerable areas. In addition it is possible to undertake mapping of sequential inundation phases, including the duration, depth of inundation, affected areas and direction of current. Similar advancements in telemetry associated with ocean and atmosphere observation systems have also contributed to upgrading in climate prediction. (Hollister and Yodamani, 2001)

3.1.2 Geographic Information Systems (GIS)

“Using a GIS, it is possible to pinpoint hazard trends and start to evaluate the consequences of potential emergencies or disasters. When hazards are viewed with other map data, such as buildings, residential areas, rivers and waterways, streets, pipelines, power lines, storage facilities, forests, etc…” (Wattegama, 2007). Geography information system is almost an all encompassing approach to managing disasters. It is detailed and actively used in warning systems and planning during disasters. It depends strongly on accuracy of data and accurately simulated models. It usually contains large database and detailed simulated features. With these characteristics, it is a tool for preparing and monitoring of disasters.

According to GIS website, “GIS can integrate and relate any data with a spatial component, regardless of the source of the data. For example, you can combine the location of mobile workers, located in real-time by GPS devices, in relation to customers' homes, located by address and derived from your customer database.” It is a powerful tool for generating information for various clients. GIS helps in predicting, monitoring and managing disasters. As an integrative tool, so much information can be generated and shared easily among stakeholders in disaster management. Expertise in the use of GIS is of essence and can be accomplished by capacity building among disaster prone communities. This will obviously go a long way in mitigating the effects of disasters. Ice blasts, cyclones, hurricanes, desert encroachment, wild fires, tsunamis and overall changes in climate are increasingly being predicted accurately. It is also a solution provider.


3.2.1 Radios: -

As already highlighted, radio communications plays a prominent role in disaster management. It offers a range of applications which is certainly critical to some aspects if not the entire disaster management process. Based on the insight offered in the previous chapter, it can be understood that natural hazards are unavoidable. Nevertheless, with radio communication technologies, it is possible to have equipped and organized setups that link in one go emergency operation centres, broadcast systems as well as emergency personnel. Using a network of radio communication facilities, it is possible to reduce a minimum, the undesirable consequences.

Radios can be used to minimize the impact of cyclones which ranks highest in occurrence for national disasters (Cf. Appendix II). According to Hollister and Yodmani [2001], radio communication is the fulcrum for the internationally recognized program for managing cyclones, Cyclone Preparedness Program as used in Bangladesh. It operates an extensive network of communication facilities namely; Radios (HF & VHF), VHF transceivers, transistor radios. This infrastructure network ensures that high-risk areas are continuously monitored. Information about potential cyclones is relayed to the communication centre which in turn communicated using appropriate broadcast avenues.

Furthermore, Obi and Iwasaki [2005] cites the introduction of some kind of digital radio receivers to aid the effectiveness of early warning systems. In this system, the radios would simply come on once emergencies are announced from terrestrial or satellite bases. This can happen even when the radio is off.

3.2.2 Amateur Radios: -

Regardless of role and impact of radios in disaster preparedness, there are circumstances in which it could turn out ineffective. Coile [1997] highlights amongst others the 1989 earthquake in Santa Cruz, California as one clear incident. On that occasion, medical helicopters from Stanford medical centre meant for emergency rescue missions couldn't be used because the helicopters couldn't establish radio communication with the hospital and as such, the former was unable to get landing clearance. This situation was overcome with the aid of amateur radios.

Amateur Radio has its roots in the 19th Century but was developed along with radio communications in the 20th Century [RSGB, 2009]. They are two-way radio system that receives on one frequency and then re-transmits simultaneously on another frequency what it hears. Amateur radio is not as dependent on terrestrial infrastructure that can fail. Amateur Radios operate using shortwave frequencies, unlike regular AM/FM broadcast frequencies and as such travel are able to travel throughout the ionosphere. Amateur radio is considered principally as a disaster communications tool. It serves as back-up when everything else fails.

Over time, the amateur radio (also referred to as hams) has remained a vital necessity in emergency rescue missions in the United States and beyond. With respect to the 1989 California earthquake, hams relayed the landing instructions to the Stanford helicopters using home bases. According to Colie [1997], it was also used to coordinate the arrival structural engineers as well as serve as a means of communication in Red Cross shelters for a week.

There exist other more recent events where the amateur radio proved priceless; the 2001attackson theWorld Trade Centre New York, the2003 North America blackoutas well as Hurricane Katrinain September 2005, In all of these American natural disasters, amateur radios was used to coordinate disaster relief activities when other systems failed.

Hams also find prominent use on the Asian continent which has already been identified without doubt as the most susceptible and vulnerable of all the continents in the world. In 2004, the earthquake and resulting tsunamiacross theIndian Oceanwiped out all communications with theAndaman Islands, except for aDX-peditionthat provided a means to coordinate relief efforts. Recently, Amateur Radio operators in thePeople's Republic of Chinaprovided emergency communications [Wikipedia, 2009]. At the time of writing this review in September, another earthquake had just hit Indonesia and Samoa and yet again the amateur radios were put to good use.



The role of broadcast systems in our everyday lives is inestimable. In disaster response scenarios, its role is even more crucial. Owing to broadcast systems, it's now increasingly possible to communicate natural disaster occurrences in real-time. Thereby, immediately bringing about response, support and relief. Nevertheless, as events in the recent Samoa earthquake on September 30, 2009 have shown, current efforts are needed in real-time assessment of the impact such natural disasters and consequently a satisfactory coordination of rescue efforts that ought to follow.

Broadcast systems support disaster management activities in a number of ways which are phenomenal. Amongst the many broadcast tools in use, there are which are feature more prominently; television and radio. As already pointed out, the Asian continent is without doubt one with the highest probability of natural disasters. According to a survey study conducted by the International Telecommunications Union (ITU), of the 800 million people living in sub-Saharan Asia, one in four own/have access to radios (200 million); one in thirteen own/have access to TV sets (50 million); one in thirty three have fixed telephone lines (24 million); one in fifteen have GSM lines (53 million); one in one hundred and thirty have a PC (6 million); and only one in one hundred and thirty use the Internet (6 million) [Azongi 2007]. In other words, TV and radio are the two most prominent sources of information relating to natural disasters.

Broadcast systems have aided in no small measure in the recovery of communities. It has been used extensively in helping the usually traumatised communities in counselling and lending hand to internally displaced persons. An example is the use of radio in Meulaboh community in Aceh province of Indonesia in post 2004 tsunami as reported by Asia Pacific Development and cited by Wattegama [2007]. The programme is sponsored by UNDP.


More than any other technological innovation, satellites have by miles raised the speed of response to natural disasters. Essentially, satellites make possible the delivery of imagery and communication. With one coverage beam, a good part of the world can be served. Indeed, Arthur Clarke, a renowned inventor and author proposed a theory in 1945 that only three satellites are needed to cover the globe at any one time. [Charlesworth, 2008]

The role satellites play in disaster management is increasingly being acknowledged by the international community. Kassianides reports in Azongi [2007] that since 1999, space agencies work together in the provision of satellite imagery to nations in need following disasters. According to the report, satellites played a vital part in recovery efforts following the tsunami disaster in 2004. It allowed for satellite imagery, taken from several locations. Such imagery provided an accurate depiction of the extent of the damage caused by the tsunami disaster.

Going further, Kassianides points out that modern satellite technology allows for a portable infrastructure. This setup offers a viable alternative to the terrestrial options such as land lines which could easily go are down or become overloaded, GSM antennas are damaged, and/or GSM networks become saturated. On the contrary, portable satellite solutions, such as VSATs, e.t.c. can save the day. This is possible because a satellite dish can be installed anywhere, even in rubbles; the only prerequisite for it to function is a direct ‘line of sight'. In other words, the terminal on ground needs to ‘see' the satellite. For scenarios where roads remain intact, satellite vans can be stationed to assist in recovery communications operations.

Satellites generally offer two key resources, imagery and communication. While the former can be regarded as unique to satellites alone at least in real time, the same cannot be said for the latter. Terrestrial infrastructure, it could be argued possibly offer greater reliability for communications. Nevertheless, given the fact that their availability in disaster times is 99% of the time not guaranteed, satellites become priceless. It can provide essential communications within hours of a crisis, which would include a range of services, from vital co-ordination of relief efforts, to giving reassurance to family and friends. Furthermore, improving technologies allow for better services such as mobile communications, Voice and broadband data e.t.c.

All efforts and project associated with telecommunications is coordinated by the ITU. Over the years, the delicate nature of disaster management is receiving greater support bringing about increased partnerships leading to significant achievements. In one particular case, one of the strategic partner organisations, working in close cooperation with a number of recovery and space-based organisations, Télécom Sans Frontières (TSF) was appointed by United Nations to deploy satellite communications in Lebanon, August 2006.

It setup two telecoms centres, which supported 22 aid organisations and over 600 displaced families. Within a month, 3 gigabytes of data were sent and 1,850 minutes of calls were made. Some of the services provided in the aftermath of the disaster include the provision of logistics support for relief agencies, telephony, email, internet access, videoconferencing, instant messaging, telemedicine to connect with world-class trauma specialists. In addition, it provided Tele-education which helped to bring normality to displaced children or destroyed schools. It facilitated media coverage event a great deal. This went a long way to raise the awareness and profile of the disaster, as well as support to continuity of normal activities. This in turn supports the community Azongi [2007].


3.4.1 Amidst the ever growing list of services that internet offers, its role in the aftermath of a disaster is essential. Ordinarily, the internet is a huge information resource. The same is the case when recovering from a disaster. Usually, reconstruction takes centre stage and nowadays support in that regard can be mobilized from any part of the world using the internet. The internet offers a medium for prospective goodwill many of which are scattered and fragmented, can be channelled to good causes overcoming bureaucratic and formal bottlenecks.

Again, Hollister and Yodmani [2001] recount another vital application of the internet in disaster recovery. Following the Gujarat earthquake in 2001, a number of private web sites also sprung up. These offered highly localized damage and needs information. The government website was no longer the only information provider to the outside world willing to offer assistance.

3.4.2 Telephony as an early warning tool is very useful. It provides speedy access and people seem to heed to it more than broadcasts because of its personalized nature. This benefit notwithstanding, its penetration in developing countries is limited to only the rich minority sections of the population. During disasters, its effectiveness is usually limited as the base infrastructure may have been damaged. In a classic case, highlighted by Submaranian in Wattegama [2007], a call from an Indian in Singapore to a research institute in Nallavadu back in India saved four communities from the 2004 deadly tsunami. Alongside the calls made on telephones are Short Message Service (SMS). The two main advantages are their ability to go through even when networks are congested and it can be sent to many people at once. However, it is still dependent on network availability.

That case in point goes to show that the new communications technologies are indeed capable of transforming the nature of disaster communication networks from a conventional "command and control" type to a highly decentralized and perhaps multi-node network. In the long run, in those areas where new web-based technologies have percolated, this has a potential to redefine the role of the governments in disaster management. In a way, this is in consonance with the increasing recognition of the importance of community-based disaster management.


There are other information and communication technology applications or infrastructure not outlined in the above sections for want of space but still relevant in disaster management. These include; PABX, sirens, facsimile machines, satellite radios, walkie-talkies, beeper, CATV, PHS, sirens, community radios, cell broadcasting, emails, mobile telephony e.t.c. All these assist in one way another in disaster management.

Furthermore, even for the technology and infrastructure discussed above, many of these find relevance in more than one area. Indeed, all possibly do. For example, mobile phones could be actively employed in preparedness or response; satellites are utilized in mitigation such as earth observation as well as in recovery activities. Broadcast systems can be used in information dissemination which will facilitate better preparedness ahead of an impending disaster.


From the preceding discussion, it is quite obvious that technology in general has impacted tremendously in disaster management. This argument holds water largely because of advances in sensing technology and given this evolving trend, it could be argued further that sensors are bound to play an increasingly pivotal role in the future much more than the modern-day personal computer [Akyildiz et al, 2002].

Already outlined are some of applications of sensing technology in disaster management. There exist other areas where the technology finds important applications. A few of these are mentioned in Mobile Radio Communications, one of the modules covered during the taught part of this Master's program. According to Izuka (2009), some of these areas include:-

Military applications:- battlefield surveillance, targeting, battle damage assessment nuclear, biological & chemical attack detection, ammunition monitoring e.t.c.

Environmental applications; Earth & environment monitoring, forest fire detection, tracking of movement of birds, meteorological and geographical research e.t.c.

Health Applications;- integrated patient monitoring, tele-monitoring of human physiological data, personnel & patient monitoring, drug administration e.t.c.

Home Applications:- automation of home appliances (vacuum cleaners, refrigerators, VCRs, micro-wave ovens), Smart environments e.t.c.

Commercial Applications: - retail merchandizing, inventory management, vehicle tracking and monitoring, environmental control in office buildings, product quality monitoring e.t.c.

As previously pointed out, sensors in themselves are not a new concept as they have been there for quite a while. However, a great degree of advancements have emerged in recent times giving rise a wider variety of applications. The earliest application of sensors was based on the ‘sense perception concept'. This implied that the sensors had to be in physical contact with the phenomenon being monitored. Furthermore, traditional sensors as they are often referred to were ‘standalone' meaning that they would usually work on their own. This is contrastingly different to sensors as we have them today. This growing dynamism in the feature and applicability of sensors and their possible application is considered in detail in the remainder of this work.


Using Unmanned Aerial Vehicles + Sensor Networks - UAVnet-NBS

Prior applications of Unmanned Aerial Vehicles (UAVs) centred largely on military applications as with many first technologies. Current efforts in recent times seek to explore other possible non-military applications. On the other hand, there has been a paradigm shift in use of sensing technology from traditional sensors (briefly talked about earlier) to sensor networks. This work in line with these two popular approaches now seeks to investigate how UAV's working together with present-day sensor technology can be applied and hence find relevance in disaster management.



Having established the opinion that technology has impacted quite significantly across the four aspects of disaster management till date, there is no refuting the fact that a logical, implementable outcome of objective set at the onset of this work will further raise its priceless and growing significance to humankind. Amongst all, it seeks to one humanitarian goal; to minimise to the barest minimum the loss of human lives by accelerating the rescue mission of trapped and vulnerable victims immediately following a disaster.

To achieve this, a network of sensors is deployed in an area identified as vulnerable possibly a building which herein could be referred as a site. These sensors seek to detect any human within its range of operation. Such sensors are by default in standby mode (non operational) and only come into operation when normal fixed infrastructure ceases to be i.e. when it becomes unavailable or is seriously hampered. Such circumstances presuppose that a disaster has just erupted and supporting amenities such as power supply, telecommunication infrastructure e.g. base stations, antennas e.t.c. is cut-off.

Following such collapse of fixed infrastructure, the sensors which where hitherto powered in stand-by mode from an external power source now switch to alternative power sources possibly on internal batteries while switching to operating mode. Communication within the sensor network is via a suitable adhoc wireless routing protocol. With the sensors in active mode, they seek to sense out any trapped victims. Ideally, motion sensors are best suited to perform this task and its operation could be limited to densely populated buildings where there deployment outlay is predefined and easy to identify.

Furthermore, the UAVs would manage and coordinate while performing a similar role. It would operate remotely in tandem with the sensor networks on ground. This implies that they are fitted with sensing elements which could include pictorial and imagery devices. The UAVs fly over the location in question. The sensors on ground as well as that on the UAVs form a sensor network with many nodes. Communication between them is made is via a wireless medium as they both actually undertake in the sensing task. On identifying a target which in this case is a victim, all information (signals) related to the discovery is communicated via the network to an external party acting as a coordinator or controller which interprets and facilitates any necessary response activity. The UAV in itself is used to relay the signals.

4.2 CONCEPT ANALYSIS - Outline of the Solution Components


The full concept of the UAV proposed for use in this investigation already exists. As such, only a brief overview is thus presented. An Unmanned Aerial Vehicle, UAV is basically defined as an aircraft without a pilot onboard. The earliest concept of the UAV dates back to the end of the 19th century even before the advent of aeroplanes. Understandably, its most notable first application was for military purposes especially during the First and Second World Wars. Nevertheless, in recent times, there is growing research into other possible civil and commercial applications.

The features of unmanned aerial vehicles are quite variant as depicted in 4B. The size of a UAV could vary from small to big depending on the intended application. According to IESSG [2009], UAVs, also referred to as Drones may be as small as a simple model aircraft or large like the WK450, the air vehicle for the UK armed forces' Watchkeeper programme, which has a wingspan of over 10 metres.

Ideally, any UAV would consist of three parts; an air vehicle, ground control station as well as product the product data dissemination system. There is a fourth component which is the human component popularly referred to as the “ground troops”. UAVs owing to their nature are capable of flying at lower heights than normal airplanes. They are also capable of flying for much longer hours. UAVs are considered advantageous largely because of the circumstances in which they can operate in UAV circles is referred to as the three ‘D' circumstances IESSG, 2009;-

* Dull - long-term surveillance/patrols, weather monitoring e.t.c.

* Dangerous - following criminals, monitoring hostage situations e.t.c.

* Dirty - oil spillage, contaminated regions e.t.c.

In line with the ongoing trend to seek out non - military applications for UAVs, the Welsh Assembly Government is driving research in this area hence this investigation. Even so, there are already on-going prominent non-military applications currently going on. These include environment monitoring and disaster management, the latter is the core focus of this report.


A sensor network is a collection of sensor nodes equipped with sensing, communication (short range radio) and processing capabilities. (Bruno et al, 2006). The nodes are deployed thickly in or close to the phenomena under scrutiny. This concept works in favour of disaster relief operations which is the case study in question. Akyildiz et al (2001) outline the basic characteristics of sensor networks include: -

§ Random deployment owing to large number of nodes usually in thousands

§ Processing ability of individual nodes

§ Cooperative capability efforts between nodes

§ Possess protocols/ algorithms with self-organizing (self-healing) capabilities

The features highlighted are basic to sensor networks. Furthermore, many more possible applications, including the one under investigation require such sensor networks to have wireless adhoc networking techniques giving rise to concepts such as Mobile Adhoc Networks, MANET and Wireless Sensor Networks, WSN.

It is imperative to outline in some detail the factors that affect the design of sensor networks as this has a large bearing on the choice of the protocol or algorithm to be used for communicating between the nodes in the network. According to Akyildiz et al (2001), there are also a number of factors considered pivotal in designing sensor networks. These include; - power consumption, reliability, scalability, production costs, hardware constraints, topology, environment, transmission media, e.t.c.


This factor has a close link to reliability and is often times considered from that perspective. Fault tolerance is defined as the ability to sustain sensor network functionalities without any interruption on the occurrence of sensor node(s) failure (Hoblos et al, 2000). Linking this to performance, a higher reliability would imply an improved quality of service. In other words, the degree of reliability of a network bears direct impact on the quality of service deliverable.

For a number of reasons such as environmental interference, physical damage, power unavailability e.t.c., sensor nodes may fail when in operation. In many cases, this does not affect the functioning of the sensor network. This is understandably so because there are alternative nodes which could suffice in any one operation. The degree to which these operational conditions can be maintained given nodal failure is a measure of its reliability.


The scalability criterion concerns itself with the magnitude of sensor nodes to be implemented in any application. Given the paradigm shift from traditional sensors to sensor networks; there is a plausible geometric leap in the number of deployable sensors. This point is further supported by the fact that sensors need not be in physical contact with the phenomena under observation.

The task of identifying possible victims given the occurrence of a disaster would only be feasible with these modern features in sensor technology. Consequently, it is only logical that the sensor to be deployed would be in the magnitude or hundreds, possibly thousands. These assertions are tied closely to production costs, yet another factor to be considered.


Evidently, sensor networks outperform traditional sensor by miles. Even then, Patel et al (2000) argue that the cost of any implementable network must be justified in order to guarantee widespread applications. In order words, the cost of producing the former must not be more expensive than the costs of deploying traditional sensors, the former's benefits notwithstanding. Accordingly, the cost of each sensor node must be kept as low as possible bearing in mind those sensor nodes to be adopted would be equipped with additional units considered hereafter.


A sensor node is ideally made up of four basic components; a sensing unit, a processing unit, a transceiver and a power unit. There are other components which are application dependent such as location finder system, mobilizer antenna e.t.c. The sensing unit consists of two parts, analog to digital converter (ADC) as well as sensors. The former converts the analog signals which the sensors obtained from the observed phenomena. The processing unit facilitates the collaboration between the sensor nodes which actually undertake the sensing. As such, it usually has some amount storage associated with it. The transceiver is responsible for connecting individual nodes to the network. The last part of the sensor node is the power unit which arguably is the most significant. As already pointed out, sensor network often operate on batteries. This makes its functioning even more critical considering the fact that in many operational circumstances, the offer no option of replenishment. Yet still, it supports all the other component parts of the node. It is for reasons such as this that the power requirements are very much a main consideration in sensor networks.


Power supply and by extension its consumption is vital in the functioning of any gadget or device. The application scenario under consideration is even more critical as a result of the prevalent circumstance which has hitherto been described. Given this background, there are two key points to remember

- Sensor node lifetimes are largely dependent on battery and as such limited

- Power replenishment is assumed impossible during operational periods.

There are three different activities within the networks which consume power :- sensing, data processing and communication. The amount of power consumed via sensors depends on the nature of its application; the more intense the nature of the sensing activity, the more the power consumed. Even so, the highest amount of power is consumed in communication of data which is far more than that consumed in processing. This is understandably so because asides the fact that communication involves use of power for data transmission as well as reception, it also takes cognizance of start up power used by the transceiver unit. Shih et al (2001) propose two approaches for computing power consumed in the communication and processing units.

EQUATIONS 3 & 4 FROM 1909a

In view of above conditions, power management and conservation would be traded in for quality in terms communication pattern. As such, the sensor network would largely work better with power-aware protocols. [Singh et al, 1998]


Transmission media is concerned with the channel in which the nodes communicate. Broadly speaking, nodes communicating in sensor networks interact using a wireless medium. Such media can be provided in one three ways; radio, optical means or infrared. The choice of any of these is dependent on its intended use. Of the three media, radio patronage is possibly the highest considering the fact that the technology precedes the other two. Nevertheless, technological advancements in them have brought about increased patronage in recent times.


The nature of environment is vital for ascertaining the most suitable sensors to deploy. Generally, sensors work on their own in a variety of areas; within an moving objects such as an enclosure (building), open air, underwater e.t.c. their working condition vary and this in turn determines what sensor could be deployed. Given that the anticipated working environment on the occurrence of a disaster, deployable sensors would be expected to work in high pressure conditions. What's more, the sensors would be looking to sense likely survivors. As such, it should be capable of identifying human sense phenomena. (Types of sensors)


This factor considers the reaction, performance and response of the sensors before, during and after their deployment. They are typically deployed numbers in their numbers, in many instances, they are inaccessible and unattended. As such, many nodes are failure prone. These circumstances make maintenance all the more tasking.


Routing is a pivotal process wireless networks and even for wired networks. In both realms, it is oversees the task information transfer. According to Jayakumar and Gopinath, (2007)

“The goal of any routing protocol is to have an efficient route establishment between any pair of nodes such that messages can be delivered in a timely manner”.

In other words, it devises means to ensure that messages, usually packaged as packets are correctly transported from one point to another, in this case from source to destination. The manner in which this is achieved distinguishes what protocol is used.

A routing protocol as applied in this context can be defined as a standard or a formula that ascertains the correct path via which data is transmitted. In routing terms, a protocol oversees how nodes (packet originators) decide which way to route packets. They also specify how nodes in a network share information and report updates. It can be seen from the definition that protocols play one and the same function in wired and wireless networks only in that in the latter circumstance, the nodes are unmindful of the network's topology. Some of the other characteristics of wireless include;

* DECENTRALIZED ADMINISTRATION - distributed processing

* INFRASTRUCTURELESS - independent of fixed or established infrastructure

* MULTI-HOP ROUTING - Every node has routing capability

* DYNAMIC TOPOLOGY - frequent changing, unpredictable route changes

* IDENTIFICATION - No universal identity e.g. IP address

* DEPLOYABILITY / COSTS - Easily deployable and cost effective

* CONNECTIVITY - flawed connectivity with increased security venerability

* POWER SUPPLY / ENERGY CONSTRINTS - often battery supported

* BANDWIDTH CONSTRAINS - owing to additional routing information



Different authors and researchers have adopted different classifications for routing protocols based on different criteria such as structure [Al-Karaki and Kamal, 2004] scheduling, state information e.t.c. [Jayakumar and Gopinath 2007], Nevertheless, the most common broad classification running through is based on routing strategy / operation [Royer and Toh, 1999]. 4D is a classification of routing pr.otocols based on the type of cast property which to a large extent incorporates some of the other classifications. A glossary of keywords alongside table of examples of different protocols is listed in the appendix III

For a number of logical reasons, this investigation shall be explored utilizing a protocol which shall be is considered solely on it routing strategy. Two strong points support this approach; asides the fact that it is the most common classification, there is increasing focus, research and study on these class of protocols in a bid to achieve standardization amongst other things. This will further enhance interoperability leading to geometric leap in possible applications. Based on routing strategy, there are basically two types; proactive or table-driven and reactive or on-demand.

Proactive routing protocols are protocols where it is required that each individual node keeps routing information. Such information are typically kept in a number of different tables which are periodically updated, hence the name Table-driven protocols. Examples of proactive routing protocols include; proactive routing protocols such as DSDV [Perkins and Bhagwat 1994], FSR [Iwata et al, 1999], WRP [Shree Murthy and J. J. Garcia-Luna-Aceves 1999, 1996], CGSR [Chiang et al, 1997], etc.

One type of proactive protocol is distinguished from another by the way the routing information is updated, detected and the type of information kept in its routing table. Furthermore, each routing protocol may maintain different number of tables [Abolhasan et al, 2003]. On the other hand, reactive or on-demand protocols seek routes as and when they are needed. Examples of these include DSR [Maltz et al, 1999], AODV [Perkins and Royer,1999], TORA [Park and Corson, 1997], ABR [Chai-Keong, 1996] etc. Some of these shall be looked at in more detail later.


PROACTIVE (Table-Driven)

REACTIVE (On-Demand)


Delay (end to end)

Minimal as routes pre-determined



Route Availability

Always available




In the range of hundreds

Usually higher than proactive


Storage Capability

Higher than on-demand

Route dependent


Updating Periodicity

Always required

Never required


Handling Mobility

Occurs at regular intervals

Uses localized route discovery


Traffic Intensity

Higher than On-Demand

Increases with active routes


Multicasting capability




Routing Metric

Shortest path

Shortest but freshest path



Large Overhead

Large delays

TABLE D1 - Comparative Analysis of Proactive And Reactive Protocols

Having provided some outline on the major classifications and decided in favour of the classification to proceed onwards, the remainder of the section will discuss reactive protocols at some length. It concludes with a particular choice of protocol following an in-depth analysis of routing protocols relying largely on secondary data e.g. simulation results drawn from a wide variety of authors herein referenced while taking into consideration the design considerations highlighted earlier.

Sevy (2009) points out that owing to the topology dynamism of reactive protocols as well as proactive protocols, there are key factors to be considered in undertaking its function;-

* Shortest path i.e. allowance of fewest hops

* Shortest time i.e. period offering lowest latency

* Shortest weighted path - utilization of available bandwidth


Leading to this point, justifications remain a plenty as to the suitability of MANETs in disaster management. Moving on, this section will consider what protocol is appropriate for the application under consideration. This discussion is all the more crucial because though MANETs are by their nature infrastructure-less, their applicability is largely dependent on the design considerations for the sensor network which have earlier been outlined in section 4.2.2. In the same vein, the protocol / algorithm to be suitably deployed will again rightly take cognizance of the nature of the sensor network which as we have highlighted for the purpose of this work will consist of hundreds, possibly thousands of sensors randomly deployed.

The comparative analysis discussed above is based to a large extent and supported by actual works undertaken by Yadav and Yadav, [2007], Jayakumar and Gopinath, [2007] and Bansal and Barua [2002]. From the foregoing, it is evidently clear that on-demand (reactive) protocols are most suitable for implementation in this application as they are more effective than Table-Driven and in turn fit the design scenario painted earlier. Nonetheless, one critical drawback and in fact the only advantage of the latter is its smaller end-to end time delays compared to the former. This factor is critical for disaster management which requires communication in real-time giving room to a consideration of Hybrid protocols. Jayakumar and Gopinath [2007] opine that such protocols generally bring together the advantages of two approaches so analysed.

The idea of hybrid solution seems plausible. A similar concept is considered by Fujiwara and Watanabe [2007] who propose a hybrid solution in a slightly different manner by using both an ad-hoc networking scheme as well as a hierarchical centralized network. Exploring the possibility of a hybrid routing protocol requires practical scenarios with huge logistic and time considerations which do not fall within the scope this work and can be considered an area of future work for a MSc. Project or Doctorate Thesis. reference.

Accordingly, this work shall be considered using reactive protocols which were introduced earlier. These can be further categorized according to how the routes are determined into Source and Hop-by-Hop. Source ‘path' routing place the complete destination path i.e. specifies the entire route in the message header. Hop-by-Hop ‘destination' routing on the other hand only specifies the destination in the message header. The intermediate nodes in turn ascertain the next hop en-route the final destination.



HOP-BY-HOP (Point to Point)


Source storage




Intermediate node storage




Routing overhead


Lower than source



Not scalable



Intermediate node Routing tables


All nodes keep localized routing tables

TABLE D2 - Features of Source and Hop by Hop Protocols

For this investigation, a number of reactive protocols shall be considered with a view to ascertaining what factors will support and to what degree, the objective in view. It will also highlight their support or drawback when juxtaposed with other factors. The protocols being considered have been identified as the most popular reactive protocols, more so as earlier pointed out, they are amongst the first considerations for the standardization work of ad-hoc protocols [Das et al, 2000], Bruno et al, 2005]. These include DSR [Maltz et al, 1999], AODV [Perkins and Royer,1999], TORA [Park and Corson, 1997

4.2.5 Destination-sequenced distance vector (DSDV)

DSDV Features

This algorithm is an example of source demand routing protocol which characterized with huge overheads. Even so, DSDV is able to limit its overhead. In order to minimize the overhead conveyed through the network, two types of update packets called “full dump” and “incremental” packets. The former carries all the available routing information while the latter takes only information which has changed since the last full dump. Consequently, the incremental update messages are sent more frequently than the full dump packets. Other features include;

• Each node knows the state and topology of the entire network

• Routes are chosen by a metric (least delay, best signal strength, etc.)

• Periodically and when triggered transmits the entire routing table to neighbours

- Full dumps

- Incremental dumps

• Avoids loops by using sequence numbers

DSDV Discovery

• When a link loss is detected at node N:

- the metric of the route to the destination through the lost link is advertised as infinity (the worst value), and

- An incremental update is flooded to the neighbours

DSDV Evaluation

• Loop avoidance

• Constant routing overhead versus mobility

- Overhead increases as the number of nodes increases

• DSDV can no longer find a route reliably when there is high mobility (< 300s pause times)

4.2.6 Demand Source Routing, DSR

DSR Features

DSR is a typical example of a source on-demand routing protocol as can be seen from 4E above. In other words, each carries the full address from source to destination. As such, the overhead increases as the size of the network increases. Other features include;

• Routes are kept in each packet

- Routes to that point in REQUEST packets

- Full routes in data packets

• Routes are cached at each node to limit flooding of REQUEST packets

- Any route that is seen through a node is cached

• Source sends out REQUEST packets

- Any node which is the destination of a node which has a route to the destination replies with a route reply

DSR Recovery

• A Route ERROR is sent to the Source

- All nodes along the path remove that route

• Source uses a cached alternate route to destination or sends out request packets for a new route

DSR Evaluation

• A Route ERROR is sent to the Source

- All nodes along the path remove that route

• Source uses a cached alternate route to destination or sends out request packets for a new route

4.2.7 Ad-Hoc On Demand Distance Vector (AODV)

AODV features

This algorithm is based on DSDV and DSR algorithms. It utilizes the periodic beaconing and sequencing approach of the former as well as a similar DSR route discovery procedure. It has the following features

• Each node only keeps next-hop information

• Source broadcasts ROUTE REQUEST packets

- Each node that sees the request and forwards it creates a reverse route to the source

- If the node knows the route to the destination, it responds with a ROUTE REPLY

• All nodes along the reply route create a forward route to the destination

AODV recovery

• When a link loss is detected at node N

- any upstream nodes that have recently sent packets through this node are notified with an UNSOLICITED ROUTE REPLY with an infinite metric for that destination

AODV Evaluation

• Routing overhead increases as mobility increases, but not as the number of nodes increases

• Sends many packets, but they are small

- Costs to access the medium (RTS/CTS packets)

• Always delivers at least 95% of packets sent in all cases (Broch, et. al.)


Role of UAV as base station e.t.c

a proposal is made for the use of either DSR, LAR or AODV with a definitive choice lying solely on implementation flexibility. The work carried out by 2311a takes particular cognizance of the a possible disaster scenario by using a large-scale topology of 500 nodes which is a sharp contrast to other works such as 2411d, 2311e whose number of nodes were considerably smaller. In conclusion, the author opts for LAR provided the additional hardware required for locating the nodes meets other suitability criteria.

Please be aware that the free essay that you were just reading was not written by us. This essay, and all of the others available to view on the website, were provided to us by students in exchange for services that we offer. This relationship helps our students to get an even better deal while also contributing to the biggest free essay resource in the UK!