Kasun De Zoysa

A public transport system based sensor network for road surface condition monitoring

A well maintained road network is a must for the economic development and the well being of people in any country. Unfortunately, most developing countries do not poses such road networks. While the lack of funds is mainly to blame for not building new road networks and maintaining the existing ones the lack of proper monitoring and reporting system is a major contributory factor for the dilapidated condition of road networks in third world countries. A case in point is the road network in Sri Lanka; Sri Lanka has an extensive road network that spans the country and new roads are being built every day, yet even the roads in the capital city are not maintained properly. The lack of a monitoring and reporting mechanism is apparent in this case. We propose a public transport system based sensor network to monitor road surface condition. We are currently building such a network called BusNet to monitor environmental pollution and that system can be extended for road surface condition monitoring by adding acceleration sensor boards to the system.

Twitter news classification using SVM

With the development of web blogs, Social Networks, many news providers used to share their news headlines in various web sites and web blogs. Now-a-days in Sri Lanka, there are many news groups whom share their news headlines in micro blogging services such as Twitter. These data may carry out much valuable information which will relevant to many social research areas. Thus, the purpose of this research is to classify news into different groups so that the user could identify the most popular news group in a given country for a given time. The short messages were extracted from Twitter micro blog. Several active news groups were chosen to extract the short messages. Each short message was classified manually into 12 groups. These classified data were used to train the machine learning techniques. Words of each short message was considered as features and a feature vector was created using bag-of-words approach in order to create the instances. The data were trained using SVM (Support Vector Machine) machine learning techniques. The main reason of using SVM for the current study is, SVM supports high dimensional data. Current research is a high dimensional problem as a large number of features will be collected using short messages. Cross validation was done in order to avoid the biasness of data. The performance of the system will be the effectiveness of the system. Thus precision and recall values are calculated to measure the performance of the system. Fβ was calculated to obtain a single value measurement. The results show that the system provides high performance for most groups. However, the group development-government does not show much performance using SVM.

Mobile ATM for developing countries

Society benefits from M-Commerce applications to a greater extent. The most attractive benefit of M-Commerce applications is the mobility. Even though users have a poor computer literacy, they will be able to use the M-Commerce applications easily. Additionally, the M-Commerce applications have the potential of reducing the distance barriers. In developing countries, especially in rural areas, accessing financial and banking services is a critical issue. This paper proposes a system called Mobile-ATM to address this problem by incorporating the mobile technology. Also it discusses the limitations of traditional ATM systems, the need of a new M-Commerce application to overcome the limitations and security related issues. In the proposed solution, people can withdraw money from a Mobile-ATM without going to a traditional ATM. The Mobile-ATM system uses even cheap mobile phones, functioning as payment terminals. It will reduce the limitations of traditional ATM and enables confidential and secured ATM transactions.

Sensor-based breakage detection for electric fences

The human-elephant conflict is one of the most severe natural problems in Sri Lanka. There are rich farmlands near the elephant habitats and elephants raid these farms in search of food. This has been the main cause for nearly 70 human deaths and over 200 elephant deaths that have been recorded each year in the recent past. To manage the problem, the government has initiated projects that secure the national wildlife parks with electric fences. However, maintaining the electric fence is a challenge, because of its large perimeter and the lack of available manpower. A particular concern is that of locating faults in electric fences since these typically span a few hundred miles. Currently, park rangers are required to travel on foot to locate the faults, which could take days to complete. In this paper, we propose a novel system architecture that considerably shortens the maintenance time for electric fences, at minimal and hence affordable cost. Our architecture benefits the park rangers of the national wildlife sanctuaries to detect and repair the breakages.

Identifying Mosquito Breeding Sites via Drone Images

Public Health Inspectors (PHIs) in Sri Lanka are facing a problem of identifying certain mosquito breeding sites since they cannot easily reach places such as roof gutters, overhead water tanks, inaccessible rooftops and cement materials which are capable of retaining water. The goal of a such inspection of suspected sites is to reduce the number of dengue patients by eradicating dengue mosquito habitats. Due to the retention of water in the aforementioned sites for a long period of time, those places tend to be full of lichens. In general, lichens are visible in dark color. This characteristic helps to identify prolonged water retention areas. With the rapid advancement of technology, the drone has been created as one of the most cost effective apparatus to capture the places that a human cannot access. With respect to the aforesaid context, this paper presents a simple and a novel approach to identify mosquito breeding sites via drone images. The proposed approach processes images captured from a drone to identify possible sites where stagnant water may retain and highlights if such areas are apparent within the image. The evaluation process found that the proposed method, produces a satisfactory level of accuracy in identification of possible water retention areas and the final results depend on the drone camera tilt angle and the effect of shadows.

Wireless Sensing: What Simplicity Has to Offer?

We argue that, a wireless sensor does not have to implement a full networking stack in order to be a part of an energy efficient sensing application as long as it can deliver its sensed data to an Internet connected device over a single hop. Current hardware industry trends indicate the possibility of implementing wireless sensors for a low cost with simple capabilities. This poster highlights our preliminary work aligned with this trend and aims to open a discussion on this topic.

A Quadcopter Controller to Maintain Radio Link Quality

The excellent maneuverability and the availability of a large number of sensors including good quality video cameras make quadcopters attractive for surveillance systems. Most video surveillance systems need a real-time high quality video stream from the cameras on the quadcopter to the base station which requires a stable and reliable radio link. Moreover, to ensure a controlled flight, it is essential to maintain a good quality radio link from the base station to the quad- copter. The quality of this link depends on the distance between base station and quadcopter as well as the ambient noise. Towards this end, we present a controller to hover a quadcopter at the maximum distance from the base station while maintaining the link quality at a given value indicated by the Signal to Noise Ratio (SNR). There are existing solu- tions in the field of robotics that incorporate the concept of link aware mobility. However, to the best of our knowledge, this is the first link-quality aware mobility implementation on a quadcopter.

The deployment of TikiriDB for monitoring palm sap production

Nowadays, the industry of harvesting palm sap in Sri Lanka is facing many problems due to theft and environmental affects. In this paper, we propose a solution for the particular problem by using a wireless sensor network based system to monitor the palm sap production of a large plantation area. We are using an enhanced version of TikiriDB which provides a database abstraction on the sensor network to make the process of data collecting and analyzing more efficient. TikiriDB can be used to reduce restrictions on accessing traditional sensor networks by enabling the user to access the sensor network using multiple queries through multiple access points simultaneously.

Listening to the Giants: Using Elephant Infra-Sound to Solve the Human-Elephant Conflict

The continuing human-elephant conflict in Sri Lanka has resulted in loss of human as well as elephant lives. Detecting and localizing elephants is an essential component of any viable solution to this problem. Currently, we conduct feasibility tests on using low cost sensors to detect elephants from a long distance, leveraging the infra-sounds emitted by them. In this paper we present the test environment that we have set up for this purpose and some preliminary, but promising results.

Towards a sensor system to tame the human elephant conflict

The human elephant conflict in Sri Lanka has been a cause of major concern over the past decade. Frequent clashes between wild elephants and villagers have resulted in severe damage to property, as well as loss of lives for both humans and elephants. Competition for space is the primary reason for conflict between humans and elephants. Elephants that escape from the wildlife national parks venture into villages creating destruction in their wake. To prevent such mishaps, a proper system is required to contain and monitor elephants in national parks. In this paper, we describe different approaches to detect elephants and possible ways of monitoring the national wildlife parks. We also elaborate on the advantages and limitations of each approach, and determine what sort of system is needed to tame the human elephant conflict.