Current Research

Research & Project


Indigenous mission payload design for onboard satellite image classification based on deep learning and transfer fine-tuning data to the ground station.

The requirements of the earth data for remote sensing products are generally very demanding in terms of data quality and coverage/revisit time. In this research, we proposed the image classification method aimed at multispectral earth observation using Commercial off-The-Shelf (COTS) instruments onboard a 1U CubeSat for automatically selecting images for downlink on a 1U CubeSat. The hardware systems will be designed and developed by us without changing the severe limitations of size, power, volume, and mass imposed by 1U CubeSat. If implemented, our research will significantly contribute to image classification and more acceptable image-finding methods.

Machine learning has been widely used in multispectral remote sensing image analysis and classification for many years. The applications for multispectral or hyperspectral satellite image classification tasks often used random forests, support vector machines,  or decision trees in earlier years. However, these approaches are not optimized for smaller hardware.
Through this research, we will try to design the mission payload of our second satellite which will be developed inside Bangladesh by the students. Here are the objectives of our Research

Primary Objective
Demonstrate the multispectral camera feasibility on 1U CubeSat with On-Board Image Classification Using Deep Learning.

Secondary Objective
Space Radiation Effect on Micro-controller.
Using software-defined radio for monitoring Radio Frequency Interference.

Addressing Food Security with Earth Observation Data

This project Idea recently started after Attending the SGAC working group in IAC 2022, Paris, France.  The problem statement was given by Planet. The focus of the Working Group was on Addressing Food Security with Earth Observation Data, identifying what role private industries and governmental agencies can play in providing access to EO-powered decision support tools which can directly impact initiatives like the UN’s SDG of Zero Hunger.

Agriculture is a key sector, crucial for economic growth and the health of the population in every country. It is also at the heart of the global climate change discussion. The sector is accountable for 25% of GHG emissions and 70% of water use globally, whilst also being subject to paramount climate-related risks. Climate change poses a massive threat to food security, threatening to cut crop yields in the world’s most food-insecure regions. Accurate and reliable agricultural data is therefore essential to ensure food security and the subsistence of the World’s population, which is estimated to reach 9.75 billion by 2050.

Against this background, we are Researching on finding the best machine learning/AI solutions for crop identification.

Ingenious Space-Based Hands-On Learning Tools To Provide Space System Engineering Knowledge

Nowadays, space engineering is a thriving sector of modern engineering and seems to have unlimited potential for exploration and innovation in the coming days. There are several graduate and postgraduate degree programs in space engineering. Moreover, online education and resources for space engineering are growing continuously in terms of availability and quality with recent cultural shifts and circumstances, which enhance the learning experience and development of new knowledge and technology of the participants regardless of geographic access. To acquire the technical expertise required for satellite technology and all cross-disciplinary skills, a learner must have hands-on experience in this field. A few satellite training kits are available globally, such as ESAT, EYASSAT, PUMPKIN CUBESAT KIT, CUBESAT KIT 2.0, ORBICRAFT-PRO 1U HEPTASAT for professors/instructors and students to learn basic satellite engineering by hands-on experience. The prices of these kits are varied from USD 8500 to USD 15500, which is quite expensive. The maker of these kits offers training to learn the kits that are also expensive to afford. To engage more students/instructors worldwide, develop the skills, and make learning hardware affordable, a satellite training kit has been built, together with a comprehensive curriculum guide that could be learned individualistically or with the help of professional instructors. The mission of satellite training KIT is to collect physical variables, such as UV index, temperature, air pressure, humidity, GPS coordinate Data orientation data, and live video feeding. This training KIT can be launched using a hydrogen balloon and collected real-time data. The training method is designed for four different (education) levels: High school students, college students, undergrads, and professionals. This paper presents how we developed a kit in Bangladesh during the pandemic to give the best satellite building experience at the lowest possible price by utilizing local components. The designed kit started attracting students and instructors in space engineering, and currently, they are taking the training. This paper will also discuss the training technique we are following on emphasizing science, technology, engineering, arts, and mathematics (STEAM) and the lesson learned from the project.

Model Rocket Based Hands-on Training KIT for
the Sustainable Space Program in Bangladesh

Launching a satellite for some major corporations in the space sector is a routine activity. Due to specialized impediments, cost, and education, reaching into and exploring space is still a big challenge for countries that are new and developing their technology. Then again, due to a lack of expertise and financial capabilities, the non-space-faring countries are not capable of producing excellent hands-on educational material to educate their students in space science. To engage more students, instructors, and young professionals nationwide, we will develop a Space-based hands-on training Kit and interactive online Learning tools with a comprehensive curriculum guide. Our Hands-on training kit will contain various model rockets and a CubeSat. This platform aims to uncover facts and insights into model rocket design, develop processes, gain new knowledge, and understanding about space technology and alleviate the bridging of the skill gap between the practice and academia.

Leverage space interests and capabilities to maximize the Nation’s engagement in the multimillion-dollar space sector for economic development, technology development and demonstration, scientific advancement, workforce development, STEAM education, and enhanced utilization of the nation’s resources and capabilities.
Establish a Space Commission to develop laws & policies to conduct space-related activity in Bangladesh safely.
Design, build and launch Bangladesh’s first rocket capable of carrying industrial payloads such as satellites and offer launch services for research organizations and commercial industries.

Design and Implementation of SDR Based portable Ground Station for Satellite Data Reception.

Complex Low Earth Orbit (LEO) satellites provide us with data on cloud dynamics, rainfall, and land- and sea-surface temperatures from the particular perspective of space. These satellites are remarkable because they transmit image signals that can be easily decoded to show what they are truly seeing.

The purpose of this research is to develop a low-cost,  portable base station that utilizes a Software  Defined Radio (SDR) and an appropriate antenna  (Double Cross Dipole Antenna, V Dipole  Antenna) to receive signals from NOAA-19  satellites and decode them using open-source software applications (SDR Sharp, WXtoimg,  Orbiton, LRPT Decoder) to obtain real-time images when the satellite is overhead. These real-time images are useful to predict weather and rainfall.

A Store and Forward CubeSat Mission Data Management and Distribution System for Ground Sensor Terminal Networks of Developing Countries

The increasing interest to use CubeSats for store-and-forward-based (S&F) satellites as remote data collection systems have been demonstrated since 1990s. The S&F technology is a key element to link sensor networks deployed on the ground for specific data collection purposes. In Kyutech, cube satellites such as the BIRDS satellites, KITSUNE/SPATIUM II on-board S&F payload which is complimented by ground sensor terminals that collect specific data for its member countries. These payloads and ground sensor terminals not only aim to demonstrate technology but also build capacity for participating developing countries. Aside from the satellite with S&F payload, the ground segment which involves the download, collection, storage, and distribution of data are also important for country-specific applications. In this paper, the design, development, and implementation of a data management and distribution mechanism for the store and forward mission of KITSUNE/SPATIUM II is presented. The online platform is designed to collect, store, and organize the S&F data so that it can be distributed to the participating countries. To achieve this, the following system requirements were established: (1) archive the processed data and shall save the data in dedicated storage and (2) provide a mechanism to share the data among members of the ground sensor terminal (GST) network. The data storage mechanism involves the acquisition of primary satellite data stream during the reception at the ground station as saved in the ground station drive, data processing, evaluation, and product generation stored in temporary storage; and archiving of data for long term preservation of data in external drive and/or cloud storage. On the other hand, data handling using the back-end database is implemented by accepting S&F data in a standard file format, encoding the data through an online webpage with user access, and sharing data through data download and report generation. Initial beta test on the ground shows that the system can be accessed through dedicated user accounts per country, and different data formats can be uploaded to populate the database. Moreover, users were able to access their own data and download them for further processing and analysis. Finally, users were also able to generate reports based on the data in the database.