LOCAL SEARCH IN MEDIA ACCESS CONTROL; EVALUATING BROADBAND INFRASTRUCTURE INVESTMENT USING EFFECTIVE ALTRUISM;

Since its invention, the internet has been changing our ways of life in a revolutionary manner and with unseen speed. Nevertheless, as is the case with almost all disruptive technologies, empowerment through internet access is not free of the side effect of inequality. My technical project is motivated by such inequality and creates a low-cost technical solution. However, while improving internet access is a tractable problem, it is secondary to the more serious problems facing humanity, such as the lack of food, shelter, and healthcare, especially in less developed regions, and investment in these regions can often impact more than investments in the developed world. One of the fundamental objectives of STS research is to determine how engineering can maximally benefit humanity, and a vital aspect of this endeavor is the cost- benefit analysis of focus areas to which engineering effort can be directed. My STS project conducts such an analysis of broadband infrastructure development.
The technical portion of my thesis proposes a software-based solution to improve internet access with minimal cost. This solution uses an observation readily available to existing hardware, transmission collision rate, as a heuristic, and a simple, efficient, and reliable technique in artificial intelligence, Simulated Annealing, to create a distributed strategy to optimize network traffic. This strategy is highly compatible with current technologies, and besides the initial cost of research and development, implementing this strategy involves minimal cost due to its software nature.
The STS part of my thesis conducts a cost-benefit analysis of broadband infrastructure investment in the US using an Effective Altruism framework. The project examines published statistics to determine the scope of the impact of such an endeavor and assesses the cost and benefits of such investment from data reported by current literature. By comparing such investments with other charitable projects and providing an argument for the responsibility of the
US government to consider projects beyond the scope of the US, this project concluded that broadband infrastructure investment is not the most effective area in which investment and engineering effort should focus on and recommends medical care in less developed regions as a highly effective alternative.
While not directly related, my technical project provides the necessary background for further research in the direction of my STS research. It proposes a way to reduce the cost of broadband infrastructure investment. Similarly, my STS research complements my technical projects with the context needed to understand its impact on a national scale and provides an estimation of the current cost-effectiveness of investment in this area, which future research in network technologies can use as a benchmark.

School of Engineering and Applied Science
Bachelor of Science in Computer Science
Technical Advisor: Daniel Graham
STS Advisor: Richard Jacques