Analysis and Mitigation of Delay Effects in Computer Networks

The primary task of traffic control algorithms employed in computer networks is to ensure the reliable transmission of data among the network’s nodes. A critical aspect is preventing network congestion i.e., ensuring that the service rate of individual network elements is greater than their incoming data rate. We approach these data traffic adjustment algorithms from a control engineering perspective. The components of the control loop − controlled process, sensing, actuation, and control law − can be easily identified in the computer networks with data traffic regulators. From the control system perspective, the computer network can be modelled as a set of interconnected queue systems. By applying this approach, the behavior of the controlled network can be analyzed, and data traffic regulators can be designed to ensure that data is transmitted in an expected manner. When the automatic control of industrial processes is implemented over computer networks, ensuring low network latency is critical for efficient operation of these automated processes. We show that, by employing a control theoretical approach, it is possible to design such data traffic controllers for wireless local networks that guarantee the expected value of the communication delay remains below a specified threshold and at the same time, also ensure the best possible data transmission rate through the communication channels. The applicability of the developed traffic control method has been validated on telerobotic systems.