Networks-On-Chip (NoC) is seen as a new network paradigm for addressing the limitation of the current bus-based communication in multi-core embedded systems. Some of these systems are designed for executing hard real-time services in flight and automotive controls. But the current state-of-the-art design space exploration techniques lack uncertainty assessments. The risks that uncertainty imposes on the system’s safety have huge impacts on its predictability. For this reason, the current work on DSE is focusing on the uncertainty analysis of the systems, and to apply it in the DSE of NoC based multi-processor systems.
The performance of DNN accelerators is heavily influenced by the efficiency of on-chip communication. This is because DNNs, particularly those used in autonomous vehicles for tasks such as object detection, path planning, and decision making, require substantial computational resources. Ensuring energy-efficient and low-latency communication while maintaining realibility is a critical challenge in mapping DNN onto PEs containing accelerators that are interconnected with NoC.
Task data needs to be received by processing nodes and stored in a data queue, awaiting processing to complete the offloaded tasks. This caching and queuing process is complex but crucial for real-time task processing. However, most existing studies consider the total task processing time as merely the sum of local processing time, transmission time, and offloading processing time, often overlooking the caching and queuing stages. Additionally, many studies on cache resource scheduling focus on caching popular content at the network edge to improve hit ratios and avoid duplicate transmissions, thereby enhancing user experience. Few studies have addressed the combined allocation of computing, communication and storage resources.
The five Maqasid al-Shariah—protection of faith (din), life (nafs), intellect (aql), progeny (nasl), and wealth (mal) are essential for ensuring the well-being and sustainability of the Ummah. Decision-makers traditionally rely on muzakarah (consultation) to align policies and strategies with these objectives. However, to enhance decision-making processes, there is a growing interest in applying stochastic and formal methods. These advanced analytical approaches can provide early insights, yet their implementation presents several challenges.
…or discuss topic of your choice.
Kindly visit here for open final year projects.