Enhanced Approach of Spectral Clustering-Hybrid Mean Cut Balanced Clustering in VLSI Circuits-(SC-HMCBC)

In VLSI power system, clustering and system design become a major key for the consumption of power. Basically, clustering approach is used to generate the groups and consider for efficiency process in raising growth of IC and design. From various approaches based on simplicity and robustness the popular approach is considered with the improvement as a proposed approach of Hybrid Mean Cut Balanced clustering approach. The proposed approach is implemented in the application of VLSI circuits. In existing, the data points are split into subsets as per the associated points and the clustering center updating. Also, it avoids the sets to balance the target data sets. But in the proposed approach, the weighted undirected graph is implemented which will split the vertices into two sets and for balancing with accuracy the hybrid mean cut approach is implemented to control the clustering outcomes degree. According to the function optimization the clustering results are balanced and the extensive simulation in large sets of data is validated the performance improvement. The clustering design is stimulated using Cadence Virtuoso tool of 180nm gpdk library functions.