నైరూప్య

Parallel and Pipeline Pattern Matching Strategy forLow Power Applications

Kiruthika.T, Brindha.P

One broadly used method for representing membership of a set of items is the simple space-efficient randomized data structure known as Bloom filters. Generally the regular Bloom filter suffers in terms of power consumption and FPR (False Positive Rate). To overcome this we proposed two methods. The pipelined Bloom filter architecture for k-stages has been proposed to attain the significant power saving. The second method is the parallel Bloom filter that reduces the FPR. Further a novel Bhsequence scheme is introduced in this pooled pipelined and parallel Bloom filter architecture to reduce the FPR. Through this method around 10%-20% of the power saving can be achieved. Bloom filters are used in network security applications such as web caches, resource routing, network monitoring.