A Swarm based Binary Decision Diagram (BDD) Reordering Optimizer for Reversible Circuit Synthesis

Abdalhaq, Baker; Awad, Ahmed; Hawash, Amjad

doi:10.1109/dtis48698.2020.9081221

Cited by 2 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Genetic Algorithm (GA) and Simulated Annealing (SA) algorithms have been well exploited in BDD reordering problem [21,23]. Recent swarm optimizaiton algorithms have been also exploited to solve BDD reordering problem as published in [22]. However, the direct impact of BDD reordering algorithms on the cost of the synthesized reversible circuits should be investigated.…”

Section: Previous Workmentioning

confidence: 99%

“…Many algorithms have been proposed in the literature to find the order of the BDD nodes that results in low BDD size once reduction rules are applied [19]. However, for such an NP-Complete problem [20], meta-heuristic based algorithms (including both swarm and evolutionary based algorithms) have demonstrated great evidence in finding near-optimal solutions if compared with other deterministic algorithms [21,22,23]. Other proposed algorithms, such as sifting [24], and dynamic programming-based [25], are faster to converge, at the cost of resulting larger BDD sizes.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Reversible Logic Synthesis Using Binary Decision Diagrams With Exploiting Efficient Reordering Operators

2020

Self Cite

View full text Add to dashboard Cite

With the continuous shrinkage of transistor sizes in very large scale integrated circuits, power consumption forms a serious concern to be tackled. With their ability to allow for zero energy dissipation, reversible circuits have been considered as promising solution to meetup with low power design requirements. Moreover, advances in their synthesis methodologies can be easily applied to quantum circuits due to the inherited reversibility of the latter. Although numerous algorithms have been proposed in the literature to synthesize reversible circuits and map them into their corresponding quantum circuits, the scalability and computational effort of such algorithms form a serious concern when synthesizing large size input functions. Binary Decision Diagram-based synthesis for reversible circuits has shown great evidence in realizing reversible circuits with low quantum cost through exploiting proper reduction rules for smaller graph size. However, the order of the variables in the decision diagram impacts its overall size, and thus, the cost of its corresponding reversible circuit. While several reordering algorithms have been proposed in this manner, their direct impact on the quantum cost has not been considered. In this paper, a Binary Decision Diagram-based algorithm for reversible circuit synthesis is proposed to synthesize reversible circuits for a given Boolean function with low quantum cost through exploiting a linearized relationship between the decision diagram size and the corresponding quantum cost. Thereafter, different decision diagram reordering algorithms have been integrated with the proposed algorithm and compared in terms of their impact on the quantum cost. Experimental results show that Genetic Algorithm-based reordering for decision diagram, supported with, cycle crossover, inverse mutation, and tournament selection, results in the least quantum cost of the output circuit if compared with other algorithms due to its property in preserving the nodes of the decision diagram in their near-optimal locations during the optimiation recipe.

show abstract

Section: Previous Workmentioning

confidence: 99%