中文摘要：

针对n变量逻辑函数在不同极性下所对应的XNOR／OR电路功耗和面积不同的特点，首先用信号概率传递算法和多输入XNOR／OR（同或／或）门的低功耗分解算法建立了XNOR／OR电路的功耗估计模型．在此基础上，将基于列表技术的极性转换算法和整体退火遗传算法相结合，提出了一种针对大规模XNOR／OR电路的低功耗最佳极性搜索算法．对8个较大规模MCNC Benchmark电路测试表明，该算法搜索到的最佳极性所对应的XNOR／OR电路与极性0时的XNOR／OR电路相比，平均节省功耗和面积分别达到了84．4％和65．2％．

英文摘要：

For n-variable logic function, the power dissipation and area of XNOR/OR circuit corresponding to each polarity are different. By investigating the propagation algorithm of signal probability and low power decomposition algorithm of multi-input XNOR/OR gate, a power estimating model of XNOR/OR circuit is established, By combining the power estimating model and the polarity conversion of tabular technique with the whole annealing genetic algorithm （WAGA）, this paper proposes a new algorithm to search the best polarity of XNOR/OR circuit for low power dissipation. The results of testing eight large-scale circuits from MCNC Benchmark indicated that this algorithm is highly effective for searching the best polarity. The XNOR/OR circuits under the polarity searched by the proposed algorithm have achieved average power-saving and area-saving by 84.4% and 65.2 %, respectively, in comparison with those circuits under polarity 0