Abstract: Convergence rate is an important standard to measure the effectiveness of the algorithm, but slow convergence is the common defect of spectrum allocation algorithm. In order to allocate the spectrum?more efficiently, this paper puts forward the concept of binary discrete velocity, by giving?a new interpretation of the speed and position update strategy of the Binary?Particle?Swarm?Optimization（BPSO）, and by using the binary discrete velocity as the XOR factor to decide the change of location parameter, to better balance the development performance and the exploration performance of the algorithm. Experimental results show that the Improved Binary Particle Swarm Optimization（IBPSO） can realize a more efficient and faster allocation of the spectrum.

Key words: Particle Swarm Optimization（PSO）, spectrum allocation, binary optimization, resource allocation

摘要： 收敛速度是衡量算法有效性的重要标准，当前频谱分配算法普遍存在收敛过慢的缺陷。为更高效地完成频谱分配，对二进制粒子群算法的速度和位置更新策略在离散域作了全新诠释，提出了二进制离散速度的概念，并以其作为决定位置参数变或不变的异或因子，更好地平衡了算法的开发性能与探索性能。实验表明，改进二进制粒子群算法能够更高效、快速地完成频谱分配。

关键词: 粒子群, 频谱分配, 离散优化, 资源分配