64-QAM传输系统的Optisystem仿真设计

64-QAM是一种高效的数字调制方式,广泛应用于各种通信系统中。首先对64-QAM的调制/解调原理进行了理论研究,然后借助Optisystem光仿真软件,搭建了64-QAM传输/接收模型,并对模型的接收和发射模块进行了详细的分析,最后对模型的传输性能进行了仿真验证。仿真结果表明,接收端64-QAM的星座图清晰,即64-QAM调制格式能有效地实现数据信号的传输,该模型能为实际的工程设计提供理论上指导。     

一种新的WLAN无线Mesh网络测试床的设计与实现

研究了现有的Mesh网络测试床,针对关键节点网络接入点（MAP）的设计,提出“分而治之”的思想,将MAP的功能分解到接入点（AP）和网格点（MP）分别实现,并引入第三大功能子网。该方案在网卡的选择、节点故障处理及后期网络的扩展性等方面很好地弥补了采用单一节点模拟方案的局限性。方案的实现表明改进后的测试床具有易搭建、易维护和易扩展的特点。     

Improved optical heterodyne methods for measuring frequency responses of photodetectors

An improved optical self-heterodyne method utilizing a distributed Bragg reflector (DBR) tunable laser and an optical fiber ring interferometer is presented in this paper. The interference efficiency can be increased by 7 dB compared with the scheme using the conventional Mach-Zehnder interferometer. The unsteady process that the beating frequency experiences in each tuning period is investigated. According to the measurement results, the wavelength and optical power of the tunable laser will be steady when the square-wave frequency is lower than 300 kHz. It has been shown that when a square-wave voltage is applied to the phase section of the tunable laser, the laser linewidths vary in a wide range, and are much larger than that under dc voltage tuning. The errors caused by the variations in the linewidth of the beat signal and optical power can be eliminated using the proposed calibration procedures, and the measurement accuracy can, therefore, be significantly improved. Experiments show that the frequency responses obtained using our method agree well with the data provided by the manufacturer, and the improved optical self-heterodyne method is as accurate as the intensity noise technique.     

信令监测系统共享平台业务订阅模块的设计

针对电信运营商的运营维护的要求,对信令监测系统的技术特点以及功能需求进行了研究,提出了一套信令监测系统共享平台的整体设计方案,对业务信令数据订阅模块的消息字段、通信流程进行了详细的设计,并且编码实现了其功能。测试表明,该业务信令数据订阅模块能够完成第三方平台对信令监测系统业务信令资源列表的查询、业务信令数据的订阅以及信令数据的传送,具有较好的实用价值。     

A combined low power SAR capacitance-to-digital analog-to-digital converter for multisensory system

A combined successive approximation (SAR) capacitance-to-digital converter (CDC)/analog-to-digital converter (ADC) for biomedical multisensory system is presented in this paper. The two converters have same circuit blocks and can be exchanged by four switches. Capacitance or voltage from different sensing elements can be measured and converted to digital output directly. This single chip takes place of separated CDC and ADC so that the power consumption of the multisensory system is reduced. The asynchronous SAR circuit has low power and small area. A dynamic comparator with zero-static power is adopted. Switches are carefully designed to reduce the non-idealities of the converter. Several techniques, such as bootstrapped switches, bottom-plate sampling, dummy switches are used to improve the performance of the circuit. The CDC/ADC is fabricated in 0.18 μm CMOS process. Measurement results show that the ENOB of this 11 bits converter is 10.15 bits and its FOM is 45 fJ/conversion-step under 200 kHz sampling. The power consumption is 9.4 μW with 1.4 V power supply voltage and the core area is 0.1764 mm².     

基于Kohonen神经网络的深度图像分割方法

提出一种基于Ｋｏｈｏｎｅｎ神经网络无监督的深度图像分割方法。首先计算了深度图像各点的导数,进而得到各点的法向,以法向和深度值作为每一点的特征矢量,引入自组织神经网络进行初始的聚类;为消除初始聚类产生的过分割现象,采取相邻表面片法向分析的方法进行再分割,得到最终的分割结果。本方法避免了通常的区域分割方法初始种子不易选取的弱点,聚类所用样本少,速度快。实验结果表现了算法良好的性能。     

非晶硅碳（a-SiC∶H）薄膜光学常数的透射谱表征

报道了一种用透射谱数据分析法计算非晶硅碳薄膜的厚度、折射率、吸收系数和光学带隙等光学常数的方法和程序．这一方法引用有效谐振子模型理论的折射率色散关系，所有公式均为解析表达式，便于进行数据处理，无须专用软件，使用Excel即可完成，适用于多种半导体薄膜材料．将这种方法应用于PECVD方法制备的非晶硅碳（a-SiC∶H）薄膜，对其光学特性进行了分析．     

实时红外目标跟踪方法

针对红外图像的特点,提出一种实时目标跟踪方法。该方法采用局部二元模式(LBP)算子提取图像的多尺度LBP编码直方图作为图像特征向量,使用卡方统计度量模板图像和样本图像的相似度。结合Kalman滤波和多尺度模板更新机制,使得算法在目标尺寸大幅变化,目标部分遮挡、短暂消失等情况下仍能保持跟踪的正常稳定进行。针对该算法,设计了专用图像处理板卡,介绍该算法在图像处理板上的硬件实现方法,使目标跟踪满足50 Hz/s的实时处理要求。最后,使用两组实录红外序列图像进行了仿真实验,验证了该算法的有效性和鲁棒性。     

A power scalable PLL frequency synthesizer for high-speed △-∑ ADC

A 35-130 MHz/300-360 MHz phase-locked loop frequency synthesizer for △-∑ analog-to-digital con- verter （ADC） in 65 nm CMOS is presented. The frequency synthesizer can work in low phase-noise mode （300-360 MHz） or in low-power mode （35-130 MHz） to satisfy the ADC＇s requirements. To switch between these two modes, a high frequency GHz LC VCO followed by a divided-by-four frequency divider and a low frequency ring VCO followed by a divided-by-two frequency divider are integrated on-chip. The measured results show that the fre- quency synthesizer achieves a phase-noise of-132 dBc/Hz at 1 MHz offset and an integrated RMS jitter of 1.12 ps with 1.74 mW power consumption from a 1.2 V power supply in low phase-noise mode. In low-power mode, the frequency synthesizer achieves a phase-noise of-112 dBc/Hz at 1 MHz offset and an integrated RMS jitter of 7.23 ps with 0.92 mW power consumption from a 1.2 V power supply.     

FFT算法的一种FPGA实现

FFT运算在OFDM系统中起调制和解调的作用。针对OFDM系统中FFT运算的要求，研究了一种易于FPGA实现的FFT处理器的硬件结构。接收单元采用乒乓RAM结构，扩大了数据吞吐量。中间数据缓存单元采用双口RAM，减少了访问RAM的时钟消耗。计算单元采用基2算法，流水线结构，可在4个时钟后连续输出运算结果。各个单元协调一致的并行工作，提高了系统时钟频率，达到了高速处理。采用块浮点机制，动态扩大数据范围，在速度和精度之间得到折衷。模块化设计，易于实现更多点数的FFT运算。