三维空间声场的计算机仿真

以声线跟踪法为理论基础，通过建立声场各要素（声源、声传播方式及接收者等）的计算机模型，构造出声源至接收点之间的滤波函数，由此模拟计算声场中不同点处的声场评价指标。用VisualC 和OpenGL编制了相应的计算机仿真程序，通过比较实测和仿真结果，验证了算法的正确性和可行性。     

An altruistic differentiated service protocol in dynamic cognitive radio networks against selfish behaviors

Cognitive Radio (CR), first proposed by Mitola, has the potential to improve spectrum utilization by allowing secondary transmission in licensed primary networks, as long as the secondary transmissions do not interfere with the operation of the primary users (PUs) in the primary network. Although it is fundamental to the viability of this burgeoning technology, security aspects of CR have so far received scant attention from researchers. Initial research on CR has largely focused on the issues of resource allocation, spectrum sensing and management while security considerations are often added as an afterthought. As a result, many proposed solutions for CR, have introduced significant security gaps, intensifying the effects of selfishness, inequity, unavailability, or even malicious behaviors. This has posed challenges to quality-of-service (QoS) provisioning in Cognitive Radio Network (CRN) in the face of various network attacks like node masquerading, packet mislabeling and deliberate packet dropping. Since existing routing protocols assume that nodes will forward packets in good faith, the presence of selfish nodes coupled with the dynamic behavior of PUs can lead to excessive packet losses, high delays or even the complete failure of routing protocols in secondary networks. In this paper, we propose a cross-layer Altruistic Differentiated Service Protocol (ADSP) for the dynamic CRNs to address the QoS provisioning issue in CRNs with selfish nodes coexistence. Simulation results demonstrate that the ADSP can achieve much better performance in terms of lower delay, higher throughput and better delivery ratio for the traffic originating from collaborative nodes compared to other cognitive routing protocols in the presence of selfish nodes.     

基于心理声学参数的水下目标识别特征提取方法

研究了响度、尖锐度、粗糙度、波动强度等主要的心理声学参数及其计算方法,利用主要的心理声学参数作为目标特征参数用于水下目标分类识别,并以正确识别率为准则对这些特征参数进行了修改。使用K-均值聚类方法对3类舰船噪声实测数据进行了目标分类识别仿真实验。实验结果表明,该方法提取的特征能够较好地反映信号本质,取得了较好的分类识别效果,特别是以修改后的心理声学参数为特征具有更高的识别率。     

Sequences within the poliovirus internal ribosome entry segment control viral RNA synthesis

The 5' untranslated region of poliovirus RNA has been reported to possess two functional elements: (i) the 5' proximal 88 nucleotides form a cloverleaf structure implicated in positive-strand RNA synthesis during viral replication, and (ii) nucleotides 134 to at least 556 function as a highly structured internal ribosome entry segment (IRES) during cap-independent, internal initiation of translation. We show here that the IRES itself is bifunctional and contains sequences necessary for viral RNA synthesis per se. For this purpose, we used a dicistronic poliovirus RNA in which the translation of the viral non-structural (replication) proteins is uncoupled from the poliovirus IRES. In this system, RNA synthesis is readily detectable in transfected cells, even when the poliovirus IRES is inactivated by point mutation. However, deletion of the major part of the poliovirus IRES renders viral-specific RNA synthesis undetectable. Using the same system, we show that a three nucleotide deletion at position 500 in the 5' untranslated region drastically affects both translation efficiency and RNA synthesis. Furthermore, disruption of the secondary structure of the IRES around nucleotide 343 has minimal effects on IRES function, but dramatically reduces viral RNA replication. Taken together, these results provide direct evidence that sequences essential for viral RNA synthesis are located in the 3' region of the poliovirus IRES.     

A network approach to large-scale experimental data acquisition and analysis

The Plasma Research Laboratory at the Australian National University has developed a sophisticated and flexible data acquisition system for its LT-4 tokamak that requires very little support to construct, maintain and expand. It is novel in that it is based on a minicomputer network and is built almost entirely of commercially available products. Benchmarks show that the network system outperforms conventional stand-alone and time-shared data systems.     

Prevalence of Salmonella in broilers at retail outlets, processing plants and farms in Malaysia

A study was conducted to estimate the prevalence of Salmonella among broilers retailed at wet-markets and processing plants. Litter and feed samples obtained from both broiler and breeder farms were also examined for Salmonella. A total of 158 out of 445 (35.5%) and 52 out of 104 (50.0%) broiler carcasses obtained from wet-markets and processing plants were contaminated with Salmonella, respectively. Salmonella was isolated from 14 out of 98 (14.3%) samples of intestinal content. Litter samples from broiler and breeder farms were positive for Salmonella, 8/40 (20%) and 2/10 (20%), respectively. Salmonella isolates (230) belonging to 15 different serovars were isolated. Predominant serovars were S. enteritidis, S. muenchen, S. kentucky and S. blockley.     

Resonant-tank control of parallel resonant converter

A control method called resonant-tank control (RTC) is proposed for a parallel resonant converter operating above resonance. Using a simple linear combination of tank variables, it has potential for high-frequency DC-DC converter applications. RTC controls the tank in a near-time-optimal manner and is shown to have better dynamics than conventional frequency control. Experimental results that confirm the superior transient performance of the RTC method are provided. The principle of operation of the RTC can be extended to operation below resonance as well as to series resonant converter control     

基于遗传算法的指向性声源自适应有源消声系统优化

详细讨论了自由声场自适应有源噪声控制（ＡＡＮＣ）中，初级源为无限大刚性障板上的圆面活塞辐射器，采用一个和二个点源做次级声源时，使ＡＡＮＣ系统获得最大降噪效果的次级声源和误差传感器（以下统称为执行元件）的布放问题，并利用遗传算法（ＧＡ）确定次级源和远场误差传感器的最优位置。通过理论分析和计算机仿真结果的比较，证明了采用遗传算法可以准确地得到传统的梯度搜索优化算法难以得到的位置优化解。     

Optimization with controlled null placement in antenna array pattern synthesis

The optimal solution to the problem of minimizing the complex weight vector norm subject to a "look direction" and null constraints in array pattern synthesis is given. It is also shown that several constrained optimization problems in pattern synthesis, viz, constrained maximization of signal-to-noise ratio (SNR), constrained pattern gain maximization, and constrained least squares pattern synthesis are, in fact, equivalent. Reduction of the maximum gain in the look direction and the occurrence of main-lobe angular shifts which are inherent in constrained pattern synthesis are also investigated.     

New formulation of dyadic Green's function applied to a microstripline

We have modified the conventional two-layer dyadic Green's function analysis to include finite metal conductivity. The modified Green's function allows us to calculate the current distributions and conductor losses in microstrip circuits. A sample calculation has been applied to a microstrip line