Ilhan Kaya

A low power lookup technique for multi-hashing network applications

Many network security applications require large virus signature sets to be maintained, retrieved, and compared against the network streams. Software applications frequently fail to identify so many signatures through comparisons at very high network speeds. Bloom filters are one of the main multi-hashing schemes utilized in hardware to support this level of security. Nevertheless Bloom filters consume significant power to store, retrieve and lookup virus signatures owing to many hash function computations required to index to the memory. We present a novel lookup technique and architecture to decrease the power consumption of multi-hashing schemes, predominantly Bloom filters, in hardware. The theoretical analysis has shown that power gain achieved through new lookup technique can go up to 90%. Simulation results with three different classes of the hash functions embedded into the Bloom filter have indicated that power consumption of the Bloom filters can be considerably decreased by employing the low power lookup technique

Low-power bloom filter architecture for deep packet inspection

Bloom filters are frequently used to identify malicious content like viruses in high speed networks. However, architectures proposed to implement bloom filters are not power efficient. In this letter, we propose a new bloom filter architecture that exploits the well-known pipelining technique. Through power analysis we show that pipelining can reduce the power consumption of bloom filters up to 90%, which leads to the energy-efficient implementation of intrusion detection systems.

Comparative assessment of freeform polynomials as optical surface descriptions

Slow-servo single-point diamond turning as well as advances in computer controlled small lap polishing enables the fabrication of freeform optics, or more specifically, optical surfaces for imaging applications that are not rotationally symmetric. Various forms of polynomials for describing freeform optical surfaces exist in optical design and to support fabrication. A popular method is to add orthogonal polynomials onto a conic section. In this paper, recently introduced gradient-orthogonal polynomials are investigated in a comparative manner with the widely known Zernike polynomials. In order to achieve numerical robustness when higher-order polynomials are required to describe freeform surfaces, recurrence relations are a key enabler. Results in this paper establish the equivalence of both polynomial sets in accurately describing freeform surfaces under stringent conditions. Quantifying the accuracy of these two freeform surface descriptions is a critical step in the future application of these tools in both advanced optical system design and optical fabrication.

Edge clustered fitting grids for φ-polynomial characterization of freeform optical surfaces.

With the recent emergence of slow-servo diamond turning, optical designs with surfaces that are not intrinsically rotationally symmetric can be manufactured. In this paper, we demonstrate some important limitations to Zernike polynomial representation of optical surfaces in describing the evolving freeform surface descriptions that are effective for optical design and encountered during optical fabrication. Specifically, we show that the ray grids commonly used in sampling a freeform surface to form a database from which to perform a φ-polynomial fit is limiting the efficacy of computation. We show an edge-clustered fitting grid that effectively suppresses the edge ringing that arises as the polynomial adapts to the fully nonsymmetric features of the surface.

A Display Framework for Visualizing Real-Time 3D Lung Tumor Radiotherapy

Medical display systems are valuable tools in enabling the clinicians in the field of radiation therapy to view a patients multi-modal information and treatment plan details. The effectiveness of display systems is further improved by including computer-based visualization systems that deliver the content comprehensively. In this paper, we present a medical display and visualization framework for radiation therapy that couples a computer-based simulation of real-time lung tumor motion and its dose accumulation during treatment with an Augmented Reality Center (ARC) based display system. The simulation framework provides insights on the variations in the effectiveness of the lung therapy for changes in the patients breathing conditions. The display system aims to enhance the clinicians understanding by enhancing the 3D depth perception of the dose accumulation in lung tumors. Thus the framework acts as a tool for presenting both pre-operative studies and intra-operative treatment efficacy analysis when coupled with a real-time respiration monitor. A first evaluation of this framework was carried out using six clinical experts. Results show that, using the ARC compared to a 2D monitor, the experts were able to more efficiently perceive the radiation dose delivered to various aspects of the moving tumor and the surrounding normal tissues, as well as more quickly detecting radiation hot spots that are critical to minimizing damage to healthy tissue.

Energy-Efficient Pipelined Bloom Filters for Network Intrusion Detection

Software-based detection techniques are commonly used to identify the predefined signatures in network streams. However, the software-based techniques can not keep up with the speeds that network bandwidth increases. Hence, hardware-based systems have started to emerge. Bloom filters are frequently used to identify malicious content like viruses in high speed networks. However, architectures proposed to implement Bloom filters are not power efficient. We propose a new Bloom filter architecture that exploits the well-known pipelining technique. Through extensive power analysis we show that pipelining can reduce the power consumption of Bloom filters up to 90 %, which leads to the energy-efficient implementation of network intrusion detection systems.

Gabor Domain Optical Coherence Microscopy

In this paper, we present a developing technology targeted at clinical imaging, Gabor Domain Optical Coherence Microscopy (GD-OCM), which combines the high resolution imaging of optical coherence microscopy (OCM), high imaging speed of Fourier domain optical coherence tomography (FD-OCT), and invariant lateral resolution of our custom designed dynamic focusing objective. A high lateral resolution optical design of a dynamic-focusing optical probe with no moving parts, which provides an invariant resolution of currently < 3 μm across a 2mm full-field of view and 2mm imaging depth, is presented. Furthermore, an acquisition scheme (using the probe) that is capable of performing automatic real time data fusion to render an in-focus high resolution image throughout the depth of sample in real time was implemented. 3D imaging of an African frog tadpole is demonstrated at cellular level resolution.

Modeling Air-flow in the Tracheobronchial Tree using Computational Fluid Dynamics

In this paper, we present a biomechanical framework to model airflow inside the bronchus and deformations across the tracheobronchial tree, pipeline for the simulator, theory and initial steps to realize this framework on a highly parallel graphical processing unit (GPU). We discuss the main challenges expected and encountered to date. By using computational fluid dynamics (CFD) and computational solid dynamics (CSD) principles, we propose a numerical simulation framework that includes a biomechanical model of the tracheobronchial tree to simulate air flow inside the tree, on GPU in real-time. The proposed 3D biomechanical model to simulate the air inside the lungs coupled with a deformation model of the tracheobronchial tree, expressed through fluid-structure interaction, can be used to predict the transformations of the voxels from a 4D computed tomography (4DCT) dataset. Additionally, the proposed multi-functional CFD and CSD based framework is suitable for clinical applications such as adaptive lung radiotherapy, and a regional alveolar ventilation estimation.

Increasing the power efficiency of Bloom filters for network string matching

Although software based techniques are widely accepted in computer security systems, there is a growing interest to utilize hardware opportunities in order to compensate for the network bandwidth increases. Recently, hardware based virus protection systems have started to emerge. This type of hardware systems work by identifying the malicious content and removing it from the network streams. In principle, they make use of string matching. Bit by bit, they compare the virus signatures with the bit strings in the network. The bloom filters are ideal data structures for string matching. Nonetheless, they consume large power when many of them used in parallel to match different virus signatures. In this paper, we propose a new type of Bloom filter architecture which exploits well-known pipelining technique

Application of radial basis functions to represent optical freeform surfaces

This paper presents the use of radial basis functions (RBF) for describing freeform optical surfaces. The RBF approximation framework along with preliminary optical design experiences will be summarized.