In this paper, we propose a memory-constrained tree search (MCTS) algorithm for the detection in multiple-input multiple-output (MIMO) systems. The MCTS algorithm offers a wide range of trade-offs between computationa...
详细信息
In this paper, we propose a memory-constrained tree search (MCTS) algorithm for the detection in multiple-input multiple-output (MIMO) systems. The MCTS algorithm offers a wide range of trade-offs between computational complexity and memory requirement, and is guaranteed to achieve the exact maximum-likelihood performance. By tuning the memory size, the MCTS algorithm ranges from being memory-efficient to being computation-efficient with abundant choices in *** show that the MCTS algorithm visits slightly fewer nodes and requires slightly less memory than the sphere decoding (SD) algorithm in the memory-efficient case, and visits similar number of nodes and requires significantly less memory than the stack algorithm in the computation-efficient case.
In this paper we introduce an efficient probabilistic neural networks (PNN) model-based voice activity detection (VAD) algorithm. The inputs for PNN are code excited linear prediction coder parameters, which are stabl...
详细信息
In this paper we introduce an efficient probabilistic neural networks (PNN) model-based voice activity detection (VAD) algorithm. The inputs for PNN are code excited linear prediction coder parameters, which are stable under background noise. The PNN network output is 1 or 0 to determine the nature of the period (speech or NonSpeech). Experimental results show that the proposed VAD algorithm achieves better performance than G.729 Annex B at any noise level. The performance compares very favorably with Adaptive MultiRate VAD, phase 2 (AMR2).
This paper presents a novel lane detection algorithm for automatic drive system. The algorithm chooses a common curved lane parameter model which can describe both straight and curved lanes. The most prominent contrib...
详细信息
This paper presents a novel lane detection algorithm for automatic drive system. The algorithm chooses a common curved lane parameter model which can describe both straight and curved lanes. The most prominent contribution of this paper is: instead of using one single method to calculate all the parameters in the lane model, both the adaptive random Hough transformation (ARHT) and the Tabu Search algorithm are used to calculate the different parameters in the lane model, according to the different demands of accuracy for different parameters. Furthermore, in order to reduce the time-consume of the whole system, the strategy of multi-resolution is proposed. At last, this paper also presents a tracking algorithm based on particle filter, which can make the system more stable. The algorithm presented in this paper is proved to be both robust and fast by a large amount of experiments in variable occasions, besides, the algorithm can extract the lanes accurately even in some bad illumination occasions.
One of the difficulties in search and rescue missions is finding a small target, such as a person, in a large cluttered area. Airborne hyperspectral cameras are now being deployed to aid in this SAR mission. Motivated...
详细信息
One of the difficulties in search and rescue missions is finding a small target, such as a person, in a large cluttered area. Airborne hyperspectral cameras are now being deployed to aid in this SAR mission. Motivated by the successes of such systems, we define a hyperspectral model of human skin in the visible and near infrared regions of the spectra so we can exploit knowledge gained during the modeling process to aid in human skin detection. Based on observations of the skin model results, an efficient and robust skin detection algorithm using channels in the near infrared region of the spectra is developed. Our algorithm is denoted the Normalized Difference Skin Index, motivated by the Normalized Difference Vegetation Index used in the literature for detecting vegetation in hyperspectral imagery. We demonstrate the capabilities of our skin detection methodology to detect skin amongst objects known to cause false detections for methodologies using three channel color data.
An intrusion detection system should support the operator of the system. Thus, in addition to producing alerts, it should allow for easy insertion of new detection algorithms. It should also support dynamic selection ...
详细信息
ISBN:
(纸本)9780769533247
An intrusion detection system should support the operator of the system. Thus, in addition to producing alerts, it should allow for easy insertion of new detection algorithms. It should also support dynamic selection and de-selection of detection algorithms, and it should adjust its resource consumption to the current need. Such a system would allow the operator to easily extend the system when new detection algorithms become available. It would also allow the operator to maintain a low-cost monitoring baseline and perform more extensive monitoring when it is required. In this paper we propose an architecture for intrusion detection which aims at providing the operator with this support. The architecture uses a modular design to promote a high degree of flexibility. This supports creation of an environment in which state-of-the-art intrusion detection algorithms easily can be inserted. The modular design also allows for detection algorithms to be enabled and disabled when required. Additionally, the architecture uses a sensor reconfiguration mechanism to affect the amount of data collected. When a detection algorithm is enabled or disabled, the sensor providing the input data to the algorithm is correspondingly reconfigured. This implies a minimum of excess collected data. To illustrate the feasibility of the architecture, we provide a proof - of-concept supporting monitoring of users for insider detection and webserver monitoring for intrusion attempts.
To real-time classify military target signal and noise under non-stationary non-Gauss noise effectively, an edge detection method based on gray correlation analysis algorithm is proposed through analyzing ZCR LPC feat...
详细信息
To real-time classify military target signal and noise under non-stationary non-Gauss noise effectively, an edge detection method based on gray correlation analysis algorithm is proposed through analyzing ZCR LPC feature from military target signal. Simulation results indicate detection rate reached 100% under the clean acoustic signal and more than 80% corresponding the noise level of -5 dB. This algorithm is effective under non-stationary non-Gauss noise and the detection efficiency and performance are superior to other algorithms.
In this work, we proposed a shot-change detection algorithm for intra-frames of H.264/AVC compressed video. The traditional video codec, such as MPEG-1, 2, H.261 or H.263, transforms the macroblocks of intra-frames in...
详细信息
In this work, we proposed a shot-change detection algorithm for intra-frames of H.264/AVC compressed video. The traditional video codec, such as MPEG-1, 2, H.261 or H.263, transforms the macroblocks of intra-frames into DCT domain, but H.264/AVC video codec requires the intra prediction process prior to DCT processing, therefore DCT coefficients of H.264/AVC represent differently from those of the traditional codec. The past algorithms of shot change detection using DCT coefficients are not suitable for H.264/AVC any more. In this paper, we compare the statistical distribution of intra prediction modes with an adaptive threshold to detect the shot change. Our algorithm also combines the luma with the chroma feature to detect shot change for color video. The experimental result shows our algorithm can achieve accurate detection rate in rapid speed.
Life detection radar is a sort of initiative life detection equipment with extensive uses. A colored Gaussian noise model and a harmonic signal are proposed by analyzing the experimental data in this paper. The capabi...
详细信息
Life detection radar is a sort of initiative life detection equipment with extensive uses. A colored Gaussian noise model and a harmonic signal are proposed by analyzing the experimental data in this paper. The capability compare is done between wavelet filter and FIR filter about the suppression of clutter and detection of signals, the experimental signals have characters of low-frequency, semi-periods, and low SNR.. The results show the delay of wavelet filter is shorter than FIR filter and SNR is improved, so new algorithm is much validity and stability.
Most existing algorithms for fault-tolerant event region detection only assume that events are spatially correlated, but we argue that events are usually both spatially and temporally correlated. By examining the temp...
详细信息
Most existing algorithms for fault-tolerant event region detection only assume that events are spatially correlated, but we argue that events are usually both spatially and temporally correlated. By examining the temporal correlation of sensor measurements, we propose a detection algorithm by applying statistical hypothesis test (SHT). SHT-based algorithm is more accurate in detecting event regions, and is more energy efficient since it avoids measurement exchanges. To improve the capability of fault recognition, we extend SHT-based algorithm by examining both spatial and temporal correlations of sensor measurements. The extended SHT-based algorithm can recognize almost all faults when sensor network is densely deployed.
Port scanning is used by malicious users to map the characteristics of a network to launch further attacks. Hence, detection of port scanning assumes paramount importance. This paper investigates the effectiveness of ...
详细信息
Port scanning is used by malicious users to map the characteristics of a network to launch further attacks. Hence, detection of port scanning assumes paramount importance. This paper investigates the effectiveness of using counts of various TCP control packets in detecting TCP SYN scanning on a single machine. The behavioural characteristics of TCP control packets are aggregated. A neural network is trained to capture this behaviour for normal as well as port scan data. It is seen from the investigation that the counts of TCP SYN, SYN-ACK and FIN packets show definite patterns in their behaviour for legitimate connections. A deviation from this behaviour is used to effectively detect TCP SYN scanning without maintaining state information.
暂无评论