Showing papers on "Blackfin published in 2009"

DSPcam: A camera sensor system for surveillance networks

Abstract: Surveillance is emerging as a key application of camera networks. In this paper, we describe our wireless smart camera system called DSPcam designed for real-time surveillance purposes. DSPcam has a Blackfin processor, CMOS image sensor and 802.11 b/g communication module. It also integrates with a standard sensor network node such as Firefly [1], through which IEEE 802.15.4-based communication is available. An open-source image processing library ported to our DSPcam platform allows for the development of many computer vision applications. We use the in-network processing ability of each DSPcam in our system to annotate video streams with meta-data information that succinctly describes key elements of the visual data being transmitted. The meta-data can be used to trigger alerts and for conducting rapid ex post facto searches. The DSPcams use a time-synchronized communication protocol called TSAM [2] that enables transmission of a large number of video flows and dynamic allocation of bandwidth for high-priority video streams. Finally, we also describe the integration of DSPcam with Sensor Andrew, a large-scale sensing network [4] deployed across the campus of Anonymous University.

Wearable wireless ECG monitoring hardware prototype for use in patients own home

Abstract: Recent miniaturization of ECG sensors and other health monitoring systems potentially allowing remote monitoring of certain diseases in the patients own home. Exploiting the full potential of this technology poses several design challenges for both the miniaturized ECG sensor, as well as the required infrastructure for data transport, storage and diagnosis. This paper proposes an ECG sensor prototype allowing experimentation on sensor technology and signal processing as well as infrastructure. The prototype is based on the Blackfin processor from Analog Devices, and uses Bluetooth for wireless communication.

A binary instrumentation tool for the Blackfin processor

Abstract: While a large number of program profiling and instrumentation tools have been developed to support hardware and software analysis on general purpose systems, there is a general lack of sophisticated tools available for embedded architectures. Embedded systems are sensitive to performance bottlenecks, memory leaks, and software inefficiencies. There is a growing need to develop more sophisticated profiling and instrumentation tools in this rapidly growing design space.In this paper we describe, DSPInst, a binary instrumentation tool for the Analog Device's Blackfin family of Digital Signal Processors (DSPs). DSPInst provides for fine-grained control over the execution of programs. Instrumentation tool users are able to gain transparent access to the processor and memory state at instruction boundaries, without perturbing the architected program state. DSPInst provides a platform for building a wide range of customized analysis tools at an instruction level granularity. To demonstrate the utility of this toolset, we provide an example analysis and optimization tool that performs dynamic voltage and frequency scaling to balance performance and power.

Software embedded implementation of real time non uniform sampling filter

Abstract: The use of digital filters is widely spread in the field of radio receivers. In literature, there is no practical realization of digital filters for non uniformly sampled data in the radio communication context. In this paper we are interested on real time filtering of non uniformly sampled signals using a Blackfin 533 Digital Signal Processor (DSP) as embedded software target. The filtering stage is not preceded by a reconstruction algorithm. As input, the filter requires signal samples and their corresponding non uniform sampling sequence. Obtained filtered data, are analyzed using specific spectrum analysis method based on zero padding and Fast Fourier Transform (FFT).

Vein recognition system based on DSP

Abstract: A design method of vein recognition system based on ADSP-BF533 is proposed.The system is composed by two function module groups: the vein image acquisition modules and the vein image processing and recognition modules.The vein images are acquired through a black-and-white analog camera under the illumination of near-infrared LEDs.The central unit for vein image processing and recognition is a Blackfin 533 DSP made by ADI(ADSP-BF533),whose work frequency is fast enough to relize the processing and recognition function.Experimental results show that the system is definitely suitable for embedded vein recognition.

Implementation of real-number parallel IFFT algorithm in Blackfin DSP

Abstract: To calculate real-number IFFT,and avoid low processing speed of existing real-number IFFT algorithm in DSP,a new program of real-number radix-2 IFFT in the Blackfin DSP is designed,which combines two real-number rows to one complex-number row.Results of experiment show that software based on parallel instruction and hardware parallel architecture improves the processing speed,can process two real-number rows(512 points per row)IFFT during 11864 cycles,which is half of time of former algorithm.The new program has been applied to the PSDOCT image pre-process system based on BF561,and satisfy the real-time imaging.

Optimization and Implementation of Embedded GUI Based on Blackfin Processor

Abstract: Aiming at the strong capability of Blackfin processors on multimedia, this paper researches the techniques related to Graphic User Interface(GUI). It implements an embedded GUI system based on uClinux, which achieves smaller size and higher performance. It is suitable for electronic products requiring display of high definition. It compares the performance between the GUI with Microwindows. Results show that the development of embedded applications, especially multimedia products with high performance can benefit greatly from the implementation of the GUI.

A nonlinear acoustic echo cancellation scheme implementation using the blackfin microcomputer

Abstract: Nonlinearity of amplifiers produces nonlinear echo in acoustic systems, which degrades the performance of audio communication. The paper presents an acoustic cancellation scheme that involve the least mean square (LMS) algorithm to update the parameters of the nonlinear function and the variable step size NLMS (VSS-NLMS) algorithm to determine the coefficient vector of the FIR filter. The performance analysis and the real time implementation using a digital signal processor (DSP) microcomputer (Blackfin BF533, form Analog Devices) are illustrated.

Design and implementation of portable GNSS baseband logger

Abstract: The design and the implementation of a portable GNSS baseband logging system which can log raw GPS L1 baseband more than one hour or longer is presented. In addition, the design covers GPS (L1, L2)/GLONASS/Galileo/Compass frequency and the maximal sample frequency is about 50MHz.The system is based on Blackfin DSP and uses CompactFlash card for storage. The data is extracted by reading the CompactFlash card to store as a data file. The recorded data file is analyzed by using the FFT to acquire the GPS satellites signals and the position is calculated by using the open source GPS positioning software. The system is very portable because of the CF+DSP design and can also be used on missile and satellite. It provides a good research and development platform for the GNSS software receiver.

Flexible Digital Display technologies using open source hardware and software for automotive applications

Abstract: With the advances in electronics, digital dashboards are now becoming available for use in the automotive industry. The main difference between an analog dashboard and a digital dashboard configuration is that the later may easily be reconfigured. To accommodate the influx of digital graphical displays in vehicles, manufacturers have started to run micro Real Time Operating Systems (RTOS) inside their vehicles. Two options are offered to manufacturers when choosing a RTOS for their project; commercial OSs or open source OSs. Commercial OS contain many overheads which include an upfront capital investment and licensing fee for each unit produced. While open source OS are royalty free and offer no such financial overheads. Any application software that is written by the manufacturer for a commercial OS, is seen as proprietary software, and hence is not accessible by other manufacturers. Whilst any software written and licensed for use with an open source OS would be accessible, therefore leading to reduction in manufacturing costs and time. The main objective of this research was to develop a flexible digital display using open source hardware and software for use in automotive applications. The development of a digital dashboard using these technologies can allow for individual customisation and in addition facilitate a significant reduction in the design cycle time. The designed display controller incorporated an Analog Devices Blackfin development board onto which an open source OS was ported. Automotive information was read from a CAN network and was used to manipulate the data displayed on the digital dashboard.

Blackfin DSP 구현을 위한 차선 검출 알고리즘의 소프트웨어 설계

Abstract: Cameras were introduced to extend a driver's view in vehicles initially. Nowdays, the cameras with image processing technologies play important roles to detect potential dangers before the driver's awareness in the field of active safety vehicle technologies. Especially, a lane departure warning system is the representative which introduced the camera with image processing technologies into the vehicles. Moreover, the needs that want extended functions over the lane detection are rising. However, limited performance in DSPs makes it difficult to implement. Thus, we present the effective software design to implement the lane detection algorithm using the Blackfin DSP. Experimental results show that the proposed method makes the processing speed faster than before.