Emotion recognition at the edge with AI specific low power architectures

Nowadays Deep Learning is applied in almost every research field and helps getting amazing results in a great number of challenging tasks. The main problem is that this kind of learning and consequently Neural Networks that can be defined deep, are resource intensive. They need specialized hardware to perform computation in a reasonable time. Many tasks are mandatory to be as much real-time as possible . It is needed to optimize many components such as code, algorithms, numeric accuracy and hardware, to make them “efficient and usable”. All these optimizations can help us to produce incredibly accurate and fast learning models. The paper reports a study in this direction for the challenging face detection and emotion recognition tasks.     

Estimation of Antenna Correlation Coefficient of N-Port Lossy MIMO Array

This paper proposes a simple yet accurate method for estimating the antenna correlation coefficient (ACC) of a high-order multiple-input multiple-output (MIMO) antenna. The conventional method employed to obtain the ACC from three-dimensional radiation patterns is costly and difficult to measure. An alternate method is to use the S-parameters, which can be easily measured using a network analyzer. However, this method assumes that the antennas are highly efficient, and it is therefore not suitable for lossy MIMO antenna arrays. To overcome this limitation, we define and utilize the non-coupled radiation efficiency in the S-parameter-based ACC formula. The accuracy of the proposed method is verified by the simulation results of a 4-port highly coupled lossy MIMO array. Further, the proposed method can be applied to N-port arrays by expanding the calculation matrix.     

孔隙水压力计埋设过程中的问题研究

孔隙水压力计在实际使用过程中,通常不能按照规范那样进行现场标定,现场常采用无套管钻孔埋设孔隙水压力计,从而可能会影响孔隙水压力计精度,且在实际使用过程中量程选择也是一个难题.针对这种情形,通过分析常压标定方式误差、量程选择问题及无套管埋设孔隙水压力计,得出采用的现场常压标定及无套管埋设是可行的.     

嵌入式Web视频服务器的设计

嵌入式Web视频服务器监控系统以ARM芯片作微处理器,摄像机视频信号压缩采用MPEG-2编码格式并在嵌入式Linux中内置服务器应用程序.数据采集压缩模块实时采集视频数据,压缩后发给网络传输模块,媒体流经ARM芯片发到Internet.该服务器直接浏览Web视频,授权用户可控制摄像机或对系统配置操作,实现远端现场监控.     

雷达改善因子测试仪的设计

基于回波模拟器原理的便携式雷达改善因子测试仪采用嵌入式Linux操作,由PC/104总线CPU、回波信号模拟器虚拟仪器卡和测试软件构成.PC/104 CPU主模块控制管理系统、产生控制目标回波及改善因子测试自动引导.该测试仪通过回波模拟器,在雷达动目标显示系统改善因子测试应用中得以验证.     

基于以太网的嵌入式步进电机控制单元

基于以太网的嵌入式步进电机控制单元由PC104总线栈接处理器和电机控制模块组成.电机控制模块采用了2块CPLD芯片,设定步进电机专用控制模式,以实现步进电机控制单元与远程主机间的数据传输、控制状态的监控及运行模式的编程.底层控制单元上的芯片驱动、板级支持包、实时操作系统及电机控制应用程序在嵌入式实时操作系统VxWorks实现.     

基于ARM的便携式网络探测器的设计与实现

本文设计了一种基于ARM和嵌入式实时操作系统的便携式网络性能测量仪,采用了分布式网络测量方案.该方案可精确测量网络时延、带宽和丢包率,可使网络性能测量系统更便携.     

嵌入式系统的网络接口模块设计

嵌入式操作系统Linux的网络接口模块,通过对DM9000的正确寻址、设置其工作模式实现DM9000与S3C2410间的连接和转换,并将接收信号线RX 、RX-和发送信号线TX 、TX-经隔离变压器与以太网水晶接头RJ45相连以实现二者的连接.接口模块包括TCP/IP协议软件和以太网芯片DM9000驱动程序两部分.     

基于无线网络的大型武器装备故障诊断系统

基于无线网络的大型武器装备故障诊断系统,其下位机在32位微处理器S3C2410A基础上,进行外围扩展.包括信号采集、信号调理电路、JTAG接口、LED显示、无线网卡接口及RS232接口,从而实现大型武器装备数据采集及和上位机的无线数据交互.上位机由配置无线网卡微机组成.并采用笔记本电脑,通过无线网卡接受各下位机采集的信息并显示.通过调用数据库,对数据进行分析,查找故障原因.