网络化嵌入式铁路调度监听系统的研究

发布时间：2018-03-11 04:37

本文选题：以太网　切入点：大容量　出处：《成都理工大学》2013年硕士论文　论文类型：学位论文

【摘要】：电话语音监听以其可备份和可回放、实时性高、方便查询等特性在几乎所有调度行业上得到了广泛的应用，如安保、交通、核电、银行、铁路等行业。随着电子技术、网络信息技术、语音编解码技术和集成芯片技术的迅猛发展，电话语音监听从最初的磁带式的监听设备发展到如今的基于以太网的大容量嵌入式监听设备，无论是在在功能上还是在存储容量等方面，语音监听都今非昔比。特别是随着嵌入式实时操作系统、片上系统以及网络技术的出现，语音监听设备也在朝着小型化、网络化的方向发展。在发展的历程中也经历了基于PC机、FLASH芯片以及移动硬盘等各种媒介的监听设备，但最终都因为成本过高、体积太大、存储容量有限等缺陷无法满足大多数用户的需要。基于上述需求，本文成功研发设计了一种网络化的嵌入式语音监听系统。相对于普通的语音监听系统，本系统具有更高的可靠性、更大的接口灵活性、更高的信息安全性、更好的电磁兼容性。网络化的嵌入式语音监听系统组成部分包括语音采集电路、AD控制电路、存储电路、3.5寸LCD电阻屏接口电路、USB2.0接口控制器、SD卡接口、振铃检测及电话解码电路、语音信息前端调理电路、以太网接口电路以及S3C2440核心电路等。基于微处理器S3C2440的嵌入式系统平台，整个系统架构采取模块化设计，这种设计方式降低了开发的难度，提高了系统的可靠性，降低了发成本。各个模块可独立调试，独立运行。本系统主要包括为主控模块、AD配置和音频采集模块、音频解码播放模块、外部存储器接口模块以及网络通讯模块五个部分，本文对这五个部分的硬件电路图进行了分析、设计并完成了裸机程序的测试。主控模块在整个系统中处于核心地位，该模块完成系统控制、数据通讯以及语音数据处理等功能。主控模块采用六层核心板设计，核心芯片S3C2440片上集成了SPI总线接口和I2S总线接口，可以直接跟FPGA以及带有I2S接口的声卡芯片进行通信。数据采集板把采集到的音频数据通过SPI总线传递给主控模块，主控模块对收到的音频数据进行特定编码，经过编码之后的数据就可以通过I2S接口给声卡播放出来，同时也在外设中存储起来作为后期查询的备份。 AD配置和音频采集模块主要由ADI的语音采集芯片AD1974和ALTERACYCLONE IV系列的EP4CE6E22C8组成。在本系统中AD1974同时对四路音频进行采集（如有需要该芯片还可以扩展成同时采集16路音频），并把采集到的数据通过SPI总线传递给FPGA进行初处理。AD采集出来的四路音频数据包含64个bit，第一、二、三、四通道的数据分别位于管脚ASDATA1和ASDATA2上，ASDATA1管脚上的数据格式是一通道32bit和二通道32bit，同样ASDATA2管脚上的数据格式就是三通道的32bit和四通道的32bit。由于我们对AD1974配置的是16bit量化，延迟一个位时钟接收数据，所以每个32bit的数据最高位都是无效位，有效数据是从次高位开始的16bit数据。根据对有效数据的计算，FPGA把四个通道的64bit有效数据取出来并按通道顺序整合到一起成为一帧数据。为了方便操作系统中数据的传递，每个通道的16bit数据格式都是低八位在前，高八位在后。考虑到不同系统之间数据传输的同步问题，特意把每个通道的最低两位替换成通道号(四个通道的通道号分别为00、01、10、11)，再通过SPI总线传递给主控模块。FPGA还负责控制语音采集芯片的配置、语音数据的存储、处理以及传输，这种采用FPGA和CPU共同控制语音采集芯片的方式，不但提高了系统的实时性和效率，同时也在设计上降低了难度。音频解码播放模块的核心芯片是飞利浦的UDA1341，UDA1341是专业级的声卡芯片，该芯片片上集成了可编程增益控制模块和滤波模块，在比较恶劣的坏境中也可以录制出比较纯净的声音，对背景噪声有很好的滤除效果，播放音频文件的效果也比较好。在本系统中，该模块的主要功能是通过I2S总线接口跟主控模块进行通信，接收主控模块编码后的语音监听信息，并把这些信息实时播放出来。外部存储器接口模块包括USB接口和SD卡接口，，USB接口方便外扩硬盘，使得存储容量大大增加。通过USB接口可以对数据进行存储、传输、处理，USB接口也可以连接键盘、无线网卡等外围设备，在该接口上使用了USB2.0OTG（On-The-Go）技术，实现了在没有Host的情况下完成从设备之间的数据传递，使本系统摆脱了主从架构的限制，实现了端对端的数据传输模式，即允许USB在没有HOST的情况下也可以实现从设备间的数据传输，这也就解决了基于PC机的监听设备所面临的成本高、体积大的问题。此外，SD卡和USB接口都支持热插拔，这对调度现场的语音监听具有实际意义。网络通讯模块设计了以太网控制器，核心芯片DM9000自适应速度10M/100M，通过控制器接入局域网和Internet，可以将监听记录下来的语音信息上传到网络终端，实现多个站点之间的信息交互与共享。网络技术的引入是本系统的一大创新，网络化的语音监听系统也是最近几年在监听行业出现的一个新的趋势，并以它独特的资源共享优势逐步替代了以往传统的监听设备。
[Abstract]:Listen to the voice call backup and playback, real-time, convenient query and other characteristics have been widely used in almost all industries such as scheduling security, transportation, nuclear power, bank, railway and other industries. With the development of electronic technology, network information technology, the rapid development of voice codec technology and integrated chip technology. The telephone voice from the initial tape monitoring equipment to the development of large capacity embedded monitoring equipment based on Ethernet now, whether in function or storage capacity, voice monitoring have shrunk. Especially with the emergence of embedded real-time operating system, system on chip and network technology, voice monitoring equipment at a small, networked direction.
In the course of development experience based on PC, FLASH chip and mobile hard disk and other media monitoring equipment, but eventually because the cost is too high, the volume is too large, the limited storage capacity and other defects can not meet the needs of most users. Based on the above requirements, this paper successfully developed the design of embedded audio monitoring system based on network compared to the normal voicemonitoring system, this system has higher reliability, greater flexibility of the interface information, higher security, better electromagnetic compatibility.
Networked embedded audio monitoring system consists of speech acquisition circuit, AD control circuit, storage circuit, 3.5 inch LCD resistivescreen interface circuit, USB2.0 interface controller, SD card interface, and the telephone ringing detection decoding circuit, voice information front-end processing circuit, Ethernet interface circuit and S3C2440 circuit. The core microprocessor embedded system platform based on S3C2440, the system adopts the modular design, this design reduces the difficulty of development, improve system reliability, reduce development costs. Each module can be independent debugging independently. The system includes the main control module, AD configuration and audio acquisition module, audio decoding module, five part of the external memory interface module, network communication module, this paper analyzes the hardware circuit diagram of the five part of the design, and completed the bare metal program Test.
The main control module is the core in the whole system, the module of system control, data communication and voice data processing and other functions. The main control module uses six layer core board design, the core chip S3C2440 chip integrated SPI bus interface and I2S bus interface, can be directly with FPGA and sound card chip with I2S interface communication the data acquisition board. Audio data is collected and transmitted to the main control module through the SPI bus, the main control module for a specific encoding of audio data received from the data after encoding after you can through the I2S interface to the sound card playing out, but also in the peripheral storage for later query backup.
The AD configuration and audio acquisition module mainly consists of ADI speech acquisition chip AD1974 and IV series ALTERACYCLONE EP4CE6E22C8. In this system AD1974 at the same time to collect the four audio (if necessary, the chip also can be extended to acquire 16 channel audio), and transmits the collected data to the four channel audio data FPGA at the beginning of.AD collected by SPI bus consists of 64 bit, first, second, three, four channel data in pin ASDATA1 and ASDATA2 respectively, the ASDATA1 pin on the data format is a 32bit channel and two channel 32bit, ASDATA2 pin on the same data format is three channel and four channel 32bit 32bit. because of our AD1974 configuration is 16bit to quantify the delay data receiving a clock, so the data for each 32bit MSB is invalid, valid data is 16bit data from the beginning of the second high. According to the calculation of the effective data, FPGA 64bit data of four channels according to the channel order taken out and integrated together into a frame of data. In order to facilitate data transfer in the operating system, 16bit data format of each channel is low in eight, high eight after considering the synchronization problem. The data transmission between different systems, specifically to each channel of the lowest two replaced channel number (four of the channel number 00,01,10,11), and then to the main control module is responsible for controlling the.FPGA voice chip configuration through SPI bus transfer, voice data storage, processing and transmission, which uses FPGA and CPU common control speech acquisition chip, not only improves the real-time and efficiency of the system, but also reduces the difficulty of design.
Audio decoding and playing module is the core chip of PHILPS UDA1341, UDA1341 is a professional grade audio chip, the chip on chip programmable gain control module and filtering module, in bad environment can also record a relatively pure sound, good filtering effect of background noise, play audio files the effect is good. In this system, the main function of this module is through the I2S bus interface with the main control module for communication, receiving the master control module after encoding voice monitoring information, and the information is played out in real time.
External memory interface module includes a USB interface and SD card interface, USB interface and convenient external hard disk, the storage capacity greatly increased. Through the USB interface can be carried out on data transmission, storage, processing, USB interface can connect the keyboard, wireless LAN and other peripheral equipment, the interface uses the USB2.0OTG (On-The-Go) technology to achieve the complete transfer from data between devices in the absence of Host, the system to get rid of the architecture constraints, realizes the end-to-end data transmission mode, which allows USB in the absence of HOST can realize the data transmission between devices, which also solves the problems of PC machine based on the monitoring equipment of high cost, large volume. In addition, the SD card and USB interface support hot swappable, it has practical significance on sitevoice monitoring. The network communication module design of Ethernet controller core Adaptive speed 10M/100M chip DM9000, the controller is connected to the LAN and Internet, can monitor the recorded voice information is uploaded to the network terminal, realize information exchange and sharing between multiple sites. The introduction of network technology is a major innovation of the system, a new trend of network-based voicemonitoring system is in recent years in the monitoring industry, and its unique resource sharing advantage gradually replaces traditional monitoring equipment.

【学位授予单位】：成都理工大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TN912.3;TP368.1

【参考文献】