First, the basic transmission process of VoIP
The traditional telephone network is to transmit the voice in the circuit Exchange Mode, the required transmission Broadband for 64kbit/s. The so-called VoIP is the IP packet switching network for the transmission platform, the analog voice signal compression, packaging and other a series of special processing, so that it can use the non-connected UDP protocol for transmission.
In order to transmit voice signals on an IP network, several elements and functions are required. The simplest form of network consists of two or more VoIP-enabled devices that are connected through an IP network. The basic structure of the VoIP model is shown in Figure 2-18. You can find out how VoIP devices convert voice signals to IP traffic and send them to IP destinations, and the IP destinations convert them back to voice signals. The network of the two sounds must support IP transmission and can be any combination of IP routers and network links. Therefore, the transmission process of VoIP can be divided into the following stages simply.
Javascript:window.open (THIS.SRC); "style=" cursor:pointer; " />
1. Voice-Data conversion
Voice signal is analog waveform, through the IP to transmit voice, whether real-time applications or non-real-time applications business, the road is the first to the voice signal analog data conversion, that is, the analog voice signal 8-bit or 6-bit quantization, and then fed into the buffer storage area, The size of the buffer can be selected based on latency and coding requirements. Many low bit-rate encoders are taken in frames to encode. The typical frame length is 10~30ms. In consideration of the cost of transmission, the inter-language package is usually composed of 60, 120 or 240ms voice data . Digitization can be achieved using a variety of speech coding schemes, the current speech coding standards are mainly ITU-T g.711. The source and destination voice encoders must implement the same algorithm so that the destination voice device helps to restore the analog voice signal.
2, the original data to the IP conversion
Once the voice signal is digitally encoded, the next step is to encode the voice packet in a specific frame length. Most encoders have a specific frame length, and if an encoder uses 15ms frames, it divides the packet from the first 60ms into 4 frames and encodes it sequentially. 120 voice samples per frame (sampling rate is 8kHz). After encoding, 4 compressed frames are synthesized into a compressed voice packet into the network processor. The network processor adds headers, markers, and other information to the voice and transmits it over the network to the other end point. The Voice network simply establishes the physical connection between the communication endpoints (a line) and transmits the encoded signals between the endpoints. Unlike a circuit-switched network, an IP network does not form a connection, it requires the data to be placed in a variable-length datagram or packet, and then sends each datagram with addressing and control information, sent over the network, and forwarded to the destination one station at a time.
3. Transmission
In this channel, all networks are treated as a receive voice packet from the input and then transmitted to the network output at a certain time (t). T can vary in a full range, reflecting jitter in the network transmission. The same node in the network examines the addressing information that accompanies each IP data and uses this information to forward the datagram to the next station on the destination path. A network link can be any topology or access method that supports IP traffic.
4. IP packet-conversion of data
The destination VoIP device receives this IP data and begins processing. The network level provides a variable-length buffer to adjust the jitter generated by the network. The buffer can accommodate many voice packets, and the user can select the size of the buffer . Small buffers produce less delay, but do not adjust for large jitter. Second, the decoder will encode the voice packet extracted after the generation of a new voice packet, the module can also be manipulated by the frame, completely and the length of the decoder is the same. If the frame length is 15ms, the voice packet of 60ms is divided into 4 frames, and then they are decoded into a 60ms voice stream into the decoding buffer. During the processing of the datagram, the address and control information is removed, the original raw data is preserved, and the original data is provided to the decoder.
5. Digital voice conversion to analog voice
The playback drive takes out the Voice sample point (480) in the buffer and feeds it into the sound card, which is aired at a predetermined frequency (e.g. 8kHz) via the speaker. In short, the transmission of voice signals over an IP network is subject toFrom analog signal to digital signal conversion, digital voice package into IP packet, IP packet transmission through the network, IP packet unpacking and digital voice restore to analog signal process。
second, the driving force to promote the development of VoIP
due to the related hardware,Software, protocols and standards, many of the developments and technological breakthroughs that make VoIP widely available will soon become a reality. Technological advances and developments in these areas have contributed to creating a more efficient, functional and interoperable VoIP network. Table 2-2 provides a brief list of the major developments in these areas. As can be seen from the table, the promotion of VoIP rapid development and even widely used technical factors can be summarized as follows.
1, digital signal processor advanced digital signal processing device (digitally Signal Processor, DSP) to perform voice and data integration requires computationally intensive Ninko. DSP processing of digital signals is primarily used to perform complex computations, otherwise these computations may have to be performed by the universal CPU. The combination of their specialized processing power and low cost makes the DSP well suited to perform signal processing functions in VoIP systems.
g.729 speech compression on a single voice stream is computationally expensive and requires 20MIPS, and if a central CPU is required to handle multiple voice streams while also performing routing and system management functions, it is unrealistic to use one or more DSPs to offload the computational tasks of the complex speech compression algorithm from the central CPU. In addition, the DSP is also suitable for voice activity detection and echo cancellation such functions, trapped in their real-time processing voice data stream, and can quickly access the board memory, so. In this section, we describe in more detail how to implement speech coding and echo cancellation functions on the TMS320C6201DSP platform.
The main technical progress in promoting VoIP
| Protocols and standards | Software | Hardware |
| The | Weighted Fair Queueing method | Dsp |
| MPLS Tag Switching | Weighted Random early detection | Advanced ASIC |
| RTP, RTCP | Double funnel universal cell rate algorithm | Dwdm |
| Rsvp | Rated access crash rate | Sonet |
| Diffserv, CAR | Cisco Express Forwarding | CPU processing power |
| g.729, G.729a:cs-acelp | Extended Access table | Adsl,radsl,sdsl |
| frf.11/frf.12 | Token bucket algorithm | |
| Multilink PPP | Frame Relay Data rectifying form | |
| Sip | Priority-based Cos | |
| Packet over SONET | Integration of IP and ATM Qos/cos |
Protocol and standard software hardware weighted Fair queueing method DSP MPLS tag switching weighted random early detection advanced ASIC RTP, RTCP dual funnel universal cell rate algorithm DWDM RSVP rated access crash rate SONET Diffserv, CAR Cisco fast Forwarding CPU processing power g.729, G.729a:cs-acelp extended Access table ADSL,RADSL,SDSL frf.11/frf.12 token bucket algorithm multilink PPP Frame Relay data rectifier SIP priority based cos Packet ov Integration of ER SONET IP and ATM Qos/cos
2. Advanced ASIC (application-specific Integrated circait, ASIC) has developed a faster, more complex and more powerful Asics. ASIC is a dedicated application chip that performs a single application or a small set of features. Because they focus on narrow application targets, they can be highly optimized for specific functions, usually with a double-universal CPU that is one or several orders of magnitude faster. Just as the thin instruction set computer (RSIC) chip concentrates on the fast execution of the throw limit number, the ASIC is preprogrammed, enabling it to perform a limited number of functions more quickly. Once developed, the cost of ASIC batch production is not high, and is used for network devices, such as routers and switches, to perform routing tabular, packet forwarding, packet classification and inspection, and queueing functions. The use of Asics enables higher performance and lower costs for devices. They provide increased broadband and better QoS support for the network, so they play a big role in promoting VoIP development.
3. Transmission of IP transmissionTelecomMost of the network uses time division multiplexing, the Internet should be used for statistical multiplexing variable-length packet switching, compared to the latter, which is high utilization of network resources, interconnection and interoperability is simple and effective, the data service is very applicable, this is the Internet to develop a rapid development of one of the important reasons. However,BroadbandIP network communication has put forward the requirements of QoS and delay characteristics, so the technology development of statistical multiplexing variable length packet switching has been paid attention to. Currently, in addition to the advent of the new generation of IP protocol--ipv6, the World Internet Engineering Task Force (IETF) proposed multi-protocol tag switching (MPLS), which is based on the network layer routing of various tag/label exchange, can improve the flexibility of routing, expand the network layer routing capability, Simplifies the integration of routers and cell-based exchange to improve network performance. MPLS can be used as an independent routing protocol, and compatible with existing network routing protocols, and supports various operation, management and maintenance functions of IP network, so the QoS, route, signaling and other performance of IP network communication are greatly improved to reach or close to the level of statistical multiplexing fixed-length packet switching (ATM). And it is simpler, more efficient, cheaper and applicable than ATM. The IETF is also seizing new packet-holding technique in order to realize QoS routing. The "tunneling technology" is being researched in order to realize the broadband transmission of unidirectional link. In addition, how to choose IP Network transmission platform is also an important field of research in recent years, has appeared IP over ATM, IP-over-SDH, IP over DWDM and other technologies
The first layer is the basic level, providing high-speed data transmission backbone. IP layer to IP users with high-quality, with a certain service guarantee of IP access services. The user layer provides access forms (IP access and broadband access) and service content forms. In the base layer, Ethernet as the physical layer of IP network is a matter of course, but IP OVERDWDM is the latest technology, and has great potential for development.
Dense wavelength division multiplexing (Dense wave division MULTIPLEXING,DWDM) injects new vitality into the fiber network and provides a staggering bandwidth in telecommunications companies laying out new fibre backbone networks. DWDM technology utilizes optical fiber capabilities and advanced optical transmission equipment. The name of wavelength division multiplexing is obtained by transmitting multiple wavelengths of light (laser) from a single strand of fiber. The current system is capable of transmitting and identifying 16 wavelengths, and future systems can support 40~96 full wavelength. This is important because each additional wavelength increases the flow of information. The 2.6GBIT/S (OC-48) network can therefore be expanded 16 times times without the need to lay out new fibers.
Most new fiber networks run OC-192 at (9.6gbit/s) speeds, producing more than 150gbit/s of capacity on a pair of fibers when combined with DWDM. In addition, DWDM provides interface protocols and speed-independent features that enable the transmission of ATM, SDH, and Gigabit Ethernet signals on a single fiber, which is compatible with all networks now built, so DWDM can protect existing capital It also provides a more powerful backbone for ISPs and telecoms companies with its huge bandwidth, and makes broadband costs lower and more accessible, providing strong support for the bandwidth requirements of VoIP solutions. The increased transfer rate not only provides a thicker pipeline, reduces the chance of blocking, and slows down a lot of delays, thus reducing the QoS requirements on the IP network to a large extent.
4. Broadband Access Technology
The user access of IP network has become the bottleneck restricting the development of the whole network. From a long-term development perspective, the ultimate goal of user access is fiber-to-home (FTTH). The optical access network includes two kinds of optical digital loop carrier system and passive optical network in a broad sense. The former mainly in the United States, combined with open mouth v5.1/v5.2, on the optical fiber transmission of its integrated system, showing a great vitality. The latter is mainly in the eyes of the book and Germany. Japan has been working tirelessly for more than 10 years, taking a series of measures to reduce the cost of passive optical networks to a level similar to that of copper and metal twisted-pair wires, and to use them extensively. In recent years, the ITU has proposed an ATM-based passive optical network (APON), the advantages of ATM and passive optical network, access rate of up to 622M bit/s, broadband IP multimedia business development is very advantageous, and can reduce the failure rate and number of nodes, expand coverage. At present, the ITU has completed the standardization work, the manufacturers are actively developing, will soon have a commodity listing, will become the 21st century-oriented broadband access technology, the main direction of development.
At present, the main use of access technology are: PSTN, Iadn, ADSL, CM, DDN, X. and Ethernet, as well as broadband wireless access system columns. These access technologies have their own characteristics, the fastest development is ADSL and cm;cm (Cable Modem) using coaxial cable, high transmission rate, strong anti-jamming ability, but not two-way transmission, no unified standard. ADSL (asymmetrical Digital Loop) exclusive access to broadband, fully benefit from the existing telephone network, provide asymmetric transmission rate, the user side of the download rate can reach 8 Mbit/s, the user side upload rate can reach 1M bit/s. ADSL provides the necessary broadband for businesses and individual users, and greatly reduces costs. With lower-cost ADSL area loops, companies can now access the Internet and Internet service provider-based VPNs at higher speeds, allowing higher VoIP call capacity.
5. Central Processing Unit Technology
The central processing unit (CPU) continues to evolve in terms of function, power, and speed. This enables the multimedia PC to be widely used and improves the performance of system functions that are subject to CPU power limitations. The ability of a PC to process streaming audio and video data is long awaited in the user, so sending voice calls on a data network is certainly the next goal. This computing feature enables advanced features in advanced multimedia desktop applications and networking components to support voice applications.
Read (31) | Comments (0) | Forwards (0) |0Previous: Basic principles of VoIP
Next: What kind of job is a programmer?
Related Popular articlesThe principle and technology of VoIP
Start building with 50+ products and up to 12 months usage for Elastic Compute Service
1 on 1 presale consultation
24/7 Technical Support 6 Free Tickets per Quarter Faster Response
Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.
A comprehensive suite of global cloud computing services to power your business
Payment Methods We Support
