NTP network time device, GPS Beidou time device-Jingzhun Electronic Technology
NTP network time device, GPS Beidou time device-Jingzhun Electronic Technology
NTP network time device, GPS Beidou time device-Jingzhun Electronic Technology
Consult me Beijing Jingzhun Electronic Technology, please + 官 微 ahjzsz
1. Introduction
As the basic supporting technology of digital communication networks, the development and evolution of clock synchronization technology has always been driven by the development of communication network technologies. In terms of network, the communication network has evolved from analog to digital, from TDM network to packet network; in terms of business, from TDM voice service to multi-service model based on packet service, from fixed The voice business is mainly developed to focus on both fixed and mobile voice services, from narrowband services to broadband services and so on. In terms of the transmission technology that is closely related to the synchronization network, it has developed from coaxial transmission to PDH, SDH, WDM and DWDM, as well as the latest OTN and PTN technologies. With the continuous development of new communication services and new technologies, its synchronization requirements are getting higher and higher. The basic clock technologies including clock sources and phase-locked loops have undergone many updates. The synchronization technology is also constantly being updated, and the time synchronization technology is more It is currently the focus of industry attention.
2. Development of clock technology
The most basic technologies involved in clock synchronization include clock source technology and phase-locked loop technology. With the continuous improvement of application requirements and the continuous improvement of technology and process, clock source technology and phase-locked loop technology have also been rapidly evolved and developed.
(1) Zhongyuan Technology
The clock oscillator is the basic component of all digital communication equipment. According to the application time, the clock source technology can be divided into ordinary crystal clocks, high-stability crystal oscillators with constant temperature baths, atomic clocks, and chip-level atomic clocks.
The accuracy of general crystal oscillators is between nE-5 ~ nE-7. Due to the advantages of cheap price, small size and low power consumption, crystal oscillators are widely used in various industries and fields. However, ordinary crystal clocks are generally greatly affected by the ambient temperature. Therefore, later, crystal clocks with constant temperature baths, and even high-stability crystal clocks with double constant temperature baths, have greatly improved their performance. With the continuous development of communication technology, higher requirements have been placed on the accuracy and stability of the clock. The crystal clock source has been difficult to meet the requirements. The atomic clock technology has been applied. The rubidium clock and the cesium clock are the most representative atomic clocks. Generally speaking, the accuracy of rubidium clock can reach or better than nE-10, while cesium clock can reach or better than 1E-12.
However, due to the large size, high power consumption, and short life, the application of atomic clocks in some fields is limited, and chip-level atomic clocks are expected to solve this problem. At present, the civilian chip-level atomic clock is basically in the experimental stage, its size is only on the order of cubic centimeters, the power consumption is only on the order of 100 milliwatts, it does not consume atoms, and it extends the service life. Good stability. The chip-level atomic clock will have broad application prospects in the fields of communications, transportation, power, finance, national defense, aerospace, and precision measurement.
(2) PLL technology
Phase-locked loop technology is a circuit technology that synchronizes the output signal with the input signal in frequency and phase, that is, when the system uses the phase-locked loop technology to enter the locked state or the synchronized state, the system's oscillator output signal and the input signal The difference is zero, or kept constant. Phase-locked loop technology is the core technology of clock synchronization. It has experienced the era of analog phase-locked loop technology and digital phase-locked loop technology until it has developed to today's intelligent phase-locked loop technology.
Each component of the analog phase-locked loop is realized by an analog circuit, which is generally composed of three parts: a phase detector, a loop filter, and a voltage-controlled oscillator. The phase detector is used to identify the phase difference between the input signal and the output signal. And the output voltage error, its noise and interference components are filtered by the low-pass loop filter to form the control voltage of the voltage-controlled oscillator. The result of the action on the voltage-controlled oscillator is to pull its output oscillation frequency toward The frequency of the loop input signal is locked when the two are equal.
Compared with the analog phase-locked loop, the error control signal in the digital phase-locked loop is a discrete digital signal rather than an analog voltage, so the controlled output voltage changes are discrete rather than continuous. In addition, the loop components are all implemented with digital circuits, which improves the problem of poor stability of the analog phase-locked loop. With the development of digital technology, intelligent phase-locked loop technology has emerged, that is, direct digital frequency synthesis (DDS-Digital Direct Frequency Synthesis) technology. The intelligent all-digital phase-locked loop can be implemented in a single FPGA. With the help of the phase-locked loop state monitoring circuit, the CPU can shorten the lock time of the phase-locked loop, and gradually improve the jitter characteristics of its output frequency to achieve the best phase-locked and frequency output effect.
3. Analysis of current status of synchronization technology
The synchronization technology includes frequency synchronization technology and time synchronization technology. The frequency synchronization technology is relatively mature and will not be described in detail. The following will analyze the needs of the time synchronization in the communication field and the existing time synchronization technology that has been applied in the communication field.
3.1 Time synchronization requirements
Time synchronization has more and more extensive requirements in the field of communication. The requirements for time synchronization of various communication systems can be divided into high-precision time requirements (microsecond and nanosecond level) and ordinary precision time requirements (millisecond level and second level ).
(1) High precision time requirements
For CDMA base stations and cdma2000 base stations, the time synchronization requirement is 10 μs; for TD-SCDMA base stations, time synchronization requirement is 3 μs; for WiMAX systems and LTE, time synchronization requirements are 1 μs or even sub-microseconds, which requires time The synchronization service level needs to reach the order of 100ns. If the time synchronization between the base station and the base station fails to meet the above requirements, it may result in instruction mismatch in the selector, resulting in the call connection cannot be established normally.
For a location-based service in a 3G network, if a mobile phone receives wireless signals from multiple nearby base stations for positioning, the base stations must be time-synchronized. Generally speaking, the time synchronization error of 10ns will cause a position error of several meters, and the time accuracy required by different precision position services is also different.
(2) Time requirement for ordinary precision
For No.7 signaling monitoring system, in order to avoid false information due to the sequence error of signaling, the time information of all signaling streams must be accurate, and the time synchronization requirement is 1ms. For the charging systems of various switching networks, in order to avoid large time deviations between switches that may cause conflicting bills, the time synchronization requirement is 0.5s. For the network management system of various services, in order to effectively analyze the source of the fault and the consequences caused by it, to locate the fault and find the cause of the fault, the time synchronization requirement is 0.5s.
For RSTP in streaming media services based on IP networks, it provides a robust protocol for streaming media to achieve multipoint delivery and single delivery by on-demand. RTSP uses a timestamp method to ensure the QoS of streaming media services. For e-commerce based on IP networks, in order to ensure the security of the SSL protocol, the "time stamp" method is used to solve the "information retransmission" attack method, and the time synchronization requirement is at least about 0.1s. A large number of computer-based devices and application systems (such as mobile business systems, integrated query systems, customer service systems, etc.) in communication networks generally support NTP, and the time synchronization requirements are in the second or minute level.
3.2 Existing time synchronization technology
To meet the time synchronization requirements of different precisions, the following time synchronization technologies are mainly applied in the communication network:
(1) IRIG-B(Inter Range Instrumentation Group)和DCLS (DC Level Shift)
IRIG coding originates from recording time information for magnetic tape, with obvious analog technical colors, and has been widely used as a time transfer standard since the 1950s. Both IRIG-A and IRIG-B were developed in 1956. They have the same principle, but they use different carrier frequencies, so their resolutions are different. IRIG-B uses a 1kHz sine wave as the carrier frequency for amplitude modulation and encodes the nearest second. The contents included in the frame of IRIG-B include days, hours, minutes, seconds and control information, etc., which can be transmitted in the building using ordinary twisted-pair wires, or can be transmitted over a long distance on the analog telephone network. In the 1990s, in order to adapt to the need for the representation of years in the turn of the century, IEEE 1344-1995 stipulated a new format for IRIG-B time codes, which required the year to be included in the code, and other aspects did not change.
DCLS is another transmission form of IRIG code, that is, it uses DC potential to carry symbol information, which is equivalent to the envelope of IRIG modulation code. DCLS technology is more suitable for intra-office twisted pair transmission. When using this technology for inter-office transmission time, it is necessary to manually compensate for the fixed delay involved in the transmission system. The accuracy of IRIG can usually only reach the order of 10 microseconds.
(2) NTP(Network Time Protocal)
There are three main protocols for transferring time in a computer network: Time Protocol, Daytime Protocol, and Network Time Protocol (NTP). In addition, there is a simple network time protocol (SNTP) that is only used on the user side. The time server on the Internet will continuously monitor the timing requirements of the above protocol on different ports and send the time code in the corresponding format to the client. Among the above network time protocols, the NTP protocol is the most complicated, and the time accuracy that can be achieved is relatively high. In RFC-1305, the network structure, data format, server authentication, weighting, and filtering algorithms for running NTP are fully specified. NTP technology can be applied in local area networks and wide area networks, and the accuracy can usually only reach milliseconds or seconds.
In recent years, improved NTP has also appeared. Unlike traditional NTP, improved NTP generates and processes timestamp marks at the physical layer, which requires hardware modifications to the existing NTP interface. The improved NTP still uses the algorithm of the NTP protocol and can communicate with the existing NTP interface. Compared with the original NTP, its time accuracy can be greatly improved. At present, there are few devices supporting improved NTP, and its accuracy and applicable scenarios need to be further studied. The improved line NTP claims to be on the order of ten microseconds.
(3) 1PPS (1 Pulse per Second) and serial port ASCII character string
The second pulse signal does not contain time information, but its rising edge marks the exact start of each second. It is usually used for local testing and can also be used for intra-office time distribution. Through the RS232 / RS422 serial communication port, the time information is encoded in the ASCII code string mode, the baud rate is generally 9600bit / s, the accuracy is not high, usually also need to use 1PPS signal. Since there is currently no unified standard for the ASCII character string of the serial port, the devices of different manufacturers cannot communicate with each other, so the method has a relatively small application range. By 2008, China Mobile stipulated the specifications of the 1PPS + ToD interface, and ToD information used a binary protocol. 1PPS + ToD technology can be used for intra-office time transmission, which requires manual compensation for transmission delay. Its accuracy can usually only reach the order of 100ns, but it cannot achieve long-distance inter-office transmission.
(4) PTP(Precision Time Protocal)
The implementation principles of PTP and NTP are based on two-way peer-to-peer transmission delay. The biggest difference is the generation and processing of time tags. PTP uses the timestamp marking of the physical layer to obtain a time accuracy much higher than NTP. The PTP technology based on IEEE-1588 was originally used for industrial control that requires strict timing coordination. In order to comply with the rapid growth of the demand for high-precision time synchronization in communication networks, IEEE-1588 has evolved from the original version 1 to version 2 and has been in synchronization. It is used in equipment, optical transmission equipment, and 3G base station equipment.
In China, PTP technology is mainly based on optical transmission systems to achieve high-precision time transmission. In recent years, domestic operators have carried out research on transmitting high-precision time through terrestrial transmission systems, and conducted a large number of experiments in laboratories and on the Internet , And has achieved certain results, has surpassed the research level of foreign related parties. At present, the domestic PTP inter-office time transmission has been realized in a certain scale network environment, and the accuracy can reach the microsecond level.
4. Outlook of Synchronous New Technology
Compared with the mature frequency synchronization technology, the time synchronization technology led by PTP technology has emerged. Emerging time synchronization and existing frequency synchronization are relatively independent of each other, but in the long run, the unification of frequency synchronization and time synchronization is an inevitable trend of development. This concept serves as a guide for readers.
A new type of timing interface is defined in the ITU-T J.211 standard, namely DTI (DOCSIS Timing Interface). DTI is applied to wired cable networks, and achieves frequency and time synchronization on a single cable through protocol interaction. The basic working principle of DTI is: a DTI cable is used for connection between the server and the client. After acquiring the accurate time stamp and reference frequency signal, the server corrects the local clock and outputs the DTI signal to the downstream DTI client. Both sides of the server and the client send and receive DTI messages through the ping-pong mechanism without interruption, so as to achieve synchronization between the DTI client and the server. DTI uses the 1, 2 pins of the RJ45 interface for ping-pong transmission of the transceiver protocol to minimize the time error introduced by the delay asymmetry of transmission in both directions and minimize crosstalk. With the continuous development of technology, DTI technology will be gradually applied to the field of communication, that is, the universal timing interface technology.
The universal timing interface technology can be directly applied to an optical fiber (rather than an optical transmission system) to achieve tens of kilometers of non-relay transmission. With the continuous development of technology, the transmission of hundreds of kilometers or even thousands of kilometers can be achieved by cascading, and the transmission of time precision of hundreds of nanoseconds or even higher order can be truly achieved. Relevant experiments show that the time transmission within 10 ns can already be achieved on 80 km of optical fiber. For the general timing interface technology directly based on optical fiber transmission, the unequal impact caused by the traditional time transmission technology based on the optical transmission system can be avoided. Moreover, after adopting the single-fiber bidirectional transmission technology, the universal timing interface technology can automatically monitor and calculate the one-way propagation delay and realize automatic compensation of the delay, thereby solving the difficulty of the traditional optical transmission system-based time transmission technology Delay automatic compensation problem.
Another advantage of the universal timing interface technology is that it can provide unified time and frequency synchronization at the same time. It can be well compatible with the existing frequency synchronization network and time synchronization network, and compatible with all the systems and devices in the existing communication network that need to be synchronized. China's traditional frequency synchronization network can only be traced to the cesium atomic clock independently operated by each operator. In the next few years, the time synchronization network can only be traced to UTC through satellite timing receivers. If the universal timing interface technology is used, even when the time signal is traced to the satellite timing system, automatic delay compensation can also be achieved in the application of satellite receiver antenna feeder delay compensation. Specifically, a fixed delay is introduced in the antenna feeder part of the satellite receiver of the time source device; for different types of antenna feeders of different lengths, the delay cannot be compensated according to a uniform empirical value (for example, 4-5 ns / m), especially After the lightning arrester, amplifier, distributor, and connector are connected in series, the delay error is more difficult to control. If the universal timing interface technology with automatic delay compensation is adopted between the mushroom head and the satellite receiver, the synchronization accuracy of the time source device can be effectively guaranteed. However, based on optical fiber and adopting the universal timing interface technology, the existing frequency reference and time reference can also be traced to the ground-level national time-frequency reference, so as to fundamentally get rid of the dependence on satellite timing system. Thus, an optical fiber time synchronization network that can simultaneously provide high reliability and high quality time and frequency services is realized.
The standardization and specific implementation of the universal timing interface technology and optical fiber time synchronization network technology need to be further studied.
5 Conclusion
In summary, the emergence of miniaturized, low-power chip-level atomic clocks is undoubtedly an epoch-making and impactful revolution in the field of clock technology; and the introduction of universal timing interface technology and optical fiber time synchronization network technology is also a synchronization network technology. Development has injected new vitality. Since China's research on high-precision time synchronization has been at the forefront of the world, follow-up research should be actively carried out on new synchronization technologies.