Abstract: Recent advances in miniaturization and low-cost, low-power design have led to active research in large-scale networks of small, wireless, low-power sensors and actuators. Time synchronization is critical in sensor networks for diverse purposes including sensor data fusion, coordinated actuation, and power-efficient duty cycling. Though the clock accuracy and precision requirements are often stricter than in traditional distributed systems, strict energy constraints limit the resources available to meet these goals.We present Reference-Broadcast Synchronization, a scheme in which nodes send reference beacons to their neighbors using physical-layer broadcasts. A reference broadcast does not contain an explicit timestamp; instead, receivers use its arrival time as a point of reference for comparing their clocks. In this paper, we use measurements from two wireless implementations to show that removing the sender's nondeterminism from the critical path in this way produces high-precision clock agreement (1.85 ± 1.28μsec, using off-the-shelf 802.11 wireless Ethernet), while using minimal energy. We also describe a novel algorithm that uses this same broadcast property to federate clocks across broadcast domains with a slow decay in precision (3.68 ± 2.57μsec after 4 hops). RBS can be used without external references, forming a precise relative timescale, or can maintain microsecond-level synchronization to an external timescale such as UTC. We show a significant improvement over the Network Time Protocol (NTP) under similar conditions.

Abstract: System and method for enabling signal acquisition in a satellite positioning system (SPS) when signals from SPS satellites are attenuated by the operating environment of a SPS receiver. A preferred embodiment comprises a communications server (for example, communications server 220) coupled to a SPS receiver (for example, SPS receiver 210) at one end and a time server (for example, time server 225) by a public network (for example, the Internet 230). Preferably, the communications server 220 is coupled to the Internet 230 via a wireless network to facilitate maximum mobility and flexibility. The communications server 220 queries the time server 225 for the current time and then provides the current time to the SPS receiver 210. The SPS receiver 210 makes use of the current time to assist it in signal acquisition.

Abstract: According to a disclosed embodiment, a communication device comprises an air interface module which is synchronized in time with a universal time source such as Global Positioning System ('GPS'). The communication device further comprises a network time server which is synchronized with the universal time source. The communication device further comprises an address server configured to provide an address of the network time server to a CPU included in the communication device. The CPU is configured to synchronize time with the network time server and provide the synchronized time to a network interface. The network interface is configured to communicate the synchronized time to a user computer. For example, the user computer can run Network Time Protocol, NTP, to facilitate updating the system clock of the user computer using the synchronized time.

Abstract: Various approaches seek to optimize the quality of service of VoIP applications. We propose a system that uses synchronized time to combine the useful characteristics of both fixed and adaptive buffer strategies, thereby improving VoIP quality of service. Using a combination of global positioning system (GPS) technologies and the network time protocol (NTP), hosts can learn the precise end-to-end delay for each packet. This information can benefit both domestic and business Internet telephony users. We outline our proposed system and discuss issues arising from the use of synchronized time.

Abstract: A Hard Real Time Control Center (HRTCC), comprised of hardware, software and firmware, with time synchronisation and time delay compensation methodologies that allows Application Hardware and/or User Input Devices to be networked together on any communications network as if there were negligible network delays in the system, is disclosed. This will allow Application Hardware and/or User Input Devices (connected to an HRTCC at one location (node) on the network) to control or operate Application Hardware and/or User Input Devices connected to another HRTCC at a remote location without the detrimental effects of network time delays. The time synchronisation of the various HRTCCs on the network can be enabled using hardware (e.g. a global positioning system (GPS)) or any other software method (e.g. Network Time Protocol). Using time stamps from the time synchronisation, the time delay of the signals (data) transferred over the network can be determined. The main embodiment of the time delay compensation methodology is an estimator/predictor algorithm. The estimator generates signal information that allows the predictor, using the time delay, to project the signal information characteristics into the future by an amount equal to the time delay. If this predicted signal is used rather than the delayed signal, there will be no readily apparent time delay in the system thereby significantly improving the stability and performance of the associated application. Any software architecture can be used such as servant-client, token ring or peer to peer.

Abstract: A method for synchronizing a time architecture of a mobile network that includes a plurality of network components, each having a time subsystem clock that is updated using a network time protocol (NTP) program The method includes operating the components in a hierarchical fashion and booting up all the network components such that execution of a NTP sub-program for making small incremental changes to the each component time subsystem clock is delayed Additionally, the method includes iteratively executing a NTP sub-program for making large changes to each component time system clock until each component time sub-system clock establishes an initial synchronization with a parent component time sub-system clock Furthermore, the method includes iteratively executing the NTP sub-program for making small incremental changes once initial synchronization is established such that each component time sub-system clock maintains synchronization with the parent component time sub-system clock

Abstract: A time information management method and system for providing time information necessary for an embedded system, without using a hardware real-time clock device. The method and system comprises accessing a time server providing real-time information through the Internet during system initialization; acquiring the real-time information from the time server through a message exchange with the time server; and defining the acquired real-time information as the system's time information and updating the time information depending on a timer operation of the real-time operating system.

Abstract: A local time of a broadband loop carrier terminal is synchronized to a network time received from a source external to the broadband loop carrier terminal by comparing the local time to the network time to generate an error offset, and adjusting the local time based on the error offset. The network time may be received in response to a Network Time Protocol (NTP) poll transmitted by the broadband loop carrier terminal. In some cases, the poll is transmitted over a communication medium using an asynchronous packet protocol, such as Ethernet or TCP/IP. In one embodiment, the local time is based on the output of a local oscillator (e.g., a voltage controlled oscillator), the frequency of which may be adjusted using control logic which uses a filtered version of the error offset along with a factory calibration and/or temperature compensation to produce a control voltage to control the oscillator frequency.

Abstract: The NTP (Network Time Protocol) is widely used in various networks, such as from LAN to WAN. In the case of Internet, the time synchronization accuracy is degraded due to the property of the asymmetric structure of the network between the NTP server and the client. We propose a new technique to reduce the effect of this problem using double packets method.

Abstract: The research discussed in this paper builds on the success of the network time protocol (NTP) which has been synchronizing clocks on hosts all across the Internet for many years. With a different end goal (NTP seeks to setup and maintain agreement of UTC time in network-connected systems) we augment NTP methods with hardware to improve the accuracy of synchronization that is possible within an ethernet subnet by more than three orders of magnitude. Our approach involves a study of the sources of nondeterminism in NTP exchanges and their removal by special hardware aids. The added hardware has a real-time synchronized clock output and a programmable 'event synch' output that can be used to synchronize attached real-time equipment. The hardware aids allow us to achieve much better synchronization precision (100 nsec) while also facilitating the convergence to synchronization in a vastly shorter time-about one minute rather than hours or days as required by NTP.

Abstract: PROBLEM TO BE SOLVED: To provide a method and a system for setting time, which can autonomously set the time in a network and synchronize the time in a network element (NE) by itself SOLUTION: The system includes an element, which is connected to an optical network and specified as a NTP client, a device, which is connected to an IP network served by a NTP server and the optical network, respectively, and a device, which is specified as a NTP proxy to relay communication between the element specified as the NTP client and the NTP server These element individually includes a broadcast means for broadcasting packets, which includes information for identity to identify whether the element is specified as the NTP client on its own or whether the element is specified as the NTP proxy on its own, over the optical network with a predetermined timing, and an acquiring means for acquiring the packet broadcasted by the other element connected to the optical network COPYRIGHT: (C)2004,JPO&NCIPI

Abstract: Nowadays many software packages allow you to keep the clock of your personal computer synchronized to time servers spread over the internet. Here we present how a didactic laboratory can evaluate, in a statistical sense, the minimum synch error of this process (the other extreme, the maximum, is guaranteed by the code itself). The measurement set-up utilizes the global positioning system satellite constellation in `common view' between two similar timing stations: one acts as a time server for the other, so the final timing difference at the second station represents the total synch error through the internet. Data recorded over batches of 10 000 samples show a typical RMS value of 35 ms. This measurement configuration allows students to obtain a much better understanding of the synch task and pushes them, at all times, to look for an experimental verification of data results, even when they come from the most sophisticated `black boxes' now readily available off the shelf.

Abstract: The network delay is one of the essential elements h)r the Quality of Service (QoS). The time delay is measured by placing devices at both ends of the network and using the machine's internal time clock. The accuracy of the measurement is dependent on the precision of tihe clock, therefore the precision of the clock is very important. Network Time Protocol (NTP) is applied for the time synchronization amnong network nodes; hence, the time accuracy at each NTP server is considered very importmLt. Moreover, the improvement of the accuracy is required for the recent, broadband network.Our objective in this research is to improve the time stability of the top NTP server (stratum1) by developing a server with higher stability. The server, which is called the high-accurate time server, was designed based on external high accurate frequency signal, insted of the crystal oscillator on the motherboard of a Personal Computer (PC). The high-accurate time server is capable of the time stability of several hundred nanoseconds. Also evaluated in this paper is the high precision measurement of network delay, which was proceeded with the server.

Abstract: Firstly, the main theory and technology of computer clock synchronization is introduced and analyzed, the three main parts of a synchronization system are studied. Secondly, the design methodology for absolute physical clock synchronization system is studied, which is followed by a detailed analysis and performance comparison of two important synchronization algorithms: CRI algorithm and PCS algorithm. In the end of the paper, we briefly introduce the status quo of synchronization technology based on NTP.

Abstract: PURPOSE: A system for setting a time and a method thereof are provided to reduce the number of key operations by setting a time by connecting to a host such as a computer or a mobile communication terminal CONSTITUTION: A real time clock system(30) is connected to a host(20) The host is connected to a time server through the Internet A memory(28) includes a program with respect to a time setting and standard time of many countries An input unit(25) selects an automation mode and a setting command of the program, and selects a wanted standard time If the program is set as an automation mode by the input unit(25), a control unit(21) sets a time of a mobile device as the selected standard time The second interface unit(33) of the real time clock system(30) is connected to the first interface unit(23) of the host(20) The second interface unit(33) makes data and a command be transmitted between the control unit(21) and a control unit(31) of the real time clock system(30) through the first interface unit(23) An external device(36) is controlled by the control unit(31)

Abstract: This paper describes the analysis, implementation and performance of a new clock synchronization algorithm, which utilizes the physical properties of synthetic quartz crystals. Thus a very high accuracy (down to a few milliseconds) can be achieved. This algorithm adjusts itself to the preferences of the local quartz oscillator and dynamically to the changing environment conditions such as temperature, and humidity via a additional phase locked loop control loop. As the computing and bandwidth requirements could be kept pretty low, this algorithm could be very well used in mobile environment.

Abstract: PURPOSE: A method for controlling standard reference time of a mobile communication server is provided to enable a mobile communication server to receive a standard reference time, set a reference time and operate as a reference time server for multiple servers. CONSTITUTION: A GPS receiver repeatedly receives GPS signals through a GPS antenna(S100). A reference time server repeatedly receives standard time data from the GPS receiver and sets standard reference time of the reference time server(S110). It is discriminated whether there is a request for the standard reference time data from multiple servers connected to the reference time server(30)(S120). If there is a request for the standard reference time data form the multiple servers, the reference time server transmits the standard reference time data to the multiple servers(S130). The multiple servers resets standard reference time for synchronization(S140).

Abstract: This paper mainly discusses how to get accurate UTC time by means of GPS receiver,and how to realize the system time consistent with UTC under the distributed realtime OS QNX.At the same time, the way of building the LAN time server and implementing an algorithm for LAN time synchronization among a number of computers is described too.A detailed application example is given at the end.

Abstract: PURPOSE: A method and a system for managing time information of an embed system are provided to supply time information necessary for an embed system without using a hardware device. CONSTITUTION: In a method for managing time information at an embed system having an Internet connection function and adopting a real time operating system, a time server for supplying real time information in a system initialization is connected through the Internet and the real time information is received from the time server by a message communication to the time server(406). The received real time information is set as time information(412) and updated in accordance with a timer operation of a real time operating system(420). The time information being updated is stored in a non-volatile storage(424) in a system terminating time by a user's request or a fixed time(422). In the case that the received real time information is later than time information stored in the non-volatile storage(410), a user input request of the current time information is displayed(414) and the current time information is received as the time information(418).

Abstract: The National Institute of Standards and Technology operates an ensemble of Internet time servers. The ensemble of servers receives about 450 million time-stamp requests per day (as of May, 2002). This demand is increasing at a compound rate of almost 9% per month, so that improving the efficiency of the time synchronization process is very important. In addition, these requests are not distributed uniformly among the servers, and the busiest servers are nearly saturated during peak periods. Therefore, the capacity of the system as a whole could be increased if the load were distributed more evenly, and we have investigated methods for achieving this active load balancing. We have also developed a number of algorithms based on the Network Time Protocol that try to make the best use of the time stamps and the available network bandwidth. In particular, the algorithms can be configured to trade off accuracy for cost (defined as the network bandwidth and computer cycles needed to realize a specific performance level), so that the many users who do not need the full accuracy of the system can receive satisfactory service at much lower cost.

Abstract: No one can stop time, but time can stop a system. Users need to be aware of the different flavors of time, depending upon their needs. Most systems require that all their components be on a common time (synchrony), but they can't keep them on time unless they are also on a common frequency (syntony). Other systems simply require syntony. The U.S. Naval Observatory (USNO) is the timing reference for GPS. Its Master Clock (MC) is based upon 72 HP5071 cesium and 17 hydrogen maser frequency standards in three buildings and two sites, and their data are used to generate UTC(USNO). The USNO disseminates and distributes the time not only via GPS, but also with Loran, Network Time Protocol (NTP), and Two-Way-Satellite Time-Transfer (TWSTT). Our emphasis is on robustness through repeated calibration and multiplicity of systems, and we recommend this for our users as well.

Abstract: Packet delay traces are important measurements for analyzing end-to-end performance and for designing traffic control algorithms in computer networks. Due to the fact that the clocks at the end systems are usually not synchronized and running at different speeds, these measurements can be quite inaccurate. We propose several algorithms to estimate and remove the relative clock skews from delay measurements based on the computation of convex hulls. Compared with existing techniques, such as linear regression and linear programming, the convex-hull approach provides better insight and allows us to handle more error metrics. We obtain algorithms which are linear in the number of measurement points for the case with no clock resets. For the more challenging case with clock resets, i.e., the clocks are reset to some reference times during the measurement period, we develop linear algorithms to identity the clock resets, and derive the best clock skew lines. We extend this analysis to environments in which at least one of the clocks is controlled by NTP (network time protocol). These algorithms can greatly improve the accuracy of the measurements, and can be used both online and offline. They can also be extended for active clock synchronization, to replace or further improve NTP. Numerical experiments are presented to demonstrate the robustness of the algorithms.

Abstract: As the Internet is shifting towards a reliable QoS-aware network, accurately synchronized clocks distributed on the Internet are becoming more significant. The network time protocol (NTP) is broadly deployed on the Internet for clock synchronization among distributed hosts, but is weak in asymmetric paths, i.e., it cannot accurately estimate the clock offset between two hosts when the forward and backward paths between them have different one-way delays. In this paper, we focus on estimating the offset and skew of a clock from one-way delay measurement between two hosts, and propose an idea for improvement of such estimations, which reduces estimation errors when the forward and backward paths have different bandwidths, a major factor in asymmetric delays.