WO2017204702A1

WO2017204702A1 - Method and first network node for maintaining time synchronization

Info

Publication number: WO2017204702A1
Application number: PCT/SE2016/050478
Authority: WO
Inventors: Henrik Egnell; Yufeng Zhao; Gang Sun; Stefan Andersson
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 2016-05-24
Filing date: 2016-05-24
Publication date: 2017-11-30

Abstract

A method and a first network node (300) for maintaining time synchronization with other network nodes in the wireless network. When the first network node (300) has access to an absolute time reference (T), a propagation delay (d) is determined for a signal (T+d) received from a neighbouring network node (302) based on the absolute time reference. The first network node (300) then maintains the time synchronization based on the determined propagation delay (d) so that the timing of further signals received from the neighbouring network node (302) corresponds to the determined propagation delay (d). This may be done when the absolute time reference is not available from an external source such as a satellite system.

Description

METHOD AND FIRST NETWORK NODE FOR MAINTAINING TIME

SYNCHRONIZATION

Technical field

The present disclosure relates generally to a method and a first network node of a wireless network, for maintaining time synchronization with other network nodes in the wireless network.

Background

In wireless networks providing radio communication for various wireless devices, it is often required that communication of radio signals and other activities are time synchronized across multiple network nodes so that the network nodes operate according to a common time reference. The term "wireless network" is used herein to denote any network comprising network nodes such as base stations or the like which are capable of wireless communication with wireless devices. Further, the term "wireless device" is used herein to denote any communication equipment that is capable of wireless communication with a wireless network. Some non-limiting examples of wireless device include mobile phone, smartphone, tablet, laptop computer and Machine-to-Machine, M2M, device.

For example, wireless networks using Long Term Evolution, LTE, technology have several features that require very accurate time synchronization so that network nodes therein, also referred to as eNodeBs, can transmit and receive radio signals in a coordinated manner in communication with wireless devices. If Time Division Duplex, TDD, is employed, e.g. in an LTE network, uplink transmissions from wireless devices to network nodes and downlink transmissions from network nodes to wireless devices are performed on the same frequency band such that the uplink and downlink transmissions are separated in time. It can be readily understood that nodes in a network using TDD need to transmit and receive in a synchronized manner so as to avoid interference between the uplink and downlink transmissions and to enable efficient use of available radio resources. Other examples of features that require accurate time synchronization include

coordinated multi-casting of information from several network nodes, positioning of wireless devices, coordination of uplink and downlink transmissions in time-limited subframes or similar, and so forth.

Different TDD configurations have been defined for uplink and downlink

transmissions in specific time intervals called subframes which are comprised in a radio frame that is repeated over time. A subframe is basically defined by a preset time period and a radio frame comprises a predefined number of consecutive subframes, e.g. 10 subframes. Fig. 1 illustrates a radio frame of 10 milliseconds comprising 10 subframes 0-9 of 1 millisecond each. A TDD subframe can also be seen as a radio resource that can be allocated for a transmission. In TDD, each subframe is reserved for uplink transmissions or downlink

transmissions such that the uplink and downlink transmissions do not occur at the same time. In a wireless network that employs TDD, it is possible to use different uplink-downlink, UL-DL, configurations of subframes, e.g. depending on the current need for uplink and downlink radio resources, respectively. The same UL- DL configuration is typically used in a synchronized manner over an extensive area with many cells and serving network nodes, sometimes even across the entire wireless network, to avoid interference between uplink and downlink transmissions. However, the UL-DL configuration may be changed in a dynamic manner depending on the traffic. A set of different UL-DL configurations predefined for LTE is shown in the table of Fig. 2, including seven UL-DL configurations 0-6 each having ten subframes 0-9 comprised in a repeatable radio frame as of Fig. 1 . Subframes reserved for downlink transmissions are denoted D and subframes reserved for uplink are denoted U. There are also "special" subframes denoted S which are divided into three parts including a downlink part, a guard period, and an uplink part. The network nodes thus need to switch between a downlink subframe and an uplink subframe simultaneously, which can be achieved if the nodes use the same time reference for time synchronization.

One way of achieving the required time synchronization in TDD and other situations is to align an internal clock in every network node to one and the same time reference, herein referred to as an "absolute time reference", so that all network nodes can operate in time according to the absolute time reference. Such an absolute time reference can usually be obtained from signals transmitted from satellites of as a Global Navigation Satellite System, GNSS, such as the Global Positioning System, GPS. The GNSS satellites transmit a clock signal as the absolute time reference, which can be received by any network nodes having GNSS coverage to enable the above time synchronization. Alternatively, the absolute time reference can be obtained by means of the Network Time Protocol, NTP or the Precision Time Protocol, PTP, both of which distribute a clock signal over a network.

However, it happens frequently that a network node loses its GNSS coverage as the satellites change their positions and/or obstacles occur. In case the absolute time reference becomes unavailable for whatever reason, the network node needs to maintain its time synchronization internally as long as possible until the absolute time reference hopefully becomes available again. The time synchronization can be maintained internally, e.g. by means of an advanced control function called Digital Controlled Crystal Oscillator, DCXO, which controls generation of the clock signal in the network node in an extrapolated manner. This operation mode is often referred to as "holdover" mode. This may work well for a short while but soon the internally generated clock signal will drift away from the correct absolute time reference, to become inaccurate and useless. If the GNSS coverage has not been recovered at that point, the network node cannot continue to operate properly when time synchronization is required and it must be shut down. It is thus a problem that a network node can only operate for a limited short time using an internally generated clock signal, when having no access to the absolute time reference. Another problem is that some network nodes and/or networks are not able to handle a clock signal using NTP or PTP. Summary

It is an object of embodiments described herein to address at least some of the problems and issues outlined above. It is possible to achieve this object and others by using a method and a first network node as defined in the attached independent claims.

According to one aspect, a method is performed by a first network node of a wireless network, for maintaining time synchronization with other network nodes in the wireless network. In this method, when the first network node has access to an absolute time reference, the first network node determines a propagation delay for a signal received from a neighbouring network node based on the absolute time reference. The first network node then maintains the time synchronization based on the determined propagation delay so that the timing of further signals received from the neighbouring network node corresponds to the determined propagation delay. Thereby, the first network node is able to obtain and maintain an accurate time synchronization for any extended period of time even if the first network node would lose access to the absolute time reference.

According to another aspect, a first network node of a wireless network is arranged to maintain time synchronization with other network nodes in the wireless network. The first network node is configured to, when the first network node has access to an absolute time reference, determine a propagation delay for a signal received from a neighbouring network node based on the absolute time reference. The first network node is also configured to maintain the time synchronization based on the determined propagation delay so that the timing of further signals received from the neighbouring network node corresponds to the determined propagation delay.

The above method and first network node may be configured and implemented according to different optional embodiments to accomplish further features and benefits, to be described below.

A computer program is also provided comprising instructions which, when executed on at least one processor in the first network node, cause the at least one processor to carry out the method described above. A carrier is also provided which contains the above computer program, wherein the carrier is one of an electronic signal, optical signal, radio signal, or a computer readable storage medium. Brief description of drawings

The solution will now be described in more detail by means of exemplary embodiments and with reference to the accompanying drawings, in which:

Fig. 1 is a schematic illustration of a regular radio frame comprising 10 subframes, which can be used for wireless communication. Fig. 2 is a table with predefined UL-DL configurations which can be used when TDD is employed for wireless communication.

Fig. 3 is a communication scenario where the solution may be employed involving a first network node and at least one neighboring network node, according to some possible embodiments. Fig. 3A is a diagram illustrating the timing of a signal propagating from the neighboring network node to the first network node in Fig. 3.

Fig. 4 is a flow chart illustrating a procedure in a first network node, according to further possible embodiments.

Fig. 5 is a flow chart illustrating an example of a procedure in the first network node for maintaining time synchronization, according to further possible

embodiments.

Fig. 6 is a flow chart illustrating an example of a procedure in the first network node for issuing an alarm, according to further possible embodiments.

Fig. 7 is a flow chart illustrating another example of a procedure in the first network node for issuing an alarm, according to further possible embodiments. Fig. 7A illustrates the first network node and two neighboring network nodes involved in the procedure of Fig. 7.

Fig. 8 is a block diagram illustrating a first network node in more detail, according to further possible embodiments. Detailed description

Briefly described, a solution is provided for a first network node to obtain accurate time synchronization even if the first network node would lose access to an absolute time reference, basically operating in the above-described holdover mode. This can be achieved as follows. First, the first network node determines a propagation delay d for a signal received from a neighboring network node when the first network node actually has access to the absolute time reference, which may be done very seldom or even just once, basically, since the distance between the first network node and the neighboring network node is typically fixed. It is assumed that the first network node and the neighboring network node belong to the same wireless network with synchronized network nodes, or possibly to two different networks which are synchronized to each other.

Then the first network node uses the propagation delay d for maintaining accurate time synchronization so that the timing of further signals received from the neighboring network node corresponds to the determined propagation delay, assuming that the neighboring network node transmits according to the correct absolute time reference. The propagation delay d from the neighboring network node may need to be updated from time to time, e.g. in case the position of one (or both) of the network nodes has been changed.

Fig. 3 is a communication scenario where the above procedure is schematically illustrated, involving the first network node 300, the neighboring network node 302 and a satellite system 304 which provides the absolute time reference denoted T. Since the two network nodes 300, 302 are located with some distance apart, a signal transmitted by the neighboring network node 302 is received at the first network node 300 with some delay due to the signal propagation from node 302 to node 300 over the "propagation distance" indicated in the figure. This propagation delay is denoted d. Fig. 3A illustrates the timing of the signal that propagates from node 302 to node 300 in Fig. 3.

Hence, if the time of transmitting a signal from the neighboring network node 302 is T according to the absolute time reference obtained from the satellite system 304, this signal is received by the first network node 300 at the time T + d and d can thus be calculated as the difference between reception time and T. At some point later, when the absolute time reference T is not available, T can be calculated when knowing d and the time of receiving a further signal from the neighboring network node 302, by subtracting the delay d from the reception time. T can then be used for setting and adjusting an internal time in the first network node so that the internal time corresponds to the absolute time reference. T may be repeatedly determined in this way when receiving successive signals from the neighboring network node 302, and T can be used for continually adjusting or updating the internal time in the first network node 300. The first network node is thereby able to perform communication of radio signals according to the internal time and maintained time synchronization when time synchronization with other network nodes in the wireless network is required.

The signal transmitted by the neighboring network node 302, which is used in the procedure described herein for determining the propagation delay d and for maintaining time synchronization, may be a known reference signal which is repeatedly transmitted at known occasions e.g. at some predefined point within a subframe. For example, when TDD is used in the wireless network, this reference signal may, without limitation, be a Cell-specific Reference Signal, CRS. The CRS is customarily transmitted by network nodes in all downlink subframes and it can be used by wireless devices for cell search, initial acquisition, downlink channel estimation, and measurements of downlink channel quality. However, it should be noted that the solution described herein is not limited to using the CRS, and any detectable signal of known occurrence in time may be used as the above- mentioned reference signal. An example of how the solution may be employed will now be described with reference to the flow chart in Fig. 4, in terms of actions performed by a first network node of a wireless network, for maintaining time synchronization with other network nodes in the wireless network. Without limiting the described features and embodiments, reference will also be made to the example scenario shown in Figs 3 and 3A. The procedure illustrated by Fig. 4 may thus be performed by a network node 300 that is operable in a wireless network e.g. as a base station, eNodeB, access point, etc., depending on the type of network and terminology used. A first action 400 illustrates that when the first network node 300 has access to an absolute time reference, the first network node 300 determines a propagation delay d for a signal received from a neighboring network node 302 based on the absolute time reference. In an example embodiment, the propagation delay d may be determined by comparing reception time of the signal with the absolute time reference. It was described above how the propagation delay d can be determined as the difference between the signal reception time and the absolute time reference. In some further example embodiments, the first network node 300 may obtain the absolute time reference e.g. by any of the above-mentioned Global Navigation Satellite System, GNSS, Network Time Protocol, NTP, and Precision Time Protocol, PTP.

A further action 402 illustrates that the first network node 300 may lose its access to the absolute time reference when the satellite signal cannot be received and detected properly, e.g. due to changed position of the satellite 304 or occurrence of some obstacle blocking the satellite signal. Action 402 means effectively that the first network node 300 needs to start operating in the above-described holdover mode by maintaining time synchronization internally without acquiring and using the absolute time reference which has thus become unavailable to the first network node 300 in this example.

A next action 404 illustrates that the first network node 300 maintains the time synchronization based on the propagation delay d that was determined in action

400, so that the timing of further signals received from the neighboring network node 302 corresponds to the determined propagation delay d. In an example embodiment, the maintaining of the time synchronization may be performed when the first network node 300 has no access to the absolute time reference, i.e. when action 402 has occurred. This way, the correct time reference can be obtained from the further signals received from the neighboring network node 302, e.g. when repeatedly receiving some known reference signal such as the above-mentioned CRS although not limited thereto. Each time the signal is received from the neighboring network node 302, the first network node 300 is thus able to adjust its internal time to coincide with the correct absolute time, basically by subtracting the delay d from the time of receiving the signal, assuming that the neighboring network node 302 is correctly synchronized. It should be noted that the first network node 300 may maintain the time synchronization in this manner based on signals received from multiple neighboring network nodes, and the solution is not limited to using just one neighboring network node.

In further example embodiments, the first network node 300 may then perform communication of radio signals according to the maintained time synchronization, as illustrated by another optional action 406. It may thus be desirable to use the procedure described herein to maintain the time synchronization when the first network node 300 has lost its access to the absolute time reference. However, it is also possible to obtain time synchronization in this way even if the first network node 300 actually has access to the absolute time reference, e.g. when it is desirable to confirm the absolute time reference and/or to check the timing accuracy of transmissions from the neighboring network node 302. In further example embodiments, this communication may involve a time-critical feature including any of: Time Division Duplex (TDD), multi-casting, coordination between network nodes of uplink and downlink transmissions, and wireless device positioning. The term "time-critical feature" thus indicates a feature which requires transmitting and receiving signals with correct timing according to the time synchronization. Some further embodiments that may be used in this procedure will now be described. In another example embodiment, if TDD is used in the wireless network the signal used for determining the propagation delay may be a CRS transmitted by the neighbouring network node 302. In another example embodiment, the time synchronization may be updated at regular intervals, e.g. whenever receiving a suitable reference signal such as the CRS from the neighbouring network node 302.

It was mentioned above that the first network node 300 is able to adjust its internal time to coincide with the correct absolute time when receiving further signals from the neighboring network node 302, basically by subtracting the previously determined propagation delay d from the time of receiving the signal. In

accordance with the latter embodiment above, this adjustment of the internal time can be frequently repeated to update and maintain the time synchronization. A more detailed example of how the first network node 300 may operate when performing action 404 in Fig. 4 will now be described with reference to the flow chart in Fig. 5.

It is assumed that the propagation delay d has been determined when the first network node 300 had access to the absolute time reference, as of action 400. In a first action 500 of Fig. 5, it is detected that access to the absolute time reference is lost, which thus corresponds to action 402 above. A following action 502 illustrates that the first network node 300 receives a signal, e.g. the CRS, that has been transmitted from the neighboring network node 302. In a further action 504, the first network node 300 derives the absolute time reference based on the received signal and the previously determined propagation delay d by subtracting d from the time of receiving the signal.

The first network node 300 then checks, in an action 506, whether the time reference derived in action 504 deviates from the internal time currently used by the first network node 300, or not. If so, the first network node 300 adjusts the internal time to coincide with the derived time reference, in an action 508. The first network node 300 then returns to action 502 and receives the next signal from the neighboring network node 302, and the following actions 504-508 are repeated for updating the internal time if necessary. If no deviation is detected in action 506, actions 508 is skipped and first network node 300 returns directly to action 502. This way, an accurate time synchronization can be maintained by frequently repeating the procedure of Fig. 5, even when the absolute time reference is not available from a satellite system or other external source.

In yet another example embodiment, the first network node 300 may issue an alarm when detecting that signals received from the neighbouring network node 302 are out of synchronization. In that case, another example embodiment, may be that the above detecting comprises determining a propagation delay d for at least one further signal received from the neighbouring network node 302 based on the absolute time reference, and detecting that the propagation delay d has changed which thus indicates that the neighbouring network node 302 may have gone out of synchronization. An example of such an alarm procedure will now be described with reference to the flow chart in Fig. 6 which illustrates an example procedure performed by the first network node 300.

In a first action 600, the first network node 300 obtains the absolute time reference, e.g. from a satellite 304 as described above. In a next action 602, the first network node 300 receives a reference signal from the neighbouring network node 302. In a next action 604, the first network node 300 determines a propagation delay d for the received signal based on the absolute time reference. Action 604 is thus performed basically in the manner described above for action 400 which will not be described here again.

A dashed arrow in the figure indicates that actions 600-604 are then repeated at least once and after obtaining the absolute time reference anew in action 600, the propagation delay d is determined once more, possibly after waiting for a period of time, not shown. The first network node 300 then checks, in a next action 606, whether the propagation delay d has changed since actions 600-604 were performed previously. If d has not changed, actions 600-606 are repeated once more, after waiting for some time in an action 608. However, if the propagation delay d has actually changed in action 606, the first network node 300 issues an alarm e.g. to an Operation and Maintenance, O&M, node or similar, in an action 610, since this indicates that the neighboring network node has gone out of synchronization with the wireless network.

The O&M node is thereby enabled to investigate the neighboring network node further to find out what has gone wrong, which is somewhat outside the scope of this solution. The procedure of Fig. 6 may be performed at predetermined intervals, e.g. according to a preset scheme or the like, to check whether the synchronization of the neighboring network node works properly by comparing the propagation delay d with last time it was checked.

In another example embodiment, the time synchronization may be maintained based on propagation delays determined for signals received from at least two neighbouring network nodes. This embodiment has the advantage that reliability can be increased by checking the synchronization with at least two neighbouring network nodes instead of only one. If the same time reference is derived from both or all neighbouring network nodes, this time reference can effectively be verified. On the other hand, different time references would indicate that either of the neighbouring network nodes may have gone out of synchronization. This can be detected by comparing the propagation delays determined for the respective at least two neighbouring network nodes.

In another example embodiment, an alarm may be issued when detecting that the difference between the propagation delays determined for the at least two neighbouring network nodes has changed, which indicates that either of the first and second neighbouring network nodes may have gone out of synchronization with the wireless network. An example of how this may be done will now be described with reference to the flow chart in Fig. 7 which illustrates another example procedure performed by the first network node 300 when maintaining time synchronization based on propagation delays determined for signals received from two neighbouring network nodes, as also illustrated by Fig. 7A. Throughout this procedure, the first network node 300 has access to the absolute time reference which is used for checking whether either of the neighbouring network nodes 302A, 302B has gone out of synchronization, as follows. In a first action 700, the first network node 300 obtains the absolute time reference, e.g. from a satellite 304 as described above. In a next action 702, the first network node 300 receives a reference signal from a first neighboring network node 302A. In a next action 704, the first network node 300 determines a propagation delay d1 for the received signal based on the absolute time reference. Action 704 is thus performed basically in the manner described above for action 400 which will not be described here again. Corresponding actions are also performed for a second neighboring network node 302B. Thus, a further action 706 illustrates that a reference signal is also received from the second neighboring network node 302B, and another action 708 illustrates that a propagation delay d2 is determined for the signal based on the absolute time reference. It might be noted that the propagation delays d1 and d2 can be different or basically the same, depending on the distance between the first network node and each respective neighboring network node 302A, 302B. A dashed arrow in the figure indicates that actions 700-708 are then repeated at least once and after obtaining the absolute time reference anew in action 700, both propagation delays d1 and d2 are determined once more, possibly after waiting for a period of time, not shown. The first network node 300 then checks, in a next action 710, whether the difference d1 -d2 has changed since actions 700- 708 were performed previously. If d1 -d2 has not changed actions 700-708 are repeated once more, after waiting for some time in an action 712. However, if the difference d1 -d2 has actually changed in action 710, the first network node 300 issues an alarm to an O&M node or similar in an action 714, since this indicates that either of the first and second neighboring network nodes 302A, 302B has gone out of synchronization with the wireless network.

The O&M node is thereby enabled to investigate the first and second neighboring network nodes 302A, 302B further to find out what has gone wrong, which is somewhat outside the scope of this solution. The procedure of Fig. 7 may be performed at predetermined intervals, e.g. according to a preset scheme or the like, to check whether the synchronization of the neighboring network nodes 302A, 302B works properly by comparing the difference d1 -d2 with last time it was checked.

The block diagram in Fig. 8 illustrates a detailed but non-limiting example of how a first network node 800 may be structured to bring about the above-described solution and embodiments thereof. The first network node 800 may be configured to operate according to any of the examples and embodiments of employing the solution as described above, where appropriate, and as follows. The first network node 800 is shown to comprise a processor P and a memory M, said memory comprising instructions executable by said processor P whereby the first network node 800 is operable as described herein. The first network node 800 also comprises a communication circuit C with suitable equipment for receiving and transmitting signals in the manner described herein.

The communication circuit C is configured for communication with wireless devices and other network nodes using suitable protocols depending on the implementation. This communication may be performed in a conventional manner over a communication network employing radio links for wireless communication, which is not necessary to describe here as such in any detail. The solution and embodiments herein are thus not limited to using any specific types of networks, technology or protocols for radio communication. The first network node 800 is operable in a wireless network and comprises means configured or arranged to perform at least some of the actions 400-306, 500-506, 600-610 and 700-714 of the flow charts in Figs 4-7, respectively. The first network node 800 is arranged to maintain time synchronization with other network nodes in the wireless network. When the first network node 800 has access to an absolute time reference, the first network node 800 is configured to determine a propagation delay d for a signal received from a neighbouring network node, not shown, based on the absolute time reference. This operation may be performed by a determining unit 800A in the first network node 800, e.g. in the manner described for action 400 above. The first network node 800 is further configured to maintain the time synchronization based on the determined propagation delay d so that the timing of further signals received from the neighbouring network node corresponds to the determined propagation delay d. This operation may be performed by a

synchronization unit 800B in the first network node 800, e.g. as described for action 404 above.

The first network node 800 may also be configured to issue an alarm when detecting that signals received from the neighbouring network node are out of synchronization. This alarm operation may be performed by an alarm unit 800C in the first network node 800, e.g. as described above for actions 606-608. The first network node 800 may also be configured to issue an alarm when detecting that the difference between propagation delays d1 , d2 determined for signals received from at least two neighbouring network nodes has changed. This alarm operation may be performed by the alarm unit 800C, e.g. as described above for actions 710-712.

It should be noted that Fig. 8 illustrates various functional units in the first network node 800, and the skilled person is able to implement these functional units in practice using suitable software and hardware. Thus, the solution is generally not limited to the shown structures of the first network node 800, and the functional units 800A-C therein may be configured to operate according to any of the features and embodiments described in this disclosure, where appropriate.

The functional units 800A-C described above can be implemented in the first network node 800 by means of suitable hardware and program modules of a computer program comprising code means which, when run by the processor P causes the first network node 800 to perform at least some of the above-described actions and procedures. The processor P may comprise a single Central

Processing Unit (CPU), or could comprise two or more processing units. For example, the processor P may include a general purpose microprocessor, an instruction set processor and/or related chips sets and/or a special purpose microprocessor such as an Application Specific Integrated Circuit (ASIC). The processor P may also comprise a storage for caching purposes. Each computer program may be carried by a computer program product in the first network node 800 in the form of a memory having a computer readable medium and being connected to the processor P. The computer program product or memory in the first network node 800 may thus comprise a computer readable medium on which the computer program is stored e.g. in the form of computer program modules or the like. For example, the memory may be a flash memory, a Random-Access Memory (RAM), a Read-Only Memory (ROM), an Electrically Erasable Programmable ROM (EEPROM) or hard drive storage (HDD), and the program modules could in alternative embodiments be distributed on different computer program products in the form of memories within the first network node 800.

The solution described herein may be implemented in the first network node 800 by means of a computer program storage product 802 comprising a computer program 804 with computer readable instructions which, when executed on the first network node 800, cause the first network node 800 to carry out the actions according to any of the above embodiments, where appropriate.

While the solution has been described with reference to specific exemplifying embodiments, the description is generally only intended to illustrate the inventive concept and should not be taken as limiting the scope of the solution. For example, the terms "network node", "propagation delay", "absolute time reference", "internal time" and "time-critical feature" have been used throughout this

disclosure, although any other corresponding entities, functions, and/or

parameters could also be used having the features and characteristics described here. The solution is defined by the appended claims.

Claims

1 . A method performed by a first network node (300) of a wireless network, for maintaining time synchronization with other network nodes in the wireless network, the method comprising: - when the first network node (300) has access to an absolute time reference, determining (400) a propagation delay (d) for a signal received from a

neighbouring network node (302) based on the absolute time reference, and

- maintaining (404) the time synchronization based on the determined propagation delay (d) so that the timing of further signals received from the neighbouring network node (302) corresponds to the determined propagation delay (d).

2. A method according to claim 1 , wherein said maintaining of the time synchronization is performed when the first network node (300) has no access to the absolute time reference.

3. A method according to claim 1 or 2, wherein the propagation delay (d) is determined by comparing reception time of the signal with the absolute time reference.

4. A method according to any of claims 1 -3, wherein the absolute time reference is obtained by any of: Global Navigation Satellite System, GNSS, Network Time Protocol, NTP, and Precision Time Protocol, PTP.

5. A method according to any of claims 1 -4, wherein Time Division Duplex,

TDD, is used in the wireless network and wherein the signal used for determining the propagation delay is a Cell-specific Reference Signal, CRS, transmitted by the neighbouring network node (302).

6. A method according to any of claims 1 -5, wherein the time

synchronization is updated at regular intervals.

7. A method according to any of claims 1 -6, further comprising issuing (608) an alarm when detecting that signals received from the neighbouring network node (302) are out of synchronization.

8. A method according to claim 7, wherein said detecting comprises determining (604) a propagation delay (d) for at least one further signal received from the neighbouring network node (302) based on the absolute time reference, and detecting (606) that the propagation delay (d) has changed.

9. A method according to any of claims 1 -8, wherein the time

synchronization is maintained based on propagation delays (d1 , d2) determined (704, 708) for signals received from at least two neighbouring network nodes.

10. A method according to claim 9, further comprising issuing (712) an alarm when detecting that the difference (d1 -d2) between the determined propagation delays (d1 , d2) has changed.

1 1 . A method according to any of claims 1 -10, further comprising performing (406) communication of radio signals according to the maintained time

synchronization.

12. A method according to claim 1 1 , wherein said communication involves a time-critical feature including any of: Time Division Duplex (TDD), multi-casting, coordination between network nodes of uplink and downlink transmissions, and wireless device positioning.

13. A first network node (800) of a wireless network, the first network node (800) being arranged to maintain time synchronization with other network nodes in the wireless network, wherein the first network node (800) is configured to:

- when the first network node (300) has access to an absolute time reference, determine (800A) a propagation delay (d) for a signal received from a

neighbouring network node (302) based on the absolute time reference, and - maintain (800B) the time synchronization based on the determined propagation delay (d) so that the timing of further signals received from the neighbouring network node (302) corresponds to the determined propagation delay (d).

14. A first network node (800) according to claim 13, wherein the first network node (800) is configured to perform said maintaining of the time

synchronization when the first network node (300) has no access to the absolute time reference.

15. A first network node (800) according to claim 13 or 14, wherein the first network node (800) is configured to determine the propagation delay (d) by comparing reception time of the signal with the absolute time reference.

16. A first network node (800) according to any of claims 13-15, wherein the first network node (800) is configured to obtain the absolute time reference by any of: Global Navigation Satellite System, GNSS, Network Time Protocol, NTP, and Precision Time Protocol, PTP.

17. A first network node (800) according to any of claims 13-16, wherein

Time Division Duplex, TDD, is used in the wireless network and wherein the signal used for determining the propagation delay is a Cell-specific Reference Signal, CRS, transmitted by the neighbouring network node (302).

18. A first network node (800) according to any of claims 13-17, wherein the first network node (800) is configured to update the time synchronization at regular intervals.

19. A first network node (800) according to any of claims 13-18, wherein the first network node (800) is configured to issue an alarm when detecting that signals received from the neighbouring network node (302) are out of

synchronization.

20. A first network node (800) according to claim 19, wherein the first network node (800) is configured to perform said detecting by determining a propagation delay (d) for at least one further signal received from the neighbouring network node (302) based on the absolute time reference, and detecting that the propagation delay (d) has changed.

21 . A first network node (800) according to any of claims 13-20, wherein the first network node (800) is configured to maintain the time synchronization based on propagation delays (d1 , d2) determined (704, 708) for signals received from at least two neighbouring network nodes.

22. A first network node (800) according to claim 21 , wherein the first network node (800) is configured to issue an alarm when detecting that the difference (d1 -d2) between the determined propagation delays (d1 , d2) has changed.

23. A first network node (800) according to any of claims 13-22, wherein the first network node (800) is configured to perform communication of radio signals according to the maintained time synchronization.

24. A first network node (800) according to claim 23, wherein said communication involves a time-critical feature including any of: Time Division Duplex (TDD), multi-casting, coordination between network nodes of uplink and downlink transmissions, and wireless device positioning.

25. A computer program comprising instructions which, when executed on at least one processor, cause the at least one processor to carry out the method according to any one of claims 1 -12.

26. A carrier containing the computer program of claim 25, wherein the carrier is one of an electronic signal, optical signal, radio signal, or computer readable storage medium.