CN106452645B - A kind of group of stars method for synchronizing network time based on cross layer design - Google Patents

Abstract

A time synchronization method for constellation network based on cross-layer design. The spread spectrum signal system is adopted in the physical layer, and the clock difference between the child node and the parent node is calculated through the semi-two-way inter-satellite ranging method in the synchronization time slot. Complete high-precision time synchronization between nodes. The MAC layer adopts network synchronization management, broadcasts synchronization status information to the network, manages the synchronization topology, and provides the result of time synchronization parent node selection to the physical layer for inter-node synchronization, thereby realizing time synchronization between nodes in the entire network. The method of the invention is suitable for the time synchronization requirements between satellites of various constellations, has no special requirements for satellite systems, has strong versatility, wide application range, simple on-satellite processing, and has good application prospects.

Description

A Time Synchronization Method for Constellation Network Based on Cross-Layer Design

technical field

The invention belongs to the field of satellite communication, and relates to a satellite network time synchronization method, which can realize the time synchronization of satellite networking in a complex space environment, and is suitable for the application of inter-satellite time synchronization of various constellations.

Background technique

Inter-satellite network interconnection can maximize the service capability of the network, expand the service scope of the network, and effectively make up for various deficiencies in single-satellite in-orbit operation, thereby building a space service system with complete functions and stable performance.

Compared with the previous constellations, the formation constellation requires more accurate mutual position measurement between the satellite platforms participating in the formation flight, and a more stable formation configuration. At the same time, due to the information exchange between each satellite platform, it is required The clocks of each satellite platform are strictly unified, and the autonomy of a single satellite platform is stronger. At the same time, due to the need for integrated transmission of satellite formations, satellite nodes need to obtain relative distance information through accurate ranging, and based on this, determine the transmission power and the length of the protection segment in the frame structure. Therefore, it is of great significance to obtain high-precision time-frequency synchronization results to adapt to space satellite networking with a large span of time and space.

In theory, only when the frequencies of the two clocks are consistent with the time of the clock face can it be called time synchronization. However, due to the instability of the frequency source with adjustable frequency, the time difference between the two clocks cannot be kept as a constant in actual work, and the frequency standard of the two clocks is allowed to have certain errors, as long as the two clocks are constantly adjusted If the absolute value of the clock face difference is kept within a tolerable range, it can be considered that the two clocks have achieved a certain precision of time synchronization.

In a wireless ad hoc network using the TDMA system, the synchronization process between each child node and the master node is usually divided into two steps: open-loop synchronization and closed-loop synchronization. The master node periodically broadcasts a synchronization beacon, and each child node immediately restarts the timing after detecting the synchronization beacon, which is open-loop synchronization. Due to the influence of propagation delay, the timing start times of each sub-node in the network should be different. If each child node only receives the broadcast from the master node and does not send data to the master node, obviously open-loop synchronization is sufficient. For the propagation delay of the link, the closed-loop synchronization method is used for calibration: the synchronization node periodically sends an ECHO (echo, a round-trip measurement frame used for delay measurement) packet to the reference node in a fixed time slot, and through The local clock counter starts counting. The reference node returns the packet immediately after receiving the ECHO packet. When the synchronization node receives the reply ECHO, the clock counter stops counting. According to the count value, the link between the synchronization node and the reference node can be estimated. The path delay, so that the node can refresh the link propagation delay in real time, and correct the open-loop synchronization. Considering that not every child node is visible to the master node, each child node first selects a node with a higher clock level as the synchronization reference, and each node will send a synchronization code in its synchronization time slot, when the node receives After reaching the code, the local clock counter is calibrated, thereby expanding the coverage area of the reference clock.

The inter-satellite network transmission scheme adopts the TDMA system, and the satellites in the cluster perform precise time synchronization through the master-slave synchronization method. The traditional synchronization method completes the open-loop synchronization and closed-loop synchronization between the master-slave nodes through the frame interaction of the MAC layer. . In the fast-sounding satellite cluster satellite networking project, the typical inter-satellite distance is 10-1300km, and the transmission rate is 128kbps-1kbps. Time synchronization between nodes is performed through frame interaction at the MAC layer. At such a low transmission rate, the synchronization accuracy will become Very poor, about 10us-1000us, which is far from meeting the needs of TDMA precise time synchronization.

Contents of the invention

The technical problem solved by the present invention is: to overcome the deficiencies of the prior art, to provide a time synchronization method based on the cross-layer design of the constellation network, the synchronous topology management between the nodes of the constellation is carried out at the MAC layer, and the spread spectrum is adopted at the physical layer The signal system uses the semi-two-way time comparison and ranging method to calculate the clock difference between nodes and realize the synchronization between nodes. Under different transmission rates, the spreading code rate of the physical layer remains unchanged, and the time synchronization accuracy will not deteriorate due to the reduction of the transmission rate. By combining the synchronization topology management of the MAC layer with the spread spectrum ranging of the physical layer, the time synchronization of the entire network can be realized, and a stable and high-precision inter-satellite synchronization solution is provided for the constellation network using the TDMA system.

The technical solution of the present invention is: a method for time synchronization of constellation network based on cross-layer design, comprising:

(1) Start the time synchronization function after each satellite node is turned on;

(2) Each satellite node listens to the network synchronization status information broadcast by other satellite nodes in orbit. If the satellite node does not listen to the network synchronization status information broadcast by other satellite nodes within a set length of time, it determines its own role is the time reference node; if the satellite node collects the network synchronization status information broadcast by the surrounding nodes within a set length of time, the node with the highest clock level is selected as the parent node of the satellite time synchronization according to the collected synchronization status information, At the same time, determine yourself as a child node;

(3) For the time reference node, receive the synchronization request reported by the new network node at the MAC layer, and respond to the time synchronization process of the new network child node at the physical layer; allocate synchronization time slots and broadcast time slots to the child nodes, and The synchronization time slot responds to the synchronization request of the child nodes, completes the synchronization of the entire network time, and performs synchronization management on each child node;

(4) For non-time reference nodes, send the selected parent node to the physical layer at the MAC layer, and initiate the time synchronization process with the parent node at the physical layer; after the time synchronization process is completed, the child node applies to the reference node for synchronization Time slot and broadcast time slot, and interact with the parent node in the synchronization time slot, complete the synchronization maintenance with the parent node, and realize the synchronization of the whole network time;

(5) After the time synchronization process of each satellite node is completed, the synchronization status information of the satellite node itself is broadcast, the synchronization time is diffused, and the synchronization management of each child node and the synchronization maintenance of each parent node are carried out.

The network synchronization status information includes node ID, reference node ID, clock level, and synchronization time slot occupancy status, wherein the node ID refers to the identification number of the star itself, and the reference node ID refers to the identification number of the satellite where the time reference of the whole network is located, and the clock The level refers to the position of the star in the time transfer. The clock level of the reference node is 0, and the clock level of the satellite node connected to it is 1, and so on. The farther away from the reference node, the lower the clock level. The synchronization time slot occupancy state refers to which synchronization time slots are occupied and which are not occupied in the entire superframe period.

The time synchronization process includes:

Step 1: After the child node fills the local clock face time t _{son_send} into the synchronization frame, it sends a synchronization frame to the parent node, and the parent node uses the received clock face time t _{son_send} of the child node to send the leading edge and the time of receiving the leading edge of the synchronization frame t _{father_arrive} calculates the local pseudorange t _α :

t _α = t _{father_arrive -} t _{son_send} ;

Step 2: The parent node fills in the local clock face time t _{father_send} and the local pseudorange t _α in the feedback frame and sends it back to the child node, and the child node uses the received parent node to send the leading clock face time t _{father_send} and the received synchronization frame leading edge Compute the local pseudorange t _β at time t _{son_arrive} :

t _β =t _{son_arrive} -t _{father_send} ;

Step 3: The child node calculates the clock difference Δt between the two nodes based on the measured local pseudorange t _β and the local pseudorange t _α received from the parent node:

Δt=(t _α -t _β )/2;

Step 4: Add Δt to the time of the child node to complete the time synchronization between the child node and the parent node.

The child nodes perform synchronous management, including:

The first step: wait for the broadcast time slot of the child node, and check whether the broadcast frame of the child node is received; if the broadcast frame of the child node is received, continue; if the broadcast frame of the child node is not received, the counter is incremented by 1 and enters the second step ;

Step 2: judge the value of the counter, if the counter is less than or equal to the maximum exit threshold MAX_exit, continue to wait for the broadcast time slot of the child node to detect the broadcast frame of the child node; if the counter is greater than MAX_exit, mark this node as a network exit node;

Step 3: Send the ID number of the off-net child node to the parent node in the service frame.

The parent node is maintained synchronously, including:

Step 1: Wait for the broadcast time slot of the parent node, and detect whether the broadcast frame of the parent node is received; if the broadcast frame of the parent node is received, the counter is cleared and continues; if the broadcast frame of the parent node is not received, the counter is incremented by 1 Enter the second step;

Step 2: Judge the value of the counter, if the counter is less than or equal to the maximum asynchronous threshold MAX_asyn, continue to wait for the broadcast time slot of the child node to detect the broadcast frame of the child node; if the counter is greater than MAX_asyn, mark the node as an asynchronous node, and restart Start the inbound synchronization process.

The advantage of the present invention compared with prior art is:

(1) In the method of the present invention, a semi-two-way spread spectrum measurement time synchronization system is adopted between nodes, which can provide high synchronization accuracy with as few interaction rounds as possible, and save scarce communication resources on the star.

(2) the inventive method carries out the synchronous topology management by the MAC layer through cross-layer design, is convenient to use the high management algorithm of processor operation complexity to optimize the topology management performance on the star;

(3) The method of the present invention adopts the modulation method of spread spectrum to measure distance at the physical layer, provides higher measurement accuracy based on the hardware circuit, avoids the influence of the space transmission rate change on the measurement accuracy, and finally completes the time synchronization of the whole network;

(4) the method of the present invention distributes exclusive synchronous time slot for each node, initiates by child node, carries out semi-two-way spread spectrum measurement to parent node, with the mode of time division multiple access, provides independent time synchronous channel for a plurality of nodes in the network, Avoid synchronization signal conflicts and effectively solve multi-node synchronization problems.

Description of drawings

Fig. 1 is a block flow diagram of the inventive method;

Fig. 2 is a schematic diagram of the frame structure of the present invention;

Fig. 3 is a schematic diagram of the broadcast frame structure of the present invention;

Fig. 4 is a schematic diagram of the synchronous frame structure of the present invention;

Fig. 5 is a schematic diagram of the feedback frame structure of the present invention;

Fig. 6 is a schematic diagram of node initialization and information collection process of the present invention;

Fig. 7 is a schematic diagram of the process of selecting a time reference node in the present invention;

Fig. 8 is a schematic diagram of the node synchronization and broadcast time slot acquisition process of the present invention;

9 is a schematic diagram of the time synchronization process of the ranging module of the present invention;

Fig. 10 is a processing flowchart of the time reference node of the present invention;

Fig. 11 is the processing flowchart of the non-time reference node (network access) of the present invention;

Fig. 12 is a flow chart of child node synchronization management in the present invention;

Fig. 13 is a flowchart of parent node synchronization maintenance in the present invention.

Detailed ways

As shown in Figure 1, it is the processing flow of the cross-layer constellation network time synchronization method of the present invention, which specifically includes the following steps:

(1) Start the time synchronization function after each satellite node is turned on;

(2) On-orbit listening to the network synchronization status information broadcast by other satellite nodes. The network synchronization status information includes: node ID, reference node ID, clock level, synchronization slot occupancy status, etc.;

Among them, the node ID refers to the identification number of the own star itself, the reference node ID refers to the identification number of the satellite where the time reference of the whole network is located, and the clock level refers to the position of the own star in the time transfer, that is, the reference node clock level is 0, The clock level of the satellite node connected to him by one hop is 1, and so on. The farther away from the reference node, the lower the clock level. The synchronization slot occupancy status refers to which synchronization slots have been occupied in the entire superframe period Not occupied.

(3) If the satellite node does not detect the network synchronization status information broadcast by other satellite nodes within a sufficiently long period of time, then determine its own role as the time reference node; if the satellite node collects the surrounding nodes within a sufficiently long period of time Broadcast network synchronization state information, then according to the synchronization state information of each node, select the node with the highest clock level as the parent node of the local time synchronization, and determine itself as the child node;

(4) For non-time reference nodes: the time synchronization topology management module of the MAC layer sends the selected parent node to the physical layer; for time reference nodes: the time synchronization topology management module receives the synchronization request reported by the new network access node;

(5) For non-time reference nodes: the physical layer initiates the time synchronization process with the parent node; for time reference nodes: the physical layer responds to the time synchronization process of the child nodes;

(6) For non-time reference nodes: apply to the reference node for synchronization time slots and broadcast time slots; for time reference nodes: allocate synchronization time slots and broadcast time slots to child nodes;

(7) For non-time reference nodes: by analogy, each satellite node interacts with its parent node in its own synchronization time slot according to the above method, completes the synchronization maintenance process with the parent node, and realizes the high precision of the entire network time Synchronize. After the time synchronization process of the satellite node is completed, the synchronization status information of the satellite node itself is broadcast, which is used for the diffusion of the synchronization time and the synchronization management of each child node. For the time reference node: by analogy, the reference satellite node responds to the synchronization request of the child nodes in the synchronization time slot of each node according to the above method, completes the high-precision synchronization of the entire network time, and performs synchronization management on each child node.

The key links are described in detail below.

To establish a TDMA system, the system frame format must first be designed. All time slot allocation in the system is based on this system frame format. Taking time as the unit, the superframe has the longest time, and M slots are divided under the superframe. Multiframe (in order to ensure that each node has an exclusive synchronization time slot, so the value of M should be greater than the total number of satellite nodes in the satellite formation), each multiframe is composed of a synchronization time slot and N business frames ( In order for each node to have sufficient service frame resources, the value of N should be much larger than the number of satellite nodes, for example, when the number of nodes is 8, the value of N can be 256 or 512).

The system frame format designed by the present invention is as shown in Figure 2, and a super frame of the system contains M multiframes; A multiframe includes a synchronization time slot and N service frames; the synchronization time slot is used for semi-two-way spread spectrum ranging; A broadcast frame is set in the service frame, which is used for broadcasting the synchronization state information of each node.

The present invention involves three kinds of frame structures

a) Broadcast frame structure

The broadcast frame is used to send information such as the node clock level and the reference node, and is used for the child node to select the parent node; the child node can send unallocated time slots in the broadcast frame to request the time slot number from the reference node. The structure of the broadcast frame is shown in Figure 3 shown.

Broadcast address: identifies the broadcast frame (5 bits)

Sending address: sending node ID (5 bits)

Clock level: the clock level of the node sending the broadcast frame (3 bits)

Reference node: Indicates the synchronization master node ID (5 bits)

Multiframe count: the multiframe count of the node sending the broadcast (5 bits)

CRC: check digit.

b) Synchronous frame structure

The synchronization frame is used for child nodes to initiate a synchronization process, and its structure is shown in Figure 4.

Destination address: the ID of the receiving node, fill in the receiving node ID of the synchronization frame;

Sending address: the ID of the sending node; fill in the ID of the sending node of the synchronization frame;

P/F: Synchronization frame and feedback frame distinguish mark, this bit of synchronization frame is 0;

Local clock face time: the time when the leading edge of the local synchronization frame is sent out, used to calculate the arrival of the synchronization frame epoch;

CRC: check digit.

c) Feedback frame structure

The feedback frame is used for the parent node to respond to the synchronization request of the child node, and to feed back time information. Its structure is shown in Figure 5.

Destination address: the ID of the receiving node, fill in the receiving node ID of the feedback frame;

Sending address: the ID of the sending node, fill in the ID of the sending node of the feedback frame;

P/F: Synchronization frame and feedback frame distinguish mark, this bit of feedback frame is 1;

Local clock face time: the time at which the leading edge of the local feedback frame is sent out, used to calculate the arrival of the feedback frame epoch;

Local pseudorange value: through local spread spectrum capture, when the synchronization frame sent by other nodes arrives at the epoch, calculate the local pseudorange corresponding to other nodes, and send the frame to the corresponding node;

CRC: check digit.

1) Node initialization and information collection process

The initialization and information collection process of the node is shown in Figure 6. After the node is turned on, the synchronization state is set to an asynchronous state, and the synchronous broadcast frame sent by the surrounding nodes can be listened to, and several superframe periods can be listened to. Determine whether there is a synchronous network around. If there is a synchronization network, the selected parent node information will be sent to the ranging module, and the ranging module will complete the time synchronization process with the parent node; if there is no synchronization network around, it will enter the clock reference node selection stage.

After the satellite group enters the working orbit, each node in the constellation does not need to be turned on in a specific order, and the start-up time can be freely arranged; after the node is turned on, its synchronization state is set to an asynchronous state, and the time synchronization function is activated.

2) Time reference node selection process

The time reference node selection process is shown in Figure 7, which mainly includes the following steps:

Step 1: Randomly back off for T superframe periods (firstly, it is necessary to listen for a long enough time to ensure that the existing network can be heard. In addition, the T value of each node is randomly determined and different, which can ensure There is always a node that will stop listening first, start establishing a reference), and listen to the channel for broadcast frames. If there is a broadcast frame, the parent node is selected, and the obtained parent node ID is sent to the ranging module, and then the synchronization time slot acquisition process is completed; if the broadcast frame is not received, the second step process is entered.

Step 2: Send a broadcast frame on the first business time slot of the first multiframe, set the synchronization time slot of the ranging module to the length of one synchronization time slot ahead of the start of the first multiframe, and set the The node is identified as a time reference node (each node will listen to T superframe periods, and T of each node is a random value generated by itself, which is not equal to each other, so there is always a node with the smallest T that will finish listening first, Set yourself as the benchmark, once a node is confirmed as the benchmark, other nodes will synchronize to this benchmark). Then start the timer (the timer length is set to 1 superframe period) to listen to the ID number of the requesting network node reported by the ranging module, and if the requesting node ID is received, the node will be assigned a broadcast time slot and synchronized with the ranging module time slot (allocation result information will be sent in unicast mode in the business time slot).

3) Node synchronization and broadcast time slot acquisition process

The acquisition of node synchronization and broadcast time slots is to send a network access application to the parent node, and the reference node allocates synchronization time slots and broadcast time slots to complete the entire network access process. The specific process is shown in Figure 8 and consists of the following steps:

Step 1: After receiving the broadcast frame of the parent node, identify this node as a child node, send the ID of the parent node to be synchronized as a parameter to the ranging module and set the time synchronization timer, if the ranging module is not within the timer time Return to the synchronous state, then proceed to the first step again; if the ranging module returns to the synchronous state, enter the second step.

Step 2: After obtaining the synchronization state of the ranging module, wait for the time slot allocation result frame sent by the parent node, record the broadcast time slot and synchronization time slot allocated by the reference node to the node, and send the synchronization time slot as a parameter to the measurement distance module

4) The time synchronization process of the ranging module

The time synchronization topology management module of the MAC layer sends the selected parent node to the physical layer, and the physical layer initiates a semi-two-way spread spectrum ranging process as shown in Figure 9:

Step 1: The master-slave synchronization process between nodes is carried out in the synchronization frame of 76ms. After the child node fills the local clock face time t _{son_send} into the synchronization frame, it sends the synchronization frame to the superior clock node. The master node uses the received child node The local pseudorange t _α is calculated by t _{son_send when the clock face of the leading edge is sent and the time t father_arrive when} the leading edge of the synchronization frame is received:

t _α =t _{father_arrive} -t _{son_send}

Step 2: The parent node fills in the local clock face time t _{father_send} and the local pseudorange t _α in the feedback frame and sends it back to the child node. The child node calculates the local pseudorange t _β by using the time t _{father_send} of the clock face when the parent node sends the front edge and the time t _{son_arrive} when it receives the front edge of the synchronization frame:

t _β =t _{son_arrive} -t _{father_send}

Step 3: The sub-node can calculate the clock difference Δt between the two nodes according to the measured local pseudo-range t _β and the local pseudo-range t _α received from the master node:

Δt=(t _α -t _β )/2

Step 4: Adjust the clock accordingly, and add Δt to the time of the child node to realize the time synchronization between the child node and the parent node of the superior clock.

5) Processing flow of time reference node

The time reference node mainly completes the allocation of broadcast time slots and synchronization time slots, and maintains time synchronization with one-hop nodes. The processing flow is shown in Figure 10, which mainly includes the following steps:

Step 1: Determine whether it is the multiframe header where the node is located. If so, the ranging module responds to the synchronization process of the new network node in the even synchronization time slot, and reports the ID number of the new network node to the CPU. The CPU sends broadcast frames in broadcast slots. If it is not the multiframe where the node is located, enter the second step.

Step 2: The ranging module of the time reference node responds to the synchronization maintenance process of the child node in the synchronization time slot of the child node.

Step 3: Receive the time slot recycling information reported by each child node, and the CPU allocates time slots for new network nodes and records them.

Step 4: The CPU sends a new time slot allocation result in the service frame.

6) Processing flow of non-time reference nodes (network access)

The request and allocation process of non-time reference node broadcast time slots and synchronization time slots is shown in Figure 11, including the following steps:

Step 1: Determine whether it is the multiframe header where the node is located. If yes, the ranging module completes the time synchronization with the parent node in the synchronization time slot of the odd frame; completes the time synchronization with the new network access node in the synchronization time slot of the even frame. Time synchronization, and report the ID number of the network node to the CPU. Broadcast frames are sent in broadcast slots. If it is not the multiframe where the node is located, enter the second step.

Step 2: The ranging module completes the time synchronization maintenance process with the child nodes in the synchronization time slot, and the CPU receives the broadcast frames of other nodes.

Step 3: Receive the time slot recovery information reported by the child node, and report the time slot recovery information to the parent node.

Step 4: The CPU issues a new time slot allocation result in the service time slot.

7) Child node synchronization management process

The synchronization management process of child nodes is shown in Figure 12, including the following processes:

The first step: wait for the broadcast time slot of the child node, and check whether the broadcast frame of the child node is received; if the broadcast frame of the child node is received, continue; if the broadcast frame of the child node is not received, the counter is incremented by 1 and enters the second step .

Step 2: Determine the value of the counter. If the counter is less than or equal to the maximum exit threshold MAX_exit, continue to wait for the broadcast time slot of the child node to detect the broadcast frame of the child node; if the counter is greater than MAX_exit, mark this node as an exit node.

Step 3: Send the ID number of the off-line child node to the parent node in the business frame

8) Parent node synchronization maintenance process

The synchronization maintenance process of the parent node is shown in Figure 13, which mainly includes the following steps:

Step 1: Wait for the broadcast time slot of the parent node, and detect whether the broadcast frame of the parent node is received; if the broadcast frame of the parent node is received, the counter is cleared and continues; if the broadcast frame of the parent node is not received, the counter is incremented by 1 Go to the second step.

Step 2: Judge the value of the counter, if the counter is less than or equal to the maximum asynchronous threshold MAX_asyn, continue to wait for the broadcast time slot of the child node to detect the broadcast frame of the child node; if the counter is greater than MAX_asyn, mark the node as an asynchronous node, and restart Start the inbound synchronization process.

The content that is not described in detail in the description of the present invention belongs to the well-known technology of those skilled in the art.

Claims (5)

1. a kind of group of stars method for synchronizing network time based on cross layer design, it is characterised in that including：

(1) start time synchronization function after each satellite node booting；

(2) the in-orbit Network Synchronization status information for listening to other satellite node broadcast of each satellite node, if satellite node is being set The Network Synchronization status information for not listening to other satellite node broadcast in the time of length, then when being determined as role self Between datum node；If satellite node has collected the Network Synchronization state letter of surroundings nodes broadcast within the time of setting length The highest node of clock stratum, then is chosen to be the father node of this satellite time synchronization by breath according to the Synchronization Status Message of collection, Determine oneself to be child node simultaneously；

(3) for time reference node, the synchronization request that new network access node reports is received in MAC layer, and new in physical layer response The time synchronization process of networking child node；Synchronization time slot and time slot are distributed to child node, and in the synchronization time slot of each node The synchronization request of child node is responded, the synchronization of the whole network time is completed, and management is synchronized to each child node；

(4) for non-temporal datum node, selected father node is issued to physical layer in MAC layer, and physical layer initiate with The time synchronization process of father node；After the completion of time synchronization process, child node is to datum node application synchronization time slot and when broadcasting Gap, and interacted in synchronization time slot and father node, maintenance synchronous with father node is completed, realizes the synchronization of the whole network time；

(5) after the completion of the time synchronization process of each satellite node, the Synchronization Status Message of broadcasting satellite node itself is synchronized The diffusion of time, and management and each father node are synchronized to each child node and synchronize maintenance.

2. a kind of group of stars method for synchronizing network time based on cross layer design according to claim 1, it is characterised in that：Institute The Network Synchronization status information stated includes node ID, datum node ID, clock stratum, synchronization time slot occupied state, interior joint ID refers to that the identifier of this star itself, datum node ID refer to that the identifier of the whole network time reference place satellite, clock stratum refer to this For star the location of in Time Transmission, datum node clock stratum is 0 grade, the clock grade with the satellite node of his a jump connection Not Wei 1, and so on, remoter with datum node, clock stratum is lower, and synchronization time slot occupied state refers in entire super frame period Which synchronization time slot is occupied, which is unoccupied.

3. a kind of group of stars method for synchronizing network time based on cross layer design according to claim 1 or 2, feature exist In：The time synchronization process, including：

The first step：Child node is by t when local clock face_{son_send}After inserting synchronization frame, synchronization frame is sent to father node, father node utilizes T when the local clock face that the child node received is sent_{son_send}With the time t for receiving synchronization frame forward position_{father_arrive}Calculate this Ground pseudorange t_α：

t_α=t_{father_arrive}-t_{son_send}；

Second step：Father node is by t when local clock face_{father_send}With local pseudorange t_αIt is filled up to feedback frame and sends back child node, son T when the local clock face that node is sent using the father node received_{father_send}With the time for receiving synchronization frame forward position t_{son_arrive}Calculate local pseudorange t_β：

t_β=t_{son_arrive}-t_{father_send}；

Third step：Child node is according to the local pseudorange t measured_βAnd receive the local pseudorange t of father node_αTwo nodes are calculated Between clock deviation Δ t：

Δ t=(t_α-t_β)/2；

4th step：The time of child node adds Δ t, completes the time synchronization of child node and father node.

4. a kind of group of stars method for synchronizing network time based on cross layer design according to claim 1 or 2, feature exist In：The child node synchronizes management, including：

The first step：The time slot for waiting child node, detects whether the broadcast frame for receiving child node；If receiving the wide of child node Frame is broadcast then to continue；Counter adds 1 to enter second step if the broadcast frame for not receiving child node；

Second step：The value for judging counter continues waiting for son if counter is less than or equal to maximum logout thresholding MAX_exit The broadcast frame of the time slot detection child node of node；If counter be greater than MAX_exit if by this vertex ticks be logout section Point；

Third step：The ID number of logout child node is sent to father node in traffic frame.

5. a kind of group of stars method for synchronizing network time based on cross layer design according to claim 1 or 2, feature exist In：The father node synchronizes maintenance, including：

The first step：The time slot for waiting father node, detects whether the broadcast frame for receiving father node；If receiving the wide of father node Frame is broadcast by counter O reset and is continued；Counter adds 1 to enter second step if the broadcast frame for not receiving father node；

Second step：The value for judging counter continues waiting for if counter is less than or equal to maximum asynchronous thresholding MAX_asyn The broadcast frame of the time slot detection child node of child node；By this node labeled as non-same if counter is greater than MAX_asyn Node is walked, and restarts networking synchronizing process.