CN103237323B - Based on multichannel parallel node acceptance method - Google Patents

Abstract

The invention discloses a multi-channel-based parallel node admission method, which mainly solves the problem of low admission efficiency of the existing node admission method. The implementation steps are: allocate channel resources and divide the admission cycle; plan time slots for different purposes in each admission cycle; use downlink and uplink detection frame time slots to form a two-way signaling channel; use downlink detection frames to complete downlink power control and downlink channel training; the slave node selects an idle two-way signaling channel and sends an uplink detection frame; the central node sends a downlink detection frame according to the received uplink detection frame; the slave node receives the downlink detection frame and queries the selected two-way signaling The state of the channel; according to the queried state, the node acceptance process is completed. The invention has the advantages of short admission time, less conflicts when there are many slave nodes, and high admission efficiency, and can be applied to various local area networks and access network fields that adopt time division multiplexing/time division multiple access mechanisms to allocate channels.

Description

Parallel node admission method based on multi-channel

technical field

The invention belongs to the technical field of communication and relates to a node admission method, which can be applied to the fields of various local area networks and access networks that adopt time division multiplexing/time division multiple access mechanisms to allocate channels.

Background technique

In LANs and access networks based on shared media, a central point-to-multipoint logical topology is often used. In this network structure, there is a central node and multiple subordinate nodes, and nodes other than the central node are called subordinate nodes. Data communication can be directly performed between the central node and the subordinate nodes, and the communication between the subordinate nodes generally needs to be forwarded by the central node. After a new slave node is powered on, initialized or reset, the process of registering with the central node and joining the existing system is called node admission. Usually, normal data communication can only be carried out after node admission is completed. In order to complete node admission, the central node and subordinate nodes usually need to exchange information multiple times. In this process, the channel is time-divided into continuous and non-overlapping admission cycles; each admission cycle contains downlink sounding frame time slots and uplink sounding frame time slots; downlink sounding frame time slots are used by the central node to generate and send The downlink sounding frame and the time slot of the uplink sounding frame are used for the slave node to generate and send the uplink sounding frame. The uplink/downlink detection frame mainly completes the channel training of the physical layer, and carries the MAC layer uplink/downlink signaling frame exchanged between the central node and the slave node during the node acceptance process. The start time interval between two adjacent downlink sounding frames is an admission cycle.

In existing network systems, a one-to-one admission method based on a single channel is usually adopted. In this method, each admission period contains only one downlink detection frame time slot and one uplink detection frame time slot, which are respectively used for the central node to send a downlink detection frame and the slave node to send an uplink detection frame; an uplink/downlink detection frame The frame is used for channel training of the physical layer and bears one uplink/downlink MAC layer signaling frame of the node acceptance process; the central node accepts each slave node in a serial manner, that is, only one slave node is accepted Only then can the admission of the next slave node be carried out.

The above-mentioned node admission method is currently widely used in local area networks and access networks based on shared media, but it has the disadvantages that when there are many slave nodes, the admission time is long, and the admission requests are prone to conflicts.

In a local area network and an access network, there are usually many slave nodes, and a central node needs to complete the process of accepting multiple slave nodes. However, in the one-to-one admission method based on a single channel, one uplink/downlink detection frame transmitted in one uplink/downlink detection frame time slot carries one uplink/downlink signaling frame. A central node can only carry out the admission process of one slave node at the same time. When multiple slave nodes need to be admitted to the network at the same time, it takes a long time to complete the entire admission process. In the case of electricity, it takes a long time for the central node to complete the admission process of all slave nodes; and after the current slave node completes the admission process, a large number of slave nodes waiting to be admitted usually compete for admission opportunities in the only uplink detection frame time slot , and send admission requests at the same time, resulting in conflicts, resulting in a waste of time, the conflicting slave nodes can only send admission requests again in the uplink detection frame time slots in the subsequent admission cycle, thus further prolonging the admission time.

Contents of the invention

The object of the present invention is to address the shortcomings of the above-mentioned prior art, and propose a multi-channel-based parallel node admission method to shorten the admission time, reduce the collision probability, improve the admission efficiency, and save channel resources at the same time.

To achieve the above object, the technical solution of the present invention comprises the following steps:

(1) Use time division multiplexing/time division multiple access mechanism to allocate channel resources, adopt time division multiplexing for downlink and time division multiple access for uplink, and divide the channel into continuous and non-overlapping admission cycles in time;

(2) Use fixed planning or dynamic planning in each admission cycle to plan the following time slots for different purposes:

Several downlink data frame time slots for the central node to send downlink data frames, and several uplink data frame time slots for the slave nodes to send uplink data frames;

Several downlink control frame time slots for the central node to send downlink control frames, and several uplink control frame time slots for the slave nodes to send uplink control frames;

A downlink sounding frame time slot for the central node to send a downlink physical layer sounding frame;

N uplink sounding frame time slots for the slave node to send N uplink physical layer sounding frames;

(3) Use one downlink sounding frame time slot and N uplink sounding frame time slots planned in step (2) to form N bidirectional signaling channels:

(3a) Using a downlink physical layer detection frame sent in a downlink detection frame time slot, divide N downlink MAC layer signaling frame channels to carry N downlink MAC layer signaling frames, and each channel carries one downlink MAC layer signaling frame In the signaling frame, the MAC layer refers to the medium access control layer;

(3b) Using an uplink physical layer detection frame sent in an uplink detection frame time slot, provide an uplink MAC layer signaling frame channel to carry one uplink MAC layer signaling frame;

(3c) Corresponding N downlink MAC layer signaling frame channels to N uplink MAC layer signaling frame channels to form N bidirectional signaling channels, that is, the i-th downlink MAC layer signaling frame channel corresponds to the i-th uplink MAC layer signaling frame channel, which constitutes the i-th bidirectional signaling channel, and the downlink MAC layer signaling frame carried by each downlink MAC layer signaling frame channel has a flag indicating that the current state of its corresponding bidirectional signaling channel is busy or free;

(4) The central node continues to send downlink detection frames in the downlink detection frame time slot, and the slave nodes continue to receive downlink detection frames, and use the downlink detection frames to perform downlink power control and downlink channel training. After completing downlink power control and downlink channel training, enter step (5);

(5) The slave node receives the downlink detection frame, queries the flags in the N downlink MAC layer signaling frames, and counts the number of idle two-way signaling channels as M. If M=0, wait for an admission cycle and repeat the steps (5); if M>0, randomly select a bidirectional signaling channel among the M idle bidirectional signaling channels, and send an admission request in its corresponding uplink sounding frame time slot;

(6) The central node updates the flags in the N-channel downlink MAC layer signaling frame, and sends a downlink detection frame:

If the central node does not receive the uplink detection frame in a certain uplink detection frame time slot, then keep the flag in the downlink MAC layer signaling frame corresponding to the time slot unchanged, and in the downlink detection frame time slot, the central node sends the downlink detection frame ;

If the central node receives the uplink sounding frame in a certain uplink sounding frame time slot, the sign in the downlink MAC layer signaling frame corresponding to the time slot is set as busy, and in the downlink sounding frame time slot, the central node uses the two-way signal Make the channel send an acceptance response;

(7) The slave node receives the downlink detection frame, and queries the sign in the downlink MAC layer signaling frame corresponding to the selected bidirectional signaling channel:

If the flag is idle, the admission request of the slave node conflicts, the slave node executes the backoff algorithm, waits for several admission cycles, and returns to step (5);

If the flag is busy, the admission request of the slave node is sent successfully, the slave node and the central node use the two-way signaling channel for uplink channel training or uplink power control, the central node uses the two-way signaling channel to broadcast channel parameters, and finally the central node The flag in the downlink MAC layer signaling frame corresponding to the bidirectional signaling channel is set as idle to complete the acceptance process of the slave node.

Compared with the one-to-one admission method based on single channel, the present invention has the following advantages:

1. The present invention realizes parallel admission, shortens the admission time, reduces the conflict probability, and improves the admission efficiency.

Based on the single-channel one-to-one admission method, one admission cycle only allocates one uplink and downlink detection frame time slot, and carries one uplink and downlink signaling frame respectively, so a central node can only accept one slave node at the same time In the process, other slave nodes can send admission requests only after the admission of the current node is completed, so the admission time is long. However, the present invention provides N bidirectional signaling channels, which can simultaneously realize the parallel admission of up to N slave nodes, shorten the admission time, and improve the admission efficiency. In addition, when many slave nodes are waiting for admission at the same time, the one-to-one admission method based on single channel sends admission requests at the same time in the only uplink detection frame time slot after the current node completes the admission process. , it is easy to conflict, which further prolongs the admission time. However, the present invention divides N two-way signaling channels, and the slave node randomly selects a channel from a plurality of idle two-way signaling channels, and sends an admission request in its corresponding uplink detection frame time slot, which reduces the number of channels in the same uplink detection frame. The number of slave nodes competing for admission opportunities in time slots reduces the probability of collisions and improves admission efficiency.

2. The present invention effectively saves channel resources while realizing parallel reception.

In the parallel reception method described in the present invention, only one downlink detection frame time slot can be used to provide N downlink signaling frame channels, and at the same time, each slave node can perform downlink channel training and downlink power by sharing one downlink detection frame time slot. control, thereby effectively saving channel resources.

Description of drawings

Fig. 1 is the realization flowchart of the first embodiment of the present invention;

Fig. 2 is a fixed planning diagram of the acceptance period in Embodiment 1 of the present invention;

Fig. 3 is a channel structure diagram of a MAC layer signaling frame in an embodiment of the present invention;

Fig. 4 is the implementation flowchart of the second embodiment of the present invention;

FIG. 5 is a dynamic planning diagram of an admission cycle in Embodiment 2 of the present invention.

detailed description

The content of the present invention will be further elaborated below in conjunction with the accompanying drawings.

Embodiment one

In this example, the network applied by the present invention is composed of a central node 0 and 64 slave nodes, among which the four slave nodes 1, 2, 3, and 4 are ready to perform the admission process after being powered on.

Referring to Figure 1, the implementation steps of this example are as follows:

Step 1. Allocate channel resources and divide the admission period.

The time division multiplexing/time division multiple access mechanism is used to allocate channel resources, that is, time division multiplexing is used for downlink and time division multiple access is used for uplink, and the channel is divided into continuous and non-overlapping admission periods in time, and each admission period contains downlink The detection frame time slot and the uplink detection frame time slot, and the start time interval between two adjacent downlink detection frames is an admission cycle. In this embodiment, the time length of the admission period is fixed and is set to 20000us.

In step 2, a fixed plan is adopted in each admission cycle, and time slots for different purposes are planned.

(2a) Allocate a downlink sounding frame time slot at the beginning of the first admission cycle for the central node to send downlink sounding frames;

(2b) Allocate an uplink sounding frame time slot in the middle of the first admission cycle, and allocate before this time slot uplink sounding frame time slots, allocated after this time slot Uplink detection frame time slots, a total of N uplink detection frame time slots, all used for slave nodes to send uplink detection frames, N≥2;

(2c) Allocate several downlink data frame time slots and several downlink control frame time slots after the downlink sounding frame time slots in the first admission cycle, and allocate several downlink data frame time slots after the N uplink sounding frame time slots in the first admission cycle Uplink data frame time slots and several uplink control frame time slots;

(2d) Repeat steps (2a)-step (2c) to plan all admission cycles.

In this embodiment, the planning result of an admission cycle is shown in FIG. 2 . Figure 2 shows several downlink and uplink data frame time slots, several downlink and uplink control frame time slots, one downlink detection frame time slot and eight uplink detection frame time slots.

Step 3, using the time slots of the downlink and uplink detection frames planned in step 2 to form a bidirectional signaling channel.

(3a) Using a downlink physical layer detection frame sent in a downlink detection frame time slot, divide N downlink MAC layer signaling frame channels, and carry N downlink MAC layer signaling frames;

(3b) Using an uplink physical layer detection frame sent in an uplink detection frame time slot, provide an uplink MAC layer signaling frame channel to carry one uplink MAC layer signaling frame;

(3c) Corresponding N downlink MAC layer signaling frame channels to N uplink MAC layer signaling frame channels to form N bidirectional signaling channels, that is, the i-th downlink MAC layer signaling frame channel corresponds to the i-th uplink MAC The layer signaling frame channel constitutes the i-th bidirectional signaling channel;

In this embodiment, 1 downlink detection frame is used to divide 8 downlink MAC layer signaling frame channels according to frequency, and 8 uplink detection frames are used to provide 8 uplink MAC layer signaling frame channels. The signaling frame channels correspond to 8 uplink MAC layer signaling frame channels, forming 8 bidirectional signaling channels, as shown in FIG. 3 .

Step 4, the central node continuously sends the downlink detection frame in the time slot of the downlink detection frame, and the slave node continuously receives the downlink detection frame, and uses the downlink detection frame to complete downlink power control and downlink channel training.

In this embodiment, the downlink detection frame sent by the central node 0 carries 8 downlink MAC layer signaling frames, wherein the flags in the 1st, 4th and 7th downlink MAC layer signaling frames are all idle, Its types are all empty frames, and the signs in other downlink MAC layer signaling frames are all busy;

The four slave nodes 1, 2, 3, and 4 continuously receive the downlink detection frame, detect the signal strength of the downlink detection frame, and adjust the gain of the receiving power according to the signal strength, so that the receiver can accurately receive the downlink detection frame and complete the downlink power control;

The four slave nodes 1, 2, 3, 4 extract the training sequence signal from the downlink sounding frame signal, and estimate the physical layer channel parameters according to the training sequence number signal, so that the slave nodes can select the modulation format according to the channel parameters and complete the downlink channel training.

Step 5: The slave node receives the downlink detection frame, queries the flags in the N downlink MAC layer signaling frames, and counts the number of idle bidirectional signaling channels.

In this embodiment, the slave nodes 1, 2, 3, and 4 query the flags in the 8-way downlink MAC layer signaling frames carried by the downlink detection frame, and find that the 1st, 4th, and 7th downlink MAC layer signaling frames The flag in the frame is idle, and it is obtained that the first, fourth and seventh bidirectional signaling channels are idle, and the number of idle bidirectional signaling channels is calculated as M=3.

Step 6, the slave node selects a channel from the bidirectional signaling channels that have been counted, sets the type of the uplink MAC layer signaling frame it carries as an admission request, and sends an uplink probe frame in its uplink probe frame time slot:

The first slave node 1 selects the 7th bidirectional signaling channel from the three bidirectional signaling channels counted out, sets the type of the 7th uplink MAC layer signaling frame as an admission request, and The uplink sounding frame time slot sends the uplink sounding frame;

The second slave node 2 selects the first bidirectional signaling channel from the three bidirectional signaling channels counted, sets the type of the first uplink MAC layer signaling frame as an admission request, and The uplink sounding frame time slot sends the uplink sounding frame;

The third slave node 3 and the fourth slave node 4 both select the fourth bidirectional signaling channel from the three bidirectional signaling channels counted, and set the type of the fourth uplink MAC layer signaling frame as an admission request , and simultaneously send the uplink sounding frame in the fourth uplink sounding frame time slot.

Step 7, the central node updates the flags in the N-path downlink signaling frame, and sends a downlink detection frame.

In this embodiment, the central node 0 receives the uplink detection frame sent by the slave node 2 in the first uplink detection frame time slot, queries the type of the uplink MAC layer signaling frame carried by the frame, and obtains the type of the signaling frame It is an admission request, so the central node 0 sets the sign in the No. 1 downlink MAC layer signaling frame as busy, and sets its type as an admission response;

Since the admission request of slave node 3 and the admission request of slave node 4 conflict in the fourth uplink detection frame time slot, central node 0 does not receive the uplink detection frame in the fourth uplink detection frame time slot, and central node 0 keeps The flag in the 4th downlink MAC layer signaling frame is idle, and the type of the signaling frame is kept as an empty frame;

The central node 0 receives the uplink detection frame sent by the slave node 1 in the seventh uplink detection frame time slot, and queries the type of the uplink MAC layer signaling frame carried by the frame, and obtains that the type of the signaling frame is an admission request, so the central node Node 0 sets the sign in the 7th downlink MAC layer signaling frame as busy, and sets its type as acceptance response;

In the downlink detection frame time slot, the central node 0 sends a downlink detection frame, which carries 8 downlink MAC layer signaling frames, among which the signs and types of the 1st, 4th and 7th downlink MAC layer signaling frames As mentioned above, the flags and types of other downlink MAC layer signaling frames remain unchanged.

Step 8, the slave node receives the downlink detection frame, and inquires about the flag in the downlink MAC layer signaling frame corresponding to the selected bidirectional signaling channel.

In this embodiment, the slave node 1 receives the downlink detection frame, and inquires that the sign in the seventh downlink MAC layer signaling frame is busy;

The slave node 2 receives the downlink detection frame, and finds that the flag in the first downlink MAC layer signaling frame is busy;

The slave node 3 and the slave node 4 both receive the downlink detection frame, and both find out that the flag in the fourth downlink MAC layer signaling frame is idle.

Step 9, perform the following operations according to the sign in the downlink MAC layer signaling frame corresponding to the inquired bidirectional signaling channel:

Slave node 2 and central node 0 use the first two-way signaling channel to complete uplink channel training and uplink power control;

Slave node 1 and central node 0 use the seventh bidirectional signaling channel to complete uplink channel training and uplink power control;

Central node 0 uses the first and seventh bidirectional signaling channels to broadcast channel parameters;

Central node 0 sets the flags in the downlink MAC layer signaling frames corresponding to the first and seventh bidirectional signaling channels as idle, and completes the admission process of slave node 1 and slave node 2, and then slave node 1 and slave node 2 Data communication using data frame time slots and control frame time slots;

Slave node 3 and slave node 4 execute the backoff algorithm, wait for several admission cycles, and return to step 5.

Embodiment two

In this example, the network applied by the present invention is composed of a central node 0 and 32 slave nodes, wherein the two slave nodes 1 and 2 are ready to execute the admission process after being powered on.

With reference to Fig. 4, the implementation steps of the present embodiment are as follows:

Step 1. Allocate channel resources and divide the admission period.

The time division multiplexing/time division multiple access mechanism is used to allocate channel resources, that is, time division multiplexing is used for downlink and time division multiple access is used for uplink, and the channel is divided into continuous and non-overlapping admission periods in time, and each admission period contains downlink The detection frame time slot and the uplink detection frame time slot, and the start time interval between two adjacent downlink detection frames is an admission period. In this embodiment, the time length of the admission period is fixed and is set to 20000us.

In step 2, use dynamic programming in each admission period to plan time slots for different purposes.

(2a) Allocate a downlink sounding frame time slot at the beginning of the first admission cycle for the central node to send downlink sounding frames;

(2b) Allocate an uplink sounding frame time slot in the middle of the first admission cycle, and allocate before this time slot uplink sounding frame time slots, allocated after this time slot Uplink detection frame time slots, a total of N uplink detection frame time slots, all used for slave nodes to send uplink detection frames, N≥2;

(2c) Allocate several downlink data frame time slots and several downlink control frame time slots after the downlink sounding frame time slots in the first admission cycle, and allocate several downlink data frame time slots after the N uplink sounding frame time slots in the first admission cycle Uplink data frame time slots and several uplink control frame time slots;

(2d) After the first admission cycle ends, the central node executes the scheduling algorithm to plan the second admission cycle. And so on, planning all admission cycles.

In this embodiment, the planning results of two admission cycles are shown in FIG. 5 . In the first acceptance cycle in Figure 5, 3000 downlink data frame time slots, 50 downlink control frame time slots, 1000 uplink data frame time slots, 50 uplink control frame time slots, and 1 downlink detection frame time slot are drawn. and 8 uplink detection frame time slots, the second acceptance period draws 2500 downlink data frame time slots, 50 downlink control frame time slots, 500 uplink data frame time slots, 50 uplink control frame time slots, 1 There are 8 downlink sounding frame time slots and 8 uplink sounding frame time slots.

Step 3, using the time slots of the downlink and uplink detection frames planned in step 2 to form a bidirectional signaling channel.

(3a) Using a downlink physical layer detection frame sent in a downlink detection frame time slot, divide N downlink MAC layer signaling frame channels, and carry N downlink MAC layer signaling frames;

(3b) Using an uplink physical layer detection frame sent in an uplink detection frame time slot, provide an uplink MAC layer signaling frame channel to carry one uplink MAC layer signaling frame;

(3c) Corresponding N downlink MAC layer signaling frame channels to N uplink MAC layer signaling frame channels to form N bidirectional signaling channels, that is, the i-th downlink MAC layer signaling frame channel corresponds to the i-th uplink MAC The layer signaling frame channel constitutes the i-th bidirectional signaling channel;

In this embodiment, 1 downlink detection frame is used to divide 8 downlink MAC layer signaling frame channels according to frequency, and 8 uplink detection frames are used to provide 8 uplink MAC layer signaling frame channels. The signaling frame channels correspond to 8 uplink MAC layer signaling frame channels, forming 8 bidirectional signaling channels, as shown in FIG. 3 .

Step 4, the central node continuously sends the downlink detection frame in the time slot of the downlink detection frame, and the slave node continuously receives the downlink detection frame, and uses the downlink detection frame to complete downlink power control and downlink channel training.

In this embodiment, the downlink detection frame sent by the central node 0 carries 8 downlink MAC layer signaling frames, wherein the signs in each downlink MAC layer signaling frame are busy;

The two slave nodes 1 and 2 continuously receive the downlink detection frame, detect the signal strength of the downlink detection frame, and adjust the gain of the receiving power according to the signal strength, so that the receiver can accurately receive the downlink detection frame and complete the downlink power control;

The two slave nodes 1 and 2 extract the training sequence signal from the downlink sounding frame signal, and estimate the physical layer channel parameters according to the training sequence number signal, so that the slave nodes can select the modulation format according to the channel parameters and complete the downlink channel training.

Step 5: The slave node receives the downlink detection frame, queries the flags in the N downlink MAC layer signaling frames, and counts the number of idle bidirectional signaling channels.

In this embodiment, the slave nodes 1 and 2 query the flags in the 8-way downlink MAC layer signaling frames carried by the downlink detection frame, and find that the flags in the 8-way downlink MAC layer signaling frames are all busy, and obtain 8 bidirectional The signaling channels are all busy, and the number of idle two-way signaling channels is calculated as M=0.

Step 6, according to the statistics of step 5, the number of idle two-way signaling channels is 0, and the slave node 1 and slave node 2 wait for the admission period to end, and then return to step 5.

Since the number of idle bidirectional signaling channels is 0, that is, in this admission cycle, the 8 bidirectional signaling channels are all undergoing the admission process of other slave nodes, and cannot be used for slave nodes 1 and 2 to send uplink detection frames, so Neither the slave node 1 nor the slave node 2 sends an uplink detection frame in this acceptance period, and returns to step 5 after the acceptance period ends.

The above descriptions are only two specific examples of the present invention, and do not constitute any limitation to the present invention. Obviously, for those skilled in the art, after understanding the content and principles of the present invention, it is possible without departing from the principles of the present invention, In the case of structure, various modifications and changes in form and details are made, but these modifications and changes based on the concept of the present invention are still within the protection scope of the claims of the present invention.

Claims (7)

1., based on a multichannel parallel node acceptance method, comprise the steps:

(1) adopt time division multiplexing/time division multiple access scheme allocation of channel resources, to descending employing time division multiplexing, to up employing time division multiple access, channel is divided in time continuously and the receiving cycle of non-overlapping copies;

(2) within each receiving cycle, adopt fixing planning or Dynamic Programming mode, cook up the time slot of following different purposes:

Send several downlink data frame slots of downlink data frame for Centroid, send several upstream data frame slots of uplink data frames for slave node;

Send several descending control frame time slots of descending control frame for Centroid, send several upload control frame slots of upload control frame for slave node;

A descending explore frame time slot of a down physical layer explore frame is sent for Centroid;

N number of uplink detection frame slot of N number of upstream physical layer explore frame is sent for slave node;

(3) the descending explore frame time slot utilizing step (2) to plan to obtain and N number of uplink detection frame slot, form N number of two-way signalling path:

(3a) the down physical layer explore frame sent at a descending explore frame time slot is utilized, mark off N number of descending MAC layer signaling frame passage, the descending MAC layer signaling frame in carrying N road, each passage carries a descending MAC layer signaling frame in road, described MAC layer, refers to medium access control layer;

(3b) utilize the upstream physical layer explore frame sent at a uplink detection frame slot, a up MAC layer signaling frame passage is provided, carries row MAC layer signaling frame on the way;

(3c) N number of descending MAC layer signaling frame passage is corresponding with N number of up MAC layer signaling frame passage, form N number of two-way signalling path, i.e. i-th corresponding i-th up MAC layer signaling frame passage of descending MAC layer signaling frame passage, form i-th two-way signalling path, with mark in the descending MAC layer signaling frame of each descending MAC layer signaling frame passage carrying, show that the current state of the two-way signalling path of its correspondence is in busy or idle;

(4) Centroid sends down physical layer explore frame at descending explore frame time slot, slave node continuous reception down physical layer explore frame, and utilize down physical layer explore frame to carry out downlink power control and down channel training, after completing downlink power control and down channel training, enter step (5);

(5) slave node receives down physical layer explore frame, the mark in the descending MAC layer signaling frame in inquiry N road, and the number of the two-way signalling path that statistics is idle is M, if M=0, then repeats step (5) after waiting for this receiving end cycle; If M>0, then Stochastic choice two-way signalling path in the two-way signalling path of this M free time, sends at the uplink detection frame slot of its correspondence and receives request;

(6) Centroid upgrades the mark in the descending MAC layer signaling frame in N road, sends down physical layer explore frame:

If Centroid does not receive at some uplink detection frame slots and receives request, then the mark in the descending MAC layer signaling frame keeping this time slot corresponding is constant, and at descending explore frame time slot, Centroid sends down physical layer explore frame;

If Centroid have received at some uplink detection frame slots and receives request, be then set to busy by the mark in descending MAC layer signaling frame corresponding for this time slot, at descending explore frame time slot, Centroid utilizes this two-way signalling path to send and receives response;

(7) slave node receives down physical layer explore frame, the mark in the descending MAC layer signaling frame that the two-way signalling path selected by inquiry is corresponding:

If be masked as the free time, then the receiving request of slave node there occurs conflict, and slave node performs back off algorithm, waits for several receivings cycle, returns step (5);

If be masked as busy, then the receiving request of slave node sends successfully, slave node and Centroid utilize this two-way signalling path to complete up channel training or uplink power control, Centroid utilizes this two-way signalling path broadcast channel parameter, finally by Centroid, the mark in descending MAC layer signaling frame corresponding for this two-way signalling path is set to the free time, completes the receiving process of this slave node.

2. node acceptance method according to claim 1, in the receiving cycle in wherein said step (1), its time length setting is fixed length or random length.

3. node acceptance method according to claim 1, wherein said step fixes planning mode in (2), refer within each receiving cycle, the quantity of data frame interval and the quantity of position, control frame time slot are all the same with position, the position of descending explore frame time slot, the quantity of uplink detection frame slot with position.

4. node acceptance method according to claim 1, Dynamic Programming mode in wherein said step (2), refer within the difference receiving cycle, quantity and position, the position of descending explore frame time slot, the quantity of uplink detection frame slot and the position of the quantity of data frame interval and position, control frame time slot are all to adjust.

5. node acceptance method according to claim 1, marks off N number of descending MAC layer signaling frame passage wherein described in step (3a), is to divide by frequency partition or code word.

6. node acceptance method according to claim 1, utilizing down physical layer explore frame to carry out downlink power control in wherein said step (4), is adjusted according to the gain of signal strength signal intensity to received power detecting the down physical layer explore frame received by slave node.

7. node acceptance method according to claim 1, down physical layer explore frame is utilized to carry out down channel training in wherein said step (4), be first from the down physical layer explore frame signal received, extract training sequence signal by slave node, then according to this training sequence signal, physical layer channel parameter estimated.