Time slot scheduling method for underwater dynamic multi-hop distribution network based on time division multiple access protocol

Abstract

The application discloses a time slot scheduling method of an underwater dynamic multi-hop distribution network based on a time division multiple access protocol, relates to the field of underwater acoustic communication, and is aimed at the underwater multi-hop distribution network, improves the traditional static time slot allocation based on the TDMA protocol, designs a data packet frame structure, comprehensively considers transmission time delays among underwater nodes, dynamically adjusts time slots allocated to the underwater nodes based on the maximum transmission time delay, considers a dynamic node random access condition, and realizes the design of the time slot scheduling method of the underwater dynamic multi-hop distribution network.Compared with the traditional static time slot allocation method, the method can effectively improve the overall throughput of the network and optimize the network operation efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of underwater acoustic communication, in particular to a time slot scheduling method for underwater dynamic multi-hop distribution networks based on a time division multiple access protocol. BACKGROUND

[0002] The ocean is an important hinterland for people's normal life activities and the realization of social development, and has high development value and significance. Nowadays, many countries in the world attach great importance to marine undertakings, and underwater wireless sensor networks (UWSNs) have become a new research hotspot. UWSNs have achieved good results in underwater resource exploration, natural disaster prevention such as earthquakes and tsunamis, underwater target positioning and other aspects. UWSNs are to place sensor nodes in a specific sea area, and these nodes establish an underwater interconnected network. They monitor specific information through sensors and transmit information to each other, and finally send the information to the base station on the sea surface. Due to the sparse placement of underwater network nodes and the long communication distance between nodes, the communication of nodes in UWSNs mostly adopts the mode of underwater acoustic communication, which is currently widely used in UWSNs. However, since most of the network is placed underwater, its environment is more complex than other working environments such as land, so whether the underwater communication network can quickly and accurately and conveniently transmit information has become a problem that needs to be further studied and solved.

[0003] The MAC protocol is a bottom layer of underwater acoustic communication network, and its role is to allocate channel resources for each node in the network. According to the different channel allocation strategies, the existing MAC protocols can be divided into three categories: contention allocation MAC protocol, fixed allocation MAC protocol and hybrid MAC protocol of the two. The fixed allocation MAC protocol allocates time, frequency, space and code symbols orthogonally, so that users can send packet data in an orderly manner. This kind of protocol mainly includes: TDMA (Time Division Multiple Access), FDMA (Frequency Division Multiple Access) and CDMA (Code Division Multiple Access) etc. Due to the narrow bandwidth of underwater acoustic communication network, it is difficult to divide the frequency band, so the FDMA protocol is not applicable. In addition, when using the CDMA protocol, it is necessary to ensure that the code symbol sequence used remains orthogonal, and the coding work is relatively complex. Compared with wireless sensor networks on land, underwater sensor networks have long propagation delay, low point-to-point transmission rate and high bit error rate. In the contention allocation MAC protocol, when there are users in the network who need to send messages, they can directly occupy the channel to send or first make a channel reservation and then send messages. Due to the long propagation delay and location uncertainty of underwater, it is difficult to implement the contention allocation MAC protocol underwater, and the reliability is low.

[0004] In summary, the TDMA protocol is very suitable for the MAC layer protocol of the underwater communication network due to its reliability and easy implementation, but the traditional TDMA protocol based on static time slot allocation has the problem of low network transmission efficiency due to the fixed time slot of each node, and is mainly used in the underwater single-hop distributed network with fixed node number, so it is necessary to improve the existing TDMA protocol to improve the network transmission efficiency and enhance the protocol adaptability. SUMMARY

[0005] The present inventors propose a time slot scheduling method for underwater dynamic multi-hop distributed network based on TDMA protocol to solve the above problems and technical needs, and the technical scheme of the present application is as follows:

[0006] The time slot scheduling method for underwater dynamic multi-hop distributed network based on TDMA protocol comprises the following steps:

[0007] Let the initial number of nodes be N, and allocate the same length of initial communication time slot to each node; after the Nth communication time slot, add a time slot as a reserved time slot for the access of dynamic nodes;

[0008] In the first two communication periods, all nodes send node information in the allocated initial communication time slot, confirm the delay information with each adjacent node in the non-communication time slot of the node, and update the allocated communication time slot length of the node in the next communication period according to the delay information;

[0009] In the subsequent communication period, if a dynamic node accesses, in the current communication period and the next communication period, the dynamic node sends node information in the reserved time slot, the remaining nodes send node information according to the newly allocated communication time slot, all nodes confirm the delay information with each adjacent node in the non-communication time slot of the node, and update the allocated communication time slot length of the node in the next communication period according to the delay information; after the N+1th communication time slot, add a time slot as a reserved time slot for the access of the next dynamic node.

[0010] A further technical scheme is that the initial communication time slot and the initial reserved time slot have the same length t0=L max / c+T g , wherein L max represents the maximum communication distance of a single node, c represents the underwater sound speed, and T g represents the guard interval.

[0011] A further technical scheme is that in the first two communication periods, the method for updating the allocated communication time slot length of the node in the next communication period comprises:

[0012] In the first communication cycle, all nodes broadcast their node information in turn according to the initial communication time slots allocated, including the total number of nodes in the current network, the adjacent node ID and hidden node ID of the node, the time slot length of each node and the total length of time slots, the time point t of sending data packets by the node r ; in the non-communication time slots, the local node information is updated according to the node information broadcast by the adjacent nodes received, to confirm the adjacent nodes of the node itself;

[0013] In the second communication cycle, all nodes broadcast their node information in turn according to the initial communication time slots allocated again, and in the non-communication time slots, the local node information is updated according to the node information broadcast by the adjacent nodes received, to confirm the adjacent nodes and hidden nodes of the node itself; at the same time, the time point t of sending data packets by the adjacent nodes received is used to update the communication time slot length allocated by the node in the next communication cycle r and the local time t l , the time delay information t d of each adjacent node is calculated l = t r , and the time delay information t d calculated is used to update the communication time slot length allocated by the node in the next communication cycle;

[0014] Among them, all nodes within the maximum communication range of a node are called adjacent nodes of the node, otherwise they are called hidden nodes of the node.

[0015] A further technical solution is that in the subsequent communication cycle, if a dynamic node accesses, the method for updating the communication time slot length allocated by the node in the next communication cycle in the current communication cycle and the next communication cycle includes:

[0016] In the i-th communication cycle, the dynamic node confirms the start time of the reserved time slot according to the adjacent node information of the dynamic node, the dynamic node broadcasts its node information according to the allocated reserved time slot, and the remaining nodes broadcast their node information in turn according to the communication time slots allocated in the i-1-th time, including the total number of nodes in the current network, the adjacent node ID and hidden node ID of the node, the time slot length of each node and the total length of time slots, and the time point t of sending data packets by the node r ; all nodes update the local node information according to the node information broadcast by the adjacent nodes received in the non-communication time slots, to confirm the adjacent nodes of the node itself;

[0017] In the i+1-th communication cycle, the dynamic node broadcasts its node information according to the allocated reserved time slot again, and the remaining nodes broadcast their node information in turn according to the communication time slots allocated in the i-1-th time, and all nodes update the local node information according to the node information broadcast by the adjacent nodes received in the non-communication time slots, to confirm the adjacent nodes and hidden nodes of the node itself; at the same time, the time point t of sending data packets by the adjacent nodes received is used to update the communication time slot length allocated by the node in the next communication cycler The local time t l , the time delay information t d between each adjacent node is calculated l = t r -t d , and the communication time slot length allocated by the node in the next communication cycle is updated according to the calculated time delay information t

[0018] Wherein, i>2; all nodes within the maximum communication range of a node are called adjacent nodes of the node, otherwise, they are called hidden nodes of the node.

[0019] The further technical solution is that the method for updating the communication time slot length allocated by the node in the next communication cycle according to the calculated time delay information t d includes:

[0020] After each node confirms the time delay information t d of all adjacent nodes, the maximum time delay t dmax is found, and the communication time slot length allocated by the node in the next communication cycle is updated as: t=t dmax +t g , wherein T g represents a guard interval.

[0021] The further technical solution is that the method for the dynamic node to calculate the start time of the reserved time slot according to the adjacent node data packet includes:

[0022] After accessing the network, the dynamic node first enters a listening state to receive the node information broadcast by the adjacent nodes, and calculates the start time of the reserved time slot according to the time slot length of the adjacent nodes, the total length of the time slots and the length of the reserved time slot, wherein the listening state means that the node can only receive the adjacent node information in the non-communication time slot.

[0023] The further technical solution is that the node information is embodied in the form of a data packet frame structure, and includes:

[0024] Length: indicating the length of the control frame;

[0025] srcAddr: source address, ID of the sending node;

[0026] dstAddr: destination address, ID of the receiving node;

[0027] NodeNum: node data, containing the total number of nodes in the current network, the adjacent node ID and the hidden node ID of the node;

[0028] SlotNum: time slot data, containing the time slot length of each node, the total number of time slots in a communication cycle and the total length of the time slots;

[0029] Time: time data, containing the time point t of the data sent by the node r ;

[0030] Data: data packet from the application layer;

[0031] CRC: redundancy check code, used for error control.

[0032] Further technical solutions are that before the method is executed, it is assumed that:

[0033] All nodes in the network are globally clock-synchronized, so that each time slot is aligned; the communication of any two adjacent nodes is not interfered; the network node attributes do not change during the entire communication period; and the maximum distance between any two nodes is two hops.

[0034] The beneficial technical effects of the application are:

[0035] Considering the underwater dynamic multi-hop distributed network in which a node dynamically changes, a network time slot scheduling method based on a TDMA protocol is designed, which considers the random access of dynamic nodes and improves the problem of low network transmission efficiency caused by long end-to-end delay in the static time slot scheduling based on the traditional TDMA protocol. In the first two communication periods, the time slots in the next communication period are adaptively and dynamically allocated according to the delay information of the node and each adjacent node, to obtain the dynamic time slot allocation of all nodes and the adjacent node information and hidden node information of each node, thereby effectively improving the network efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 is a schematic diagram of an underwater dynamic multi-hop distributed network topology provided by the application.

[0037] Figure 2 is a static time slot allocation diagram based on a TDMA protocol.

[0038] Figure 3 is a dynamic time slot allocation diagram based on a TDMA protocol.

[0039] Figure 4 is a data packet frame structure provided by the application.

[0040] Figure 5 is a flowchart of an underwater dynamic multi-hop distributed network time slot scheduling method provided by the application.

[0041] Figure 6 is a schematic diagram of an underwater dynamic multi-hop distributed network topology in an embodiment of the application.

[0042] Figure 7 is a throughput simulation performance comparison in an embodiment of the application. Detailed Implementation

[0043] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0044] This embodiment provides a time slot scheduling method for an underwater dynamic multi-hop distributed network based on a time-division multiple access protocol. In this method, the number of nodes dynamically changes, and at most one dynamic node randomly accesses the network during each communication cycle. Hidden nodes exist within the network, such as… Figure 1 As shown. In this network, the following definitions exist:

[0045] 1) All nodes within a node's maximum communication range are called the node's neighboring nodes; otherwise, they are called the node's hidden nodes.

[0046] 2) Time slot allocation is divided into static time slot allocation and dynamic time slot allocation. Static time slot allocation assigns a fixed time slot to each node, and the node only transmits data within the allocated time slot, such as... Figure 2 As shown; dynamic time slot allocation dynamically allocates time slots to each node based on the maximum transmission delay, such as... Figure 3 As shown.

[0047] 3) The data packet frame structure transmitted between nodes is as follows: Figure 4 As shown, it includes:

[0048] Length: Indicates the length of the control frame;

[0049] srcAddr: Source address, ID of the sending node;

[0050] dstAddr: Destination address, ID of the receiving node;

[0051] NodeNum: Node data, including the total number of nodes in the current network, the IDs of the neighboring nodes of this node, and the ID of the hidden node;

[0052] SlotNum: Slot data, including the slot length of each node, the total number of slots in one communication cycle, and the total slot length;

[0053] Time: Time data, including the time t when this node sent the data. r ;

[0054] Data: Represents data packets originating from the application layer;

[0055] CRC: Redundancy Check Code, used for error control.

[0056] It should be noted that the following assumptions are made before implementing this method:

[0057] 1) All nodes in the network are precisely synchronized with a global clock, ensuring alignment of each time slot; 2) Communication between any two adjacent nodes is interference-free; 3) The attributes of network nodes do not change during the entire communication period, i.e., adjacent nodes will not become hidden nodes, and hidden nodes will not become adjacent nodes; 4) The maximum distance between any two nodes is two hops, such as... Figure 1 As shown, communication between node 1 and node 2 requires node 4 as a hop point.

[0058] like Figure 5 As shown, the underwater dynamic multi-hop distributed network time slot scheduling method includes the following steps:

[0059] Step 1: Let the initial number of nodes be N, and the maximum communication distance of a single node be L. max ( Figure 1 If the radius of the middle circle is L, then each node is allocated an initial communication time slot of the same length t0, where t0 = L. max / c+T g Where c represents the speed of sound underwater, T g This represents the protection interval. An additional time slot is added after the Nth communication time slot, resulting in a total of N+1 time slots. The first N time slots are for communication between the N nodes, and the last time slot is a reserved time slot for dynamic node access, with the same length t0. All nodes transmit data within the predetermined time slots. During their non-communication time slots, they enter a listening state and can only receive information from neighboring nodes.

[0060] Step 2: During the first communication cycle, all nodes take turns broadcasting their own information according to the allocated initial communication time slots. During non-communication time slots, they update their local node information based on the node information broadcast by neighboring nodes to confirm their own neighboring nodes.

[0061] In this embodiment, the broadcast node information is represented in the form of a data packet frame structure. The information to be broadcast includes Length, srcAddr, dstAddr, NodeNum, SlotNum, and Time. When updating local node information, the node information broadcast by neighboring nodes is stored in the corresponding position according to the format of the data packet frame structure, such as identifying neighboring nodes in the NodeNum part of the data packet. After this communication, each node knows its neighboring nodes, but since node 1 is located in the first position of the carousel, node 1's node information has not yet been broadcast in this round. Except for the last node, the hidden node information of the other nodes is also uncertain, so two rounds of communication are required.

[0062] Step 3: During the second communication cycle, all nodes take turns broadcasting their own information according to the allocated initial communication time slots. In non-communication time slots, they update their local node information based on the node information broadcast by neighboring nodes to confirm their neighboring nodes and hidden nodes. Simultaneously, they update their information based on the time point t of the data packets sent by neighboring nodes. r With local time t l Calculate the time delay information t with each adjacent node. d =t l -t r And based on the calculated time delay information t d Update the communication time slot length allocated to this node in the next communication cycle. Specifically, each node confirms the delay information t of all neighboring nodes. d Then, find the maximum time delay t. dmax Then the length of the communication time slot allocated to this node in the next communication cycle is updated as follows: t = t dmax +T g The time slot data is then embedded into the SlotNum portion of the data packet.

[0063] Step 4: In the subsequent communication cycle, all nodes transmit data according to the newly allocated communication time slots, and update the data of adjacent nodes and time slot data in real time based on the method in Step 3. The reserved time slot length in each communication cycle remains unchanged and is kept at the initial time slot length t0.

[0064] If a dynamic node connects during a subsequent communication cycle, its time slot is allocated as a reserved time slot, and step 5 is executed.

[0065] Step 5: During the i-th communication cycle (i > 2), N+1 dynamic nodes connect. Each dynamic node determines the start time of its reserved time slot based on its neighboring node information. The dynamic node broadcasts its own information according to its allocated reserved time slot. The remaining nodes take turns broadcasting their own information according to the (i-1)-th allocated communication time slot. All nodes update their local node information during non-communication time slots based on the node information broadcast by their neighboring nodes to confirm their own neighboring nodes. The broadcast node information is also represented in the form of a data packet frame structure, as described in Step 2.

[0066] The start time for a dynamic node to confirm the reserved time slot based on data packets from its neighboring nodes includes:

[0067] After a dynamic node connects to the network, it first enters a listening state, receives node information broadcast by neighboring nodes, and calculates the start time of the reserved time slot based on the time slot length of neighboring nodes, the total time slot length, and the reserved time slot length.

[0068] In this step, the dynamic node starts broadcasting node information in the reserved time slot, its adjacent nodes receive the information of the dynamic node and save it to update their own adjacent node information, and broadcast in the next communication period to ensure that all nodes can confirm the information of the dynamic node accessing the network in the next round of communication period.

[0069] Step 6: In the i+1th communication period, the dynamic node broadcasts its node information according to the allocated reserved time slot, and the remaining nodes broadcast their node information according to the i-1th allocated communication time slot in turn, and all nodes update their local node information according to the received adjacent node broadcasted node information in the non-communication time slot to confirm their own adjacent nodes and hidden nodes; at the same time, according to the received adjacent node sending data packet time point t r and the local time t l , the time delay information t d of each adjacent node is calculated, t l = t r -t d , and the time delay information t dmax is used to update the allocated communication time slot length of the node in the next communication period, and the expression is: t = t g + T dmax , where t d is the maximum time delay in the time delay information t max of all adjacent nodes, and the time slot data is embedded in the SlotNum part of the data packet.

[0070] At this time, the total number of nodes becomes N+1, and one time slot is added after the N+1th communication time slot, that is, N+2 time slots are set in total, of which the first N+1 time slots are the communication time slots of the N+1 nodes, and the last time slot is a reserved time slot for the access of the next dynamic node, and the length of the reserved time slot is also t0.

[0071] In an embodiment, as shown in Figure 6 , there are 5 initial nodes in the network, and the nodes in the same dashed circle can communicate with each other, that is, node 1, node 3, and node 4 can communicate with each other, node 2, node 3, and node 4 can communicate with each other, and node 3 and node 5 can communicate with each other, and the nodes that cannot communicate with each other are hidden nodes. In this embodiment, the maximum communication distance L max of a single node is set to 1000m, and the maximum transmission time delay t dmax between nodes is calculated as 0.67s at a sound speed c of 1500m / s, and the guard interval T g is set to 0.2s, so the initial communication time slot length t0 of this embodiment is set to 0.87s. Assuming that in the fifth communication period, a dynamic node 6 accesses the network, the network throughput changes with time as Figure 7The throughput, which is the total amount of data exchanged per unit of time, is often used to measure network performance.

[0072] Figure 7 The simulation comparison results of network average throughput of dynamic time slot TDMA protocol and static time slot TDMA protocol can be seen from the figure that in the first two communication periods, the network is still in the node information confirmation stage, and there is no data exchange, so the throughput is 0. From the third communication period, the throughput of the two networks gradually increases and tends to be stable. It can be seen that the throughput of the dynamic time slot TDMA protocol is higher than that of the traditional TDMA protocol, because the time slot of the dynamic time slot TDMA protocol is dynamically determined according to the time delay between nodes, the utilization rate of the channel is improved, the time length occupied by each period is greatly shortened, and the network average throughput is improved. In the fifth communication period, the network throughput decreases temporarily and then gradually increases and finally tends to be stable, because the dynamic node access needs to first exchange node information, which affects data transmission, and when the node information is updated, the data transmission returns to normal, and the throughput gradually increases. It can be seen that the dynamic time slot allocation proposed in the application has higher throughput than the traditional static allocation, and effectively improves the network efficiency.

[0073] The above only describes the preferred embodiments of the present application, and the present application is not limited to the above embodiments. It can be understood that other improvements and changes directly derived or thought by those skilled in the art without departing from the spirit and concept of the present application should be considered to be included in the protection scope of the present application.

Claims

1. A time slot scheduling method for underwater dynamic multi-hop distribution networks based on time division multiple access protocol, characterized in that, The method comprises: Let the initial node number be N and assign the same length of initial communication time slot for each node; after the first N communication time slot, increase one time slot as a reserved time slot for dynamic node access; In the first two communication periods, all nodes send node information in the assigned initial communication time slot, confirm time delay information with each adjacent node in the non-communication time slot of the node, and update the length of the communication time slot assigned by the node in the next communication period according to the time delay information; If the dynamic node accesses in the following communication cycle, the dynamic node sends its node information in the reserved time slot in the current communication cycle and the next communication cycle, the rest of the nodes send their node information according to the newly allocated communication time slot, all nodes confirm the time delay information with each adjacent node in the non-communication time slot of the node, and update the length of the communication time slot allocated by the node in the next communication cycle according to the time delay information; after the first N+ 1 communication time slot, a time slot is added as a reserved time slot for the access of the next dynamic node; If the dynamic node accesses, the method for updating the length of the communication time slot assigned by the node in the next communication period in the current communication period and the next communication period comprises: In the first i communication cycle, the dynamic node confirms the starting time of the reserved time slot according to its adjacent node information, and broadcasts its node information according to the allocated reserved time slot, while the rest of the nodes broadcast their node information in turn according to the first allocated communication time slot, including the total number of nodes in the current network, the adjacent node ID and hidden node ID of the node, the time slot length of each node and the total length of the time slot, the time point of transmitting data packet by the node i ; all nodes update the local node information according to the received adjacent node broadcasted node information in the non-communication time slot to confirm their adjacent nodes; ​ In the i+ Within one communication cycle, the dynamic node broadcasts its information according to the allocated reserved time slot, and the other nodes broadcast their information according to the allocated time slot. i- Each node broadcasts its own information in turn during the allocated communication time slots. During non-communication time slots, all nodes update their local information based on the node information broadcast by their neighboring nodes to identify their neighbors and hidden nodes. Simultaneously, they adjust their information based on the time points when neighboring nodes send data packets. With local time Calculate the time delay information with each adjacent node. And based on the calculated time delay information Update the communication time slot length allocated to this node in the next communication cycle; wherein i >2; all nodes within the maximum communication range of a node are called its neighbors, otherwise they are called its hidden nodes.

2. The time slot scheduling method for underwater dynamic multi-hop distributed networks based on time division multiple access protocol according to claim 1, characterized in that, The initial communication time slot is the same as the initial reserved time slot length, both of which are wherein represents the maximum communication distance of a single node, c represents the underwater sound velocity, represents the protection interval.

3. The method of claim 1, wherein, In the first two communication periods, the method for updating the length of the communication time slot assigned by the node in the next communication period comprises: In the first communication cycle, all nodes broadcast their node information in turn according to the assigned initial communication time slot, including the total number of nodes in the current network, the adjacent node ID and hidden node ID of the node, the time slot length of each node and the total length of the time slot, the time point of sending data packet by the node ; and updating the local node information according to the received node information broadcasted by the adjacent nodes in the non-communication time slot to confirm the adjacent nodes of itself During the second communication cycle, all nodes again take turns broadcasting their own information according to the allocated initial communication time slots. In non-communication time slots, they update their local node information based on the node information broadcast by neighboring nodes to identify their neighboring nodes and hidden nodes. Simultaneously, they adjust their information based on the time points when neighboring nodes send data packets. With local time Calculate the time delay information with each adjacent node. And based on the calculated time delay information Update the communication time slot length allocated to this node in the next communication cycle; All nodes within the maximum communication range of a node are called adjacent nodes of the node, otherwise, they are called hidden nodes of the node.

4. The method of claim 1, wherein, The calculated delay information The method for updating the length of the communication time slot allocated by the node in the next communication period comprises: Each node confirms the time delay information of all adjacent nodes Then, find the maximum time delay in them Then, find the maximum time delay in them The length of the communication time slot allocated by the node in the next communication period is updated as: The protection interval is represented.

5. The method of claim 1, wherein, The method for confirming the start time of the reserved time slot by the dynamic node according to the adjacent node data packet comprises: The dynamic node enters a listening state after accessing the network, receives node information broadcast by adjacent nodes, and calculates the start time of the reserved time slot according to the time slot length of the adjacent node, the total length of the time slot and the length of the reserved time slot, wherein the listening state means that the node can only receive adjacent node information in the non-communication time slot.

6. The method of claim 1-5, wherein, The node information is in the form of a data packet frame structure, and comprises: Length: indicating the length of the control frame; srcAddr: source address, ID of the sending node; dstAddr: destination address, ID of the receiving node; NodeNum: node data, containing the total number of nodes in the current network, adjacent node ID and hidden node ID of the node; SlotNum: time slot data, containing the time slot length of each node, the total number of time slots in a communication period and the total length of the time slot; Time: time data, containing the time point at which the data is sent by the node ; Data: indicating the data packet from the application layer; CRC: redundancy check code, used for error control.

7. The method of claim 1-5, wherein, Before the method is executed, it is assumed that: All nodes in the network are globally clock-synchronized so that each time slot is aligned; communication between any two adjacent nodes does not interfere with each other; the network node attributes do not change during the entire communication period; and the maximum distance between any two nodes is two hops.

Images