CN117098192A - Urban rail ad hoc network resource allocation method based on capacity and time delay optimization - Google Patents
Urban rail ad hoc network resource allocation method based on capacity and time delay optimization Download PDFInfo
- Publication number
- CN117098192A CN117098192A CN202310986895.XA CN202310986895A CN117098192A CN 117098192 A CN117098192 A CN 117098192A CN 202310986895 A CN202310986895 A CN 202310986895A CN 117098192 A CN117098192 A CN 117098192A
- Authority
- CN
- China
- Prior art keywords
- node
- transmission
- time
- network
- nodes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000013468 resource allocation Methods 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000005457 optimization Methods 0.000 title claims abstract description 18
- 230000005540 biological transmission Effects 0.000 claims abstract description 122
- 238000004891 communication Methods 0.000 claims abstract description 34
- 238000012549 training Methods 0.000 claims abstract description 8
- 230000000737 periodic effect Effects 0.000 claims abstract description 6
- 238000012544 monitoring process Methods 0.000 claims abstract description 5
- 230000007246 mechanism Effects 0.000 claims abstract description 4
- 238000004364 calculation method Methods 0.000 claims description 18
- 230000009471 action Effects 0.000 claims description 11
- 238000011156 evaluation Methods 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 4
- 230000009191 jumping Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 230000002787 reinforcement Effects 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 claims description 2
- 238000004220 aggregation Methods 0.000 claims description 2
- 230000001174 ascending effect Effects 0.000 claims description 2
- 230000008569 process Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 241000854291 Dianthus carthusianorum Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
- H04W28/18—Negotiating wireless communication parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
- H04W4/44—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
- H04W4/46—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for vehicle-to-vehicle communication [V2V]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/24—Connectivity information management, e.g. connectivity discovery or connectivity update
- H04W40/32—Connectivity information management, e.g. connectivity discovery or connectivity update for defining a routing cluster membership
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
- H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention provides a capacity and time delay optimization-based urban rail ad hoc network resource allocation method which is characterized by comprising the following steps of S1, establishing transmission paths of vehicles and multiple nodes of the vehicles and the places which are not intersected; s2, analyzing the space multiplexing condition of the time-frequency resource blocks of the nodes in the network and the interference condition when the nodes on the path share the time-frequency resource blocks, and calculating the network capacity; step S3, setting transmission rates of different time slots of the node; step S4, calculating average transmission time delay based on the characteristics of periodic communication between the vehicle and the ground; s5, obtaining an optimal communication path set and a resource allocation scheme of the vehicles and the vehicle ground through a training Q network; and S6, aiming at the obtained optimal strategy, designing an implementation mechanism to inform nodes in the cluster, and monitoring the state of the nodes by utilizing the capacity and time delay performance indexes of the nodes. On the basis of the clustering ad hoc network architecture, optimal resource allocation is realized, and the low-time-delay and high-throughput communication requirements of vehicles and vehicles of the urban rail train control system are met.
Description
Technical Field
The invention relates to the field of rail transit, in particular to a urban rail ad hoc network resource allocation method based on capacity and time delay optimization.
Background
The self-organizing network is applied to a next generation urban rail train control system, a clustering networking architecture is adopted, the network is divided into different clusters, and gateway nodes, mobile nodes and relay nodes are deployed beside the rails; the gateway node is placed at a station, is connected with a wireless network and a wired network, and is a transmitting end or a receiving end for the ground communication of the comprehensive service carrier; the gateway node is used as a cluster head and is responsible for management of members in the cluster and interaction of information among the clusters; when the tracking train and the front train position cannot directly communicate with different clusters, the gateway node can provide data transfer service; the train is used as a mobile node to send or receive vehicle-mounted data; the relay node is responsible for forwarding the data of the adjacent nodes, so as to realize multi-hop communication between the vehicle and the ground; the allocation of radio resources is an important factor affecting the communication service quality of the ad hoc network, so how to meet the communication requirements of low latency and high throughput by allocating and adjusting radio resources is a key of applying the wireless ad hoc network in the urban rail system.
The existing wireless resource allocation method is mainly divided into static resource allocation and dynamic resource allocation; static resource allocation can lead to service which cannot adapt to real-time change, and the resource utilization rate is low; the dynamic resource allocation needs to collect node information of the whole network and then allocate the resources, so that the service demand change of the nodes cannot be responded in time.
Accordingly, there is a need for optimizing throughput and latency to meet the communication needs of urban rail-mounted systems.
Disclosure of Invention
The invention aims to provide a capacity and time delay optimization-based urban rail ad hoc network resource allocation method which is used for solving the problems.
The technical scheme of the invention is as follows: a urban rail ad hoc network resource allocation method based on capacity and time delay optimization comprises the following steps:
step S1, based on the position of each train in a cluster, establishing a transmission path where a plurality of nodes of the train and the train ground are not intersected;
s2, according to a protocol interference model, analyzing the space multiplexing condition of a node time-frequency resource block in a network and the interference condition when nodes on a path share the time-frequency resource block, and calculating the capacity of a receiving node, the capacity of the path and the network capacity in a cluster according to a shannon formula;
step S3, setting the transmission rate of each node in different transmission time slots as the channel capacity of the time slot;
s4, analyzing queuing conditions of the data packets based on the characteristics of periodic communication between the vehicle and the ground, and calculating queuing length, waiting time delay and average transmission time delay of the data packets;
step S5, combining deep reinforcement learning, designing intelligent agents, states, actions and rewarding functions in a model, obtaining an optimal communication path set and a resource allocation scheme of a vehicle and a vehicle ground through a training Q network, and updating a transmission strategy according to the received signal strength RSSI value of a first-hop relay node;
and S6, aiming at the obtained optimal strategy, designing an implementation mechanism to inform nodes in the cluster, and monitoring the state of the nodes by utilizing the capacity and time delay performance indexes of the nodes.
Preferably, step S1 specifically includes:
in order to ensure that the damage of part of nodes in the paths does not affect the transmission of other paths and avoid the queuing delay of data packets at a relay node, m node-disjoint concurrent transmission paths from a source to a destination are respectively established for n trains in a cluster, and the transmission path set in a network is omega= { omega 1 ,Ω 2 ,…,Ω n Each train S i The concurrent transmission path set is Ω i ＝{M 1 ,M 2 ,…,M m Node set ofPath M k The node set is asz is the number of relay nodes on the path; the establishment of the disjoint paths of the vehicle and the vehicle-ground nodes needs to meet the following conditions:
where Ω is a set of transmission paths in the network, Ω i And omega e Respectively is train S i And S is e And N is equal to the sum of the concurrent transmission paths of the (a) Ωi Andrespectively, path sets Ω i And omega e Node set, M k And M q Is a train S i Is>Andfor path M k And M q A set of nodes thereon.
Preferably, step S2 specifically includes:
assuming that the total bandwidth W of the system is divided into a mutually orthogonal sub-channels, each sub-channel has a bandwidth ofTime period T f Dividing into b time slots, each time slot having a length of +>The set of time-frequency resource blocks is k= { (f) n ,t m )|n∈[1,a],m∈[1,b],n,m∈N * And (f), where n ,t m ) For a time-frequency resource block, N and m are subscripts of the sub-channel and the time slot, respectively, N * Is a positive integer set;
according to the protocol interference model, considering the spatial multiplexing of the time-frequency resource blocks, when a certain distance is met between nodes on a path, the resource blocks can be shared without causing interference;
defining resource block occupancy factor for node iThe method comprises the following steps:
if node i occupies resource block (f) n ,t m ) Resource block occupation factor1, otherwise, < >>Is 0;
the spatial multiplexing condition of node i and node k or q is:
d ij ＜R c ,d kj ＞R i or d ij ＜R c ,d qj ＞R i (4)
where k represents a node in the same transmission path set as node i, q represents a node in a different transmission path set from node i, d ij Representing the Euclidean distance, d, between node i and node j kj Represents the Euclidean distance, d, between node k and node j qj Representing the Euclidean distance, R, between node q and node j c Representing the transmission range of the node, R i Representing the interference range of the node.
Preferably, in S2, the interference condition analysis and the network capacity calculation when the nodes share the time-frequency resource block specifically include:
(1) Interference condition analysis:
if a certain node i in the transmission path set occupies a resource block to perform data transmission, when a receiving node j is positioned in the interference range of another sending node k and the node k occupies the same resource block to perform transmission, the node k generates interference to the receiving node j; the receiving node j on the path may be interfered by the nodes occupying the same resource block in the same transmission path set and different transmission path sets;
wherein,occupying f for receiving node j n ,t m Signal-to-noise ratio of resource block,/->For the transmission path set Ω i Total interference of other transmitting nodes k to receiving node j +.>Aggregation Ω for other transmission paths e The total interference of the transmitting node q to the receiving node j, P c For the transmission power of the node, N 0 G is the noise power ij Represents the channel gain, g, between node i and node j kj Representing the channel gain, g, between node k and node j qj Representing the channel gain between node q and node j;
considering only free space path loss, the channel gain isGamma is the path loss constant, x i And y i Respectively the abscissa and the ordinate of the node i; let it be assumed that at t m Time slotsThe running of the train can be approximately regarded as uniform linear motion, and the running speed is upsilon, then t m The position coordinates of the slot trains can be expressed as(x s ,y s ) At t for the train m Position coordinates at the beginning of a slot;
(2) Network capacity calculation:
by using shannon's formula, node j occupies f n ,t m Capacity of resource blockThe calculation is as follows:
wherein ω is f n ,t m The channel bandwidth of the resource block,occupy f for node j n ,t m Signal to noise ratio of resource block;
considering that the node adopts a multi-interface multi-channel technology, the node j is at t m Capacity of time slotThe calculation is as follows:
wherein,occupy f for node j n ,t m Capacity of resource blocks;
path M i During a time period T f Average capacity inThe calculation is as follows:
wherein,for node j at t m Capacity of time slot>For path M i A set of nodes on the table;
the average capacity realized by the transmission path set of n trains in the cluster is calculated as:
wherein,for path M i During a time period T f Average volume in omega i For each source node S i And Ω is a set of transmission paths in the network.
Preferably, step S4 specifically includes:
assume that the communication periods of the vehicle and the ground are T s The length of the data packet is h, and in order to meet the time delay requirement of the service, the source node must complete a complete data transmission and receiving process in each communication period; the queuing situation of the data packet is analyzed, and the average transmission delay is calculated as follows:
(1) And (3) queuing condition analysis:
according to the characteristic of periodic communication between the train and the ground, when the train or gateway node receives information sent by a plurality of source nodes, a new data packet is generated after short processing, and the new data packet is used as the source node to send information to a plurality of destination nodes, so that the phenomenon that the data packet is queued in a buffer area of the source node can occur, and therefore, queuing delay exists at the source node; the source node transmits data in the second period after completing data transmission and receiving in one period, so that the phenomenon of queuing of the data packet at the relay node is avoided, and therefore, queuing delay of the data packet at the relay node is avoided; since the node can only send data in the allocated time slot, there is scheduling delay at both the source node and the relay node; in summary, the waiting time delay of the data packet at the source node includes queuing time delay and scheduling time delay, and the waiting time delay at the relay node only has scheduling time delay;
(2) Calculating average transmission delay:
let t of the packet at the previous node a m The time slot is transmitted in a time slot, at t m Time slot start time,/->At t m The end time of the time slot, the transmission time is kappa a The transmission delay of the data packet at node a is +.>h is the packet length, < >>T at node a for a data packet m The sending rate of the time slot reaches the destination node at the time of kappa a +Δ a The delay of generating new data packet after processing at destination node is ignored, and the node is considered to adopt multi-interface multi-channel technology, so the queuing length of new data packet at source node S is ∈ ->Expressed as:
wherein,representing the source node at t m-1 Queue length at the end of a slot, κ a Indicating the moment of transmission of the data packet at node a, < >>At t m Time slot start time,/->Representing the total transmission rate of the nodes in a, a representing at t m The set of all nodes whose time slots are sent to the destination node, is->T representing the packet at the source node s m Transmission rate of time slot,/">Representing the source node at t m Queue length at the end of the time slot;
because the data transmission of the source node needs to occupy a plurality of time slots, the time slot ascending sequence set of the source node is defined asWherein->n represents the number of occupied time slots, x represents the time slot sequence number +.>Is->Time slot start time,/->Is->The end time of the time slot; according to different generation moments of the data packet at the source node, the waiting time delay at the source node is divided into the following two cases:
case one: when (when)Or->When the first packet of the current queue will be in +.>Sending at the moment;
and a second case: when (when)When the first data packet of the current queue is being transmitted;
wherein Y represents the transmission delay of the preceding data packet in the transmission time slot of the data packet,indicating that the data packet is at the sourceQueuing length of node s, < >>Indicating that the data packet is +.>Transmission rate of time slot,/">Indicating that the data packet is +.>Transmission rate of time slot,/">Is->Time slot start time, κ a Delta for the transmission time of the data packet at node a a For the transmission delay of the data packet at node a, < >>Indicating that the data packet is +.>The transmission rate of the time slot; if 0.ltoreq.Y < τ, the packet will be in +.>Time slot transmission;
combining the two cases, the waiting time delay q of the data packet at the source node s The calculation is as follows:
wherein,is->Time slot start time, κ a Delta for the transmission time of the data packet at node a a The transmission delay of the data packet at the node a is represented by Y, wherein Y represents the transmission delay of the data packet in front of the transmission time slot of the data packet;
single-hop delay d i,j The calculation is as follows:
wherein, kappa j For the moment of sending the data packet at node j, κ i T is the sending time of the data packet at the node i f Is the length of the time period;
path M i End-to-end delay of (c)The calculation is as follows:
wherein q s D, waiting time delay of data packet at source node i,j In the form of a one-hop delay,for path M i Node set on->Node for data packet>Transmission delay at the location;
average transmission delay D of n trains in the cluster Ω The calculation is as follows:
wherein m is i Omega for the number of concurrent transmission paths per train i For each source node S i Is a set of transmission paths in the network,for path M i End-to-end delay of (c).
Preferably, step S5 specifically includes:
firstly, designing an intelligent body, a state, actions and rewarding functions in a model, wherein each train is regarded as an intelligent body, the intelligent body searches an optimal communication path set and a resource allocation scheme of the train and the train place according to the current network state, the state is an abstraction of the urban rail self-organizing network environment, the actions comprise two aspects of path selection and resource block allocation, and rewarding consists of two parts of network capacity and transmission delay; the status, action and reward functions are each specifically represented as follows:
s t ＝{U t ,Q t ,H t ,G t ,V t } (20)
a t ＝{Ω t ,K t } (21)
r t ＝λ r R Ω -λ d D Ω (22)
in the formula (20), U t Represents the position of the train, Q t Representing a network topology including locations of nodes and connectivity relationships between nodes, H t Representing available node and resource block information, G t Channel gain information representing first hop, V t Spatial multiplexing information representing a network;
in the formula (21), Ω t Represents the node disjoint transmission path set selected by each vehicle at the moment t, K t A set of resource blocks representing the allocation of each vehicle to nodes on the transmission path;
in the formula (22), R Ω And D Ω Respectively representing the sum of network capacities of all concurrent transmission pathsAverage time delay lambda r And lambda (lambda) d Respectively, the weights of the two.
Preferably, in step S5, the optimal communication path set and resource allocation scheme of the vehicle and the vehicle ground are obtained through the training Q network, and the specific steps are as follows:
(1) Initializing a network communication environment, evaluating a parameter theta of a network and a parameter theta' of a target network, and inputting relevant training parameters;
(2) The train interacts with the environment according to the state s t Executing action a t Obtain rewards r t And enter the next state s t+1 And obtain tuple < s t ,a t ,r t ,s t+1 Storing the data in an experience memory pool D;
(3) After enough tuples exist in the experience pool, randomly extracting the mini-batch tuple from the experience pool, respectively generating an evaluation Q value and a target Q value through an evaluation network and a target network, then calculating a loss function between the evaluation Q value and the target Q value, and updating network parameters of the evaluation network by adopting gradient descent;
(4) Assigning the parameter values of the evaluation network to the target network at intervals, and updating the parameters of the target network;
(5) Judging whether the round is ended; if not, jumping to the step (2); if the method is finished, outputting a Q network, an optimal communication path set and a resource allocation scheme;
(6) Judging the received signal strength RSSI value of the first-hop relay node; if the train is smaller than the given threshold delta and the train is not in the data transmission stage, jumping to the step (1); if the threshold delta is greater than or equal to the given threshold delta, the original scheme is preserved.
The invention has the beneficial effects that:
according to the invention, optimal resource allocation is realized on the basis of a clustering ad hoc network architecture, on one hand, the communication networking architecture is simplified without depending on fixed infrastructure, the anti-destruction performance of the network is improved, and the deployment and maintenance cost is reduced; on the other hand, the communication requirements of low time delay and high throughput of the urban rail train control system train and the train-ground can be met, and the running safety and the running efficiency of the train are improved.
Drawings
Fig. 1 is an overall step diagram of a metro ad hoc network resource allocation method based on capacity and time delay optimization provided by an embodiment of the present invention;
fig. 2 is a schematic diagram of a vehicle-to-vehicle and vehicle-to-vehicle communication scene in a method for allocating urban rail ad hoc network resources based on capacity and time delay optimization according to an embodiment of the present invention;
fig. 3 is a diagram of a node resource allocation information issuing step in a metro ad hoc network resource allocation method based on capacity and time delay optimization according to an embodiment of the present invention;
fig. 4 is a step diagram of node status monitoring in a metro ad hoc network resource allocation method based on capacity and time delay optimization according to an embodiment of the present invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and specific examples so that those skilled in the art may better understand the present invention and practice it, and the embodiments of the present invention are not limited thereto.
Example 1
As shown in fig. 1, the overall step diagram of a metro ad hoc network resource allocation method based on capacity and time delay optimization includes:
step S1, based on the position of each train in a cluster, establishing a transmission path where a plurality of nodes of the train and the train ground are not intersected;
s2, according to a protocol interference model, analyzing the space multiplexing condition of a node time-frequency resource block in a network and the interference condition when nodes on a path share the time-frequency resource block, and calculating the capacity of a receiving node, the capacity of the path and the network capacity in a cluster according to a shannon formula;
step S3, setting the transmission rate of each node in different transmission time slots as the channel capacity of the time slot;
s4, analyzing queuing conditions of the data packets based on the characteristics of periodic communication between the vehicle and the ground, and calculating queuing length, waiting time delay and average transmission time delay of the data packets;
step S5, combining deep reinforcement learning, designing intelligent agents, states, actions and rewarding functions in a model, obtaining an optimal communication path set and a resource allocation scheme of a vehicle and a vehicle ground through a training Q network, and updating a transmission strategy according to the received signal strength RSSI value of a first-hop relay node;
and S6, aiming at the obtained optimal strategy, designing an implementation mechanism to inform nodes in the cluster, and monitoring the state of the nodes by utilizing the capacity and time delay performance indexes of the nodes.
As shown in fig. 2, a communication scene of vehicles and vehicles in a cluster is established according to a cluster network architecture of the urban rail system, gateway nodes are deployed at the central position of each cluster, and relay nodes are distributed beside the rail at equal intervals. Each train in the cluster needs to communicate with the ground through the forwarding of the gateway node in addition to the traffic through the multi-hop relay nodes, and each row of relay nodes in the lateral direction can only communicate with the relay nodes in its neighboring row.
As shown in fig. 3, when a train transmits train information to a gateway node, node resource configuration information including an optimal transmission path set, resource blocks occupied by nodes on a path, power and transmission rate is added to a transmitted data packet, and then the node resource configuration information is transmitted along with data along a communication path. And after receiving the data packet sent by the previous node, the relay node judges whether to forward the information, if the resource configuration information contains the information of the relay node, the relay node forwards the data with the designated resource block at the set power and transmission rate, and if not, the relay node chooses to discard the information which is not forwarded.
As shown in fig. 4, when the station equipment sends the trackside information to the train through the gateway node, the receiving node on the path evaluates the state of the sending node according to the throughput and the time delay performance index, adds the state information of the last node in the forwarded data packet, then forwards the state information of the node and the data to the next node, and finally the train can obtain the state information of all the nodes on the transmission path. When the performance index of a certain node on the path is continuously smaller than the designated threshold, the train will re-perform path selection and resource allocation and avoid selecting the problem node, thereby ensuring the reliability of communication and improving the survivability of the network.
Those of ordinary skill in the art will appreciate that: the drawings are schematic representations of one embodiment only and the flow in the drawings is not necessarily required to practice the invention.
Claims (7)
1. A urban rail ad hoc network resource allocation method based on capacity and time delay optimization is characterized by comprising the following steps of
Step S1, based on the position of each train in a cluster, establishing a transmission path where a plurality of nodes of the train and the train ground are not intersected;
s2, according to a protocol interference model, analyzing the space multiplexing condition of a node time-frequency resource block in a network and the interference condition when nodes on a path share the time-frequency resource block, and calculating the capacity of a receiving node, the capacity of the path and the network capacity in a cluster according to a shannon formula;
step S3, setting the transmission rate of each node in different transmission time slots as the channel capacity of the time slot;
s4, analyzing queuing conditions of the data packets based on the characteristics of periodic communication between the vehicle and the ground, and calculating queuing length, waiting time delay and average transmission time delay of the data packets;
step S5, combining deep reinforcement learning, designing intelligent agents, states, actions and rewarding functions in a model, obtaining an optimal communication path set and a resource allocation scheme of a vehicle and a vehicle ground through a training Q network, and updating a transmission strategy according to the received signal strength RSSI value of a first-hop relay node;
and S6, aiming at the obtained optimal strategy, designing an implementation mechanism to inform nodes in the cluster, and monitoring the state of the nodes by utilizing the capacity and time delay performance indexes of the nodes.
2. The urban rail ad hoc network resource allocation method based on capacity and time delay optimization of claim 1, wherein step S1 specifically comprises:
in order to ensure that the damage of part of nodes in the path does not affect the transmission of other paths and avoid queuing delay of data packets at the relay node, slave sources are respectively built for n trains in the clusterThe m nodes to the destination do not intersect concurrent transmission paths, and the set of the transmission paths in the network is Ω= { Ω 1 ,Ω 2 ,…,Ω n Each train S i The concurrent transmission path set is Ω i ＝{M 1 ,M 2 ,…,M m Node set ofPath M k The node set is asz is the number of relay nodes on the path; the establishment of the disjoint paths of the vehicle and the vehicle-ground nodes needs to meet the following conditions:
where Ω is a set of transmission paths in the network, Ω i And omega e Respectively is train S i And S is e Is provided with a set of concurrent transmission paths,and->Respectively, path sets Ω i And omega e Node set, M k And M q Is a train S i Is>And->For path M k And M q A set of nodes thereon.
3. The method for using the urban rail ad hoc network resource allocation method based on capacity and time delay optimization according to claim 1, wherein,
step S2, specifically comprising:
assuming that the total bandwidth W of the system is divided into a mutually orthogonal sub-channels, each sub-channel has a bandwidth ofTime period T f Dividing into b time slots, each time slot having a length of +>The set of time-frequency resource blocks is k= { (f) n ,t m )|n∈[1,a],m∈[1,b],n,m∈N * And (f), where n ,t m ) For a time-frequency resource block, N and m are subscripts of the sub-channel and the time slot, respectively, N * Is a positive integer set;
according to the protocol interference model, considering the spatial multiplexing of the time-frequency resource blocks, when a certain distance is met between nodes on a path, the resource blocks can be shared without causing interference;
defining resource block occupancy factor for node iThe method comprises the following steps:
if node i occupies resource block (f) n ,t m ) Resource block occupation factor1, otherwise, < >>Is 0;
the spatial multiplexing condition of node i and node k or q is:
d ij ＜R c ,d kj ＞R i or d ij ＜R c ,d qj ＞R i (4)
where k represents a node in the same transmission path set as node i, q represents a node in a different transmission path set from node i, d ij Representing the Euclidean distance, d, between node i and node j kj Represents the Euclidean distance, d, between node k and node j qj Representing the Euclidean distance, R, between node q and node j c Representing the transmission range of the node, R i Representing the interference range of the node.
4. The method for using the urban rail ad hoc network resource allocation method based on capacity and time delay optimization according to claim 3, wherein the interference condition analysis and network capacity calculation when the nodes share the time-frequency resource block in S2 specifically comprise:
(1) Interference condition analysis:
if a certain node i in the transmission path set occupies a resource block to perform data transmission, when a receiving node j is positioned in the interference range of another sending node k and the node k occupies the same resource block to perform transmission, the node k generates interference to the receiving node j; the receiving node j on the path may be interfered by the nodes occupying the same resource block in the same transmission path set and different transmission path sets;
wherein,occupying f for receiving node j n ,t m Signal-to-noise ratio of resource block,/->For the transmission path set Ω i Total interference of other transmitting nodes k to receiving node j +.>Aggregation Ω for other transmission paths e The total interference of the transmitting node q to the receiving node j, P c For the transmission power of the node, N 0 G is the noise power ij Represents the channel gain, g, between node i and node j kj Representing the channel gain, g, between node k and node j qj Representing the channel gain between node q and node j;
considering only free space path loss, the channel gain isGamma is the path loss constant, x i And y i Respectively the abscissa and the ordinate of the node i; let it be assumed that at t m The running of the train can be approximately regarded as uniform linear motion in time slots, and the running speed is upsilon, then t m The position coordinates of the slot trains can be expressed as(x s ,y s ) At t for the train m Position coordinates at the beginning of a slot;
(2) Network capacity calculation:
by using shannon's formula, node j occupies f n ,t m Capacity of resource blockThe calculation is as follows:
wherein ω is f n ,t m The channel bandwidth of the resource block,occupy f for node j n ,t m Signal to noise ratio of resource block;
considering that the node adopts a multi-interface multi-channel technology, the node j is at t m Capacity of time slotThe calculation is as follows:
wherein,occupy f for node j n ,t m Capacity of resource blocks;
path M i During a time period T f Average capacity inThe calculation is as follows:
wherein,for node j at t m Capacity of time slot>For path M i A set of nodes on the table;
the average capacity realized by the transmission path set of n trains in the cluster is calculated as:
wherein,for path M i During a time period T f Average volume in omega i For each source node S i And Ω is a set of transmission paths in the network.
5. The method for using the urban rail ad hoc network resource allocation method based on capacity and time delay optimization according to claim 1, wherein step S4 specifically comprises:
assume that the communication periods of the vehicle and the ground are T s The length of the data packet is h, and in order to meet the time delay requirement of the service, the source node must complete a complete data transmission and receiving process in each communication period; the queuing situation of the data packet is analyzed, and the average transmission delay is calculated as follows:
(1) And (3) queuing condition analysis:
according to the characteristic of periodic communication between the train and the ground, when the train or gateway node receives information sent by a plurality of source nodes, a new data packet is generated after short processing, and the new data packet is used as the source node to send information to a plurality of destination nodes, so that the phenomenon that the data packet is queued in a buffer area of the source node can occur, and therefore, queuing delay exists at the source node; the source node transmits data in the second period after completing data transmission and receiving in one period, so that the phenomenon of queuing of the data packet at the relay node is avoided, and therefore, queuing delay of the data packet at the relay node is avoided; since the node can only send data in the allocated time slot, there is scheduling delay at both the source node and the relay node; in summary, the waiting time delay of the data packet at the source node includes queuing time delay and scheduling time delay, and the waiting time delay at the relay node only has scheduling time delay;
(2) Calculating average transmission delay:
let t of the packet at the previous node a m The time slot is transmitted in a time slot, at t m The starting time of the time slot is the time of day,at t m The end time of the time slot, the transmission time is kappa a The transmission delay of the data packet at node a is +.>h is the packet length, < >>T at node a for a data packet m The sending rate of the time slot reaches the destination node at the time of kappa a +Δ a The delay of generating new data packet after processing at destination node is ignored, and the node is considered to adopt multi-interface multi-channel technology, so the queuing length of new data packet at source node S is ∈ ->Expressed as:
wherein,representing the source node at t m-1 Queue length at the end of a slot, κ a Indicating the moment of transmission of the data packet at node a, < >>At t m Time slot start time,/->Representing the total transmission rate of the nodes in a, a representing at t m The set of all nodes whose time slots are sent to the destination node, is->T representing the packet at the source node s m Transmission rate of time slot,/">Representing the source node at t m Queue length at the end of the time slot;
because the data transmission of the source node needs to occupy a plurality of time slots, the time slot ascending sequence set of the source node is defined asWherein->n represents the number of occupied time slots, x represents the time slot sequence number +.>Is->Time slot start time,/->Is->The end time of the time slot; according to different generation moments of the data packet at the source node, the waiting time delay at the source node is divided into the following two cases:
case one: when (when)z=0 or->z e N, the first packet in the current queue will be +.>Sending at the moment;
and a second case: when (when)z epsilon N, the first data packet of the current queue is being sent;
wherein Y represents the transmission delay of the preceding data packet in the transmission time slot of the data packet,representing the queuing length of the data packet at the source node s, is->Indicating that the data packet is +.>Transmission rate of time slot,/">Indicating that the data packet is +.>Transmission rate of time slot,/">Is->Time slot start time, κ a Delta for the transmission time of the data packet at node a a For the transmission delay of the data packet at node a, < >>Indicating that the data packet is +.>The transmission rate of the time slot; if 0.ltoreq.Y < τ, the packet will be in +.>Time slot transmission;
combining the two cases, the waiting time delay q of the data packet at the source node s The calculation is as follows:
wherein,is->Time slot start time, κ a Delta for the transmission time of the data packet at node a a The transmission delay of the data packet at the node a is represented by Y, wherein Y represents the transmission delay of the data packet in front of the transmission time slot of the data packet;
single-hop delay d i,j The calculation is as follows:
wherein, kappa j For the moment of sending the data packet at node j, κ i T is the sending time of the data packet at the node i f Is the length of the time period;
path M i End-to-end delay of (c)The calculation is as follows:
wherein q s D, waiting time delay of data packet at source node i,j In the form of a one-hop delay,for path M i A set of nodes on the upper-level node,node for data packet>Transmission delay at the location;
average transmission delay D of n trains in the cluster Ω The calculation is as follows:
wherein m is i Omega for the number of concurrent transmission paths per train i For each source node S i Is a set of transmission paths in the network,for path M i End-to-end delay of (c).
6. The method for using the urban rail ad hoc network resource allocation method based on capacity and time delay optimization according to claim 1, wherein,
the step S5 specifically comprises the following steps:
firstly, designing an intelligent body, a state, actions and rewarding functions in a model, wherein each train is regarded as an intelligent body, the intelligent body searches an optimal communication path set and a resource allocation scheme of the train and the train place according to the current network state, the state is an abstraction of the urban rail self-organizing network environment, the actions comprise two aspects of path selection and resource block allocation, and rewarding consists of two parts of network capacity and transmission delay; the status, action and reward functions are each specifically represented as follows:
s t ＝{U t ,Q t ,H t ,G t ,V t } (20)
a t ＝{Ω t ,K t } (21)
r t ＝λ r R Ω -λ d D Ω (22)
in the formula (20), U t Represents the position of the train, Q t Representing a network topology including locations of nodes and connectivity relationships between nodes, H t Representing available node and resource block information, G t Channel gain information representing first hop, V t Spatial multiplexing information representing a network;
in the formula (21), Ω t Represents the node disjoint transmission path set selected by each vehicle at the moment t, K t A set of resource blocks representing the allocation of each vehicle to nodes on the transmission path;
in the formula (22), R Ω And D Ω Respectively representing network capacity and average delay of all concurrent transmission paths lambda r And lambda (lambda) d Respectively, the weights of the two.
7. The method for using the urban rail ad hoc network resource allocation method based on capacity and time delay optimization according to claim 6, wherein,
in step S5, an optimal communication path set and a resource allocation scheme of the vehicle and the vehicle ground are obtained through the training Q network, and the specific steps are as follows:
(1) Initializing a network communication environment, evaluating a parameter theta of a network and a parameter theta' of a target network, and inputting relevant training parameters;
(2) The train interacts with the environment according to the state s t Executing action a t Obtain rewards r t And enter the next state s t+1 And obtain tuple < s t ,a t ,r t ,s t+1 Storing the data in an experience memory pool D;
(3) After enough tuples exist in the experience pool, randomly extracting the mini-batch tuple from the experience pool, respectively generating an evaluation Q value and a target Q value through an evaluation network and a target network, then calculating a loss function between the evaluation Q value and the target Q value, and updating network parameters of the evaluation network by adopting gradient descent;
(4) Assigning the parameter values of the evaluation network to the target network at intervals, and updating the parameters of the target network;
(5) Judging whether the round is ended; if not, jumping to the step (2); if the method is finished, outputting a Q network, an optimal communication path set and a resource allocation scheme;
(6) Judging the received signal strength RSSI value of the first-hop relay node; if the train is smaller than the given threshold delta and the train is not in the data transmission stage, jumping to the step (1); if the threshold delta is greater than or equal to the given threshold delta, the original scheme is preserved.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310986895.XA CN117098192B (en) | 2023-08-07 | 2023-08-07 | Urban rail ad hoc network resource allocation method based on capacity and time delay optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310986895.XA CN117098192B (en) | 2023-08-07 | 2023-08-07 | Urban rail ad hoc network resource allocation method based on capacity and time delay optimization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117098192A true CN117098192A (en) | 2023-11-21 |
| CN117098192B CN117098192B (en) | 2024-04-26 |
Family
ID=88772880
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310986895.XA Active CN117098192B (en) | 2023-08-07 | 2023-08-07 | Urban rail ad hoc network resource allocation method based on capacity and time delay optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117098192B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118118409A (en) * | 2024-04-30 | 2024-05-31 | 国网浙江省电力有限公司信息通信分公司 | Metropolitan area optical transport network node position and user allocation planning method, system and equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113543271A (en) * | 2021-06-08 | 2021-10-22 | 西安交通大学 | Effective capacity-oriented resource allocation method and system |
| CN114615730A (en) * | 2022-03-22 | 2022-06-10 | 西安电子科技大学 | Content coverage oriented power distribution method for backhaul limited dense wireless network |
| CN115866787A (en) * | 2022-10-28 | 2023-03-28 | 国网吉林省电力有限公司 | Network resource allocation method integrating terminal direct transmission communication and multi-access edge calculation |
| CN116405904A (en) * | 2023-04-06 | 2023-07-07 | 上海大学 | TACS network resource allocation method based on deep reinforcement learning |
-
2023
- 2023-08-07 CN CN202310986895.XA patent/CN117098192B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113543271A (en) * | 2021-06-08 | 2021-10-22 | 西安交通大学 | Effective capacity-oriented resource allocation method and system |
| CN114615730A (en) * | 2022-03-22 | 2022-06-10 | 西安电子科技大学 | Content coverage oriented power distribution method for backhaul limited dense wireless network |
| CN115866787A (en) * | 2022-10-28 | 2023-03-28 | 国网吉林省电力有限公司 | Network resource allocation method integrating terminal direct transmission communication and multi-access edge calculation |
| CN116405904A (en) * | 2023-04-06 | 2023-07-07 | 上海大学 | TACS network resource allocation method based on deep reinforcement learning |
Non-Patent Citations (2)
| Title |
|---|
| MOHAMMED LAROUI等: "Scalable and Cost Efficient Resource Allocation Algorithms Using Deep Reinforcement Learning", IEEE * |
| 张青苗等: "智能轨道交通中无线通信技术应用与展望", 无线电通信技术, vol. 48, no. 5 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118118409A (en) * | 2024-04-30 | 2024-05-31 | 国网浙江省电力有限公司信息通信分公司 | Metropolitan area optical transport network node position and user allocation planning method, system and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117098192B (en) | 2024-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104994583A (en) | Multi-channel MAC protocol method based on cluster mechanism in vehicular Ad hoc network | |
| US7756096B2 (en) | Transmission resource reservation management in wireless network | |
| CN101626575B (en) | Method, device and system for performing frequency planning in wireless Mesh returning network | |
| CN117098192B (en) | Urban rail ad hoc network resource allocation method based on capacity and time delay optimization | |
| Mammu et al. | Cluster-based MAC in VANETs for safety applications | |
| CN101223794A (en) | Method and system of wireless communication between devices | |
| Lin et al. | Integrated message dissemination and traffic regulation for autonomous VANETs | |
| Zhang et al. | EVC-TDMA: An enhanced TDMA based cooperative MAC protocol for vehicular networks | |
| Mammu et al. | Direction aware cluster-based multi channel MAC protocol for vehicular ad hoc networks | |
| Gupta et al. | A D2D multicast network architecture for vehicular communications | |
| CN105554107B (en) | Optimal cooperation point selection method based on TDMA agreements in high dynamic vehicular ad hoc network network | |
| Li et al. | MAC-SCC: Medium access control with a separate control channel for multihop wireless networks | |
| Liu et al. | An elastic-segment-based V2V/V2I cooperative strategy for throughput enhancement | |
| Li et al. | ASTSMAC: Application suitable time-slot sharing MAC protocol for vehicular ad hoc networks | |
| Nasrallah et al. | Distributed time synchronization mechanism for large-scale vehicular networks | |
| CN116528313A (en) | Unmanned aerial vehicle low-energy-consumption rapid routing method for task collaboration | |
| Diab et al. | Enhanced multi-channel MAC protocol for multi-hop wireless sensor networks | |
| CN107105388A (en) | A kind of vehicle-mounted network route method of cross-layer based on link transmission ability | |
| CN108055667B (en) | Cognitive vehicle network combined routing and resource allocation method | |
| CN107592213B (en) | Infrastructure-based vehicle networking large-scale heterogeneous network capacity expansion rate model construction method | |
| CN113347606B (en) | V2I/V2V network cooperative transmission method and system based on elastic partition | |
| Gao et al. | Cluster-based interference-free MAC protocol with load aware in software defined VANET | |
| CN112672310B (en) | Orthogonal frequency division and time division combined service channel dividing and distributing method | |
| Li et al. | A MAC protocol for multi-channel multi-interface wireless mesh network using hybrid channel assignment scheme | |
| Khan et al. | NA-SMT: A Network-Assisted Service Message Transmission Protocol for Reliable IoV Communications |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |