CN109089273B - Relay selection method based on state transition probability in Ad-Hoc network - Google Patents
Relay selection method based on state transition probability in Ad-Hoc network Download PDFInfo
- Publication number
- CN109089273B CN109089273B CN201811122242.2A CN201811122242A CN109089273B CN 109089273 B CN109089273 B CN 109089273B CN 201811122242 A CN201811122242 A CN 201811122242A CN 109089273 B CN109089273 B CN 109089273B
- Authority
- CN
- China
- Prior art keywords
- state
- probability
- relay
- node
- neighbor
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- 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
- H04W40/22—Communication route or path selection, e.g. power-based or shortest path routing using selective relaying for reaching a BTS [Base Transceiver Station] or an access point
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
The invention provides a relay selection method based on state transition probability in an Ad-Hoc network, and innovatively provides a relay selection strategy based on the state transition probability. And calculating the possible arrival probability of the neighbor node by combining the node geographic information and the channel environment information, and selecting the node with the larger probability for data transmission, thereby improving the data transmission performance. Meanwhile, in order to reduce the system operation complexity and save the system energy, a transformed metropolis selection criterion is adopted to simulate greedy search of annealing to gradually remove the transmission states with small probability, so that the operation space is greatly reduced. The simulation gives the influence of the algorithm parameters on the operation speed and the success rate. Meanwhile, the algorithm is shown to have better control on the increase of the system energy consumption and the failure probability when the network topology changes.
Description
Technical Field
The invention relates to the technical field of information, in particular to a relay selection method based on state transition probability in an Ad-Hoc network.
Background
The Ad-Hoc network is a distributed wireless network, does not depend on any fixed communication infrastructure, can be rapidly networked through a self-organizing network technology in emergency, has strong survivability, and has good performance in the fields of modern digital wars, earthquake and flood, field exploration and the like. However, Ad-Hoc networks also pose a number of serious problems as the number of hops increases. Firstly, the network topology and the transmission route are dynamically changeable due to the mobile characteristics of the network nodes, thereby resulting in complex operation and poor stability of the system. Meanwhile, the Ad-Hoc network nodes have limited energy, cannot be supplemented or replaced in time in many application occasions, and the service life of the network is seriously threatened[1-2]. Therefore, the problem of energy consumption of Ad-Hoc network has become a restriction to its developmentThe main bottleneck. How to reduce the energy consumption of the nodes and prolong the service life of the network, students conduct deep research from different angles. In practical application, selecting a proper relay for the multi-hop Ad-Hoc network to perform next-hop transmission is an effective means for saving energy and shortening time delay. Conventionally, a relay selection method based on location information has been favored because the coordinates of nodes can be randomly generated in a limited scene area during simulation, so that the distance between the nodes can be calculated. Literature reference[3]The optimal relay node is selected by deducing a curve equation of the envelope of the optimal relay node area, dividing the area into concentric rings with the same error rate performance and combining the positions of the relay node and the destination node. The method fully utilizes the space diversity characteristic to improve the performance of the system error rate. Literature reference[4]In order to prolong the life cycle of the network, a cost formula consisting of node transmitting energy consumption, receiving energy consumption and residual energy is constructed based on node position information, so that optimal relay routing selection under the balance of three types of energy is realized. Literature reference[5-7]The relay is selected by mainly considering factors such as the node position and the distance from the target node. Therefore, the methods consider the influences of the position, the transmission distance and the like, but ignore the factors of important path interference and the like. Later, some scholars focused on the influence of channel conditions. Literature reference[8-10]A relay selection method based on average signal-to-noise ratio is researched, namely, a node with the average signal-to-noise ratio larger than a specified threshold on each link of a transmitting node and a relay node is selected as a relay, but the prior knowledge of a channel is required to be known, and the average signal-to-noise ratio of all relay nodes is comprehensively estimated, so that the extra cost is large. Efficiency may be lower under the influence of dynamically moving network topology and battery life. Literature reference[11-13]Instantaneous channel state information based on local measurements is employed to select a suitable relay. An optimal relay transmission path is decided from a plurality of selectable relay nodes, and the optimal relay transmission path does not need any prior knowledge including network topology information and geographical position information and does not need to be communicated among the relay nodes. But whether the selection of the optimal relay transmission path is successful or not depends on the node to the instant system of the current wireless channelThe method is an opportunistic end-to-end optimal relay selection method and has instability characteristics. In recent years, relay node selection algorithm aiming at energy saving has become a research hotspot in wireless networks[14-18]。
Reference documents:
[1] study of energy-based routing protocols in wireless ad hoc networks [ D ]. university of north-and-river science 2015.
Gao H Y,Research on the routing protocols based energy in wireless ad hoc networks[D].Hebei University Of Science and Technology,2015
[2]Wang Y L,Mei S,Wei Y F,et al.Improved ant colony-based multi-constrained QoS energy-saving routing and throughput optimization in wireless Ad-hoc networks[J].The Journal of China Universities of Posts and Telecommunications,2014,21(1):43-53.
[3] The dynamic geography cooperative routing algorithm [ J ] selected by the cooperative relay node of the Sun Lijuan, Zhang Ying, computer engineering and design, 2017,38(2):281 and 286.
Sun L J,Zhang G L.Dynamic geographic cooperative routing algorithm for cooperative relay node selection[J].Computer Engineer and Design,2017,38(2):281-286.
[4] Zhou Lei, Su hong, Tang Hao, etc. wireless network cooperation routing algorithm [ J ] based on location information electronic measurement and instrument report 2015(5) 708-.
Zhou L,Su H,Tang H,etc.Wireless network cooperative routing algorithm based on location information[J].Journal of electronic measurement and instrument,2015(5):708-716.
[5] Sunweisheng, Yaojinsong, Tang Happy. recent Relay cooperative Algorithm research [ C ]. Zhejiang institute of electronics 2014 academic annual meeting collection 2014.
Sun W S,Yao J S,Tang X L.The nearest relay collaboration algorithms research[C].Zhejiang Institute of Electronics 2014academic annual conference papers,2014.
[6]Nielsen J J,Olsen R L,Madsen T K,et al.Optimized Policies for Improving Fairness of Location-Based Relay Selection[C]//2013IEEE 77th Vehicular Technology Conference,2013,14(2382):1-5.
[7]
B.Distance based selection for user-relaying in OFDMA-based wireless networks[C]//Signal Processing and Communication Application Conference.IEEE,2016:989-992.
[8] Chen Jiu, Ruixian Wei SR threshold Relay selection Algorithm research [ J ] computer engineering and application, 2014,50(7): 221-.
Chen Q,Rui X Y.Research on the relay selection algorithm based on SR threshold[J].Computer Engineering and Application,2014,50(7):221-224.
[9] Shohain, huyue, yankun, etc. relay selection and power distribution for multi-source-multi-relay cooperative communications [ J ]. proceedings of beijing post and telecommunications university, 2017,40(2):73-78.
Xiao H L,Hu Y,Yan K,etc.Relay selection and power allocation in multi source-multi relay cooperative communication[J].Journal of Beijing University of Posts and Telecommunications,2017,40(2):73-78.
[10] Plum wave relay selection strategy and power optimization allocation algorithm study [ D ] Chongqing university, 2016.
Li Y B.Research on the relay selection strategy and power optimal allocation algorithm[D].Chongqing University,2016.
[11] Yanli monarch, Cao Zhang Hua, Zhang Shi, etc. Relay selection scheme of single eavesdropping double-hop cooperative network and performance analysis thereof [ J ] Chongqing postand telecommunication university journal (Nature science edition), 2016,28(5):648 657.
Yang L J,Cao Z H,Zhang S B,etc.Relay selection schemes for dual-hop cooperative networks with an eavesdropper and their performance analysis[J].Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition),2016,28(5):648-657.
[12] Zhaoyeli, Guo li, Zhu Shi Liang, et al. design of a relay node selection scheme in cooperative communications [ J ] computer applications, 2015,35(1):1-4.
Zhao Y L,Guo L,Zhu Z L,etc.Relay node selection scheme in cooperative communication[J].Journal of Computer Application,2015,35(1):1-4.
[13] The relay selection algorithm and performance analysis [ J ] based on the predicted signal-to-noise ratio, 2016,37(12): 3196-.
Qin C L,Xiao K.Relay selection algorithm based on predictive signal-to-noise ratio and performance analysis[J].Computer Engineering and Design,2016,37(12):3196-3200.
[14]Wang L J,Han T.The Relay Selection Algorithm Based on Minimizing the User Terminal Energy Consumption[C]//International Conference on Wireless Communication and Sensor Network.IEEE,2014:270-275.
[15] Qixiajust, dong hai jun. the wireless sensor network based on dormancy opportunistically routes the relay node to select [ J ] signal processing, 2017(s1):58-64.
Qi X G,Dong H J.Opportunistic routing relay node selection based on hibernation in wireless sensor network[J].Journal of Signal Processing,2017(s1):58-64.
[16]Luo J,Hu J,Wu D,et al.Opportunistic Routing Algorithm for Relay Node Selection in Wireless Sensor Networks[J].IEEE Transactions on Industrial Informatics,2017,11(1):112-121.
[17] Liujie, Chenniu, Ningchun, et al Relay selection strategies in energy-harvesting full-duplex Relay networks research [ J ] Signal processing, 2017,33(1): 116-.
Liu J Q,Chen J,Ren G C,etc.Research on Relay Selection Strategy of Energy Harvesting Full-Duplex Relay Network[J].Journal of Signal Processing,2017,33(1):116-125.
[18] An energy-efficient joint relay selection and power distribution algorithm [ J ] in Tngwen, Yanglin, Lihaxiang and energy collection bidirectional relay network, 2017,45(5):1124 and 1129.
Ding C W,Yang L,Li G X.Energy efficient relay selection and power allocation for energy harvesting two-way relay network[J],Acta Electronica Sinica,2017,45(5):1124-1129.
[19]Rossi M,Tapparello C,Tomasin S.On Optimal Cooperator Selection Policies for Multi-Hop Ad Hoc Networks[J],IEEE Transactions on Wireless Communications,2011,10(2):506-518.
[20]Yu T,Lin-Hua M A,Hong T,et al.Power Allocation and Routing Algorithm in Multihop Cooperative Relaying Networks[J].Transactions of Beijing Institute of Technology,2013,33(12):1247-1252.
[21] Tangjiashan, queuing theory and its application [ M ], scientific Press, 2016.06
Tang J S,Queuing theory and its application[M],Science Press,2016.06
[22] Monte Carlo full waveform analysis method for simulated tempered Markov chain [ J ] Physics, 2014,63(16): 164205-.
Yin W Y,He W J,Gu G H,etc.A full waveform analysis approach based on the simulated tempering Markov chain Monte Carlo method[J].Acta Physica Sinica,2014,63(16):164205-164215.
Disclosure of Invention
The invention aims to solve the defects in the prior art, comprehensively considers the problems of mutual interference among channels, transmission distance between nodes, energy loss and the like under the actual condition, provides a relay selection algorithm based on network state transition probability and aims to better realize the transmission performance of a multi-hop network.
A relay selection method based on state transition probability in an Ad-Hoc network comprises the following steps:
the information source s firstly sends a message to the optimal neighbor subset, the neighbor subsets respectively continue to forward to the respective optimal neighbor subsets after receiving the message, and so on until the message is finally merged to reach the information destination t;
in an Ad-Hoc network, selecting an optimal relay from a plurality of relay nodes as the optimal neighbor subset to send a message by constraint of network state transition probability, wherein the calculation method of the network state transition probability comprises the following steps:
let the channel gain between any two node links be hi,jWith a transmission power of piConsidering the mutual interference of adjacent links, the signal-to-noise ratio can be expressed as γ ═ pihi,j)/(σi,j 2+∑k∈ψ,k≠i(pkhk,j))，σ2For additive white gaussian noise power, the probability density function PDF is expressed as:
then:
therefore, the calculation formula of the outage probability in the Rayleigh fading channel model DF mode can be derived as follows:
setting a node set in a current state x as A (x), enabling a to belong to A (x), enabling a terminal state to be t, enabling x to be not equal to t, deriving from the definition of interrupt probability, and enabling the successful transmission probability from a cooperative node subset a of the current hop to any node n of the next hop to be 1-pout(n, a), if θ is:
then, according to the nature of the Markov chain, the transition probability of the system from state x to state y can be defined as:
further, according to the method, when the optimal relay is selected from the plurality of relay nodes as the optimal neighbor subset to send the message, unnecessary neighbor sets and behavior sets with low contribution rate are reduced, and the reduction method comprises the following steps:
calculating the transition probability p of each state according to the formula (7)xy(a) And (3) clipping the neighbor nodes with lower transmission probability and leaving the paths with higher transmission probability, thereby optimizing the state space used by the algorithm for operation, and the following steps are carried out:
where S represents the set of all states of the system, the maximum pruning probability for all pruned neighbor nodes during the transition from state x to state y of the system can be defined as:
then the maximum difference in energy consumption values before and after the curtailment is:
wherein the content of the first and second substances,
is an overhead estimate for state x. During the execution of the algorithm, when the delta (x) is 0, the M (x) is 0, namely no node change exists between the states x and y, pruning is not carried out, iterative pruning is carried out when the delta (x) is greater than 0,
the upper bound of state reduction is obtained by the (16) type of transformation:
in order to make the number of nodes in N' (x) appropriately large, an upper bound M of the formula (17) is determined_Provided that M (x) does not exceed M_All can be used.
Further, as described above, the method removes the transmission states with small probabilities step by using greedy search, and the greedy search method is:
step 1: when k is 0, the current solution of the Ad-Hoc network is S (0) ═ x, and the following steps are performed at the temperature T:
step 2: generating a neighborhood subset based on the state x of the current solution S (k)
Randomly obtaining a new state y from N (S (k)), and calculating the difference of the upper bound of the pruning and the energy consumed by reaching the new state y: Δ C ═ M_-C(y)
And 3, step 3: if the delta C is less than or equal to 0, the energy consumption exceeds the upper limit, and the direct rejection is performed; if Δ C > 0, a random number rand ∈ random [0,1) is generated, if the probability of reception is greater than
Then receive y is the next current solution. If a new state y is received, S (k +1) is made y, otherwise S (k +1) is made x.
And 4, step 4: and (5) executing the next iteration when k is k +1, judging whether the algorithm should be ended according to a given convergence criterion, if so, turning to the step 5, and otherwise, turning to the step 2.
And 5, step 5: and returning.
Has the advantages that:
in a multi-hop Ad-Hoc network topology structure in practical application, the number of relay nodes of each hop capable of cooperatively transmitting messages is changed, the invention provides that a proper neighbor node subset and optimal activities are selected from an available neighbor node set to serve as a relay transmission target of the next hop, and meanwhile, the energy consumption and the time delay of the network are considered. As the network dynamically changes, its state space is also constantly changing and increasing. When the state space is increased, a greedy search and simulated annealing algorithm is simulated, and nodes and states which have a low probability of reaching in the next transmission link are removed, so that the state space for operation is simplified, and the purpose of optimizing the relay transmission performance is achieved.
Drawings
FIG. 1 is a multi-hop Ad-Hoc network channel model;
FIG. 2 is a model of time delay and queuing at a relay node along a transmission route;
FIG. 3 is a relationship between scene size and system energy consumption;
fig. 4 is a graph of scene size versus number of cooperating nodes.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention are described below clearly and completely, and it is obvious that the described embodiments are some, not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the Ad-Hoc network, the optimal relay is selected from a plurality of relay nodes through the constraint of the network state transition probability, therefore, the invention firstly provides a method for calculating the interruption probability, and then calculates the network state transition probability through the interruption probability and determines the optimal relay. In order to reduce the complexity of relay selection, an effective operation method is provided, namely the operation speed is improved by limiting the number of relay nodes participating in calculation, and meanwhile, the quality of the relay nodes participating in calculation is ensured so as to optimize the relay transmission performance.
Consider a multi-hop Ad-Hoc network with a total number of nodes of ψ, one node having only one antenna interface. In the conventional relay cooperative transmission process, the basic process is as follows: information is sent from a source node s to a target node t, the source node s firstly broadcasts a message to all neighbor nodes thereof, and all the nodes receiving the message forward the message to the next hop. And in the same way, cooperative forwarding is continuously forwarded until the terminal node t is reached. It can be seen that this conventional broadcasting method will generate a large redundancy and overhead. The invention optimizes on the basis, obtains the 'optimal subset' of the neighbor nodes by adopting a proper pruning strategy as a forwarding target, and forwards the optimal subset through a Decode and Forward (DF) relay protocol. Assuming that the number of "optimal subset" nodes participating in cooperative transmission is a, a transmission channel model from the nodes in the set a to any relay receiving node under the conditions of path loss and fading is shown in fig. 1. In fig. 1, a source s first sends a message to its optimal neighbor subset, and the neighbor subsets continue forwarding to their respective optimal neighbor subsets after receiving the message, and so on until the last message reaches a destination t after being merged. In each hop of transcoding forwarding, the optimal neighbor subset always selects nodes closer to the sink t as relays, which helps to complete data communication faster.
Probability of state transition
The invention assumes that the network environment is Ad-Hoc multi-hop network, and transmits messages in a Rayleigh channel by decoding and forwarding relay protocol, and the transmitting power of the wireless nodes is the same. Due to the difference of the quality and loss of each link, various phenomena such as delay or flash may occur in the network. Therefore, in order to facilitate the selection of an appropriate network state, the present invention defines transition probabilities between network states to describe dynamic changes of the Ad-Hoc network.
Assuming that a transmitting node transmits at a unit bandwidth rate R, let the capacity of a wireless channel be C, and the signal-to-noise ratio of the channel be γ, then when the channel capacity is:
C＝log2(1+γ) (1)
an interrupt event occurs when the channel capacity falls below rate R[19]. Let the channel gain between any two node links in FIG. 1 be hi,jWith a transmission power of piConsidering the mutual interference of adjacent links, the SNR can be expressed
Is gamma ═ pihi,j)/(σi,j 2+∑k∈ψ,k≠i(pkhk,j)) (2)
σ2For additive white gaussian noise power, the Probability Density Function (PDF) can be expressed as:
then:
therefore, the calculation formula of the outage probability in the Rayleigh fading channel model DF mode can be derived as follows:
setting a node set in a current state x as A (x), enabling a to belong to A (x), enabling a terminal state to be t, enabling x to be not equal to t, deriving from the definition of interrupt probability, and enabling the successful transmission probability from a cooperative node subset a of the current hop to any node n of the next hop to be 1-pout(n, a), if θ is:
then, according to the nature of the Markov chain, the transition probability of the system from state x to state y can be defined as:
if x and y both belong to the terminal state t, i.e. when x and y are equal to t, it is obvious that there is pxy(a)＝ptt(a) 1. In the actual network transmission process, the transition probability depends on the relative position of the transmission node and the transmission channel and other comprehensive factors.
Computation of network reward values
The energy consumption of the relay network node mainly comprises three parts: transmit, receive, and idle. Typically, transmit power consumption is a major consideration, while power consumption at reception and idle is often considered constant or negligible[20]. In practice, the transmission power consumption depends on the number of cooperating nodes and the environmental parameters of the transmitted message. The invention assumes that the energy is normalized and the sending energy consumption is set as VEDelay consumption of VDA is the set of nodes which analyze the message and cooperate, m is the noncoordination of other analyzed messagesMake node set, at this time, the energy consumption VECan be expressed as:
VE＝a+ωm(8)
the parameter ω is not less than 0, which is an overhead coefficient when the extra node analyzes the message, and the formula (8) shows that if the number of nodes transmitted simultaneously in a certain hop is more, the total transmission energy consumption is larger, otherwise, the transmission power of the system can be saved.
Delay consumption VDCan be considered to consist of two parts. The delay on the transmission route and the delay of the queuing process at the relay node. In fig. 2, this is the queuing model.
In fig. 2, data to be transmitted randomly comes to a service organization, and waits for service according to a certain queuing rule. In the relay system, the relay nodes are single antennas, the queuing system is a single-channel single service, the data arrival meets Poisson distribution, the service time obeys exponential distribution, and the waiting time of the data packets of the relay transmission system on each relay node is distributed as[21]：
Fq(w)＝1-ρe-(μ-λ)w (9)
Average wait period of
Where μ is the sending rate, λ is the arrival rate, and ρ is the system utilization, i.e.
Thus, the total queuing processing time is:
the invention sets the time delay on the transmission path as a constant qtAnd then:
the network report value of the system state x to the state y via a relay nodes is defined as follows:
assuming that the current state x reaches the state y through all possible paths, probability weighting is performed on all possible paths, and the equation that takes the least time to find the end point can be defined as:
J*(x)＝mina∈A[∑y∈N(x)pxy(a)(V(x,a,y)+ξJ(y))],ξ∈[0,1] (13)
relay selection optimization strategy
Complexity reduction technique
The invention reduces the complexity of the system by adopting a strategy of pruning the network state space. All nodes in the state x are known as A (x), the neighbor set is known as N (x), and unnecessary and low-contribution-rate neighbor sets and behavior sets are reduced, so that new sets A '(x) and N' (x) can be obtained. Obviously having a reduced set of nodes
Neighbor set
In the actual complexity reduction process, the transition probability p of each state is calculated according to the formula (7)xy(a) And cutting off the neighbor nodes with lower transmission probability and leaving the paths with higher transmission probability, thereby optimizing the state space used by the algorithm for operation. Thus, the formula (13) becomes:
where S represents the set of all states of the system. The maximum pruning probability of all pruned neighbor nodes during the transition of the system from state x to state y can be defined as:
then, in combination of equations (13) and (15), the maximum difference between the energy consumption values before and after the reduction is:
wherein the content of the first and second substances,
is an overhead estimate for state x. During the execution of the algorithm, when Δ (x) > 0, m (x) is 0, that is, no node change exists between states x and y, pruning is not performed, and iterative pruning is performed when Δ (x) > 0.
The upper bound of state reduction is obtained by the (16) type of transformation:
(15) the smaller M (x), the lower the transition probability pxyThe smaller the sum of the probabilities of all the nodes except N' (x) obtained, the higher the probability pxyHas been selected into N' (x). Therefore, in order to appropriately increase the number of nodes in N' (x), the upper bound M of equation (17) is determined_Provided that M (x) does not exceed M_All can be used.
The pruning Algorithm is shown as Algorithm 1.
The pruning strategy realizes simplified operation on a large number of state spaces, namely M (x) obtained by each iterative operation only exceeds an upper limit M_The switched state y is eliminated, otherwise, the switched state y is accepted, and the pruned neighbor subset N '(x) and the active subset a' (x) are obtained. In the scheme, the upper bound value directly influences the number of pruning, and when the upper bound value is too large or is close to the sum of the probabilities of all neighbor nodes in the state x, the relay cooperation subset is rare, and the failure probability is increased; when in useIf the upper bound value is too small, the pruning algorithm will be substantially ineffective. Therefore, in order to reduce the system complexity under the premise of ensuring the success rate, the invention refers to the simulated annealing algorithm (SA) for further optimization[22]. Through the modification and application of Metropolis criterion, the method is proposed that M (x) < M_In the acceptable state intervals, the conversion state is received with a certain probability, so that the state space is further optimized, the system redundancy is reduced, the convergence speed is improved, and the system energy is saved.
Optimization algorithm design
Considering the difference value of probability statistics of the node of the remaining neighbor after the reduction upper bound of the multi-hop relay Ad-Hoc network state and the conversion into the new state, if the reduction upper bound is less than or equal to the probability total of the node of the remaining neighbor after the network is converted into the new state, namely the difference value delta C of the two is less than 0, the state is not accepted; when Δ C ≧ 0, it is accepted with a certain probability. Since the smaller Δ C, the greater the likelihood of acceptance in the traditional Metropolis guidelines. In the relay transmission network studied by the invention, the larger the Δ C, the greater the probability of acceptance should be. Therefore, the present invention utilizes the transformation rule of Metropolis to make df equal to M_-MXThen the transformed acceptance probability is:
wherein q represents the acceptance probability, x, y represents the system state, TkRepresenting the temperature value at time k. The transformed Metropolis criteria are performed as follows:
step 1: when k is 0, the current solution of the Ad-Hoc network is S (0) ═ x, and the following steps are performed at the temperature T:
step 2: generating a neighborhood subset based on the state x of the current solution S (k)
Randomly obtaining a new state y from N (S (k)), and calculating the difference of the upper bound of the pruning and the energy consumed by reaching the new state y: Δ C ═ M_-C(y)
And 3, step 3: if the delta C is less than or equal to 0, the energy consumption exceeds the upper limit, and the direct rejection is performed; if Δ C > 0, a random number rand ∈ random [0,1) is generated, if the probability of reception is greater than
Then receive y is the next current solution. If a new state y is received, S (k +1) is made y, otherwise S (k +1) is made x.
And 4, step 4: and (5) executing the next iteration when k is k +1, judging whether the algorithm should be ended according to a given convergence criterion, if so, turning to the step 5, and otherwise, turning to the step 2.
And 5, step 5: and returning.
Based on this, the present invention proposes isa (evolved unified modeling) Algorithm based on Metropolis transformation criteria as shown in algorithmm 2. The ISA (s (a)) aims to calculate whether the next state is accepted, and its working process is as follows: setting an initial value init _ T, if M (S (a)i) Does not exceed the upper bound, the reception probability q calculated by the equation (19)k(x, y) is accepted at 1-qk(x, y) rejecting; the part exceeding the upper bound is directly rejected. And changing the value of T in each iteration process, setting the upper boundary of T, and changing T into init _ T multiplied by beta when T increases according to a certain slope and exceeds the initial set terminal _ T. When the next iteration comes, it will continue to grow with the previous slope until the terminal _ T is reached again. By repeatedly transforming the value of T, T will approach to terminal _ T when the iteration number is increased through mathematical analysis.
The parenchyma of T is subjected to a process of repeated progressive increase, and when T is increased, q is expressed by the formula (19)k(x, y) will be smaller, i.e. the culled branches will be gradually larger. When T becomes terminal _ T, it degrades into a greedy algorithm that seeks smaller branches.
Numerical simulation
Setting a network scene as a square area, wherein a starting point s and an end point t are respectively positioned at two ends of the scene, and coordinates of other nodes are randomly generated and uniformly distributed in the network scene of 150m × 150 m. In the simulation process, the initial values of the relevant parameters are set as n ═ 10, ψ ═ 15, ω ═ 0.5, μ ═ 0.9, λ ═ 0.4, ξ ═ 0.8, initT0＝.01,termianl T1＝0,β1.＝2。
Considering the global topology of the Ad-Hoc network as an undirected complete graph, when the number of initial nodes of the network is n, the network has n (n-1)/2 edges. Given n 10, the resulting network can vary in state up to several trillion times. In order to illustrate the effectiveness of the ISA algorithm in reducing the state space, the present invention verifies that the average number of access state spaces calculated by 1000 iterations of the algorithm and the average failure probability of the algorithm are shown in table 1 under different values of Δ. In table 1, the number of actual access state spaces represents the computational complexity or the computational speed of the algorithm, and the average failure probability represents the probability that the algorithm of the present invention cannot find the optimal relay. As can be seen from table 1, the value of Δ directly affects two important indexes, i.e., the complexity of system operation and the failure probability of the algorithm. In the normal case: the larger the value of delta is, the smaller the number of the state spaces actually accessed is, and the failure probability of the algorithm is continuously increased; the smaller the value of delta, the more the number of the state spaces actually accessed, and the more the failure probability of the algorithm is reduced. The overall situation is: as Δ decreases, the number of access states tends to increase and the probability of failure tends to decrease, because as equation (17) indicates that Δ increases, the upper bound on the reduction complexity increases, and thus the more state space is culled and the less state space remains for actual access. However, the probability of failure increases inversely when Δ < 1, indicating that Δ is too small, which can dramatically increase the number of states and instead cause some decisions to fail. Therefore, in the actual network design, a trade-off needs to be made in terms of reducing the system operation complexity and reducing the algorithm failure probability according to the requirement. In order to balance the relationship between the two, the invention takes Δ equal to 1 to perform the following simulation.
TABLE 1 number of Access State spaces and failure rates
Tab.1Number of access state spaces and failure rate
In order to verify various influences on system performance caused by mobility characteristics of a multi-hop Ad-Hoc network, fig. 3 shows a change situation caused by system energy consumption when a node moves. Due to the change of the scene size, the distance between the nodes is changed, and under the condition that the total number of the nodes is not changed, the farther the distance between the source node and the target node is, the larger the total energy consumption of the system is. As can be seen from fig. 3, the energy consumption of the system crosses at 67 meters when the total node number n is 10 and n is 15, and when the network scenarios are all within a range less than 67 meters, the more densely distributed the node number is, the less energy consumption is. This is because, in a small range, if n is 15, the number of neighbor nodes of each node increases, and optional relay from the source node to the destination node becomes easy. When the scene is larger than 67 meters, the distribution of the nodes becomes sparse, the system energy consumption is increased when n is 15 compared with when n is 10 due to the influence of the transmission power and the number of the nodes for analyzing the message, and at this time, when n is 15, more relay nodes are selected to compensate the expansion of the scene.
Fig. 4 shows the relationship between the scene size change and the number of cooperative nodes. Observing fig. 4, in the case of being smaller than the scene at about 67 meters, the number of cooperative nodes when n is 15 is less than when n is 10. However, as the scene is increased, the number of the cooperative nodes will increase after the total number of the nodes is increased, and thus the energy consumption obtained according to the formula (12) will also increase, which is consistent with the energy consumption relation diagram of fig. 3.
The relay selection strategy based on the network state transition probability solves the problem of operation complexity caused by high redundancy state space in a multi-hop Ad-Hoc network environment to a certain extent. The strategy carries out appropriate transformation on the Metropolis criterion, and carries out state pruning and greedy search on a network state space with a certain probability, thereby realizing the preferred selection of the relay node. Experimental results show that indexes such as system operation amount, energy consumption and success rate are obviously optimized by the ISA algorithm provided by the invention, and the scheme can provide reliable technical support for further realizing optimal relay transmission in the later period.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (1)
1. A relay selection method based on state transition probability in an Ad-Hoc network is characterized by comprising the following steps:
the information source s firstly sends a message to the optimal neighbor subset, the neighbor subsets respectively continue to forward to the respective optimal neighbor subsets after receiving the message, and so on until the message is finally merged to reach the information destination t;
in an Ad-Hoc network, selecting an optimal relay from a plurality of relay nodes as the optimal neighbor subset to send a message by constraint of network state transition probability, wherein the calculation method of the network state transition probability comprises the following steps:
let the channel gain between any two node links be hi,jWith a transmission power of wiConsidering the mutual interference of adjacent links, the signal-to-noise ratio can be expressed as γ ═ (w)ihi,j)/(σi,j 2+∑k∈ψ,k≠i(wkhk,j))，σ2For additive white gaussian noise power, the probability density function PDF is expressed as:
then:
therefore, the calculation formula of the outage probability in the Rayleigh fading channel model DF mode can be derived as follows:
setting a node set in a current state x as A (x), enabling a to belong to A (x), enabling a terminal state to be t, enabling x to be not equal to t, deriving from the definition of interrupt probability, and enabling the successful transmission probability from a cooperative node subset a of the current hop to any node n of the next hop to be 1-pout(n, a), if θ is:
then, according to the nature of the Markov chain, the transition probability of the system from state x to state y can be defined as:
when the best relay is selected from a plurality of relay nodes as the best neighbor subset to send messages, unnecessary neighbor sets and behavior sets with low contribution rate are reduced, and the reduction method comprises the following steps:
assuming that the current state x reaches the state y through all possible paths, probability weighting is performed on all possible paths, and the equation that takes the least time to find the end point can be defined as:
J*(x)＝mina∈A[∑y∈N(x)pxy(a)(V(x,a,y)+ξJ(y))],ξ∈[0,1] (13)；
knowing that all nodes in the state x are A (x), the neighbor set is N (x), and unnecessary neighbor sets and behavior sets with low contribution rate are reduced, so that new sets A '(x) and N' (x) can be obtained; with a bitReduced node set
Neighbor set
In the actual complexity reduction process, the transition probability p of each state is calculated according to the formula (7)xy(a) Cutting off neighbor nodes with lower transmission probability and leaving paths with higher transmission probability, thereby optimizing the state space of the algorithm for operation; thus, the formula (13) becomes:
where S represents the set of all states of the system, the maximum pruning probability for all pruned neighbor nodes during the transition from state x to state y of the system can be defined as:
then the maximum difference in energy consumption values before and after the curtailment is:
wherein the content of the first and second substances,
is an overhead estimate for state x; in the algorithm execution process, when Δ (x) ═ 0, it means that m (x) is 0, i.e. there is no node change between states x and y, then no pruning is performed, when Δ (x) > 0, iterative pruning is performed, and the upper bound of state reduction is obtained by the formula (16) deformation:
in order to make the number of nodes in N' (x) appropriately large, an upper bound M of the formula (17) is determined-Provided that M (x) does not exceed M-All can be used;
further optimizing by referring to a simulated annealing algorithm; through the modification and application of Metropolis criterion, the method is proposed that M (x) < M-Within the acceptable state intervals, the conversion state is accepted with a certain probability; the greater Δ C, the greater the likelihood of acceptance; therefore, using Metropolis's transformation criterion, let df be M--MXThen the transformed acceptance probability is:
wherein q represents the acceptance probability, x, y represents the system state, TkRepresents the temperature value at the k-th time;
removing the transmission states with small probability step by adopting greedy search, wherein the greedy search method comprises the following steps:
step 1: when k is 0, the current solution of the Ad-Hoc network is S (0) ═ x, and the following steps are performed at the temperature T:
step 2: generating a neighborhood subset based on the state x of the current solution S (k)
Randomly obtaining a new state y from N (S (k)), and calculating the difference of the upper bound of the pruning and the energy consumed by reaching the new state y: Δ C ═ M--C(y)；
And 3, step 3: if the delta C is less than or equal to 0, the energy consumption exceeds the upper limit, and the direct rejection is performed; if Δ C > 0, a random number rand ∈ random [0,1) is generated, if the probability of reception is greater than
Receiving y as the next current solution; if a new state y is received, making S (k +1) equal to y, otherwise, making S (k +1) equal to x;
and 4, step 4: executing next iteration when k is k +1, judging whether the algorithm should be ended according to a given convergence criterion, if so, turning to the 5 th step, and otherwise, turning to the 2 nd step;
and 5, step 5: and returning.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811122242.2A CN109089273B (en) | 2018-09-26 | 2018-09-26 | Relay selection method based on state transition probability in Ad-Hoc network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811122242.2A CN109089273B (en) | 2018-09-26 | 2018-09-26 | Relay selection method based on state transition probability in Ad-Hoc network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109089273A CN109089273A (en) | 2018-12-25 |
| CN109089273B true CN109089273B (en) | 2021-08-20 |
Family
ID=64842450
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811122242.2A Active CN109089273B (en) | 2018-09-26 | 2018-09-26 | Relay selection method based on state transition probability in Ad-Hoc network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109089273B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111132075B (en) * | 2019-12-30 | 2021-08-31 | 西北工业大学 | Air-ground integrated vehicle networking relay selection method based on state transition probability |
| CN111641991B (en) * | 2020-05-07 | 2022-02-08 | 西北工业大学 | Multi-relay two-hop network secure transmission method based on data caching |
| CN111711931B (en) * | 2020-06-11 | 2021-12-07 | 西南科技大学 | Optimal channel selection method for rapid convergence |
| CN112188583B (en) * | 2020-10-08 | 2022-08-02 | 上海海事大学 | Ocean underwater wireless sensing network opportunistic routing method based on reinforcement learning |
| CN112272380B (en) * | 2020-10-28 | 2022-09-20 | 中原工学院 | Online industrial wireless sensor network deployment method facing complex deployment environment |
| CN112367692B (en) * | 2020-10-29 | 2022-10-04 | 西北工业大学 | Air-ground integrated vehicle networking relay selection method based on link service quality |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102523588A (en) * | 2011-12-08 | 2012-06-27 | 东南大学 | Method for reducing interruption probability of large-scale wireless self-organizing network |
| CN103067285A (en) * | 2012-12-25 | 2013-04-24 | 北京银易通网络科技有限公司 | Energy-saving and reliable multicast method based on relay clique in mobile ad-hoc network and sensor network |
| WO2016161603A1 (en) * | 2015-04-09 | 2016-10-13 | Telefonaktiebolaget Lm Ericsson (Publ) | Trigger conditions for measurement reports for relay selection |
| CN106131918A (en) * | 2016-08-12 | 2016-11-16 | 梁广俊 | The associating Path selection of energy acquisition node and power distribution method in wireless sense network |
| CN106993320A (en) * | 2017-03-22 | 2017-07-28 | 江苏科技大学 | Wireless sensor network cooperation transmission method for routing based on many relay multi-hops |
| CN107911184A (en) * | 2017-08-04 | 2018-04-13 | 常州工学院 | A kind of method of the opportunity cognition route based on frequency spectrum perception |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7593376B2 (en) * | 2005-12-07 | 2009-09-22 | Motorola, Inc. | Method and apparatus for broadcast in an ad hoc network using elected broadcast relay nodes |
-
2018
- 2018-09-26 CN CN201811122242.2A patent/CN109089273B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102523588A (en) * | 2011-12-08 | 2012-06-27 | 东南大学 | Method for reducing interruption probability of large-scale wireless self-organizing network |
| CN103067285A (en) * | 2012-12-25 | 2013-04-24 | 北京银易通网络科技有限公司 | Energy-saving and reliable multicast method based on relay clique in mobile ad-hoc network and sensor network |
| WO2016161603A1 (en) * | 2015-04-09 | 2016-10-13 | Telefonaktiebolaget Lm Ericsson (Publ) | Trigger conditions for measurement reports for relay selection |
| CN106131918A (en) * | 2016-08-12 | 2016-11-16 | 梁广俊 | The associating Path selection of energy acquisition node and power distribution method in wireless sense network |
| CN106993320A (en) * | 2017-03-22 | 2017-07-28 | 江苏科技大学 | Wireless sensor network cooperation transmission method for routing based on many relay multi-hops |
| CN107911184A (en) * | 2017-08-04 | 2018-04-13 | 常州工学院 | A kind of method of the opportunity cognition route based on frequency spectrum perception |
Non-Patent Citations (2)
| Title |
|---|
| Robust Relay Selection and Outage Probability Analysis for Cooperative Communications in Aircraft Approach;Hua Li 等;《2012 8th International Conference on Mobile Ad-hoc and Sensor Networks》;20121216;全文 * |
| 有扰环境协作自组织网络的中继区域选择及系统性能分析;徐艳丽 等;《东南大学学报（英文版）》;20120831;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109089273A (en) | 2018-12-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109089273B (en) | Relay selection method based on state transition probability in Ad-Hoc network | |
| Jabbar et al. | MEQSA-OLSRv2: A multicriteria-based hybrid multipath protocol for energy-efficient and QoS-aware data routing in MANET-WSN convergence scenarios of IoT | |
| Feng et al. | Green information and communication technology: energy efficiency in a motorway model | |
| Cengiz et al. | A review on the recent energy-efficient approaches for the Internet protocol stack | |
| Liu et al. | Balance energy-efficient and real-time with reliable communication protocol for wireless sensor network | |
| Yazdinejad et al. | Increasing the performance of reactive routing protocol using the load balancing and congestion control mechanism in manet | |
| Mishra et al. | Power aware routing in mobile ad hoc networks-a survey | |
| Cheng et al. | PADCP: Power-aware dynamic clustering protocol for wireless sensor network | |
| Feng et al. | Energy saving geographic routing in ad hoc wireless networks | |
| Katiravan et al. | Energy efficient link aware routing with power control in wireless ad hoc networks | |
| Banu et al. | A New Multipath Routing Approach for Energy Efficiency in Wireless Sensor Networks | |
| Loganathan et al. | Improved Load-Balanced Clustering for Energy-Aware Routing (ILBC-EAR) in WSNs. | |
| Liu et al. | A novel routing algorithm based on probability prediction for mobile opportunistic networks | |
| Cheikh et al. | Energy efficient relay selection method for clustered wireless sensor network | |
| Lu et al. | Energy aware dynamic task allocation in mobile ad hoc networks | |
| Ting et al. | Status adaptive routing with delayed rebroadcast scheme in AODV-based MANETs | |
| Cai et al. | A load-balanced minimum energy routing algorithm for wireless ad hoc sensor networks | |
| Prajapati et al. | Cross layer design with extensive virtual MIMO: FS-MUP optimization model for wireless sensor network | |
| Hu et al. | Performance analysis of routing protocols in UAV ad hoc networks | |
| Scientific | OPPORTUNISTIC ROUTING USING RELAY NODE SELECTION IN ACOUSTIC WIRELESS SENSOR NETWORKS FOR EFFICIENT AND RELIABLE NETWORK PERFORMANCE | |
| Vir et al. | Traffic Generator Based Power Analysis of Different Routing Protocol For Mobile Nodes in Wireless Sensor Network Using QualNet | |
| Gautam et al. | Reduction of Power Consumption and Transmission Delay by Optimization Technique for WN | |
| Elewely et al. | A proposed routing scheme for power reduction in wireless networks | |
| Gengsheng et al. | The research of clustering protocol based on chain routing in WSNs | |
| Yang et al. | An energy-efficient data gathering protocol for wireless sensor networks |
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 |