CN102083101B - Information transmission method for cognitive radio sensor network - Google Patents

Abstract

The invention relates to an information transmission method for a cognitive radio sensor network by combining a routing protocol and dynamic frequency spectrum allocation, comprising following five steps of: (1) periodically sensing a frequency spectrum to obtain available frequency spectrum information in real time by all sensor nodes in the network; (2) requesting a route to inform adjacent nodes to carry out negotiation on the available frequency spectrum and communication power through broadcasting by a source node when data are required to be forwarded to a sink node by the source node; (3) carrying out negotiation on the available frequency spectrum and the communication power with the node after the request is received by the neighbor node to minimize the use cost of links between two nodes; (4) calculating and saving the use cost of all the links by the source node when multiple links are available, and sequentially forwarding the routing request to the sink node by all the nodes of the available links; and (5) initiating the route to establish the request when the RREQ (Routing Request) from the source node is received by the sink node, and determining a path with the minimum forwarding cost in a distributed way according to the saved cost of all the nodes to obtain maximization of the network throughput and the life cycle.

Description

A kind of information transmission method for cognitive radio sensor network

Technical field

The present invention has effectively utilized the usable spectrum of sensor node real-time perception in the network, by the forward-path of dynamic frequency spectrum deployment foundation from data source nodes to aggregation node, to obtain the maximization of network throughput and life cycle, be applied to the transmission of information in the cognitive radio sensor network (CRSN, Cognitive Radio Sensor Network).

Background technology

Cognitive radio (CR, Cognitive Radio) is radio communication concept and the technology of rising in recent years, it can the present communication environment of perception in all frequency spectrums use chance.Wireless sensor network (WSN, Wireless Sensor Network) is another study hotspot of message area in recent years, the change that it has brought information Perception with characteristics such as its low-power consumption, low cost, distributed and self-organizings.The superiority that cognitive radio is brought, as improve the availability of frequency spectrum, improve communication quality etc., combine with multi-hop WSN in the resource-constrained situation, can effectively utilize frequency spectrum, have broad application prospects.Therefore, very naturally cognitive radio technology can be combined with wireless sensor network, thereby produce a kind of novel sensor network structure, be i.e. cognitive radio sensor network (CRSN, Cognitive Radio Sensor Network).

Because the frequency spectrum resource that CRSN uses is to obtain by the mode that detects, so CRSN has the frequency spectrum dynamic variation characteristic in essence.Simultaneously, as wireless sensor network, must pay close attention to the maximization of network lifetime, this mainly is reflected in the aspects such as balancing energy and minimum energy consumption.Routing Protocol among tradition Ad Hoc network and the traditional WSN can not adapt to the above-mentioned characteristic of CRSN well, and in the existing research, has plenty of the problem that Design of Routing Protocol should be noted under the CR environment of discussing, but do not provide corresponding solution, what have only is to provide routing algorithm for CRN or WSN, but and not exclusively be fit to the CRSN environment, so design can reflect that CRSN characteristics and the Routing Protocol that is suitable for working seem very urgent in CESN.

Summary of the invention

Technical problem: the objective of the invention is to provide in CRSN a kind of information transmission method for cognitive radio sensor network, the method can obtain the maximization of network throughput and life cycle.

Technical scheme: the present invention is a kind of information transmission method for cognitive radio sensor network, it comprises five processes: cycle frequency spectrum perception process, broadcasting or forwarding route requests process, usable spectrum and power of communications negotiations process, Route establishment process and route maintenance procedure.

Distributed frequency spectrum is distributed for joint route agreement of the present invention and information transferring method and Route Selection combines.At first, with broadcast request (RREQ, Routing Request), its neighbor node is being received after this request and this node carries out the negotiation of usable spectrum and power of communications when source node need to send data to aggregation node; Then, but arranged the multilink time spent, source node calculates the trace utilization cost of each link and stores, and all nodes in the available link are transmitted route requests successively until aggregation node; At last, receive the RREQ of source node when aggregation node after, initiate the Route establishment request, each node is according to the cost of storage, distributed earth determines to have minimum path of transmitting cost, to obtain the maximization of network throughput and life cycle.The method comprises following 5 processes and is undertaken by described order:

1) cycle frequency spectrum perception process: each sensor node periodically carries out frequency spectrum perception in the network, and this perception can independently be carried out, and also can cooperate and finish,

2) broadcasting or forwarding route requests process: when source node need to send data to aggregation node, will broadcast RREQ, its neighbor node is being received after this request and this node carries out the negotiation of usable spectrum and power of communications, but there is being the multilink time spent, source node calculates the trace utilization cost of each link and stores, all nodes in the available link are transmitted route requests successively until aggregation node

3) usable spectrum and power of communications negotiations process: this process is take maximization network throughput and life cycle as target, and by linear weighted function formation cost function, after node is received neighbor node RREQ, to minimize point-to-point transmission link trace utilization cost as target, carry out the negotiation of usable spectrum and power of communications with this neighbor node

4) Route establishment process: receive the RREQ of source node when aggregation node after, initiate request of route discovery, each node is according to the cost that stores, to minimize step 3) described in cost function be target, distributed earth is determined forward-path,

5) route maintenance procedure: adopt the routing error grouping and confirm to divide into groups to carry out route maintenance, when route can not be used because of the change of network topology, the node of chain rupture place is take oneself as data source nodes, again broadcast request, consult usable spectrum and power of communications until aggregation node to transmit remaining data.

Two aspects of described forwarding route requests process: on the one hand, and if only if this neighbor node to the distance of aggregation node greater than self to aggregation node apart from the time, node is just accepted this route requests, otherwise abandons request; On the other hand, if consult to show between two nodes there is not available frequency spectrum, namely in the known constantly inaccessible situation of link, node is not transmitted route requests.The route requests forwarding process as shown in Figure 2.

Communication spectrum and power of communications machinery of consultation are take maximization network throughput and life cycle as target, provided the multiple-objection optimization function, with linear weight sum method multiple objective function is converted into single-goal function, is out of shape this single-goal function by abbreviation and draws the forward-path cost function.Each node carries out the negotiation of usable spectrum and power of communications to minimize point-to-point transmission link trace utilization cost as target.

In the described route finding process, receive the RREQ of source node when aggregation node after, initiate the Route establishment request, each node is according to stored cost, utilize the distributed dijkstra's algorithm in the graph theory, distributed earth is looked for the path with minimal path cost function and is transmitted, and this path is an efficient solution of described multiple-objection optimization function.

Beneficial effect: the CRSN information transferring method that the present invention proposes joint route agreement and dynamic frequency spectrum deployment.Advantage is that cognitive radio is combined with sensor network, the frequency spectrum of improving the occasion access reduces collision and the competition time delay of bringing owing to the sensor node very dense, sensor node distributed earth in the network carries out dynamic frequency spectrum deployment and routing procedure, with the maximization of acquisition network throughput and the maximization of life cycle.

Description of drawings

Fig. 1 is the overall procedure of method.

Fig. 2 is the route requests forwarding process.

Fig. 3 is that communication spectrum and power of communications are consulted flow process.

Fig. 4 is distributed Route Selection flow process.

Embodiment

Distributed frequency spectrum is distributed information transferring method of the present invention and Route Selection combines.At first, with broadcast request (RREQ, Routing Request), its neighbor node is being received after this request and this node carries out the negotiation of usable spectrum and power of communications when source node need to send data to aggregation node; Then, but arranged the multilink time spent, source node calculates the trace utilization cost of each link and stores, and all nodes in the available link are transmitted route requests successively until aggregation node; At last, receive the RREQ of source node when aggregation node after, initiate the Route establishment request, each node is according to the cost of storage, distributed earth determines to have minimum path of transmitting cost, to obtain the maximization of network throughput and life cycle.The method comprises following 5 processes and is undertaken by described order:

1) cycle frequency spectrum perception process: each sensor node periodically carries out frequency spectrum perception in the network, and this perception can independently be carried out, and also can cooperate and finish,

2) broadcasting or forwarding route requests process: when source node need to send data to aggregation node, will broadcast RREQ, its neighbor node is being received after this request and this node carries out the negotiation of usable spectrum and power of communications, but there is being the multilink time spent, source node calculates the trace utilization cost of each link and stores, all nodes in the available link are transmitted route requests successively until aggregation node

3) usable spectrum and power of communications negotiations process: this process is take maximization network throughput and life cycle as target, and by linear weighted function formation cost function, after node is received neighbor node RREQ, to minimize point-to-point transmission link trace utilization cost as target, carry out the negotiation of usable spectrum and power of communications with this neighbor node

4) Route establishment process: receive the RREQ of source node when aggregation node after, initiate request of route discovery, each node is according to the cost that stores, to minimize step 3) described in cost function be target, distributed earth is determined forward-path,

5) route maintenance procedure: adopt the routing error grouping and confirm to divide into groups to carry out route maintenance, when route can not be used because of the change of network topology, the node of chain rupture place is take oneself as data source nodes, again broadcast request, consult usable spectrum and power of communications until aggregation node to transmit remaining data.

The transmitting power of node is retrained by two aspects: be the available maximum transmission power of perception user on the one hand, this is mainly determined by perception user hardware condition; Be the activity of other users in the network on the other hand, node one will overcome the interference that self is subject to and noise guaranteeing proper communication, and two will reduce transmitting power to guarantee the proper communication of user around the impact within reason.Thus, the transmitting power of each node exists a supremum and an infimum, if the some time is carved infimum less than supremum, then node perceives a frequency spectrum cavity-pocket constantly at this.

Because the dynamic variation characteristic of frequency spectrum and the otherness of the usable spectrum that each node real-time perception arrives need to be carried out spectrum allocation may and Route Selection dynamically.According to the target of maximization network throughput and life cycle, the CRSN information transferring method of a kind of joint route agreement and dynamic frequency spectrum deployment has been proposed.The method has provided the multiple-objection optimization function take maximization network throughput and life cycle as target, with linear weight sum method multiple objective function is converted into single-goal function, is out of shape this single-goal function by abbreviation and draws the forward-path cost function.Sensor node in the network is after receiving route requests RREQ, to minimize point-to-point transmission link trace utilization cost as target, distributed earth carries out the negotiation of usable spectrum and power of communications, has the minimum cost forward-path as target to obtain on this basis, carries out routing procedure.Dynamic frequency spectrum deployment can with effectively utilizing frequency spectrum access chance, be carried out Route Selection to obtain the maximization of network throughput and life cycle on this basis.

By reference to the accompanying drawings, the present invention program is designed work further concrete analysis and description.

(1) mathematical expression of network throughput and life cycle

The throughput of path P is the data volume of transmitting in the unit interval on this path, is equivalent to the actual average data transmission rate in this path, and in the given situation of data to be transferred amount, the maximization network throughput is equivalent to and minimizes the path delay of time.The path delay of time T _PComprise propagation delay time t _Tran, processing delay t _ProcWith queuing delay t _Line, establish data packet transfer, respectively be path delay of time of grouping:

$T_P=K\·(\underset{i=0}{\overset{H-1}{\mathrm{\Σ}}}\frac{N_0/K}{R_{i,i+1}}+\underset{i=0}{\overset{H-1}{\mathrm{\Σ}}}\frac{N_0/K}{R_p}+\underset{i=0}{\overset{H-1}{\mathrm{\Σ}}}(\frac{N_0(K-1)/2K+Q_i}{R_p}+\frac{N_0(K-1)/2K+Q_i}{R_{i,i+1}}))$

$\≈K\·\underset{i=0}{\overset{H-1}{\mathrm{\Σ}}}(N_0/2+Q_i)(\frac{1}{R_p}+\frac{1}{R_{i,i+1}})$

In the following formula, N ₀Be the data volume of transmitting, K is packet count, R _{I, i+1}Be the message transmission rate on the link (i, i+1), R _pBe the data processing rate of node, Q _iBe the data to be transferred sequence length of node i, H is total jumping figure of path P.

The maximization problems of network lifetime can be converted into the average residual ENERGY E on the selected path P of maximization _P, its formulation is the maximization following formula:

$E_P=\frac{1}{H}(\underset{i=0}{\overset{H-1}{\mathrm{\Σ}}}{e}_i-K\·(\underset{i=0}{\overset{H-1}{\mathrm{\Σ}}}\frac{P_{i,i+1}{(N_0/K)}^2}{R_{i,i+1}}+{\mathrm{HE}}_p{N}_0/K))$

$=-\frac{1}{H}\underset{i=0}{\overset{H-1}{\mathrm{\Σ}}}\frac{P_{i,i+1}{N}_0^2}{{\mathrm{KR}}_{i,i+1}}+\frac{1}{H}\underset{i=0}{\overset{H-1}{\mathrm{\Σ}}}{e}_i-E_p{N}_0$

In the following formula, e _iBe the dump energy of node i, E _pFor the node single-bit is processed energy consumption, P _{I, i+1}Be the upper power that sends the required consumption of bit of link (i, i+1).

(2) method of dynamic frequency spectrum deployment

The CRSN information transferring method of joint route agreement of the present invention and dynamic frequency spectrum deployment is take maximization network throughput and life cycle as target, and it is as follows to provide the multiple-objection optimization function:

Look for:

P _{I, i+1}, P

Minimize: T _P

Maximization: E _P

Interval

Communications band for link (i, i+1).

Utilize linear weight sum method that multiple objective function is converted into single-goal function, be out of shape this single-goal function by abbreviation and draw the forward-path cost function:

$\mathrm{\Φ}=\sqrt{\frac{1}{H}}\·\underset{i=0}{\overset{H-1}{\mathrm{\Σ}}}(2{\mathrm{\ρ}}_i-\frac{e_i}{\mathrm{\ξ}})+{\mathrm{\λ}}_2{E}_p{N}_0$

And have

$\frac{{\mathrm{\λ}}_2{N}_0^2{P}_{i,i+1}}{{\mathrm{\λ}}_1(N_0/2+Q_i)(1+R_{i,i+1}/R_p)}=\mathrm{\η}$

Parameter ξ and network size among the cost function Φ (comprising monitored area size, leader cluster node number etc.) and data source nodes be to the distance dependent between aggregation node, in addition,

${\mathrm{\ρ}}_i=\sqrt{{\mathrm{\λ}}_1{\mathrm{\λ}}_2{N}_0^2(N_0/2+Q_i)\frac{P_{i,i+1}}{R_{i,i+1}}(\frac{1}{R_{i,i+1}}+\frac{1}{R_p})}$

Cost 2 ρs of method for allocating dynamic frequency spectrums to minimize the point-to-point transmission link _i-e _i/ ξ is target.All adopt the BPSK modulation system if establish each node, by calculating the optimum results that can obtain first on each frequency be:

$R_{i,i+1}\left(f\right)=\min \{\frac{{\mathrm{\λ}}_1\mathrm{\η}(N_0/2+Q_i)}{{\mathrm{\λ}}_2{P}_{i,i+1}^{\min }\left(f\right)N_0^2-{\mathrm{\λ}}_1\mathrm{\η}(N_0/2+Q_i)/R_p},\mathrm{w\Δ}{f}_B\}$

$P_{i,i+1}\left(f\right)=\frac{{\mathrm{\λ}}_1\mathrm{\η}(N_0/2+Q_i)(1+R_{i,i+1}\left(f\right)/R_p)}{{\mathrm{\λ}}_2{R}_{i,i+1}\left(f\right)N_0^2}$

Then, node utilizes following target function frequency range that negotiation communication uses on the basis of above-mentioned optimum results.Require selected transmission rate, should be the optimum results on certain frequency in the selected frequency range, and the transmitted power on this frequency should not drop on outside the frequency spectrum cavity-pocket, this can finish by simple search.Provide the method for allocating dynamic frequency spectrums flow process as shown in Figure 3.

(3) method of joint route selection

Route selection method of the present invention adopts the distributed dijkstra's algorithm in the graph theory, receive the RREQ of source node when aggregation node after, initiate the Route establishment request, each node is according to stored cost, and distributed earth is looked for the path with minimal path cost function and transmitted.In the distributed dijkstra's algorithm, be connection charge from X to Y if establish Cost (X, Y), Cost_min (X, Y) is the expense of economizing route most from X to Y.Suppose A and Node B, C ..., D has direct connection.Algorithm is followed following principle so:

$\mathrm{Cost}\_\min (A,Z)=\left\{\begin{array}{c}\mathrm{Cost}(A,B)+\mathrm{Cost}\_\min (B,Z)\\ \mathrm{Cost}(A,C)+\mathrm{Cost}\_\min (C,Z)\\ M\\ \mathrm{Cost}(A,D)+\mathrm{Cost}\_\min (D,Z)\end{array}\right.$

According to this principle, as long as node A knows that it arrives the expense of each neighbor node, each neighbour knows the province route of the own Z of arriving simultaneously, and node A just can determine that it arrives the province route of Z.Node A just can carry out above-mentioned calculating like this, and determines to have economized route most.

Each node is only known expense and available all information of neighbours that arrive each neighbour.Therefore, in Distributed Calculation, each node is to its information of adjacent office node broadcasts known to it.Each node is received new information and is correspondingly upgraded its routing table.Because neighbours are broadcast message termly constantly, final network node is connected with connection and is spread all over for information about whole network.The information that enters node allows them to find new node and the shortest path that leads to other nodes.Algorithm flow as shown in Figure 4.

Claims (3)

1. information transmission method for cognitive radio sensor network, it is characterized in that the method is a kind of information transmission method for cognitive radio sensor network based on joint route agreement and dynamic frequency spectrum deployment, at first, with the broadcast request, the neighbor node of aggregation node is being received after this request and aggregation node carries out the negotiation of usable spectrum and power of communications when source node need to send data to aggregation node; Then, but arranged the multilink time spent, source node calculates the trace utilization cost of each link and stores, and all nodes in the available link are transmitted route requests successively until aggregation node; At last, receive the route requests of source node when aggregation node after, initiate the Route establishment request, each node is according to the cost of storage, distributed earth determines to have minimum path of transmitting cost, to obtain the maximization of network throughput and life cycle; The method specifically comprises following process:

1) cycle frequency spectrum perception process: each sensor node periodically carries out frequency spectrum perception in the network, this perception is independently to carry out, or cooperation finishes,

2) broadcasting or transmit the route requests process: when source node need to send data to aggregation node with the broadcast request, the neighbor node of aggregation node is being received after this request and aggregation node carries out the negotiation of usable spectrum and power of communications, but there is being the multilink time spent, source node calculates the trace utilization cost of each link and stores, all nodes in the available link are transmitted route requests successively until aggregation node

3) usable spectrum and power of communications negotiations process: this process is take maximization network throughput and life cycle as target, and by linear weighted function formation cost function, after node is received the neighbor node route requests, to minimize point-to-point transmission link trace utilization cost as target, carry out the negotiation of usable spectrum and power of communications with the neighbor node of aggregation node

4) Route establishment process: receive the route requests of source node when aggregation node after, initiate request of route discovery, each node is according to the cost that stores, and to minimize cost function described in the step 3) as target, distributed earth is determined forward-path,

5) route maintenance procedure: adopt the routing error grouping and confirm to divide into groups to carry out route maintenance, when route can not be used because of the change of network topology, the node of chain rupture place is take oneself as data source nodes, again broadcast request, consult usable spectrum and power of communications until aggregation node to transmit remaining data.

2. information transmission method for cognitive radio sensor network as claimed in claim 1, it is characterized in that usable spectrum and power of communications negotiations process, this process is take maximization network throughput and life cycle as target, provided the multiple-objection optimization function, with linear weight sum method multiple objective function is converted into single-goal function, be out of shape this single-goal function by abbreviation and draw the forward-path cost function, each node carries out the negotiation of usable spectrum and power of communications to minimize point-to-point transmission link trace utilization cost as target.

3. information transmission method for cognitive radio sensor network as claimed in claim 1, it is characterized in that in the described Route establishment process, receive the route requests of source node when aggregation node after, initiate the Route establishment request, each node is according to stored cost, utilize the distributed dijkstra's algorithm in the graph theory, distributed earth is looked for the path with minimal path cost function and is transmitted, and this path is an efficient solution of multiple-objection optimization function.

Images